From: Larry__Weiss Newsgroups: alt.folklore.computers Subject: CDC6600 - just how powerful a machine was it? Date: Sun, 30 Jun 2002 08:05:57 -0500 Organization: Airnews.net! at Internet America Lines: 9 Message-ID: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> X-Orig-Message-ID: <3D1F0235.696B59C4@airmail.net> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Sun Jun 30 08:15:22 2002 NNTP-Posting-Host: ![b&q1k-XM[fdKm,?O.1 (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.77 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!news.airnews.net!cabal10.airnews.net!cabal1.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:110908 I just stumbled across some pics of the CDC6600: http://www.bambi.net/computer_museum/cdc6600_and_console.jpg http://online.sfsu.edu/~hl/mmm.html Just how powerful was that machine? How would it compare to modern-day machines, especially PC's? - LarryW ###### From: "Eric S. Harris" Newsgroups: alt.folklore.computers,comp.sys.cdc, Subject: Re: CDC6600 - just how powerful a machine was it? Date: Sun, 30 Jun 2002 14:07:26 -0500 Organization: MindSpring Enterprises Lines: 91 Distribution: inet Message-ID: <3D1F56EE.DAA8337E@mindspring.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: eric_harris_76@yahoo.com NNTP-Posting-Host: a8.bf.6d.32 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Server-Date: 30 Jun 2002 19:28:49 GMT X-Mailer: Mozilla 4.79 [en]C-CCK-MCD (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!news-out.nuthinbutnews.com!propagator-sterling!news-in.nuthinbutnews.com!newsfeed.wirehub.nl!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!newsfeed0.news.atl.earthlink.net!news.atl.earthlink.net!news.mindspring.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:110929 Perhaps someone in news:comp.sys.cdc has something to add to this? Larry__Weiss wrote: > > I just stumbled across some pics of the CDC6600: > http://www.bambi.net/computer_museum/cdc6600_and_console.jpg > http://online.sfsu.edu/~hl/mmm.html > > Just how powerful was that machine? How would it compare to modern-day > machines, especially PC's? For its day, incredibly powerful. Compared to today's PCs, puny in speed, memory and disk storage. You could probably go to Wal-Mart or wherever and for less than $1000 buy something that could emulate one. I'll give what little else I can, off the top of my head. Some of it may apply to later machines, and not to the 6600. Don't take anything below too uncritically, unless you get confirmation from someone who really DOES know, or it's really easy to get right and hard to get wrong. Come to think of it, that just leaves the 17-bit comment. Emulating a CDC 6600 on a PC would be an interesting and probably pointless programming challenge. The CDC 6600 and its compatible successors and predecessors used 60-bit words (only word-addressable) and had 6-bit characters. Well, sometimes 6-bit characters. That was what the hardware and the operating system supported. There were also other schemes for representing more than 63 (or 64) characters: 6-12 character sets, 8-in-12 and maybe others. (I seem to recall something about 7.5 characters/word, but that may just be a horrible, horrible dream.) Learning of the one-byte/two-byte characters in Unicode brought back memories of the 6-12 character set, where 7604 (octal) and 00 both represent a colon. (Or was it 7402?) There was some operating system support for 6-12, but it was limited. The CPU was I don't know how many orders of magnitude slower. I seem to recall something around then about memory speeds in the 200 nanosecond range, but that may have been in a different context. The address space was limited to 18 bits (17 for users), and I think the mainframe memory could be ordered (or upgraded to) that size. Later machines in the series (or a later series compatible with the CDC 6600) could go higher, but the 17-bit limit on user programs remained. They had an external kind of storage, Extended Core Storage, which went up to (I think) 22 or 24 bits of addressing. (Later ECS was semiconductor, IIRC.) It was used for several purposes. Storage more capacious than memory but faster than disk. Communication between mainframes in Multimainframe mode (a la DEC's VAXcluster) to coordinate access to shared disks, and perhaps other things. Swapping storage, for timesharing (and batch) jobs that needed to be moved out of memory for a brief while. There was no virtual storage or segmented or paged memory; if your process had any memory at all, your entire address space was in memory. There were Peripheral Processors which did the I/O. There were 10 or 20 of them, each of which had its own memory (12 bits of addressing?) and full access to the central processor's memory. One of them was special (PP 0), and contained the PP-resident part of the operating system. The rest were mostly or completely interchangeable. It's possible that some devices were attached to certain PPs, so only they could access them. Or maybe all could access all; I forget. If software in a Peripheral Processor encountered a completely unexpected situation, the convention was to program it to "hang" (spin in a tight loop), rather than go on. As long as it wasn't PP 0, the computer could continue, only slightly degraded. Sometimes more than one would get hung up. Rebooting the machine -- in the dead of night, normally -- would fix it, and provide a dump for later analysis. [1] Despite their shortcomings, there were some nice things about the CDC 6600 series machines, and their operating systems. Compared to the dominant mainframe of the day -- IBM 360s, 370s -- they had wonderful operating systems. To copy a file, you used the copy command. One-liner. To make a file bigger, you wrote to it. And so on. None of this IEBGENER/IEBCOPY foolishness. Files were just a series of bytes. (OK, 6-bit bytes, and there were those EOR marks, and oddball line terminators, but still...) I must be getting old. Nostalgia is setting in again. -Eric [1] The central processor was called the CP, the peripheral processors, PPs. Where I worked, at the daily change control meeting -- normally chaired by an IBM type, usually a woman -- the previous day's hardware and software problems were also reviewed. Some of female attendees -- and occasionally the chair -- would be nearly smirking when they and the CDC types would discuss a "PP hang" or say that machine A had to be restarted because it had too many hung PPs. -ESH -- E-mail privacy: http://www.theclairefiles.com/Personal/libspriv.html (Perhaps a bit over-cautious, she is. Perhaps.) ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Sun, 30 Jun 2002 19:44:15 -0500 Organization: Airnews.net! at Internet America Lines: 26 Message-ID: X-Orig-Message-ID: <3D1FA5DF.442E4147@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Sun Jun 30 19:53:28 2002 NNTP-Posting-Host: !Yha^1k-XNKa/n`/i'[3 (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.77 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news.airnews.net!cabal11.airnews.net!cabal1.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111021 "Eric S. Harris" wrote: > Perhaps someone in news:comp.sys.cdc has something to add to this? > Larry__Weiss wrote: > > I just stumbled across some pics of the CDC6600: > > http://www.bambi.net/computer_museum/cdc6600_and_console.jpg > > http://online.sfsu.edu/~hl/mmm.html > > Just how powerful was that machine? How would it compare to modern-day > > machines, especially PC's? > > The CPU was I don't know how many orders of magnitude slower. > I started digging around Google, and found that this was already asked back in 1999 and got 82 replies including an estimate that the 6600 was rated at about 1 MIPS. http://groups.google.com/groups?&q=%22Comparative+Speed+of+CDC+6600%22 The CDC 6600 was the very first machine that I ever programmed on (1969). The physicists for sure loved its speed at that time. I sure wish I still had the sources for those matrix crunching programs to just see how fast they can be run today. It's surely amazing how much "idle" computing capacity the world has right now. - LarryW ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Reply-To: mschaef_ng@mschaef.com Organization: mschaef.com X-Newsreader: trn 4.0-test76 (Apr 2, 2001) From: mschaef@eris.io.com (MSCHAEF.COM) Originator: mschaef@eris.io.com (MSCHAEF.COM) X-GC-Trace: gv1-U9yeIpUkd11QS/HyWXPHE6JdVhlRTu0KIrGjD527TouaPodf9POZfYg/g== Lines: 35 Message-ID: <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> NNTP-Posting-Date: Sun, 30 Jun 2002 20:17:54 CDT X-Trace: sv3-E2jtu77sc2ki14AANVOr744gEw49mSW8xBbZtKDOKP1yvHIfHGu8i/OzgLglbDP78euyylOwB+NJ9Pv!sFmtscW+wPVx0PiWn+S2LVVVOR8Qrvx9BE8NlJ8M6eW0rKSthD/OAsWzt8rXZUV+3z+lYDEZrymd!KQsLBbMzyzrOqYglAhg= X-Complaints-To: abuse@GigaNews.Com X-DMCA-Notifications: http://www.giganews.com/info/dmca.html X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Mon, 01 Jul 2002 01:17:54 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!border1.nntp.aus1.giganews.com!nntp.giganews.com!nntp3.aus1.giganews.com!bin3.nnrp.aus1.giganews.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111007 In article , Larry__Weiss wrote: >"Eric S. Harris" wrote: > >I started digging around Google, and found that this was already asked >back in 1999 and got 82 replies including an estimate that the 6600 was rated at >about 1 MIPS. > > http://groups.google.com/groups?&q=%22Comparative+Speed+of+CDC+6600%22 > >The CDC 6600 was the very first machine that I ever programmed on (1969). >The physicists for sure loved its speed at that time. I sure wish I still >had the sources for those matrix crunching programs to just see how fast >they can be run today. It's not the CDC 6600, but this site has information on one of its almost contemporaries: http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&selm=92218.123445MACKINNO%40QUCDN.QueensU.CA has information on an IBM 360/91's relative performance to a number of modern (for 1992) machines running the same Fortran code: >>IBM RS/6000 550 (FORTRAN 2.2): 0.07 seconds >>IBM 3081/G (Fortran 1.3): 0.25 seconds >>Sparcstation 1+ (FORTRAN 1.4): 0.38 seconds >>486/25 (Lahey F77L-EM/32): 0.48 seconds >>IBM 360/91 (Fortran H): 0.65 seconds >>386/33 with 387 (Lahey F77L-EM/32): 0.90 seconds It's also contains some exposition on the subject. -Mike -- http://www.mschaef.com ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: 1 Jul 2002 01:26:42 GMT Organization: The National Capital FreeNet Lines: 146 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 1025486802 2396 134.117.136.30 (1 Jul 2002 01:26:42 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 1 Jul 2002 01:26:42 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sunqbc.risq.qc.ca!news-peer-east1.sprintlink.net!news.sprintlink.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:111045 Larry__Weiss (lfw@airmail.net) writes: ... > I sure wish I still > had the sources for those matrix crunching programs to just see how fast > they can be run today. /* Gauss-Seidel Iteration for simultaneous linear equations. */ /* Load the problem. */ /*TRACE ?R;*/ problem = ARG( 1 ); SAY problem; PARSE VALUE LINEIN( problem ) WITH n itmax eps; SAY n itmax eps; np1 = n + 1; /* Read the problem file to extract the A[n,np1] matrix, followed by the X[n] vector. The blank delimited numbers are stored in an arbitrary number of records, and blank records are ignored. In the following finite state machine, state 0 initializes the read routine; state 1 gets a non-empty line of data; state 2 inserts one element into the A matrix; and state 3 inserts one element into the X vector. (Limited error testing yet.) */ /* State 0: */ reading = 1; /* Assume there's data. */ loading_a = 1; i = 1; j = 0; /* Set up for matrix A. */ state = 1; /* Switch state. */ DO WHILE reading; /* Unrequired trailing records are not read.*/ SELECT /* state */ ; WHEN state = 1 THEN /* Looking for a non-blank record. */ DO; IF LINES( problem ) = 0 THEN DO; SAY 'Premature end of data' loading_a i j; EXIT; /*reading = 0; */ END; ELSE DO; data_line = LINEIN( problem ); wc = WORDS( data_line ); /* word count */ IF wc > 0 THEN DO; wn = 0; /*word number */ IF loading_a THEN state = 2; ELSE state = 3; END; END; END; WHEN state = 2 THEN /* Loading a.i.j */ DO; j = j + 1; /* Figure out next A subscript, if any. */ IF j > np1 THEN DO; j = 0; i = i + 1; IF i > n THEN DO; i = 0; SAY 'End of A load.'; loading_a = 0; state = 3; END; END; ELSE DO; wn = wn + 1; /* Use the next number. */ a.i.j = WORD( data_line, wn ); IF wn = wc THEN state = 1; END; END; WHEN state = 3 THEN /* Loading x.i */ DO; i = i + 1; /* Figure out next X subscript, if any. */ IF i > n THEN DO; SAY 'End of X load.'; reading = 0; END; ELSE DO; wn = wn + 1; /* Use the next number. */ x.i = WORD( data_line, wn ); IF wn = wc THEN IF i < n THEN state = 1; END; END; OTHERWISE DO; SAY 'State error' state; EXIT; END; END; END; CALL LINEOUT problem; /* Normalize diagonal elements in each row. */ DO i = 1 TO n; astar = a.i.i; IF ABS( astar ) < 1e-10 THEN DO; SAY 'Normalization failed ' i; EXIT; END; DO j = 1 TO np1; a.i.j = a.i.j / astar; END; END; /* Begin iterations until success or limit. */ DO iter = 1 TO itmax UNTIL converged; converged = 1; /* Be optimistic. */ DO i = 1 TO n; xstar = x.i; x.i = a.i.np1; DO j = 1 TO n; IF i ^= j THEN x.i = x.i - a.i.j * x.j; END; IF ABS( xstar - x.i ) > eps THEN converged = 0; END; END; /***/ SAY 'Converged ' converged 'Iteration' iter; DO i = 1 TO n; SAY 'X' i '=' x.i; END; EXIT converged; ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 22 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:ywk/vmE7Pibf3hrfVcSHVTwAiWg= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 Jul 2002 01:41:55 GMT NNTP-Posting-Host: 63.211.247.85 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025487715 63.211.247.85 (Sun, 30 Jun 2002 18:41:55 PDT) NNTP-Posting-Date: Sun, 30 Jun 2002 18:41:55 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!netnews.com!feed2.news.rcn.net!rcn!dca6-feed2.news.algx.net!dfw3-feed1.news.algx.net!allegiance!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111030 mschaef@eris.io.com (MSCHAEF.COM) writes: > It's not the CDC 6600, but this site has information on one of its > almost contemporaries: original long ago ... reposting from http://www.garlic.com/~lynn/2000d.html#0 Is a VAX a mainframe: rain/rain4 158 3031 4341 Rain 45.64 secs 37.03 secs 36.21 secs Rain4 43.90 secs 36.61 secs 36.13 secs also times approx; 145 168-3 91 145 secs. 9.1 secs 6.77 secs rain/rain4 was from Lawrence Radiation lab ... and ran on cdc6600 in 35.77 secs. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Sun, 30 Jun 2002 23:34:37 -0400 Lines: 19 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d1fcdce_2@news.iglou.com> X-Trace: news.iglou.com 1025494478 204.250.0.238 (30 Jun 2002 23:34:38 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newspump.monmouth.com!newspeer.monmouth.com!news-out.visi.com!hermes.visi.com!uunet!ash.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111018 "Larry__Weiss" wrote in message = news:DD4E567B212BBDB5.A7ACAD2D46E9C366.E1F4F926E281E2EF@lp.airnews.net...= >=20 > The CDC 6600 was the very first machine that I ever programmed on = (1969). > The physicists for sure loved its speed at that time. I sure wish I = still > had the sources for those matrix crunching programs to just see how = fast > they can be run today. Although the Wheelers refer to RAIN/RAIN4, LINPACK was also used for relative speed comparisons, and has been ported to modern platforms and other languages. You should be able to find any number of different versions all over the net... -dq ###### Message-ID: <3D1FD2A7.6690F451@attbi.com> From: Joe Yuska X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 32 NNTP-Posting-Host: 24.91.67.33 X-Complaints-To: abuse@attbi.com X-Trace: rwcrnsc52.ops.asp.att.net 1025495522 24.91.67.33 (Mon, 01 Jul 2002 03:52:02 GMT) NNTP-Posting-Date: Mon, 01 Jul 2002 03:52:02 GMT Organization: AT&T Broadband Date: Mon, 01 Jul 2002 03:52:02 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!fr.usenet-edu.net!usenet-edu.net!proxad.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!wn4feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!rwcrnsc52.ops.asp.att.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111072 "Douglas H. Quebbeman" wrote: > "Larry__Weiss" wrote in message news:DD4E567B212BBDB5.A7ACAD2D46E9C366.E1F4F926E281E2EF@lp.airnews.net... > > > > The CDC 6600 was the very first machine that I ever programmed on (1969). > > The physicists for sure loved its speed at that time. I sure wish I still > > had the sources for those matrix crunching programs to just see how fast > > they can be run today. > > Although the Wheelers refer to RAIN/RAIN4, LINPACK was also > used for relative speed comparisons, and has been ported to > modern platforms and other languages. You should be able to > find any number of different versions all over the net... > > -dq This raises another question in my mind. In the late sixties and early seventies one set of working codes for reactor design were called the PDQ-(version-number) series. These were used by several of the (then) AEC labs as well as companies like Combustion Engineering. They were originally in Fortran, but had evolved in the CDC world into assembly code hybrids that even had specialized disk drivers. Actually, then, two questions. !. Were these unique to CDC machines or were there e. g., IBM versions of PDQ-7? 2. Have any of these dusty decks survived? Joe Yuska ###### From: "John Keeney" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 1 Jul 2002 01:57:03 -0400 Lines: 171 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 NNTP-Posting-Host: lou-ts9-4.iglou.com X-Original-NNTP-Posting-Host: lou-ts9-4.iglou.com Message-ID: <3d1fef08_1@news.iglou.com> X-Trace: news.iglou.com 1025502984 lou-ts9-4.iglou.com (1 Jul 2002 01:56:24 -0400) X-Original-NNTP-Posting-Host: 204.255.236.19 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!ash.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111010 Eric S. Harris wrote in message news:3D1F56EE.DAA8337E@mindspring.com... > Perhaps someone in news:comp.sys.cdc has something to add to this? > > Larry__Weiss wrote: > > > > I just stumbled across some pics of the CDC6600: > > http://www.bambi.net/computer_museum/cdc6600_and_console.jpg > > http://online.sfsu.edu/~hl/mmm.html > > > > Just how powerful was that machine? How would it compare to modern-day > > machines, especially PC's? > > For its day, incredibly powerful. And for a few days after. > Compared to today's PCs, puny in speed, memory and disk storage. You could > probably go to Wal-Mart or wherever and for less than $1000 buy something that > could emulate one. No doult about it. > I'll give what little else I can, off the top of my head. Some of it may apply > to later machines, and not to the 6600. Don't take anything below too > uncritically, unless you get confirmation from someone who really DOES know, or > it's really easy to get right and hard to get wrong. Come to think of it, that > just leaves the 17-bit comment. > > Emulating a CDC 6600 on a PC would be an interesting and probably pointless > programming challenge. The CDC 6600 and its compatible successors and > predecessors used 60-bit words (only word-addressable) and had 6-bit characters. > > Well, sometimes 6-bit characters. That was what the hardware and the operating > system supported. There were also other schemes for representing more than 63 > (or 64) characters: 6-12 character sets, 8-in-12 and maybe others. (I seem to > recall something about 7.5 characters/word, but that may just be a horrible, > horrible dream.) Learning of the one-byte/two-byte characters in Unicode > brought back memories of the 6-12 character set, where 7604 (octal) and 00 both > represent a colon. (Or was it 7402?) There was some operating system support > for 6-12, but it was limited. Depends on the operating system. As 8 bit became popular it became supported. > The CPU was I don't know how many orders of magnitude slower. I seem to recall > something around then about memory speeds in the 200 nanosecond range, but that > may have been in a different context. No, no, no. A 6600 never in its wildest dreams had a 200ns memory access: this was the days of real core memory, made of magnetic donuts on wires. Well, you could get a new word from memory every 100ns but that word has been desired for 800ns or more. Lots of parrallism going on what with the PPs and different parts of the CP all able to make memory request. Nearly everything was done in registers: 8 60-bit, 8 18-bit A(ddressing registers and another 8 18-bit general purpose registers. Mustn't forget the "stack" which held enough program space for a tight loop. Memory was slow, so they tried not to use it too much. > The address space was limited to 18 bits (17 for users), and I think the > mainframe memory could be ordered (or upgraded to) that size. Later machines in > the series (or a later series compatible with the CDC 6600) could go higher, but > the 17-bit limit on user programs remained. The 6600 came in 32k, 64k (64 in Microsoft speak, 65k to CDC) and 128/131k versions. Just remember, this is 60 bit words, not bytes. Oh yea, you may get as many a four instructions to the word which could be issued as quickly as every 100ns. > They had an external kind of storage, Extended Core Storage, which went up to (I > think) 22 or 24 bits of addressing. (Later ECS was semiconductor, IIRC.) It > was used for several purposes. Storage more capacious than memory but faster > than disk. Communication between mainframes in Multimainframe mode (a la DEC's > VAXcluster) to coordinate access to shared disks, and perhaps other things. > Swapping storage, for timesharing (and batch) jobs that needed to be moved out > of memory for a brief while. But then, you could buy an additional CP to go in your 6600 making it a 6700. > There was no virtual storage or segmented or paged memory; if your process had > any memory at all, your entire address space was in memory. Sure there was, they were called overlays and if you wanted them you wrote the code to do them. > There were Peripheral Processors which did the I/O. There were 10 or 20 of > them, each of which had its own memory (12 bits of addressing?) and full access > to the central processor's memory. One of them was special (PP 0), and > contained the PP-resident part of the operating system. The rest were mostly or > completely interchangeable. It's possible that some devices were attached to > certain PPs, so only they could access them. Or maybe all could access all; I > forget. If software in a Peripheral Processor encountered a completely > unexpected situation, the convention was to program it to "hang" (spin in a > tight loop), rather than go on. As long as it wasn't PP 0, the computer could > continue, only slightly degraded. Sometimes more than one would get hung up. > Rebooting the machine -- in the dead of night, normally -- would fix it, and > provide a dump for later analysis. [1] > > Despite their shortcomings, there were some nice things about the CDC 6600 > series machines, and their operating systems. Compared to the dominant > mainframe of the day -- IBM 360s, 370s -- they had wonderful operating systems. > To copy a file, you used the copy command. One-liner. To make a file bigger, > you wrote to it. And so on. None of this IEBGENER/IEBCOPY foolishness. Files > were just a series of bytes. (OK, 6-bit bytes, and there were those EOR marks, > and oddball line terminators, but still...) > > I must be getting old. Nostalgia is setting in again. -Eric > > > [1] The central processor was called the CP, the peripheral processors, PPs. > Where I worked, at the daily change control meeting -- normally chaired by an > IBM type, usually a woman -- the previous day's hardware and software problems > were also reviewed. Some of female attendees -- and occasionally the chair -- > would be nearly smirking when they and the CDC types would discuss a "PP hang" > or say that machine A had to be restarted because it had too many hung PPs. > -ESH > > -- > E-mail privacy: http://www.theclairefiles.com/Personal/libspriv.html (Perhaps a > bit over-cautious, she is. Perhaps.) ###### From: couperusNOSPAM@znet.com (Jitze Couperus) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 01 Jul 2002 08:08:12 GMT Organization: Posted via Supernews, http://www.supernews.com Message-ID: <3d200c9e.299409694@sd.znet.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> X-Newsreader: Forte Free Agent 1.21/32.243 X-Complaints-To: newsabuse@supernews.com Lines: 34 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111011 On Sun, 30 Jun 2002 19:44:15 -0500, Larry__Weiss wrote: >"Eric S. Harris" wrote: >> Perhaps someone in news:comp.sys.cdc has something to add to this? >> Larry__Weiss wrote: >> > I just stumbled across some pics of the CDC6600: >> > http://www.bambi.net/computer_museum/cdc6600_and_console.jpg >> > http://online.sfsu.edu/~hl/mmm.html >> > Just how powerful was that machine? How would it compare to modern-day >> > machines, especially PC's? >> >> The CPU was I don't know how many orders of magnitude slower. >> > >I started digging around Google, and found that this was already asked >back in 1999 and got 82 replies including an estimate that the 6600 was rated at >about 1 MIPS. > A more interesting measurement than MIPS for these machines was megaflops - given their scientifc orientation. And that was generaly measured by a standard benchmark that was run at every major realease of OS or the Fortran compiler. I believe the benchmark program used was LINPAK but I can't be sure. A table of the achieved megaflop rates for various CDC machines with different release levels of OS and Fortran compiler may be found at http://pages.sbcglobal.net/couperusj/Megaflops.html Jitze ###### From: kent@nettally.com (Kent Olsen) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: 1 Jul 2002 09:31:32 -0700 Organization: http://groups.google.com/ Lines: 142 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> NNTP-Posting-Host: 204.89.69.20 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1025541094 9735 127.0.0.1 (1 Jul 2002 16:31:34 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 1 Jul 2002 16:31:34 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!postnews1.google.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111064 Expanding on a good set of answers: > Compared to today's PCs, puny in speed, memory and disk storage. You could > probably go to Wal-Mart or wherever and for less than $1000 buy something that > could emulate one. WallyWorld is selling an Athlon 850 system for about $300. It can simulate a floating point instruction mix faster than any 6000 or Cyber 70 machine ever built. My GS160 Alpha with 730Mhz CPUs can run the same simulated FP mix at over 13 MFlop. At most, only the 2 fastest 60 bit CPUs that CDC ever made could execute FP instructions that fast. > Emulating a CDC 6600 on a PC would be an interesting and probably pointless > programming challenge. Some of us would (do) consider it a "labor of love". :) > Well, sometimes 6-bit characters. That was what the hardware and the > operating > > system supported. There were also other schemes for representing more > than 63 > > (or 64) characters: 6-12 character sets, 8-in-12 and maybe others. (I > seem to > > recall something about 7.5 characters/word, but that may just be a > horrible, > > horrible dream.) Learning of the one-byte/two-byte characters in Unicode > > brought back memories of the 6-12 character set, where 7604 (octal) and 00 > both > > represent a colon. (Or was it 7402?) There was some operating system > support > > for 6-12, but it was limited. Though 6-bit characterset werepraised early as a significant cost reduced, as the price of systems came down, CDC heard more and more complaints about 6-bit (not to be confused with 2-bit) characters sets. It was an unfair bias, as many vendors of the day ran or supported them. > The CPU was I don't know how many orders of magnitude slower. I seem to recall > something around then about memory speeds in the 200 nanosecond range, but that > may have been in a different context. Unless some chip designer gets really bored, we'll never know just how fast a 6000 style CPU would be if it were built with modern electronics. Any EEs care to venture a guess? > No, no, no. A 6600 never in its wildest dreams had a 200ns memory access: > this was the days of real core memory, made of magnetic donuts on wires. > > Well, you could get a new word from memory every 100ns but that word > has been desired for 800ns or more. Lots of parrallism going on what with > the PPs and different parts of the CP all able to make memory request. > > Nearly everything was done in registers: 8 60-bit, 8 18-bit A(ddressing > registers and another 8 18-bit general purpose registers. Mustn't forget > the "stack" which held enough program space for a tight loop. > Memory was slow, so they tried not to use it too much. Memory was 800ns. Though Cybers of that era phased memory into 4 banks so that consecutive memory references didn't conflict. This gave an effective rate near 200ns. CPUs on the 6000s came in two flavors. The monolithic 6400 and the phased 6600. You could get a system with one CPU of either type, a system with two 6400 CPUs, or a system with one of each. Two 6600 style CPUs in the same cabinet wasn't available as the memory speed wouldn't keep up. IIRC, idle PPs on NOS systems checked their input registers (IR) about 8000 times per second. The IR was the first word of the PP's Message Buffer, which was 8 words long, so the way that memory was phased, all IRs wound up in the same memory bank. Of the possible 1,250,000/second that the bank could muster, about 12% of them went to idle PPs busily searching for something to do. > There were Peripheral Processors which did the I/O. There were 10 or 20 of > them, each of which had its own memory (12 bits of addressing?) and full access > to the central processor's memory. Most 6000s were configured with a full complement of 20 PPs. Certain pricing options allowed for the purchase of machines with less, but throughput suffered pretty badly. Many were field upgraded to 20 PPs. The PP was a 12 bit processor. It had 4095.5 words. (Inside joke. It had 4096 words, but word 7777 was rarely used as it wasn't directly addressable.) Math in the PP was 18 bit. This allowed it to perform address calculations on the full length of central memory. > One of them was special (PP 0), and > contained the PP-resident part of the operating system. The rest were mostly or > completely interchangeable. Correct. On most CDC operating systems, the PP portion of the system monitor (MTR) ran in PP00. By convention, the console driver (DSD) usually ran in PP01. All other PPs were "interchangeable". > It's possible that some devices were attached to > certain PPs, so only they could access them. Devices were connected to channels. PPs could access any channel so any PP could access any device. > Despite their shortcomings, there were some nice things about the CDC 6600 > series machines, and their operating systems. Compared to the dominant > mainframe of the day -- IBM 360s, 370s -- they had wonderful operating > systems. They were so much easier to use than VM, MVS, CICS, etc... Depending on the application, the 6000s outran the IBMs, sometimes the IBMs outran the 6000s. They were definitely the two biggest horses in the race. > To copy a file, you used the copy command. One-liner. To make a file bigger, > you wrote to it. And so on. None of this IEBGENER/IEBCOPY foolishness. > Files were just a series of bytes. (OK, 6-bit bytes, and there were those EOR > marks, and oddball line terminators, but still...) Even the early DOS operating systems understood that to COPY a file meant simply that the contents of one file should be duplicated under a different name. I wonder why it took IBM so long to figure this out? :) Though technically, the PP could write a partial word to disk, and tape files could certainly contain something other that an even number of CPU words, when translated to memory, files were always an exact number of words. I too miss these old beauties. They certainly had heart! Kent ###### Message-ID: <3D209F50.A6CAA51E@attbi.com> From: Joe Yuska X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 167 NNTP-Posting-Host: 24.91.67.33 X-Complaints-To: abuse@attbi.com X-Trace: rwcrnsc53 1025547959 24.91.67.33 (Mon, 01 Jul 2002 18:25:59 GMT) NNTP-Posting-Date: Mon, 01 Jul 2002 18:25:59 GMT Organization: AT&T Broadband Date: Mon, 01 Jul 2002 18:25:59 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!fr.usenet-edu.net!usenet-edu.net!teaser.fr!easynet-quince!easynet.net!cox.net!cyclone1.gnilink.net!wn1feed!worldnet.att.net!204.127.198.204!attbi_feed4!attbi.com!rwcrnsc53.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111006 Kent Olsen wrote: > Expanding on a good set of answers: > > > Compared to today's PCs, puny in speed, memory and disk storage. You could > > probably go to Wal-Mart or wherever and for less than $1000 buy something that > > could emulate one. > > WallyWorld is selling an Athlon 850 system for about $300. It can > simulate a floating point instruction mix faster than any 6000 or > Cyber 70 machine ever built. My GS160 Alpha with 730Mhz CPUs can run > the same simulated FP mix at over 13 MFlop. At most, only the 2 > fastest 60 bit CPUs that CDC ever made could execute FP instructions > that fast. > > > Emulating a CDC 6600 on a PC would be an interesting and probably pointless > > programming challenge. > > Some of us would (do) consider it a "labor of love". :) > > > Well, sometimes 6-bit characters. That was what the hardware and the > > operating > > > system supported. There were also other schemes for representing more > > than 63 > > > (or 64) characters: 6-12 character sets, 8-in-12 and maybe others. (I > > seem to > > > recall something about 7.5 characters/word, but that may just be a > > horrible, > > > horrible dream.) Learning of the one-byte/two-byte characters in Unicode > > > brought back memories of the 6-12 character set, where 7604 (octal) and 00 > > both > > > represent a colon. (Or was it 7402?) There was some operating system > > support > > > for 6-12, but it was limited. > > Though 6-bit characterset werepraised early as a significant cost > reduced, as the price of systems came down, CDC heard more and more > complaints about 6-bit (not to be confused with 2-bit) characters > sets. It was an unfair bias, as many vendors of the day ran or > supported them. > > > The CPU was I don't know how many orders of magnitude slower. I seem to recall > > something around then about memory speeds in the 200 nanosecond range, but that > > may have been in a different context. > > Unless some chip designer gets really bored, we'll never know just how > fast a 6000 style CPU would be if it were built with modern > electronics. Any EEs care to venture a guess? > > > No, no, no. A 6600 never in its wildest dreams had a 200ns memory access: > > this was the days of real core memory, made of magnetic donuts on wires. > > > > Well, you could get a new word from memory every 100ns but that word > > has been desired for 800ns or more. Lots of parrallism going on what with > > the PPs and different parts of the CP all able to make memory request. > > > > Nearly everything was done in registers: 8 60-bit, 8 18-bit A(ddressing > > registers and another 8 18-bit general purpose registers. Mustn't forget > > the "stack" which held enough program space for a tight loop. > > Memory was slow, so they tried not to use it too much. > > Memory was 800ns. Though Cybers of that era phased memory into 4 > banks so that consecutive memory references didn't conflict. This > gave an effective rate near 200ns. Perhaps in some later Cybers. Idon't recall. But for 6000 class machines the memory modules ran at 1 mike, with a ten-way interleave, giving a max rate of 100 ns. Somewhere in here (comp.sys.cdc) someone posted a URL to an authorized PDF of Jim Thornton's book on the design of the 6600. It has disappeared from my server, but should be googlable. Just about all is revealed there. > > > CPUs on the 6000s came in two flavors. The monolithic 6400 and the > phased 6600. You could get a system with one CPU of either type, a > system with two 6400 CPUs, or a system with one of each. Two 6600 > style CPUs in the same cabinet wasn't available as the memory speed > wouldn't keep up. More significant was that you couldn't fit two 6600 CPU's and max memory in the cruciform cabinetry of the 6600. There was also a 6200, a de-clocked 6400 set up for lowball marketing purposes. > > > IIRC, idle PPs on NOS systems checked their input registers (IR) about > 8000 times per second. The IR was the first word of the PP's Message > Buffer, which was 8 words long, so the way that memory was phased, all > IRs wound up in the same memory bank. Of the possible > 1,250,000/second that the bank could muster, about 12% of them went to > idle PPs busily searching for something to do. > > > > There were Peripheral Processors which did the I/O. There were 10 or 20 of > > them, each of which had its own memory (12 bits of addressing?) and full access > > to the central processor's memory. > > Most 6000s were configured with a full complement of 20 PPs. Certain > pricing options allowed for the purchase of machines with less, but > throughput suffered pretty badly. Many were field upgraded to 20 PPs. I'll take serious issue with this. Virtually all of the 6000's and Cyber 70's I worked on during a 17-year period had only 10 ppu. The extra 10-ppu option was expensive and required a second cabinet. Most of the 20 ppu configurations I was aware of were connected to exotic type peripherals some of which you could be killed for knowing about. I did extensive benchmarking of these machines, and in most cases you would saturate CPU and memory long before you would run out of PPU. There were of course some exceptions, like one configuration with 32 tape drives. > > > The PP was a 12 bit processor. It had 4095.5 words. (Inside joke. > It had 4096 words, but word 7777 was rarely used as it wasn't directly > addressable.) Math in the PP was 18 bit. This allowed it to perform > address calculations on the full length of central memory. > > > One of them was special (PP 0), and > > contained the PP-resident part of the operating system. The rest were mostly or > > completely interchangeable. > > Correct. On most CDC operating systems, the PP portion of the system > monitor (MTR) ran in PP00. By convention, the console driver (DSD) > usually ran in PP01. All other PPs were "interchangeable". > > > It's possible that some devices were attached to > > certain PPs, so only they could access them. > > Devices were connected to channels. PPs could access any channel so > any PP could access any device. > > > Despite their shortcomings, there were some nice things about the CDC 6600 > > series machines, and their operating systems. Compared to the dominant > > mainframe of the day -- IBM 360s, 370s -- they had wonderful operating > > systems. > > They were so much easier to use than VM, MVS, CICS, etc... Depending > on the application, the 6000s outran the IBMs, sometimes the IBMs > outran the 6000s. They were definitely the two biggest horses in the > race. > > > To copy a file, you used the copy command. One-liner. To make a file bigger, > > you wrote to it. And so on. None of this IEBGENER/IEBCOPY foolishness. > > Files were just a series of bytes. (OK, 6-bit bytes, and there were those EOR > > marks, and oddball line terminators, but still...) > > Even the early DOS operating systems understood that to COPY a file > meant simply that the contents of one file should be duplicated under > a different name. I wonder why it took IBM so long to figure this > out? :) > > Though technically, the PP could write a partial word to disk, and > tape files could certainly contain something other that an even number > of CPU words, when translated to memory, files were always an exact > number of words. > > I too miss these old beauties. They certainly had heart! > > Kent ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 01 Jul 2002 13:33:49 -0500 Organization: Airnews.net! at Internet America Lines: 10 Message-ID: <9CB29FCC3F41C59A.EF703648D867E408.8B870FA1781CCB87@lp.airnews.net> X-Orig-Message-ID: <3D20A08D.A82F195E@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Mon Jul 1 13:35:46 2002 NNTP-Posting-Host: !Z.OU1k-Y(4RI)>1buWB (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!howland.erols.net!news.airnews.net!cabal10.airnews.net!cabal2.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111019 Larry__Weiss wrote: > The CDC 6600 was the very first machine that I ever programmed on (1969). > The physicists for sure loved its speed at that time. I sure wish I still > had the sources for those matrix crunching programs to just see how fast > they can be run today. > Does anyone know when UT Austin retired their CDC-6600 ? - LarryW ###### From: kent@nettally.com (Kent Olsen) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: 1 Jul 2002 11:37:08 -0700 Organization: http://groups.google.com/ Lines: 142 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> NNTP-Posting-Host: 204.89.69.20 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1025548628 13988 127.0.0.1 (1 Jul 2002 18:37:08 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 1 Jul 2002 18:37:08 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!postnews1.google.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111061 Expanding on a good set of answers: > Compared to today's PCs, puny in speed, memory and disk storage. You could > probably go to Wal-Mart or wherever and for less than $1000 buy something that > could emulate one. WallyWorld is selling an Athlon 850 system for about $300. It can simulate a floating point instruction mix faster than any 6000 or Cyber 70 machine ever built. My GS160 Alpha with 730Mhz CPUs can run the same simulated FP mix at over 13 MFlop. At most, only the 2 fastest 60 bit CPUs that CDC ever made could execute FP instructions that fast. > Emulating a CDC 6600 on a PC would be an interesting and probably pointless > programming challenge. Some of us would (do) consider it a "labor of love". :) > Well, sometimes 6-bit characters. That was what the hardware and the > operating > > system supported. There were also other schemes for representing more > than 63 > > (or 64) characters: 6-12 character sets, 8-in-12 and maybe others. (I > seem to > > recall something about 7.5 characters/word, but that may just be a > horrible, > > horrible dream.) Learning of the one-byte/two-byte characters in Unicode > > brought back memories of the 6-12 character set, where 7604 (octal) and 00 > both > > represent a colon. (Or was it 7402?) There was some operating system > support > > for 6-12, but it was limited. Though 6-bit characterset werepraised early as a significant cost reduced, as the price of systems came down, CDC heard more and more complaints about 6-bit (not to be confused with 2-bit) characters sets. It was an unfair bias, as many vendors of the day ran or supported them. > The CPU was I don't know how many orders of magnitude slower. I seem to recall > something around then about memory speeds in the 200 nanosecond range, but that > may have been in a different context. Unless some chip designer gets really bored, we'll never know just how fast a 6000 style CPU would be if it were built with modern electronics. Any EEs care to venture a guess? > No, no, no. A 6600 never in its wildest dreams had a 200ns memory access: > this was the days of real core memory, made of magnetic donuts on wires. > > Well, you could get a new word from memory every 100ns but that word > has been desired for 800ns or more. Lots of parrallism going on what with > the PPs and different parts of the CP all able to make memory request. > > Nearly everything was done in registers: 8 60-bit, 8 18-bit A(ddressing > registers and another 8 18-bit general purpose registers. Mustn't forget > the "stack" which held enough program space for a tight loop. > Memory was slow, so they tried not to use it too much. Memory was 800ns. Though Cybers of that era phased memory into 4 banks so that consecutive memory references didn't conflict. This gave an effective rate near 200ns. CPUs on the 6000s came in two flavors. The monolithic 6400 and the phased 6600. You could get a system with one CPU of either type, a system with two 6400 CPUs, or a system with one of each. Two 6600 style CPUs in the same cabinet wasn't available as the memory speed wouldn't keep up. IIRC, idle PPs on NOS systems checked their input registers (IR) about 8000 times per second. The IR was the first word of the PP's Message Buffer, which was 8 words long, so the way that memory was phased, all IRs wound up in the same memory bank. Of the possible 1,250,000/second that the bank could muster, about 12% of them went to idle PPs busily searching for something to do. > There were Peripheral Processors which did the I/O. There were 10 or 20 of > them, each of which had its own memory (12 bits of addressing?) and full access > to the central processor's memory. Most 6000s were configured with a full complement of 20 PPs. Certain pricing options allowed for the purchase of machines with less, but throughput suffered pretty badly. Many were field upgraded to 20 PPs. The PP was a 12 bit processor. It had 4095.5 words. (Inside joke. It had 4096 words, but word 7777 was rarely used as it wasn't directly addressable.) Math in the PP was 18 bit. This allowed it to perform address calculations on the full length of central memory. > One of them was special (PP 0), and > contained the PP-resident part of the operating system. The rest were mostly or > completely interchangeable. Correct. On most CDC operating systems, the PP portion of the system monitor (MTR) ran in PP00. By convention, the console driver (DSD) usually ran in PP01. All other PPs were "interchangeable". > It's possible that some devices were attached to > certain PPs, so only they could access them. Devices were connected to channels. PPs could access any channel so any PP could access any device. > Despite their shortcomings, there were some nice things about the CDC 6600 > series machines, and their operating systems. Compared to the dominant > mainframe of the day -- IBM 360s, 370s -- they had wonderful operating > systems. They were so much easier to use than VM, MVS, CICS, etc... Depending on the application, the 6000s outran the IBMs, sometimes the IBMs outran the 6000s. They were definitely the two biggest horses in the race. > To copy a file, you used the copy command. One-liner. To make a file bigger, > you wrote to it. And so on. None of this IEBGENER/IEBCOPY foolishness. > Files were just a series of bytes. (OK, 6-bit bytes, and there were those EOR > marks, and oddball line terminators, but still...) Even the early DOS operating systems understood that to COPY a file meant simply that the contents of one file should be duplicated under a different name. I wonder why it took IBM so long to figure this out? :) Though technically, the PP could write a partial word to disk, and tape files could certainly contain something other that an even number of CPU words, when translated to memory, files were always an exact number of words. I too miss these old beauties. They certainly had heart! Kent ###### From: J. Clarke Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 1 Jul 2002 14:50:18 -0400 Organization: http://extra.newsguy.com Lines: 46 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> NNTP-Posting-Host: p-456.newsdawg.com X-Newsreader: MicroPlanet Gravity v2.50 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews3 Xref: chonsp.franklin.ch alt.folklore.computers:111040 In article <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com>, mschaef@eris.io.com says... > In article , > Larry__Weiss wrote: > >"Eric S. Harris" wrote: > > > >I started digging around Google, and found that this was already asked > >back in 1999 and got 82 replies including an estimate that the 6600 was rated at > >about 1 MIPS. > > > > http://groups.google.com/groups?&q=%22Comparative+Speed+of+CDC+6600%22 > > > >The CDC 6600 was the very first machine that I ever programmed on (1969). > >The physicists for sure loved its speed at that time. I sure wish I still > >had the sources for those matrix crunching programs to just see how fast > >they can be run today. > > It's not the CDC 6600, but this site has information on one of its > almost contemporaries: > > http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&selm=92218.123445MACKINNO%40QUCDN.QueensU.CA > > has information on an IBM 360/91's relative performance to a number of > modern (for 1992) machines running the same Fortran code: > > >>IBM RS/6000 550 (FORTRAN 2.2): 0.07 seconds > >>IBM 3081/G (Fortran 1.3): 0.25 seconds > >>Sparcstation 1+ (FORTRAN 1.4): 0.38 seconds > >>486/25 (Lahey F77L-EM/32): 0.48 seconds > >>IBM 360/91 (Fortran H): 0.65 seconds > >>386/33 with 387 (Lahey F77L-EM/32): 0.90 seconds > > It's also contains some exposition on the subject. You might find it interesting to check out . Hercules is a nearly full System/390 emulator that runs on a PC. It's reputed to run 360 code about ten times faster than a 360 (which specifict model I forget). has a link to a machine running Hercules that is accessible online. -- -- --John Reply to jclarke at ae tee tee global dot net (used to be jclarke at eye bee em dot net) ###### From: Joe Pfeiffer Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: 01 Jul 2002 13:14:39 -0600 Organization: New Mexico State University Lines: 24 Message-ID: <1b65zz9s68.fsf@cs.nmsu.edu> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> NNTP-Posting-Host: viper.cs.nmsu.edu Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: bubba.NMSU.Edu 1025550879 2522 128.123.64.113 (1 Jul 2002 19:14:39 GMT) X-Complaints-To: usenet@bubba.NMSU.Edu NNTP-Posting-Date: 1 Jul 2002 19:14:39 GMT User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!snoopy.risq.qc.ca!msc1.onvoy!onvoy.com!hardy.tc.umn.edu!nunki.unm.edu!news.NMSU.Edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111014 kent@nettally.com (Kent Olsen) writes: > > WallyWorld is selling an Athlon 850 system for about $300. It can > simulate a floating point instruction mix faster than any 6000 or > Cyber 70 machine ever built. My GS160 Alpha with 730Mhz CPUs can run > the same simulated FP mix at over 13 MFlop. At most, only the 2 > fastest 60 bit CPUs that CDC ever made could execute FP instructions > that fast. Which two, and when were they built? FWIW, I'm very surprised that *any* machine ever built by CDC could go that fast. > Most 6000s were configured with a full complement of 20 PPs. Certain > pricing options allowed for the purchase of machines with less, but > throughput suffered pretty badly. Many were field upgraded to 20 PPs. 20? I thought the full complement was ten (actually a single PPU with ten register sets). -- Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605 Department of Computer Science FAX -- (505) 646-1002 New Mexico State University http://www.cs.nmsu.edu/~pfeiffer Southwestern NM Regional Science and Engr Fair: http://www.nmsu.edu/~scifair ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 23 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:Sej6GCffhJgE3xKeMGecaX3Ljl0= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 Jul 2002 20:38:50 GMT NNTP-Posting-Host: 65.58.59.232 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025555930 65.58.59.232 (Mon, 01 Jul 2002 13:38:50 PDT) NNTP-Posting-Date: Mon, 01 Jul 2002 13:38:50 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp1.phx1.gblx.net!nntp.gblx.net!nntp.gblx.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111031 kent@nettally.com (Kent Olsen) writes: > They were so much easier to use than VM, MVS, CICS, etc... Depending > on the application, the 6000s outran the IBMs, sometimes the IBMs > outran the 6000s. They were definitely the two biggest horses in the > race. note that CMS on VM ... had copyfile command ... effectively inherited from CTSS (aka some of the CMS people had worked on CTSS) ... although huge number of parameters were added to the copyfile over time .... eventually endowing it with all sorts of extra capability (as opposed to just simply doing a file copy). In the same bldg (545 tech sq) some other people that had worked on CTSS were working on Multics. Both CMS and unix trace some common heritage back to CTSS. here is page giving command correspondance between cms, vax, pc-dos, and unix: http://www.cc.vt.edu/cc/us/docs/unix/cmd-comp.html -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: greg@lewis.ucs.indiana.edu (Gregory Travis) Newsgroups: alt.folklore.computers,comp.sys.cdc, Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 1 Jul 2002 20:40:06 +0000 (UTC) Organization: A poorly-installed InterNetNews site Lines: 102 Distribution: inet Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> NNTP-Posting-Host: lewis.ucs.indiana.edu X-Trace: wilson.uits.indiana.edu 1025556013 4174 156.56.103.3 (1 Jul 2002 20:40:13 GMT) X-Complaints-To: news-admin@indiana.edu NNTP-Posting-Date: Mon, 1 Jul 2002 20:40:13 +0000 (UTC) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.uchicago.edu!vixen.cso.uiuc.edu!news.indiana.edu!lewis.ucs.indiana.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111050 In article <3D1F56EE.DAA8337E@mindspring.com>, Eric S. Harris wrote: >Perhaps someone in news:comp.sys.cdc has something to add to this? > >Larry__Weiss wrote: > > I just stumbled across some pics of the CDC6600: > http://www.bambi.net/computer_museum/cdc6600_and_console.jpg > http://online.sfsu.edu/~hl/mmm.html > > Just how powerful was that machine? How would it compare to modern-day > machines, especially PC's? Main memory was available in sizes up to 256K 60-bit words (about 2MB of 8-bit bytes). You could have up to 20 independent Peripheral Processors (PP), each with 4K 12-bit words (total 128KB 8-bit bytes of PP memory). Memory cycle time was 1 microsecond per bank, up to 32 banks per machine. However addresses could only be issued every 0.1 microsecond giving a maximum memory bandwith of 10,000,000 * 60 / 8 = 75MB/s Each PP could drive one of up to 24 channels at up to 1 12-bit word per microsecond (half-duplex) giving an aggregate I/O capacity of 20 * 12 * 1,000,000 / 8 = 30MB/s Three sets of CPU registers: 8 "A" registers, A0 through A7. 18 bits each. Used for address arithmetic and load/stores from/to memory. 8 "B" registers, B0 through B7. 18 bits each. Used for indexing, loop counts, etc. 8 "X" registers, X0 through X7. 60 bits each. Used for general-purpose arithmetic (including floating point) and loads/stores from/to memory. Some representative CPU instructions and timing SXi Xj (Set Xi to equal Xj.) .3 microseconds SXi Xj+Xk (60-bit floating point sum Xj and Xk to Xi) .4 microseconds SXi Xj*Xk (60-bit floating product Xj and Xk to Xi) 1 microsecond SXi Xj/Xk (60-bit floating divide Xj and Xk to Xi) 2.9 microseconds NO (No operation, pass) .1 microseconds As you can see, the machine could execute pass instructions at a rate of approx 10MIPS. Most instructions on the machine had a .3 microsecond cycle giving a nominal rating of 3MIPS overall. In addition, the machine had multiple function units, each capable of executing in parallel assuming no register conflict. The units were: BRANCH BOOLEAN SHIFT ADD (60-bit Floating point adds) LONG ADD (60-bit Integer adds) MULTIPLY (60-bit) (duplexed unit, two per machine) DIVIDE (60-bit) INCREMENT (18-bit, duplexed, two per machine) Because of the multiple functional units and register scoreboard, several operations could be in process simultaneously, further increasing throughput. Consider the following set of instructions: SB2 B3+B4 (INCREMENT UNIT) .3 microseconds LX2 4 (SHIFT UNIT) .3 microseconds SA0 B5+B6 (INCREMENT UNIT) .3 microseconds On a serial CPU the total time to execute that sequence would be .3+.3+.3 microseconds = .9 microseconds. However, since there are no register conflicts between instructions (no instruction uses a register used by any other instruction) and because each instruction can be dispatched to a separate functional unit (remember there are two increment units), the actual time to execute is .3 + .1 + .1 = .5 microseconds (instructions can be issued at a maximum rate of one per .1 microseconds). That would be a rate of 6MIPS (three instructions executed in .5 microseconds). Commodity hardware (meaning things you can I could afford to buy and take home) did not reach these levels of performance until the early-mid 1990s. The 6600 out-performed even most minicomputers (including top-of-the line VAXs) throughout the 1980s. 6600: Design begun 1960. First shown, running, to the public in Septemberof 1963 with production deliveries in fall of 1964. Fastest computer in the world from 1964 to 1969 when it was superceded by its upward-compatible big brother the 7600. 7600 was fastest computer in the world until introduction of Cray-1 in 1975. The 6600 architecture is one of the longest-lived on the planet, exceeded I believe only by the IBM 360. Several Cyber 170 systems exist today running legacy applications. Those Cyber 170s are code-compatible with the original 6600. That gives a 38-year lifespan for the architecture and instruction set. greg ###### From: greg@lewis.ucs.indiana.edu (Gregory Travis) Newsgroups: alt.folklore.computers,comp.sys.cdc, Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 1 Jul 2002 20:42:12 +0000 (UTC) Organization: A poorly-installed InterNetNews site Lines: 105 Distribution: inet Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> NNTP-Posting-Host: lewis.ucs.indiana.edu X-Trace: wilson.uits.indiana.edu 1025556139 4187 156.56.103.3 (1 Jul 2002 20:42:19 GMT) X-Complaints-To: news-admin@indiana.edu NNTP-Posting-Date: Mon, 1 Jul 2002 20:42:19 +0000 (UTC) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!newsfeed.stanford.edu!news.uchicago.edu!vixen.cso.uiuc.edu!news.indiana.edu!lewis.ucs.indiana.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111054 In article <3D1F56EE.DAA8337E@mindspring.com>, Eric S. Harris wrote: >Perhaps someone in news:comp.sys.cdc has something to add to this? > >Larry__Weiss wrote: > > I just stumbled across some pics of the CDC6600: > http://www.bambi.net/computer_museum/cdc6600_and_console.jpg > http://online.sfsu.edu/~hl/mmm.html > > Just how powerful was that machine? How would it compare to modern-day > machines, especially PC's? Main memory was available in sizes up to 256K 60-bit words (about 2MB of 8-bit bytes). You could have up to 20 independent Peripheral Processors (PP), each with 4K 12-bit words (total 128KB 8-bit bytes of PP memory). Memory cycle time was 1 microsecond per bank, up to 32 banks per machine. However addresses could only be issued every 0.1 microsecond giving a maximum memory bandwith of 10,000,000 * 60 / 8 = 75MB/s Each PP could drive one of up to 24 channels at up to 1 12-bit word per microsecond (half-duplex) giving an aggregate I/O capacity of 20 * 12 * 1,000,000 / 8 = 30MB/s Three sets of CPU registers: 8 "A" registers, A0 through A7. 18 bits each. Used for address arithmetic and load/stores from/to memory. 8 "B" registers, B0 through B7. 18 bits each. Used for indexing, loop counts, etc. 8 "X" registers, X0 through X7. 60 bits each. Used for general-purpose arithmetic (including floating point) and loads/stores from/to memory. Some representative CPU instructions and timing SXi Xj (Set Xi to equal Xj.) .3 microseconds SXi Xj+Xk (60-bit floating point sum Xj and Xk to Xi) .4 microseconds SXi Xj*Xk (60-bit floating product Xj and Xk to Xi) 1 microsecond SXi Xj/Xk (60-bit floating divide Xj and Xk to Xi) 2.9 microseconds NO (No operation, pass) .1 microseconds As you can see, the machine could execute pass instructions at a rate of approx 10MIPS. Most instructions on the machine had a .3 microsecond cycle giving a nominal rating of 3MIPS overall. In addition, the machine had multiple function units, each capable of executing in parallel assuming no register conflict. The units were: BRANCH BOOLEAN SHIFT ADD (60-bit Floating point adds) LONG ADD (60-bit Integer adds) MULTIPLY (60-bit) (duplexed unit, two per machine) DIVIDE (60-bit) INCREMENT (18-bit, duplexed, two per machine) Because of the multiple functional units and register scoreboard, several operations could be in process simultaneously, further increasing throughput. Consider the following set of instructions: SB2 B3+B4 (INCREMENT UNIT) .3 microseconds LX2 4 (SHIFT UNIT) .3 microseconds SA0 B5+B6 (INCREMENT UNIT) .3 microseconds On a serial CPU the total time to execute that sequence would be .3+.3+.3 microseconds = .9 microseconds. However, since there are no register conflicts between instructions (no instruction uses a register used by any other instruction) and because each instruction can be dispatched to a separate functional unit (remember there are two increment units), the actual time to execute is .3 + .1 + .1 = .5 microseconds (instructions can be issued at a maximum rate of one per .1 microseconds). That would be a rate of 6MIPS (three instructions executed in .5 microseconds). Commodity hardware (meaning things you can I could afford to buy and take home) did not reach these levels of performance until the early-mid 1990s. The 6600 out-performed even most minicomputers (including top-of-the line VAXs) throughout the 1980s. 6600: Design begun 1960. First shown, running, to the public in Septemberof 1963 with production deliveries in fall of 1964. Fastest computer in the world from 1964 to 1969 when it was superceded by its upward-compatible big brother the 7600. 7600 was fastest computer in the world until introduction of Cray-1 in 1975. The 6600 architecture is one of the longest-lived on the planet, exceeded I believe only by the IBM 360. Several Cyber 170 systems exist today running legacy applications. Those Cyber 170s are code-compatible with the original 6600. That gives a 38-year lifespan for the architecture and instruction set. Other 6600 "firsts" included the use of a video display operator's console in place of the old "flashing lights" console. greg ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 201 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:aTWk638hFHgKiDokJKvL6yBcCSM= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 Jul 2002 21:05:00 GMT NNTP-Posting-Host: 65.58.59.232 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025557500 65.58.59.232 (Mon, 01 Jul 2002 14:05:00 PDT) NNTP-Posting-Date: Mon, 01 Jul 2002 14:05:00 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp1.phx1.gblx.net!nntp.gblx.net!nntp.gblx.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111036 "Douglas H. Quebbeman" writes: > Although the Wheelers refer to RAIN/RAIN4, LINPACK was also > used for relative speed comparisons, and has been ported to > modern platforms and other languages. You should be able to > find any number of different versions all over the net... linpack numbers (including 6600): http://ap01.physik.uni-greifswald.de/~ftp/bench/linpack.html bunch of stuff from above Computer N=100(Mflops) ------------------------------------ --- ------------- Cray T916 (1 proc. 2.2 ns) 522 Hitachi S-3800/180(1 proc 2 ns) 408 Cray-2/4-256 (1 proc. 4.1 ns) 38 IBM RISC Sys/6000-580 (62.5MHz) 38 IBM ES/9000-520 (1 proc. 9 ns) 38 SGI CHALLENGE/Onyx (6.6ns, 2 proc) 38 DEC 4000-610 Alpha AXP(160 MHz) 36 NEC SX-1 36 FPS 510S MCP707 (7 proc. 25 ns) 33 CDC Cyber 2000V 32 Convex C-3430 (3 proc.) 32 NEC SX-1E 32 SGI Indigo2 (R4400/200MHz) 32 Alliant FX/2800-200 (14 proc) 31 IBM RISC Sys/6000-970 (50 MHz) 31 IBM ES/9000-511 VF(1 proc 11ns) 30 DEC 3000-500 Alpha AXP(150 MHz) 30 Alliant FX/2800-200 (12 proc) 29 HP 9000/715 (75 MHz) 29 Sun Sparc 20 90 MHz, (1 proc) 29 Alliant FX/2800 210 (1 proc) 25 ETA 10-P (1 proc. 24 ns) 27 Convex C-3420 (2 proc.) 27 Cray-1S (12.5 ns) 27 DEC 2000-300 Alpha AXP 6.7 ns 26 IBM RISC Sys/6000-950 (42 MHz) 26 SGI CHALLENGE/Onyx (6.6ns, 1 proc) 26 Alliant FX/2800-200 (8 proc) 25 NAS AS/EX 60 VPF 25 HP 9000/750 (66 MHz) 24 IBM ES/9000-340 VF (14.5 ns) 23 Meiko CS2 (1 proc) 24 Fujitsu M1800/20 23 DEC VAX 9000 410VP(1 proc 16 ns) 22 IBM ES/9000-320 VF (1 proc 15 ns) 22 IBM RISC Sys/6000-570 (50 MHz) 22 Multiflow TRACE 28/300 22 Convex C-3220 (2 proc.) 22 Alliant FX/2800-200 (6 proc) 21 Siemens VP400-EX (7 ns) 21 IBM ES/9221-211 (16 ns) 21 FPS Model 522 20 Fujitsu VP-400 20 IBM RISC Sys/6000-530H(33 MHz) 20 Siemens VP200-EX (7 ns) 20 Amdahl 1400 19 Convex C-3410 (1 proc.) 19 IBM ES/9000 Model 260 VF (15 ns) 19 IBM RISC Sys/6000-550L(42 MHz) 19 Cray S-MP/11 (1 proc. 30 ns) 18 Fujitsu VP-200 18 HP 9000/720 (50 MHz) 18 IBM ES/9221-201 (16 ns) 18 NAS AS/EX 50 VPF 18 SGI 4D/480(8 proc) 40MHz 18 Siemens VP100-EX (7 ns) 18 Sun 670MP Ross Hypersparc(55Mhz) 18 Alliant FX/2800-200 (4 proc) 17 Amdahl 1100 17 CDC CYBER 205 (4-pipe) 17 CDC CYBER 205 (2-pipe) 17 Convex C-3210 (1 proc.) 17 Convex C-210 (1 proc.) 17 Cray XMS (55 ns) 17 Hitachi S-810/20 17 IBM ES/9000 Model 210 VF (15 ns) 17 Siemens VP50-EX (7 ns) 17 Multiflow TRACE 14/300 17 Hitachi S-810/10 16 IBM 3090/180J VF (1 proc, 14.5 ns) 16 Fujitsu VP-100 16 Amdahl 500 16 Hitachi M680H/vector 16 SGI Crimson(1 proc 50 MHz R4000) 16 FPS Model 511 15 Hitachi M680H 15 IBM RISC Sys/6000-930 (25 MHz) 15 Kendall Square (1 proc) 15 NAS AS/EX 60 15 SGI 4D/440(4 proc) 40MHz 15 Siemens H120F 15 Cydrome CYDRA 5 14 Fujitsu VP-50 14 IBM ES/9000 Model 190 VF(15 ns) 14 IBM POWERPC 250 (66 MHz) 13 IBM 3090/180E VF 13 SGI 4D/340(4 proc) 33MHz 13 CDC CYBER 990E 12 Cray-1S (12.5 ns, 1983 run) 12 Gateway 2000 P5-100XL 12 IBM RISC Sys/6000-520H(25 MHz) 12 SGI Indigo 4000 50MHz 12 Stardent 3040 12 CDC 4680InfoServer (60 MHz) 11 Cray S-MP/MCP101 (1 proc. 25 ns) 11 FPS 510S MCP101 (1 proc. 25 ns) 11 IBM ES/9000 Model 340 11 Meiko Comp. Surface (1 proc) 11 Gateway 2000 P5-90(90 MHz Pentium) 11 SGI Power Series 50MHz R4000 11 Stardent 3020 11 Sperry 1100/90 ext w/ISP 11 Multiflow TRACE 7/300 11 DEC VAX 6000/410 (1 proc) 1.2 ELXSI 6420 1.2 Gateway 2000/Micronics 486DX/33 1.2 Gateway Pentium (66HHz) 1.2 IBM ES/9000 Model 120 1.2 IBM 370/168 Fast Mult 1.2 IBM 4381 90E 1.2 IBM 4381-13 1.2 MIPS M/800 (12.5MHz) 1.2 Prime P6350 1.2 Siemans 7580-E 1.2 Amdahl 470 V/6 1.1 Compaq Deskpro 486/33l-120 w/487 1.1 SUN 4/260 1.1 ES1066 (1 proc. 80 ns Russian) 1.0 CDC CYBER 180-840 .99 Solbourne .98 IBM 4381-22 .97 IBM 4381 MG2 .96 ICL 3980 w/FPU .93 IBM-486 33MHz .94 Siemens 7860E .92 Concurrent 3280XP .87 MIPS M800 w/R2010 FP .87 Gould PN 9005 .87 VAXstation 3100-76 .85 IBM 370/165 Fast Mult .77 Prime P9955II .72 DEC VAX 8530 .73 HP 9000 Series 850 .71 HP/Apollo DN4500 (68030 + FPA) .60 Mentor Graphics Computer .60 MIPS M/500 ( 8.3HHz) .60 Data General MV/20000 .59 IBM 9377-80 .58 Sperry 1100/80 w/SAM .58 CDC CYBER 930-31 .58 Russian PS-2100 .57 Gateway 486DX-2 (66HHz) .56 Harris H1200 .56 HP/Apollo DN4500 (68030) .55 Harris HCX-9 .50 Pyramid 9810 .50 HP 9000 Series 840 .49 DEC VAX 8600 .48 Harris HCX-7 w/fpp .48 CDC 6600 .48 IBM 4381-21 .47 SUN-3/260 + FPA .46 CDC CYBER 170-835 .44 HP 9000 Series 840 .43 IBM RT 135 .42 Harris H1000 .41 microVAX 3200/3500/3600 .41 Apple Macintosh IIfx .41 Apollo DN5xxT FPX .40 microVAX 3200/3500/3600 .40 IBM 9370-60 .40 Sun-3/160 + FPA .40 Prime P9755 .40 Ridge 3200 Model 90 .39 IBM 4381-11 .39 Gould 32/9705 mult acc .39 NORSK DATA ND-570/2 .38 Sperry 1100/80 .38 Apple Mac IIfx .37 CDC CYBER 930-11 .37 Sequent Symmetry (386 w/fpa) .37 CONCEPT 32/8750 .36 Celerity C1230 .36 IBM RT PC 6150/115 fpa2 .36 IBM 9373-30 .36 CDC 6600 .36 IBM 370/158 .22 IBM PS/2-70 (16 MHz) .12 IBM AT w/80287 .012 IBM PC w/8087 .012 IBM PC w/8087 .0069 Apple Mac II .0064 Atari ST .0051 Apple Macintosh .0038 -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <9CB29FCC3F41C59A.EF703648D867E408.8B870FA1781CCB87@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 20 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:rLP+1SyRtCK0tpKoUow1OFw4mzI= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 Jul 2002 21:12:50 GMT NNTP-Posting-Host: 65.58.59.232 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025557970 65.58.59.232 (Mon, 01 Jul 2002 14:12:50 PDT) NNTP-Posting-Date: Mon, 01 Jul 2002 14:12:50 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!eusc.inter.net!newsfeed.stueberl.de!cox.net!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111038 Larry__Weiss writes: > > Does anyone know when UT Austin retired their CDC-6600 ? > don't know about campus ... but balcones research had a cray and my wife and I manage to donate a bunch of HYPERchannel equipment to them for interconnecting various stuff. thorton after working on 6600 left cdc and founded NSC and built HYPERChannels. random past stuff http://www.garlic.com/~lynn/99.html#119 Computer, supercomputers & related http://www.garlic.com/~lynn/2001.html#19 Disk caching and file systems. Disk history...people forget http://www.garlic.com/~lynn/2001.html#20 Disk caching and file systems. Disk history...people forget -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Message-ID: <3D20E051.3D18A58B@attbi.com> From: Joe Yuska X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <9CB29FCC3F41C59A.EF703648D867E408.8B870FA1781CCB87@lp.airnews.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 24 NNTP-Posting-Host: 24.91.67.33 X-Complaints-To: abuse@attbi.com X-Trace: rwcrnsc51.ops.asp.att.net 1025564622 24.91.67.33 (Mon, 01 Jul 2002 23:03:42 GMT) NNTP-Posting-Date: Mon, 01 Jul 2002 23:03:42 GMT Organization: AT&T Broadband Date: Mon, 01 Jul 2002 23:03:42 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!wn1feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!rwcrnsc51.ops.asp.att.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111070 Anne & Lynn Wheeler wrote: > Larry__Weiss writes: > > > > Does anyone know when UT Austin retired their CDC-6600 ? > > > > don't know about campus ... but balcones research had a cray and my > wife and I manage to donate a bunch of HYPERchannel equipment to them > for interconnecting various stuff. > > thorton after working on 6600 left cdc and founded NSC and built > HYPERChannels. In between the 6600 and departure did the bulk of the work on the 6400 CPU. Then ran a research group somewhat in competition with Chippewa Falls where they did the initial work on the STAR pipeline project that eventually turned into the CYBER 200 series. Joe Yuska ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fef08_1@news.iglou.com> <3D209F50.A6CAA51E@attbi.com> Organization: JLW Consulting X-Newsreader: trn 4.0-test76 (Apr 2, 2001) From: woolsey@jlw.com (Jeff Woolsey) Originator: woolsey@jlw.com (Jeff Woolsey) X-Original-NNTP-Posting-Host: arglebargle.jlw.com Message-ID: <3d2154fe$1@news.meer.net> X-Original-Trace: 2 Jul 2002 00:23:42 -0700, arglebargle.jlw.com Lines: 40 Date: 2 Jul 2002 00:23:42 -0700 NNTP-Posting-Host: 209.157.152.10 X-Complaints-To: abuse@verio.net X-Trace: dfw-read.news.verio.net 1025594695 209.157.152.10 (Tue, 02 Jul 2002 07:24:55 GMT) NNTP-Posting-Date: Tue, 02 Jul 2002 07:24:55 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news-lond.gip.net!news.gsl.net!gip.net!kibo.news.demon.net!demon!dfw-peer!news.verio.net!dfw-read.news.verio.net.POSTED!news.meer.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111074 In article <3D209F50.A6CAA51E@attbi.com>, Joe Yuska wrote: >Kent Olsen wrote: >> > There were Peripheral Processors which did the I/O. There were 10 or 20 of >> > them, each of which had its own memory (12 bits of addressing?) and >full access >> > to the central processor's memory. >> >> Most 6000s were configured with a full complement of 20 PPs. Certain >> pricing options allowed for the purchase of machines with less, but >> throughput suffered pretty badly. Many were field upgraded to 20 PPs. > >I'll take serious issue with this. Virtually all of the 6000's and Cyber 70's I >worked on during a 17-year period had only 10 ppu. The extra 10-ppu option was >expensive and required a second cabinet. Most of the 20 ppu >configurations I was >aware of were connected to exotic type peripherals some of which you >could be killed >for knowing about. And I take issue with that. Of the dozen or so systems I worked on, only the 6400 and one Cyber 73 only had 10 PPs. All of the rest were either ordered with more, or upgraded shortly after delivery. The other Cyber 73 has 17 PPs, the Cyber 74 had 14 and then 17, and all the rest were 170s or later, all with at least 14 PPs. None had the full complement, though. >> The PP was a 12 bit processor. It had 4095.5 words. (Inside joke. >> It had 4096 words, but word 7777 was rarely used as it wasn't directly >> addressable.) It's addressible via instructions which do not involve the 12-bit incrementer... Limits uesfulness a bit. -- Jeff Woolsey {woolsey,jlw}@{jlw,jxh}.com "I didn't get a 'Harrumph!' out of _that_ guy." -Gov Le Petomaine "Delete! Delete! OK!" -Dr. Bronner on disk space management "A toy robot!!!!" -unlucky Japanese scientist ###### Newsgroups: alt.folklore.computers,comp.sys.cdc, Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Distribution: inet Organization: JLW Consulting X-Newsreader: trn 4.0-test76 (Apr 2, 2001) From: woolsey@jlw.com (Jeff Woolsey) Originator: woolsey@jlw.com (Jeff Woolsey) X-Original-NNTP-Posting-Host: arglebargle.jlw.com Message-ID: <3d215922$1@news.meer.net> X-Original-Trace: 2 Jul 2002 00:41:22 -0700, arglebargle.jlw.com Lines: 39 Date: 2 Jul 2002 00:41:22 -0700 NNTP-Posting-Host: 209.157.152.10 X-Complaints-To: abuse@verio.net X-Trace: dfw-read.news.verio.net 1025595753 209.157.152.10 (Tue, 02 Jul 2002 07:42:33 GMT) NNTP-Posting-Date: Tue, 02 Jul 2002 07:42:33 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.media.kyoto-u.ac.jp!sjc-peer.news.verio.net!news.verio.net!dfw-read.news.verio.net.POSTED!news.meer.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111134 In article , Gregory Travis wrote: Oh, I just love nitpicking on my favorite subject: >Some representative CPU instructions and timing > > SXi Xj (Set Xi to equal Xj.) .3 microseconds Well, Set Xi to the lower 18 bits of Xj sign-extended. > SXi Xj+Xk (60-bit floating point sum Xj and Xk to Xi) .4 microseconds > SXi Xj*Xk (60-bit floating product Xj and Xk to Xi) 1 microsecond > SXi Xj/Xk (60-bit floating divide Xj and Xk to Xi) 2.9 microseconds The rest of these you can get away with through the miracle of OPDEF and make them the same as the FX and/or RX instructions (pick one). >The 6600 architecture is one of the longest-lived on the planet, exceeded >I believe only by the IBM 360. Several Cyber 170 systems exist today running >legacy applications. Those Cyber 170s are code-compatible with the original >6600. That gives a 38-year lifespan So far. >for the architecture and instruction >set. >Other 6600 "firsts" included the use of a video display operator's console "Video" is a smidge inaccurate. It's a very glorified oscilloscope (Tek scopes of the day had similar calligraphic characters for readouts). -- Jeff Woolsey {woolsey,jlw}@{jlw,jxh}.com "I didn't get a 'Harrumph!' out of _that_ guy." -Gov Le Petomaine "Delete! Delete! OK!" -Dr. Bronner on disk space management "A toy robot!!!!" -unlucky Japanese scientist ###### From: Bruce Hoult Newsgroups: alt.folklore.computers,comp.sys.cdc, Subject: Re: CDC6600 - just how powerful a machine was it? Distribution: inet References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> User-Agent: MT-NewsWatcher/3.2 (PPC Mac OS X) Message-ID: Lines: 18 Date: Tue, 02 Jul 2002 19:46:55 +1200 NNTP-Posting-Host: 203.79.123.131 X-Complaints-To: abuse@tsnz.net X-Trace: news02.tsnz.net 1025596015 203.79.123.131 (Tue, 02 Jul 2002 19:46:55 NZST) NNTP-Posting-Date: Tue, 02 Jul 2002 19:46:55 NZST Organization: TelstraClear Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!enews.sgi.com!news.xtra.co.nz!newsfeed01.tsnz.net!news02.tsnz.net!bruce Xref: chonsp.franklin.ch alt.folklore.computers:111130 In article <3d215922$1@news.meer.net>, woolsey@jlw.com (Jeff Woolsey) wrote: > >Other 6600 "firsts" included the use of a video display operator's console > > "Video" is a smidge inaccurate. It's a very glorified oscilloscope > (Tek scopes of the day had similar calligraphic characters for > readouts). At the university I went to in the early 80's there was a Tek terminal on the VAX. It was a pain because when your text output got to the bottom of the screen it would start again from the first line and overprint (without erasing) whatever was there previously. So you had to hit some sort of "clear screen" key all the time. Also, it just had some sort of very long timeconstant phosphor, because the text (and graphics) would fade away after a few minutes. -- Bruce ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <3D209F50.A6CAA51E@attbi.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 07:06:13 -0400 Lines: 42 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d218927$1_3@news.iglou.com> X-Trace: news.iglou.com 1025607975 204.250.0.238 (2 Jul 2002 07:06:15 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed1.cidera.com!Cidera!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111079 "Joe Yuska" wrote in message = news:3D209F50.A6CAA51E@attbi.com... > > Most 6000s were configured with a full complement of 20 PPs. = Certain > > pricing options allowed for the purchase of machines with less, but > > throughput suffered pretty badly. Many were field upgraded to 20 = PPs. >=20 > I'll take serious issue with this. Virtually all of the 6000's and = Cyber 70's I > worked on during a 17-year period had only 10 ppu. The extra 10-ppu = option was > expensive and required a second cabinet. Most of the 20 ppu = configurations I was > aware of were connected to exotic type peripherals some of which you = could be killed > for knowing about. The 6000 Series Introduction & Peripheral Processor Training Manual shows space for the second set of 10 PPs inside the 6600 mainframe in: Wing 3 Chassis 9 between (Core) Banks 11 & 12; Wing 3 Chassis 10 between (Core) Banks 15 & 16; Wing 3 Chassis 12 between the two display controllers; Wing 1 Chassis 4 between (Core) Banks 5 & 6; Wing 4 Chassis 13 between (Core) Banks 21 & 22; Wing 4 Chassis 14 between (Core) Banks 25 & 26; Wing 4 Chassis 15 between (Core) Banks 31 & 32; Wing 4 Chassis 16 between (Core) Banks 35 & 36. Perhaps those eyes-only systems to which you refer had yet even still more exotic hardware inside the cruciform mainframe. Some other eyes-only sites used CDC 1700s, with the 6600 running the Time-Critical OS Monitor as the operating system... ...or so I hear... ;) -dq ###### From: kent@nettally.com (Kent Olsen) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: 2 Jul 2002 05:46:41 -0700 Organization: http://groups.google.com/ Lines: 39 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <1b65zz9s68.fsf@cs.nmsu.edu> NNTP-Posting-Host: 204.89.69.20 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1025614002 20650 127.0.0.1 (2 Jul 2002 12:46:42 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 2 Jul 2002 12:46:42 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!newsfeed.stanford.edu!postnews1.google.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111126 > Which two, and when were they built? FWIW, I'm very surprised that > *any* machine ever built by CDC could go that fast. The 2000 and 990. When the 990 came out it was the fastest scalar CPU in the world. (The scalar portion of the Cray and 205 CPUs notwithstanding.) It also could perform I/O with anyone. The Mormon church bought a pair of them to build their geneology database after the 990 beat a comparable 390/3090 in a database benchmark. The 2000 was the fastest commercial machine that CDC ever built. It's CPU performance is about twice the 990s. > > Most 6000s were configured with a full complement of 20 PPs. Certain > > pricing options allowed for the purchase of machines with less, but > > throughput suffered pretty badly. Many were field upgraded to 20 PPs. > > 20? I thought the full complement was ten (actually a single PPU with > ten register sets). There seems to be some disagreement on the board about this. A bank of up to 10 PPs was part of an "I/O subsystem" that consisted of the PPs and channels. If you had one bank you had up to 10 PPs and the bank's full complement of channels. If you had two banks you got the second set of channels and up to 10 additional PPs. IIRC, 6000s came with 7, 10, 14, 17, or 20 PPs. I seem to remember that a 7 PP system was delivered, but I don't recall the details (does anyone else?). As the 6000s became more popular 10 PPs weren't enough. I worked on at least a dozen sites with 6000s or Cyber 70s (basically, the same thing with slightly different packaging) and recall only one of them with less than 17 PPs. Of course, most of these customers were military or education. Commercial accounts may have behaved differently due to costs. Kent ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 02 Jul 2002 07:57:33 -0500 Organization: Airnews.net! at Internet America Lines: 11 Message-ID: X-Orig-Message-ID: <3D21A33D.C28FA166@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library1-aux.airnews.net NNTP-Posting-Time: Tue Jul 2 08:06:59 2002 NNTP-Posting-Host: !_NYB1k-VOgT5`I,?O:5 (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news.airnews.net!cabal11.airnews.net!cabal1.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111081 Jeff Woolsey wrote: > "Video" is a smidge inaccurate. It's a very glorified oscilloscope > (Tek scopes of the day had similar calligraphic characters for > readouts). > The video monitors on the CDC-6600 were very functional. I remember seeing chessboards displayed (UT Austin had their 6600 console on public display behind a glass wall in the computation center). - LarryW ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 02 Jul 2002 08:03:34 -0500 Organization: Airnews.net! at Internet America Lines: 22 Message-ID: <593184F5F1D3C2F2.F6682B4E87BEFEBF.072C191C5CB2CFEF@lp.airnews.net> X-Orig-Message-ID: <3D21A4A6.A7650282@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library1-aux.airnews.net NNTP-Posting-Time: Tue Jul 2 08:13:00 2002 NNTP-Posting-Host: !YU&01k-X"K.A/u1bs(O (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news.airnews.net!cabal11.airnews.net!cabal1.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111091 Bruce Hoult wrote: > Jeff Woolsey wrote: > > "Video" is a smidge inaccurate. It's a very glorified oscilloscope > > (Tek scopes of the day had similar calligraphic characters for > > readouts). > > At the university I went to in the early 80's there was a Tek terminal > on the VAX. It was a pain because when your text output got to the > bottom of the screen it would start again from the first line and > overprint (without erasing) whatever was there previously. So you had > to hit some sort of "clear screen" key all the time. Also, it just had > some sort of very long timeconstant phosphor, because the text (and > graphics) would fade away after a few minutes. > That was probably a TEK 4014 "storage tube" based terminal. The CDC-6600's console displays were not terminals of that sort. They were refreshed calligraphic CRT's. Very nice to have at that time. Does anyone have a picture of a CDC-6600 console with an image displayed? - LarryW ###### Newsgroups: alt.folklore.computers,comp.sys.cdc From: "Douglas A. Gwyn" Subject: Re: CDC6600 - just how powerful a machine was it? X-Nntp-Posting-Host: swctd.arl.army.mil Content-Type: text/plain; charset=us-ascii Message-ID: <3D21C487.5C6B355A@null.net> Sender: news@arl.army.mil (News Administration ) Content-Transfer-Encoding: 7bit Organization: U.S. Army Research Laboratory X-Accept-Language: en References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> Mime-Version: 1.0 Date: Tue, 2 Jul 2002 15:19:35 GMT X-Mailer: Mozilla 4.79 [en] (X11; U; SunOS 5.8 sun4u) Lines: 16 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!newsfeed.vmunix.org!news2.euro.net!transit.news.xs4all.nl!skynet.be!skynet.be!oanews!arlnews!blaze!news Xref: chonsp.franklin.ch alt.folklore.computers:111175 Bruce Hoult wrote: > At the university I went to in the early 80's there was a Tek terminal > on the VAX. It was a pain because when your text output got to the > bottom of the screen it would start again from the first line and > overprint (without erasing) whatever was there previously. So you had > to hit some sort of "clear screen" key all the time. Also, it just had > some sort of very long timeconstant phosphor, because the text (and > graphics) would fade away after a few minutes. Those are Tektronix model 40xx series and weren't really meant for extended use as text terminals. They were "storage" refresh devices which should have had the screen reflooded (erased) every 20 minutes or so for longevity. I wish I knew where I could find a 4014 in pristine condition.. Anyway, not the same as the calligraphic scopes being talked about. ###### From: tmm@spamfilter.asns.tr.unisys.com (Tim McCaffrey) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 17:31:46 +0000 (UTC) Organization: A series networking Lines: 42 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <3D209F50.A6CAA51E@attbi.com> NNTP-Posting-Host: tr-mccafftm.na.uis.unisys.com X-Trace: trsvr.tr.unisys.com 1025631106 25647 192.63.212.94 (2 Jul 2002 17:31:46 GMT) X-Complaints-To: usenet@tr.unisys.com NNTP-Posting-Date: Tue, 2 Jul 2002 17:31:46 +0000 (UTC) X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!bbnews1.unisys.com!trsvr.tr.unisys.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111167 In article <3D209F50.A6CAA51E@attbi.com>, jyuska@attbi.com says... > > > >Perhaps in some later Cybers. Idon't recall. But for 6000 class machines the memory >modules ran at 1 mike, with a ten-way interleave, giving a max rate of 100 ns. > The CDC 6000 series used 32 way interleave, 1 microsecond major cycle (memory cycle time) and 100ns minor cycle (instruction execution rate). >I'll take serious issue with this. Virtually all of the 6000's and Cyber 70's I >worked on during a 17-year period had only 10 ppu. The extra 10-ppu option was >expensive and required a second cabinet. Most of the 20 ppu configurations I was >aware of were connected to exotic type peripherals some of which you could be killed >for knowing about. > >I did extensive benchmarking of these machines, and in most cases you would saturate >CPU and memory long before you would run out of PPU. > >There were of course some exceptions, like one configuration with 32 tape drives. > I would be surprised if there was a 10 PPU add on option, at least for the 6500. Michigan State would have probably bought it, if available. When the Cyber 170/750 came in we could only use 10 of the 20 PP's, because our customized OS needed to be updated. When the additional 10 PPs were usable, it made a significant difference. One reason may have been because the OS dedicated (I think) 4 PPs to certain tasks (monitor, Argus (terminal support), Console, and something else). - Tim ###### From: Joe Pfeiffer Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: 02 Jul 2002 12:08:13 -0600 Organization: New Mexico State University Lines: 12 Distribution: inet Message-ID: <1b4rfim29e.fsf@cs.nmsu.edu> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> NNTP-Posting-Host: viper.cs.nmsu.edu Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: bubba.NMSU.Edu 1025633292 16713 128.123.64.113 (2 Jul 2002 18:08:12 GMT) X-Complaints-To: usenet@bubba.NMSU.Edu NNTP-Posting-Date: 2 Jul 2002 18:08:12 GMT User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!opentransit.net!news.tele.dk!small.news.tele.dk!137.192.6.2!msc1.onvoy!onvoy.com!hardy.tc.umn.edu!nunki.unm.edu!news.NMSU.Edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111156 woolsey@jlw.com (Jeff Woolsey) writes: > > "Video" is a smidge inaccurate. It's a very glorified oscilloscope > (Tek scopes of the day had similar calligraphic characters for > readouts). Didn't PDP-1 do that? -- Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605 Department of Computer Science FAX -- (505) 646-1002 New Mexico State University http://www.cs.nmsu.edu/~pfeiffer Southwestern NM Regional Science and Engr Fair: http://www.nmsu.edu/~scifair ###### From: Joe Pfeiffer Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: 02 Jul 2002 12:21:05 -0600 Organization: New Mexico State University Lines: 13 Message-ID: <1bznxakn3i.fsf@cs.nmsu.edu> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <1b65zz9s68.fsf@cs.nmsu.edu> NNTP-Posting-Host: viper.cs.nmsu.edu Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: bubba.NMSU.Edu 1025634064 17955 128.123.64.113 (2 Jul 2002 18:21:04 GMT) X-Complaints-To: usenet@bubba.NMSU.Edu NNTP-Posting-Date: 2 Jul 2002 18:21:04 GMT User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!upp1.onvoy!msc1.onvoy!onvoy.com!hardy.tc.umn.edu!nunki.unm.edu!news.NMSU.Edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111160 kent@nettally.com (Kent Olsen) writes: > > Which two, and when were they built? FWIW, I'm very surprised that > > *any* machine ever built by CDC could go that fast. > > The 2000 and 990. Wow -- I'd thought they'd pretty much abandoned that market by 1980... -- Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605 Department of Computer Science FAX -- (505) 646-1002 New Mexico State University http://www.cs.nmsu.edu/~pfeiffer Southwestern NM Regional Science and Engr Fair: http://www.nmsu.edu/~scifair ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d215922$1@news.meer.net> <593184F5F1D3C2F2.F6682B4E87BEFEBF.072C191C5CB2CFEF@lp.airnews.net> Organization: JLW Consulting X-Newsreader: trn 4.0-test76 (Apr 2, 2001) From: woolsey@jlw.com (Jeff Woolsey) Originator: woolsey@jlw.com (Jeff Woolsey) X-Original-NNTP-Posting-Host: arglebargle.jlw.com Message-ID: <3d21f3f7$1@news.meer.net> X-Original-Trace: 2 Jul 2002 11:41:59 -0700, arglebargle.jlw.com Lines: 20 Date: 2 Jul 2002 11:41:59 -0700 NNTP-Posting-Host: 209.157.152.10 X-Complaints-To: abuse@verio.net X-Trace: dfw-read.news.verio.net 1025635396 209.157.152.10 (Tue, 02 Jul 2002 18:43:16 GMT) NNTP-Posting-Date: Tue, 02 Jul 2002 18:43:16 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!iad-peer.news.verio.net!news.verio.net!dfw-read.news.verio.net.POSTED!news.meer.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111142 In article <593184F5F1D3C2F2.F6682B4E87BEFEBF.072C191C5CB2CFEF@lp.airnews.net>, Larry__Weiss wrote: >The CDC-6600's console displays were not terminals of that sort. >They were refreshed calligraphic CRT's. Very nice to have at that time. > >Does anyone have a picture of a CDC-6600 console with an image displayed? The Scope CLock (posted elsewhere) is close, but the characters are far too precise, and they're too big (biggest on console is 16 chars per line). While these are 170s, the concept is identical. The last four photos on this page are representative. Small (64/line) and medium (32/line) characters are displayed. http://berlin.ccc.de/~hans/vcfe2/index_4.htm -- Jeff Woolsey {woolsey,jlw}@{jlw,jxh}.com "I didn't get a 'Harrumph!' out of _that_ guy." -Gov Le Petomaine "Delete! Delete! OK!" -Dr. Bronner on disk space management "A toy robot!!!!" -unlucky Japanese scientist ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <593184F5F1D3C2F2.F6682B4E87BEFEBF.072C191C5CB2CFEF@lp.airnews.net> Distribution: inet Organization: JLW Consulting X-Newsreader: trn 4.0-test76 (Apr 2, 2001) From: woolsey@jlw.com (Jeff Woolsey) Originator: woolsey@jlw.com (Jeff Woolsey) X-Original-NNTP-Posting-Host: arglebargle.jlw.com Message-ID: <3d21f42c$1@news.meer.net> X-Original-Trace: 2 Jul 2002 11:42:52 -0700, arglebargle.jlw.com Lines: 15 Date: 2 Jul 2002 11:42:52 -0700 NNTP-Posting-Host: 209.157.152.10 X-Complaints-To: abuse@verio.net X-Trace: dfw-read.news.verio.net 1025635443 209.157.152.10 (Tue, 02 Jul 2002 18:44:03 GMT) NNTP-Posting-Date: Tue, 02 Jul 2002 18:44:03 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!iad-peer.news.verio.net!news.verio.net!dfw-read.news.verio.net.POSTED!news.meer.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111140 In article , Stephen O. Lidie wrote: >Larry__Weiss wrote: > >> Does anyone have a picture of a CDC-6600 console with an image displayed? > >I have a home movie of "pac" running. Not a 6600, maybe a 6400, but more >likely a single tube machine. GIF! GIF! er... MPEG! MPEG! -- Jeff Woolsey {woolsey,jlw}@{jlw,jxh}.com "I didn't get a 'Harrumph!' out of _that_ guy." -Gov Le Petomaine "Delete! Delete! OK!" -Dr. Bronner on disk space management "A toy robot!!!!" -unlucky Japanese scientist ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d215922$1@news.meer.net> Organization: JLW Consulting X-Newsreader: trn 4.0-test76 (Apr 2, 2001) From: woolsey@jlw.com (Jeff Woolsey) Originator: woolsey@jlw.com (Jeff Woolsey) X-Original-NNTP-Posting-Host: arglebargle.jlw.com Message-ID: <3d21f4bb$1@news.meer.net> X-Original-Trace: 2 Jul 2002 11:45:15 -0700, arglebargle.jlw.com Lines: 21 Date: 2 Jul 2002 11:45:15 -0700 NNTP-Posting-Host: 209.157.152.10 X-Complaints-To: abuse@verio.net X-Trace: dfw-read.news.verio.net 1025635585 209.157.152.10 (Tue, 02 Jul 2002 18:46:25 GMT) NNTP-Posting-Date: Tue, 02 Jul 2002 18:46:25 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.cwix.com!iad-peer.news.verio.net!news.verio.net!dfw-read.news.verio.net.POSTED!news.meer.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111144 In article , Larry__Weiss wrote: >Jeff Woolsey wrote: >> "Video" is a smidge inaccurate. It's a very glorified oscilloscope >> (Tek scopes of the day had similar calligraphic characters for >> readouts). >The video monitors on the CDC-6600 were very functional. I remember seeing >chessboards displayed (UT Austin had their 6600 console on public display >behind a glass wall in the computation center). I was trying to distinguish what the consoles displayed from video. Video is a moving raster-scan image. The console are definitiely not raster-scan. (These days, raster scan can emulate vectors pretty well; the MAME arcade-game emulator does it.) -- Jeff Woolsey {woolsey,jlw}@{jlw,jxh}.com "I didn't get a 'Harrumph!' out of _that_ guy." -Gov Le Petomaine "Delete! Delete! OK!" -Dr. Bronner on disk space management "A toy robot!!!!" -unlucky Japanese scientist ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3d21ee5c$1@news.ucsc.edu> Date: 2 Jul 2002 11:18:04 -0800 X-Trace: 2 Jul 2002 11:18:04 -0800, sundance.cse.ucsc.edu Lines: 18 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:111145 In article <3d1fcdce_2@news.iglou.com>, Douglas H. Quebbeman wrote: >Although the Wheelers refer to RAIN/RAIN4, LINPACK was also >used for relative speed comparisons, and has been ported to >modern platforms and other languages. You should be able to >find any number of different versions all over the net... Yeah, but Jack didn't write the LINPACK package until the late 1970s. And the benchmark wasn't until the early 80s. For the comparison to be meaningful the 6600 had to be taken in 1964. I think LLNL had decommissioned its last 6600 before 1978 when Jack published his ACM SIGNUM Paper although I had run on a 6400 in 1977 so 6600s were clearly still running in the late 1970s. Be warned with Lynn and the other IBM folk, (I like Lynn), but they have a tendency to quote 32-bit benchmarks against 60 and 64-bit CDC and Cray benchmarks, and this is why Bailey's first rule in benchmark is the way it is. ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fcdce_2@news.iglou.com> <3D1FD2A7.6690F451@attbi.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3d21ef6e$1@news.ucsc.edu> Date: 2 Jul 2002 11:22:38 -0800 X-Trace: 2 Jul 2002 11:22:38 -0800, sundance.cse.ucsc.edu Lines: 13 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-hog.berkeley.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:111146 In article <3D1FD2A7.6690F451@attbi.com>, Joe Yuska wrote: >This raises another question in my mind. >In the late sixties and early seventies one set of working codes for >reactor design were called the PDQ-(version-number) series. >These were used by several of the (then) AEC labs as well as companies like > >2. Have any of these dusty decks survived? Not answering specifically for reactor codes, I can assure you that AEC-era lab codes exist and are running on ASCI machines, just no longer in Fortran or CVC (to my surprise some one actually sat down and rewrote them in C). I'll know a bit more next week when I go over to LLNL. ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d200c9e.299409694@sd.znet.com> <3d21f188$1@news.ucsc.edu> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 15:25:39 -0400 Lines: 19 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d21fe34$1_1@news.iglou.com> X-Trace: news.iglou.com 1025637940 204.250.0.238 (2 Jul 2002 15:25:40 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111154 "Eugene Miya" wrote in message = news:3d21f188$1@news.ucsc.edu... > >A table of the achieved megaflop rates for various CDC machines > >with different release levels of OS and Fortran compiler may be found > >at > > > >http://pages.sbcglobal.net/couperusj/Megaflops.html >=20 > What was a Cyber 2000? > Unfortunately, the FTN compiler got too old for its time. Jitze can correct me if I'm wrong, but I thought the 2000 was the final design iteration of the 6000 Series architecture. Not many built, I'm guessing... Jitze? -dq ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 15:30:40 -0400 Lines: 30 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d21ff60$1_1@news.iglou.com> X-Trace: news.iglou.com 1025638240 204.250.0.238 (2 Jul 2002 15:30:40 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111153 "Eugene Miya" wrote in message = news:3d21ee5c$1@news.ucsc.edu... > In article <3d1fcdce_2@news.iglou.com>, > Douglas H. Quebbeman wrote: > >Although the Wheelers refer to RAIN/RAIN4, LINPACK was also > >used for relative speed comparisons, and has been ported to > >modern platforms and other languages. You should be able to > >find any number of different versions all over the net... >=20 > Yeah, but Jack didn't write the LINPACK package until the late 1970s. > And the benchmark wasn't until the early 80s. For the comparison to = be > meaningful the 6600 had to be taken in 1964. One interesting historical point... The "6000 Introduction and Peripheral Processor Training Manual" has a little propaganda at the beginning... says that the AEC had been needing a machine to crunch those "glowing" numbers, and attempts to do so kept falling short. All of those designs had been based on the use of the germanium transistor, and it had been pushed to the limits of its switching speed. CDC (read: Cray?) throught an innovative design might get around that limitation, namely, the emergent parallelism found in the use of the PPs for I/O and the manifold functional units... Then, just in time, came the silicon transistor, which the manual describes as "frosting on thge cake". -dq ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d200c9e.299409694@sd.znet.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3d21f188$1@news.ucsc.edu> Date: 2 Jul 2002 11:31:36 -0800 X-Trace: 2 Jul 2002 11:31:36 -0800, sundance.cse.ucsc.edu Lines: 31 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:111137 In article <3d200c9e.299409694@sd.znet.com>, Jitze Couperus wrote: >A more interesting measurement than MIPS for these machines was >megaflops - given their scientifc orientation. And that was generaly >measured by a standard benchmark that was run at every major >realease of OS or the Fortran compiler. I believe the benchmark >program used was LINPAK but I can't be sure. Gawd, we have to live with the MFLOPS albatross to this day. It's a terrible hold over from the era when floating point operations took significantly longer than a a clock cycle. Not even Knuth's empirical study of Fortran programs could show managers the importance of other aspects of system balance like memory bandwidth. And it's worse as people moved with 2-D codes to 3-D codes and higher Ds. I appreciate the machine, but the political measurement of MFLOPS is possibly one of the most harmful in computing done in the 60s. >A table of the achieved megaflop rates for various CDC machines >with different release levels of OS and Fortran compiler may be found >at > >http://pages.sbcglobal.net/couperusj/Megaflops.html What was a Cyber 2000? Unfortunately, the FTN compiler got too old for its time. Maybe we can dredge Peglar or McCalpin into this. ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <3D209F50.A6CAA51E@attbi.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 15:37:04 -0400 Lines: 27 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d220101_2@news.iglou.com> X-Trace: news.iglou.com 1025638657 204.250.0.238 (2 Jul 2002 15:37:37 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111147 "Tim McCaffrey" wrote in message = news:afso22>=20 > I would be surprised if there was a 10 PPU add on option, at least for = the=20 > 6500. Michigan State would have probably bought it, if available. = When the=20 > Cyber 170/750 came in we could only use 10 of the 20 PP's, because our = > customized OS needed to be updated. When the additional 10 PPs were = usable,=20 > it made a significant difference. One reason may have been because = the OS=20 > dedicated (I think) 4 PPs to certain tasks (monitor, Argus (terminal = support),=20 > Console, and something else). You *would* post this on a day I'm not carrying the 6000 Series=20 Hardware refman with me, so I can't quote the option number, but yeah, it was a "standard option". If you were running the latest OS from CDC, they had the mods to make it work. If not, you had to find a site who was running something close to the same PSR level as you were, and get the code from them. We got ours from a Navy site, the name of which escapes me... -dq ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3d21f46b$1@news.ucsc.edu> Date: 2 Jul 2002 11:43:55 -0800 X-Trace: 2 Jul 2002 11:43:55 -0800, sundance.cse.ucsc.edu Lines: 48 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:111139 >> Despite their shortcomings, there were some nice things about the CDC 6600 >> series machines, and their operating systems. Compared to the dominant >> mainframe of the day -- IBM 360s, 370s -- they had wonderful operating >> systems. I think this depended on the OS and the management of the site. It was still a batch versus interactive world back then. I'd prefer an IBM SPF site to batch oriented CDC shop, just as I would have likely prefered a Livermore CDC machine over an IBM batch shop. In article , Kent Olsen wrote: >They were so much easier to use than VM, MVS, CICS, etc... Depending >on the application, the 6000s outran the IBMs, sometimes the IBMs >outran the 6000s. They were definitely the two biggest horses in the >race. Those are most 370 terms. You have to go back to the MVT era on 360s. >> To copy a file, you used the copy command. One-liner. To make a file bigger, >> you wrote to it. And so on. None of this IEBGENER/IEBCOPY foolishness. >> Files were just a series of bytes. (OK, 6-bit bytes, and there were those EOR >> marks, and oddball line terminators, but still...) > >Even the early DOS operating systems understood that to COPY a file >meant simply that the contents of one file should be duplicated under >a different name. I wonder why it took IBM so long to figure this >out? :) IBM had a DOS OS on 360s. Disk was expensive back then. In IBM circles, there was great pride in knowing exactly how many bytes fit on a 2314 (or 2311, etc.) or 3330 disk track. PC DOS and CDC systems had internal fragementation and even you could get better machine I/O efficiency reading each track or cylinder. It was the same logic, of that era, which made efficient use of punch cards like reading every single Hollerith field. Glad that era is over. Thank God I don't remember how many bytes that was, but I might have it in a note book of that era at home. ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> <1b4rfim29e.fsf@cs.nmsu.edu> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 15:53:16 -0400 Lines: 15 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d2204ae_3@news.iglou.com> X-Trace: news.iglou.com 1025639598 204.250.0.238 (2 Jul 2002 15:53:18 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111150 "Joe Pfeiffer" wrote in message = news:1b4rfim29e.fsf@cs.nmsu.edu... > woolsey@jlw.com (Jeff Woolsey) writes: > >=20 > > "Video" is a smidge inaccurate. It's a very glorified oscilloscope > > (Tek scopes of the day had similar calligraphic characters for > > readouts). >=20 > Didn't PDP-1 do that? Yes, it did. Does. -dq ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 02 Jul 2002 15:01:06 -0500 Organization: Airnews.net! at Internet America Lines: 28 Message-ID: <97D08AF47210D8DD.9418905C83458C49.64B2F12515C8EE5D@lp.airnews.net> X-Orig-Message-ID: <3D220682.2A32C6F8@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d215922$1@news.meer.net> <3d21f4bb$1@news.meer.net> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library1-aux.airnews.net NNTP-Posting-Time: Tue Jul 2 15:06:16 2002 NNTP-Posting-Host: !aFq81k-X6E2G`],?N_% (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!newsfeed.cwix.com!howland.erols.net!news.airnews.net!cabal10.airnews.net!cabal2.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111152 Jeff Woolsey wrote: > In article , > Larry__Weiss wrote: > >Jeff Woolsey wrote: > >> "Video" is a smidge inaccurate. It's a very glorified oscilloscope > >> (Tek scopes of the day had similar calligraphic characters for > >> readouts). > > >The video monitors on the CDC-6600 were very functional. I remember seeing > >chessboards displayed (UT Austin had their 6600 console on public display > >behind a glass wall in the computation center). > > I was trying to distinguish what the consoles displayed from video. Video > is a moving raster-scan image. The console are definitiely not raster-scan. > (These days, raster scan can emulate vectors pretty well; the MAME arcade-game > emulator does it.) > Sure! They certainly were not raster-scan based. The price of memory had to come down many orders-of-magnitude before that technology was feasible. And I suppose you are correct that a person not aquainted with this older gear would take the wrong meaning from the simple word "video". That's why I'm eager to see images of anyone's photographs (or movies!) of that 'scope in action. - LarryW ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 56 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:K4gboeJiHJzm0gxXToBMmsHCvfQ= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Jul 2002 20:03:21 GMT NNTP-Posting-Host: 209.244.79.78 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025640201 209.244.79.78 (Tue, 02 Jul 2002 13:03:21 PDT) NNTP-Posting-Date: Tue, 02 Jul 2002 13:03:21 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed-east.nntpserver.com!nntpserver.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111164 eugene@cse.ucsc.edu (Eugene Miya) writes: > Be warned with Lynn and the other IBM folk, (I like Lynn), but they have a > tendency to quote 32-bit benchmarks against 60 and 64-bit CDC and Cray > benchmarks, and this is why Bailey's first rule in benchmark is the way it is. I don't have proof anymore but i believe the RAIN4 numbers were for 32 bit and the "RAIN" numbers wree for "double precision" 64bit. "fast double precision" was introduced for 168-3 (not on initial 168-1 machines) ... and so the 9.1 secs should be for RAIN ... as was the 6.77 secos for the 91. the interesting numbers are the 3031 and 158 numbers. The processor engine in the 3031 and 158 were the same; however in the case of the 158 .... there were "integrated channels" ... aka there was two sets of microcode running on the same 158 processor engine .... the microcode that implemented the 370 instruction set ... and the microcode that implemented the I/O support ... and the processor engine basically had to time-slice between the two sets of microcode. For the 3031, there were two "158" processor engines ... one processor engine dedicated to the 370 function (i.e. the 3031) and a second "158" processor engine (i.e. the "channel director") that implemented all the I/O function outboard. The dates for some of the machines (note 4341 and 3031 were about the same time): CDC 6600 63-08 64-09 LARGE SCIENTIFIC PROCESSOR IBM S/360-67 65-08 66-06 10 MOD 65+DAT; 1ST IBM VIRTUAL MEMORY IBM S/360-91 66-01 67-11 22 VERY LARGE CPU; PIPELINED AMH=AMDAHL 70-10 AMDAHL CORP. STARTS BUSINESS IBM S/370 ARCH. 70-06 71-02 08 EXTENDED (REL. MINOR) VERSION OF S/360 IBM S/370-145 70-09 71-08 11 MEDIUM S/370 - BIPOLAR MEMORY - VS READY IBM S/370-195 71-07 73-05 22 V. LARGE S/370 VERS. OF 360-195, FEW SOLD Intel, Hoff 71 Invention of microprocessor Intel DRAM 73 4Kbit DRAM Chip IBM 168-3 75-03 76-06 15 IMPROVED MOD 168 IBM 3031 77-10 78-03 05 LARGE S/370+EF INSTRUCTIONS and to repeat the numbers for rain/rain4: 158 3031 4341 Rain 45.64 secs 37.03 secs 36.21 secs Rain4 43.90 secs 36.61 secs 36.13 secs also times approx; 145 168-3 91 145 secs. 9.1 secs 6.77 secs rain/rain4 was from Lawrence Radiation lab ... and ran on cdc6600 in 35.77 secs. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: 2 Jul 2002 20:41:56 GMT Organization: The National Capital FreeNet Lines: 17 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 1025642516 16047 134.117.136.30 (2 Jul 2002 20:41:56 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 2 Jul 2002 20:41:56 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!netnews.com!xfer02.netnews.com!iad-peer.news.verio.net!news.verio.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:111168 Anne & Lynn Wheeler (lynn@garlic.com) writes: ... > > the interesting numbers are the 3031 and 158 numbers. The processor > engine in the 3031 and 158 were the same; however in the case of the > 158 .... there were "integrated channels" ... aka there was two sets > of microcode running on the same 158 processor engine .... the > microcode that implemented the 370 instruction set ... and the > microcode that implemented the I/O support ... and the processor > engine basically had to time-slice between the two sets of microcode. ... Betcha the 158 was microprogrammed by the 360/50 crowd. ("Looks like a duck, walks like a duck...") So what was the story behind the 3033? A 168 on steroids? I recall a service bureau using one of those as a primary workhorse. ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 18 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:mTTPlZNEuSK/T+//LBbUpVZgW+U= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Jul 2002 20:51:29 GMT NNTP-Posting-Host: 209.244.79.78 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025643089 209.244.79.78 (Tue, 02 Jul 2002 13:51:29 PDT) NNTP-Posting-Date: Tue, 02 Jul 2002 13:51:29 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!proxad.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111163 ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: > Betcha the 158 was microprogrammed by the 360/50 crowd. > ("Looks like a duck, walks like a duck...") > > So what was the story behind the 3033? A 168 on steroids? > I recall a service bureau using one of those as a primary workhorse. the story is that a 3031 was 158 with channel director and 3032 was 168 with channel director. The 3033 started out to be the 168 wiring diagram mapped to chips that were about 20 percent faster than 168 chips ... and with about ten times the circuits/chip. The initial remapping resulted in the 3033 being about 20 percent than 168/303s. Before any machines shipped, there was then an effort to redo the logic and utilize higher onchip density that got the 3033 up to about 50 percent faster than 168/3032. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: greg@lewis.ucs.indiana.edu (Gregory Travis) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 21:04:05 +0000 (UTC) Organization: A poorly-installed InterNetNews site Lines: 44 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fcdce_2@news.iglou.com> NNTP-Posting-Host: lewis.ucs.indiana.edu X-Trace: wilson.uits.indiana.edu 1025643852 12361 156.56.103.3 (2 Jul 2002 21:04:12 GMT) X-Complaints-To: news-admin@indiana.edu NNTP-Posting-Date: Tue, 2 Jul 2002 21:04:12 +0000 (UTC) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!vixen.cso.uiuc.edu!news.indiana.edu!lewis.ucs.indiana.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111169 In article , Anne & Lynn Wheeler wrote: >"Douglas H. Quebbeman" writes: >> Although the Wheelers refer to RAIN/RAIN4, LINPACK was also >> used for relative speed comparisons, and has been ported to >> modern platforms and other languages. You should be able to >> find any number of different versions all over the net... > >linpack numbers (including 6600): >http://ap01.physik.uni-greifswald.de/~ftp/bench/linpack.html > >bunch of stuff from above > >Computer N=100(Mflops) >------------------------------------ --- ------------- >CDC CYBER 205 (4-pipe) 17 >CDC CYBER 205 (2-pipe) 17 >CDC CYBER 990E 12 >Cray-1S (12.5 ns, 1983 run) 12 >IBM 4381 90E 1.2 >IBM 4381-13 1.2 >CDC CYBER 180-840 .99 >DEC VAX 8600 .48 >CDC 6600 .48 >CDC 6600 .36 I had seen the .36 number for the 6600 but never the .48 one. I'm not sure I really believe either. Years ago I got the chance to run some old FORTRAN code on a VAX 8600 -- code that was brought over from the 6600. The 8600 ran it about half as fast as the 6600 had. Also, IIRC, the 7600 was roughly 5x the 6600 in floating point performance and the Cray-1 was about 2x a 7600. Working backwards from the Cray-1S numbers, the 6600 should have come in around 1.2Mflops. Are we seeing a difrerence in compiler technology? The early 6600 FORTRAN compilers (RUN) were pretty poor while the last ones (FTN4, etc.) kicked ass. greg ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> <3d21ff60$1_1@news.iglou.com> <3d220b30$1@news.ucsc.edu> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 17:14:07 -0400 Lines: 21 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d22179f$1_1@news.iglou.com> X-Trace: news.iglou.com 1025644447 204.250.0.238 (2 Jul 2002 17:14:07 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-xfer2.newshosting.com!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111151 "Eugene Miya" wrote in message = news:3d220b30$1@news.ucsc.edu... > >Then, just in time, came the silicon transistor, which the manual > >describes as "frosting on the cake". >=20 > Cute: > The golden days of the discrete component era (I can say that because = I > am more in software). >=20 > Parallelism is on average at best an O(n) solution to performance = problems. >=20 > There are interesting technologies on the horizon. Hope some pan out. Quantum computers? I think the same powers that keep down fusion research that doesn't involve tokamaks and laser confinement systems might quash that, too. -dq ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> <3d21ff60$1_1@news.iglou.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3d220b30$1@news.ucsc.edu> Date: 2 Jul 2002 13:21:04 -0800 X-Trace: 2 Jul 2002 13:21:04 -0800, sundance.cse.ucsc.edu Lines: 32 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:111178 In article <3d21ff60$1_1@news.iglou.com>, Douglas H. Quebbeman wrote: >One interesting historical point... The "6000 Introduction and >Peripheral Processor Training Manual" has a little propaganda >at the beginning... says that the AEC had been needing a machine >to crunch those "glowing" numbers, and attempts to do so kept >falling short. The DOE, the DTRA, and various other three letter agencies (lurking in the group) STILL need, and their problems still fall short. Something about computational complexity, even in the face of Moore's law which sounds so impressive to people. 8^) >All of those designs had been based on the use >of the germanium transistor, and it had been pushed to the limits >of its switching speed. CDC (read: Cray?) throught an innovative >design might get around that limitation, namely, the emergent >parallelism found in the use of the PPs for I/O and the manifold >functional units... > >Then, just in time, came the silicon transistor, which the manual >describes as "frosting on the cake". Cute: The golden days of the discrete component era (I can say that because I am more in software). Parallelism is on average at best an O(n) solution to performance problems. There are interesting technologies on the horizon. Hope some pan out. ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3d220d4c$1@news.ucsc.edu> Date: 2 Jul 2002 13:30:04 -0800 X-Trace: 2 Jul 2002 13:30:04 -0800, sundance.cse.ucsc.edu Lines: 35 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:111138 In article , Anne & Lynn Wheeler wrote: >eugene@cse.ucsc.edu (Eugene Miya) writes: >> Be warned with Lynn and the other IBM folk, (I like Lynn), but they have a >> tendency to quote 32-bit benchmarks against 60 and 64-bit CDC and Cray >> benchmarks, and this is why Bailey's first rule in benchmark is the way it is. > >I don't have proof anymore but i believe the RAIN4 numbers were for 32 >bit and the "RAIN" numbers wree for "double precision" 64bit. 4 sounds like a good guess from that era. Gee, I tend to think of Double Precision as meaning 128 bits...... ;^) But then I am seeing this in the cdc group, noticing the cross post. >"fast double precision" was introduced for 168-3 (not on initial 168-1 >machines) ... and so the 9.1 secs should be for RAIN ... as was the >6.77 secos for the 91. It's unfortunate that if if any 370/168s are running in open area. Just as Kahan researched interesting floating point behavior, I'd be curious about certain speed implications. I talk on odd days with Alan Karp about this. Too much marketeering. It's unfortunate. It would be fun to try to deduce these behaviors, but I've learned all too well that even the guys who built these machines were as certain. >rain/rain4 was from Lawrence Radiation lab ... and ran on cdc6600 in >35.77 secs. Probaby LLL, but LRL has gone bye bye by way of terminology. It's their 50th this year anyways. ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D209F50.A6CAA51E@attbi.com> From: robert@bonomi.invalid Sender: robert@bonomi.invalid Originator: robert@bonomi.invalid Organization: Not Much X-Newsreader: trn 4.0-test69 (20 September 1998) Originator: bonomi@c-ns. (Robert Bonomi) Lines: 29 Message-ID: Date: Tue, 02 Jul 2002 22:02:00 GMT NNTP-Posting-Host: 207.241.52.60 X-Complaints-To: abuse@verio.net X-Trace: dfw-read.news.verio.net 1025647320 207.241.52.60 (Tue, 02 Jul 2002 22:02:00 GMT) NNTP-Posting-Date: Tue, 02 Jul 2002 22:02:00 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!HSNX.atgi.net!sjc-peer.news.verio.net!news.verio.net!dfw-read.news.verio.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111172 In article , Tim McCaffrey wrote: >In article <3D209F50.A6CAA51E@attbi.com>, jyuska@attbi.com says... >> >> >> > >>Perhaps in some later Cybers. Idon't recall. But for 6000 class machines >the memory >>modules ran at 1 mike, with a ten-way interleave, giving a max rate of 100 >ns. >> >The CDC 6000 series used 32 way interleave, 1 microsecond major cycle (memory >cycle time) and 100ns minor cycle (instruction execution rate). The 1 microsecond was a 'nominal' timing spec. on the 6600, 'read' cycle was circa 800ns, 'write' was more like 1.7 microsec. The 'stunt box' coordinated buffering/sequencing/serialization of actual memory access, and CPU references. It was capable of starting to process a reference every 100ns, thus matching 'scoreboard' instruction dispatch rate. Memory was _5-way_ ported. Two CPU paths (one for each fetch/store functional unit path), and, as I recall, two PP paths (one for each barrel), and one for ECS moves. I believe the 32-way interleave was used only in a 'max' configuration. That smaller configs used 16-way. ###### From: "Charlie Gibbs" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: 02 Jul 02 19:57:10 -0800 Organization: http://extra.newsguy.com Lines: 33 Message-ID: <1352.948T514T11973615@sky.bus.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <3d21f46b$1@news.ucsc.edu> NNTP-Posting-Host: p-384.newsdawg.com X-Newsreader: THOR 2.5a (Amiga;TCP/IP) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews4 Xref: chonsp.franklin.ch alt.folklore.computers:111189 In article <3d21f46b$1@news.ucsc.edu> eugene@cse.ucsc.edu (Eugene Miya) writes: >Disk was expensive back then. In IBM circles, there was great pride >in knowing exactly how many bytes fit on a 2314 (or 2311, etc.) or 3330 >disk track. PC DOS and CDC systems had internal fragementation and >even you could get better machine I/O efficiency reading each track >or cylinder. > >It was the same logic, of that era, which made efficient use of punch >cards like reading every single Hollerith field. > >Glad that era is over. > >Thank God I don't remember how many bytes that was, but I might have it >in a note book of that era at home. 3625 for the 2311, 7294 for the 2314, and (IIRC) 13,030 for the 3330. What's really scary is that I still remember it in hex for the 2311 and 2314: 0e29 and 1c7e respectively. I once wrote a little program to help calculate optimum block sizes for a given record size. (Let's see, gaps were 101 bytes if no key, 145 if there was a key, plus some forgotten percentage of the sum of record and key sizes... well, I suppose there are worse things that I could lie awake thinking about...) -- cgibbs@sky.bus.com (Charlie Gibbs) Remove the first period after the "at" sign to reply. I don't read top-posted messages. If you want me to see your reply, appropriately trim the quoted text and put your reply below it. ###### From: "Ken Hunter" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 17:20:36 -0600 Organization: CompuServe Interactive Services Lines: 17 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> NNTP-Posting-Host: 128.198.11.183 X-Trace: nntp-m01.news.aol.com 1025652039 13655 128.198.11.183 (2 Jul 2002 23:20:39 GMT) X-Complaints-To: newsmaster@compuserve.com NNTP-Posting-Date: Tue, 2 Jul 2002 23:20:39 +0000 (UTC) X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!howland.erols.net!ngpeer.news.aol.com!news.compuserve.com!news-master.compuserve.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111221 "Bruce Hoult" wrote in message news:bruce-4FEEB2.19465502072002@copper.ipg.tsnz.net... Yeah, I remember encountering that same thing on a Tek terminal hooked to the CYBER 170/750 at Michigan State. Also in the early '80s. Ken Hunter ###### From: "Ken Hunter" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 17:34:23 -0600 Organization: CompuServe Interactive Services Lines: 31 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> NNTP-Posting-Host: 128.198.11.183 X-Trace: nntp-m01.news.aol.com 1025652866 15359 128.198.11.183 (2 Jul 2002 23:34:26 GMT) X-Complaints-To: newsmaster@compuserve.com NNTP-Posting-Date: Tue, 2 Jul 2002 23:34:26 +0000 (UTC) X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.stueberl.de!cox.net!cyclone1.gnilink.net!ngpeer.news.aol.com!news.compuserve.com!news-master.compuserve.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111224 "Eric S. Harris" wrote in message news:3D1F56EE.DAA8337E@mindspring.com... That 7.5 characters/word would have been when using actual 8 bit ASCII characters, packed as tightly into 60-bit words as possible. Probably not something which was frequently done. "Eric S. Harris" wrote in message news:3D1F56EE.DAA8337E@mindspring.com... Actually, there was an add on product for NOS which raised the user limit up to 18-bits as well called Memory Usage Enhancement. Still available, actually, though it's owned by General Dynamics rather than Syntegra. Still used by certain DoD sites. Ken Hunter ###### From: greg@lewis.ucs.indiana.edu (Gregory Travis) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 2 Jul 2002 23:40:32 +0000 (UTC) Organization: A poorly-installed InterNetNews site Lines: 23 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D209F50.A6CAA51E@attbi.com> NNTP-Posting-Host: lewis.ucs.indiana.edu X-Trace: wilson.uits.indiana.edu 1025653239 13232 156.56.103.3 (2 Jul 2002 23:40:39 GMT) X-Complaints-To: news-admin@indiana.edu NNTP-Posting-Date: Tue, 2 Jul 2002 23:40:39 +0000 (UTC) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!vixen.cso.uiuc.edu!news.indiana.edu!lewis.ucs.indiana.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111239 In article , wrote: >In article , >Tim McCaffrey wrote: >>In article <3D209F50.A6CAA51E@attbi.com>, jyuska@attbi.com says... >>> >>> >>> >> >>>Perhaps in some later Cybers. Idon't recall. But for 6000 class machines >>the memory >>>modules ran at 1 mike, with a ten-way interleave, giving a max rate of 100 >>ns. >>> >>The CDC 6000 series used 32 way interleave, 1 microsecond major cycle (memory >>cycle time) and 100ns minor cycle (instruction execution rate). > >The 1 microsecond was a 'nominal' timing spec. on the 6600, 'read' cycle was >circa 800ns, 'write' was more like 1.7 microsec. Read was destructive (required a write-back) and thus slower than write. greg ###### Message-ID: <3D226048.9012D949@earthlink.net> From: Walter Spector Organization: Not much. X-Mailer: Mozilla 4.79 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 43 Date: Wed, 03 Jul 2002 02:24:20 GMT NNTP-Posting-Host: 158.252.219.147 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025663060 158.252.219.147 (Tue, 02 Jul 2002 19:24:20 PDT) NNTP-Posting-Date: Tue, 02 Jul 2002 19:24:20 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.stueberl.de!cox.net!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!7c693709!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111190 Ken Hunter wrote: > > "Eric S. Harris" wrote in message > news:3D1F56EE.DAA8337E@mindspring.com... That was what the hardware and the operating system supported. There were > also other schemes for representing more than 63 (or 64)characters: 6-12 > character sets, 8-in-12 and maybe others. (I seem to recall something about > 7.5 characters/word, but that may just be a horrible, horrible dream.)... > > That 7.5 characters/word would have been when using actual 8 bit ASCII > characters, packed as tightly into 60-bit words as possible. Probably not > something which was frequently done. The 7.5 chars/word mode was quite frequently used when dealing with 9-track 'stranger' tapes. CDC definately believed in diversity. When using ASCII with NOS, you had your choice of: 1.) 8-in-12 - For printers and plotters 2.) Packed 7.5 chars to a word - For 9-track 'stranger' tapes 3.) 6/12 display code - For tty I/O 4.) 6-bit display code - translated to ASCII in various hardware devices It was a mess. It really helps to consider the 6600 and followons as having 12-bit bytes. Then packing two 6-bit characters per byte. (Some older documentation even states this explicitly.) This due to the fact that PPs had 12-bit memory, the I/O channels were 12-bits wide, and even the central memory was composed of 12-bit memory modules 'glued' together. If some retrocomputing person ever bothers to write a C compiler for the CDC instruction set, a C 'char' would almost certainly be 12 bits. Walt -...- Walt Spector (w6ws at earthlink dot net) ###### Message-ID: <3D22699B.F741E111@attbi.com> From: Joe Yuska X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fcdce_2@news.iglou.com> <3D1FD2A7.6690F451@attbi.com> <3d21ef6e$1@news.ucsc.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 25 NNTP-Posting-Host: 24.91.67.33 X-Complaints-To: abuse@attbi.com X-Trace: rwcrnsc53 1025665443 24.91.67.33 (Wed, 03 Jul 2002 03:04:03 GMT) NNTP-Posting-Date: Wed, 03 Jul 2002 03:04:03 GMT Organization: AT&T Broadband Date: Wed, 03 Jul 2002 03:04:03 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!howland.erols.net!news-out.worldnet.att.net.MISMATCH!wn4feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!rwcrnsc53.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111266 Eugene Miya wrote: > > In article <3D1FD2A7.6690F451@attbi.com>, Joe Yuska wrote: > >This raises another question in my mind. > >In the late sixties and early seventies one set of working codes for > >reactor design were called the PDQ-(version-number) series. > >These were used by several of the (then) AEC labs as well as companies like > > > >2. Have any of these dusty decks survived? > > Not answering specifically for reactor codes, I can assure you that AEC-era > lab codes exist and are running on ASCI machines, just no longer in > Fortran or CVC (to my surprise some one actually sat down and rewrote > them in C). I'll know a bit more next week when I go over to LLNL. The main reason I was looking for the sources for the codes from the sixties and the seventies was that there were some neat adaptations to the 6600 architecture including hand coded algorithms that took advantage of the functional units. In addition there were specialized disk drivers for the 808/6638 disks that used multiple ppu's "ping-ponging" to avoid disk interleaving. I thought they would be cool to examine. Joe Yuska ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Followup-To: alt.folklore.computers Date: Wed, 3 Jul 2002 01:46:32 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 14 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!news.maxwell.syr.edu!newsfeed.stanford.edu!sn-xit-05!sn-xit-02!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111267 In article , Anne & Lynn Wheeler wrote: > original long ago ... reposting from > http://www.garlic.com/~lynn/2000d.html#0 Is a VAX a mainframe: > > rain/rain4 > 158 3031 4341 Why do I see so many notes about the 4341, but not the 4381. I programmed a 4381 in 1986-1989 in college. Was the '81 much faster than the '41? I ask because the '81 I used felt a lot faster than what I've been told about the CDC 6600, but then again, maybe it was just I/O speed. ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Followup-To: alt.folklore.computers Date: Wed, 3 Jul 2002 01:47:39 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 16 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!isdnet!sn-xit-02!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111270 In article , J Clarke wrote: > You might find it interesting to check out >. Hercules is a nearly full > System/390 emulator that runs on a PC. It's reputed to run 360 code > about ten times faster than a 360 (which specifict model I forget). > has a link to a machine running > Hercules that is accessible online. Does anyone know if it is possible to get MVS/XA to run on an emulator, and to get a copy of that OS for "personal education"? I used to use MVS/XA on a 4381, and I'd like to see it again, just for old time sake if nothing else. ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Followup-To: alt.folklore.computers Date: Wed, 3 Jul 2002 02:03:18 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!isdnet!sn-xit-02!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111263 In article , Anne & Lynn Wheeler wrote: > For the 3031, there were two "158" processor engines ... one processor > engine dedicated to the 370 function (i.e. the 3031) and a second > "158" processor engine (i.e. the "channel director") that implemented > all the I/O function outboard. How did the two stay in sync? I mean, I have vague and fuzzy recollections of doing assembler on a 4381 and since the programs are a mix of I/O and calculation, I assume this is very important. ###### From: couperusNOSPAM@znet.com (Jitze Couperus) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 03 Jul 2002 08:34:12 GMT Organization: Posted via Supernews, http://www.supernews.com Message-ID: <3d22ae06.92058870@sd.znet.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d200c9e.299409694@sd.znet.com> <3d21f188$1@news.ucsc.edu> X-Newsreader: Forte Free Agent 1.21/32.243 X-Complaints-To: newsabuse@supernews.com Lines: 31 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.freenet.de!newsfeed.news2me.com!logbridge.uoregon.edu!tethys.csu.net!nntp!sn-xit-05!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111185 On 2 Jul 2002 11:31:36 -0800, eugene@cse.ucsc.edu (Eugene Miya) wrote: > >What was a Cyber 2000? >Unfortunately, the FTN compiler got too old for its time. > The Cyber 2000 was indeed the last, biggest, and baddest of the Mohicans. The top end of the dual-state machines had been the 990 (also known as Theta during its development) but this was still a water-cooled machine. Then in the final spasm of CDC as a manufacturer of knuckle-dragging machines, they came out with the 2000 which was air- cooled and had an optional vector unit - but only ran in 180 (NOS/VE) state. The irony of this machine is that the last few sold were not used for Fortran number crunching at all, but filled specialised heavy-duty computing niches in other domains completely. One was used to maintain the mother of all databases (both very large for its day and involving lots of very complex logic) and another site used theirs as the core engine for the mother of all financial transaction systems. The overall muscle that these machines could bring to bear on these problems (not just CPU power, but e.g. bandwidth and address space) made both of them extrememly difficult to replace at their sites, one of which I understand is still trucking to this day. Jitze ###### From: Bruce Hoult Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <3d21f46b$1@news.ucsc.edu> <1352.948T514T11973615@sky.bus.com> User-Agent: MT-NewsWatcher/3.2 (PPC Mac OS X) Message-ID: Lines: 22 Date: Thu, 04 Jul 2002 01:25:10 +1200 NNTP-Posting-Host: 203.79.123.131 X-Complaints-To: abuse@tsnz.net X-Trace: news02.tsnz.net 1025702710 203.79.123.131 (Thu, 04 Jul 2002 01:25:10 NZST) NNTP-Posting-Date: Thu, 04 Jul 2002 01:25:10 NZST Organization: TelstraClear Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.media.kyoto-u.ac.jp!newsfeed01.tsnz.net!news02.tsnz.net!bruce Xref: chonsp.franklin.ch alt.folklore.computers:111251 In article <1352.948T514T11973615@sky.bus.com>, "Charlie Gibbs" wrote: > 3625 for the 2311, 7294 for the 2314, and (IIRC) 13,030 for the 3330. > What's really scary is that I still remember it in hex for the 2311 > and 2314: 0e29 and 1c7e respectively. What *I* find scary is that I still remember $20, $60 and $4C (JSR/RTS/JMP), $A9 and $AD and $A5 (LDA with immediate, absolute or zero page), $18 and $38 (clear and set carry) and bunches of other opcodes. Also special addresses such as $3D0 (warm restart), $FDED (output character routine), $C000 and $C010 (keyboard data and strobe registers), $C051 (switch to text mode). The scary part is of course that I haven't touched an Apple ][ in nearly twenty years, since I switched to 68k based machines (4e71 ... arrrrghh). I'm a high level language programmer ... really I am. In fact I prefer Lisp-family languages. None of that assembly language stuff spoken *here*, by choice... -- Bruce ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 3 Jul 2002 09:54:17 -0400 Lines: 30 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d23020b_1@news.iglou.com> X-Trace: news.iglou.com 1025704459 204.250.0.238 (3 Jul 2002 09:54:19 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed1.cidera.com!Cidera!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111182 "Walter Spector" wrote in message=20 > > The 7.5 chars/word mode was quite frequently used when dealing with > 9-track 'stranger' tapes. >=20 > CDC definately believed in diversity. When using ASCII with NOS, you = had > your choice of: >=20 > 1.) 8-in-12 - For printers and plotters >=20 > 2.) Packed 7.5 chars to a word - For 9-track 'stranger' tapes >=20 > 3.) 6/12 display code - For tty I/O >=20 > 4.) 6-bit display code - translated to ASCII in various hardware = devices >=20 > It was a mess. Did NOS drop the really strange *other* 8-in-12 format that was for TELEX, that had a special code (4000B?) in the first word, then the 8-in-12 in the rest of the block? With Kronos, If you had *any* interruption in the stream to the terminal, you'd end up with gobbledegook because when it resumed, it didn't have the prefix code to tell the controller (6671?) what was coming... -dq ###### From: "George R. Gonzalez" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 3 Jul 2002 10:11:54 -0500 Organization: University of Minnesota, Twin Cities Campus Lines: 51 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: "George R. Gonzalez" X-Trace: laurel.tc.umn.edu 1025709440 4942 160.94.124.25 (3 Jul 2002 15:17:20 GMT) X-Complaints-To: usenet@laurel.tc.umn.edu X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!snoopy.risq.qc.ca!msc1.onvoy!onvoy.com!hardy.tc.umn.edu!laurel.tc.umn.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111204 Some random notes on the CDC consoles: It was a dual-scope display. The display tubes were rather primitive electrostatic deflection CRT's. They were very loooong, thereby necessitating that huge deep cabinet. The deflection plates were driven by some really hefty tubes! Your typical tv of that screen size would have a medium-power 6CU6, each CRT of the console had a pair of 40-watt blast-air cooled 3C29's, one of the hottest tubes of WWII. The repair guys were scheduled to replace these every 2000 hours or so, thereby providing a continual source of still usable transmitting tubes for ham radio guys. The console was an analog stroke-vector device basically, but it had some smarts so it could accept a stream of 6-bit codes and automatically draw them. The character vector data was stored in, IIRC a not easily changeable rope-memory ROM. Thereby the term DISPLAY CODE, and that's one of reasons the poor machines were stuck with that unusual 6-bit code scheme instead of ASCII. I don't think the early consoles had anyway of doing the 12 bit extended codes either. The keyboard was mighty cheap looking for a $million dollar computer. It had ugly gray styrene keys. It had to generate 6-bit display code of course, so no lower case. At first the console was a pretty good match, technically and economically for the CPU. Later on, when 6/12 and 8/12 codes became popular, and people might have wanted to use the console from a remote location, the console became a bit of an millstone around the neck. Finally by the time the integrated circuit 17x series came out, rumor had it they whole CPU cost less to build than the cost of the old console! That explains why there was a quick move to design a newer ,smaller, cheaper console. Only one CRT, split/screen display, much smaller and cheaper, altho I don't know if it kept the vector graphics. One fella in their research lab came up with "the six pack"-- a bizarre contraption: a gaggle of 10 PPU's, a disk, no CPU, from one to 6 consoles, all running a simple word-processing screen editor (probably a descendant of the "O26" console editor! IIRC the marketing guys could not figure out how to sell such a bizarre device. Regards, George ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 33 Date: Wed, 03 Jul 2002 15:30:32 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news2.calgary.shaw.ca 1025710232 24.71.223.147 (Wed, 03 Jul 2002 09:30:32 MDT) NNTP-Posting-Date: Wed, 03 Jul 2002 09:30:32 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-han1.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news2.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111252 On Wed, 3 Jul 2002 01:46:32 -0400, Charles Shannon Hendrix wrote: >In article , Anne & Lynn Wheeler wrote: > >> original long ago ... reposting from >> http://www.garlic.com/~lynn/2000d.html#0 Is a VAX a mainframe: >> >> rain/rain4 >> 158 3031 4341 > >Why do I see so many notes about the 4341, but not the 4381. I programmed >a 4381 in 1986-1989 in college. > >Was the '81 much faster than the '41? I ask because the '81 I used >felt a lot faster than what I've been told about the CDC 6600, but >then again, maybe it was just I/O speed. IIRC about 2.0-2.5MIPS instead of about 0.5-1.0MIPS depending on the models, and the 4381-3 was a dual (4.5MIPS), could have up to 64MB memory I think, and run VM/HPO, VM/XA, or MVS realistically. Can't remember if it could have a second set of six channels or not? -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 27 Date: Wed, 03 Jul 2002 15:33:06 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news2.calgary.shaw.ca 1025710386 24.71.223.147 (Wed, 03 Jul 2002 09:33:06 MDT) NNTP-Posting-Date: Wed, 03 Jul 2002 09:33:06 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!netnews.com!xfer02.netnews.com!newsfeed1.cidera.com!Cidera!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news2.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111246 On Wed, 3 Jul 2002 02:03:18 -0400, Charles Shannon Hendrix wrote: >In article , Anne & Lynn Wheeler wrote: > >> For the 3031, there were two "158" processor engines ... one processor >> engine dedicated to the 370 function (i.e. the 3031) and a second >> "158" processor engine (i.e. the "channel director") that implemented >> all the I/O function outboard. > >How did the two stay in sync? I mean, I have vague and fuzzy >recollections of doing assembler on a 4381 and since the programs are >a mix of I/O and calculation, I assume this is very important. SIO from CPU sent CCW to channel director, results came back; other systems used CPU to handle CCWs with different ucode. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: 3 Jul 2002 15:56:55 GMT Organization: The National Capital FreeNet Lines: 10 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <3d21f46b$1@news.ucsc.edu> <1352.948T514T11973615@sky.bus.com> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 1025711815 11234 134.117.136.30 (3 Jul 2002 15:56:55 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 3 Jul 2002 15:56:55 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!194.25.134.126.MISMATCH!newsfeed01.sul.t-online.de!t-online.de!feed.news.nacamar.de!newsfeed.icl.net!newsfeed.fjserv.net!newspeer.clara.net!news.clara.net!Quza.UK.peer!nntp.gblx.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:111220 "Charlie Gibbs" (cgibbs@sky.bus.com) writes: > ... > I once wrote a little program to help calculate optimum block sizes > for a given record size. ... (Pardon me if this is a repeat) The first new JCL feature in MVS that I met in years was BLKSIZE=0 and the bugger did the math! The last track size I memorised, or had tables to consult, was for a 3350; so this addition was "a good thing". ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 82 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:94idlqLrbIKoR6e1txdtc3brrp0= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Jul 2002 16:13:46 GMT NNTP-Posting-Host: 65.59.18.130 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025712826 65.59.18.130 (Wed, 03 Jul 2002 09:13:46 PDT) NNTP-Posting-Date: Wed, 03 Jul 2002 09:13:46 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.stueberl.de!cox.net!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111201 Charles Shannon Hendrix writes: > Why do I see so many notes about the 4341, but not the 4381. I programmed > a 4381 in 1986-1989 in college. > > Was the '81 much faster than the '41? I ask because the '81 I used > felt a lot faster than what I've been told about the CDC 6600, but > then again, maybe it was just I/O speed. these particular notes (that i happened to have laying around) came from some work that i was doing for some endicott 4341 performance engineers. they wanted a benchmark run between 3031 and 4341 (this is pre-4381 ... aka after the endicott engineers produced the 4341 ... they later went on to produce the 4381). The endicott performance engineers were having trouble getting machine time to do the benchmark (also at the time, rain/rain4 were one of the few widely run benchmarks). Basically the processor hardware engineers got the first machine built and the disk engineering/product-test labs got the second machine built, aka in addition to developing new disk drives they validated existing disks against new processors as they became available. The processors in the disk engineering lab had been running "stand-alone" applications (FRIEND, couple others) for the testing .... problem was that the testcell disks under development tended to sometimes deviate from normal operational characteristics (MTBF for a standard MVS when operating a single testcel was on the order of 15 minutes). As something of a hobby, i rewrote the I/O supervisor to make it absolute bullet proof, aka no kind of i/o glitches could make the system crash. As a result it was installed in all the "test" processors in the disk engineering and product test labs .... and they were able to do concurrent, simulataneous testing of 6-12 testcells (instead of scheduling stand-alone time for one testcell at a time) on each processor (as needed). I then got the responsibility of doing system support on all those machines and periodically would get blaimed when things didn't work correctly and so had to get involved in debugging their hardware (as part of prooving that the software wasn't at fault). One such situation was the weekend they replaced the 3830 control unit for a 16-drive string of 3350s (production timesharing) with a "new" 3880 control unit and performance went into the can on that machine. Fortunately this was six months before first customer ship of the 3880 controller so there were times to make some hardware adjustments (I make this joke at one point of working 1st shift at research, 2nd shift in the disk labs, and 3rd shift down at STL, and also couple times a month supporting the operating system for the HONE complex in palo alto). In any case, at that particular point that there were two 4341s in existance, one in edicott and one in san jose disk machines. Since I supported the operating system for san jose disk ... and since while the machines might be i/o intensive ... the workload rarely exceeded 5 percent cpu utilization. They had 145, 158, 3031, 3033, 4341, ect. machines that I could worry about and had some freedom in doing other types of things with. So i ran the rain/rain4 benchmarks for the endicott performance engineers and got 4341 times (aka they couldn't get time on the machine in endicott because it was booked solidly for other things), 3031 times, and 158 times. They previously had collected numbers for the 168-3 and 91 times for rain/rain4 ... and of course rain had been run on 6600 (numbers they sent to me along with the benchmarks to run). There may have been other benchmark runs made by other people ... but I didn't do the runs and didn't have the data sitting around conveniently. I posted some numbers that I had conveniently available. misc. disk engineer related posts: http://www.garlic.com/~lynn/subtopic.html#disk misc. hone related posts: http://www.garlic.com/~lynn/subtopic.html#hone random post about working 4-shift work weeks (24hrs, 7days): http://www.garlic.com/~lynn/2001h.html#29 checking some myths -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 51 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:Iwbl5lPivbq4lPpqkLp7mkr6YOo= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Jul 2002 16:27:15 GMT NNTP-Posting-Host: 65.59.18.130 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025713635 65.59.18.130 (Wed, 03 Jul 2002 09:27:15 PDT) NNTP-Posting-Date: Wed, 03 Jul 2002 09:27:15 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111200 Charles Shannon Hendrix writes: > How did the two stay in sync? I mean, I have vague and fuzzy > recollections of doing assembler on a 4381 and since the programs are > a mix of I/O and calculation, I assume this is very important. basically instruction processors and I/O processors were architected to be asyncronous .... effectively in much the same way that machines with multiple instruction processors (SMP) are typically architected so that the multiple instruction processors operate asyncronously. The instructions for th3 360/370 i/o processors were, in fact called "channel programs". You could write a channel program ... and signal one of the asycnronous "channel processors" to begin asyncronous execution of that channel program. The channel program could cause asyncronous interrupts back to the instruction processor signling various kinds of progress/events. On some of the machines, the i/o processors were real, honest to goodness independent asyncronous processors. On other machines, a common microcode engine was used to emulate both the instruction processor and multiple i/o (channel) processors. Machines where a common processor engine was used to emulate multiple processors (cpus, channels, etc) where typically described as having "integrated" channels. 158s, 135, 145, 148, 4341, etc ... were "integrated" channel machines (aka the native microcode engine had microcode for both emulating 370 processing and microcode for performing the channel process function and executing "channel programs"). 168 machines had outboard channels (independent hardware boxes that implement the processing of channel programs). Channels processors and instruction processors had common access to the same real storage (in much the same way that multiple instruction processors have common access to the same real storage). For the 303x line of machines .... they took a 158 integrated channel machine .... and eliminated the 370 instruction emulation microcode ... creating a dedicated channel program processing machine called a "channel director". The "channel director" was then a common component used for 3031, 3032, and 3033 machines ... aka they were all "outboard channel" machines (having dedicated hardware processing units for executing channel programs) ... as opposed to "integrated channel" machines. A 3031 was then a 158 with just the 370 instruction emulation microcode and reconfigured for "outboard channel" operation rather than "integrated channel" operation. A 3032 was then a 168 that was reconfigured to use "channel director" for outbarod channels (rather than the 168 outboard channel hardware boxes). -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 34 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:RWSrcKGS7kVo8lSUm2JhQgTMULM= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Jul 2002 16:56:15 GMT NNTP-Posting-Host: 65.59.18.130 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025715375 65.59.18.130 (Wed, 03 Jul 2002 09:56:15 PDT) NNTP-Posting-Date: Wed, 03 Jul 2002 09:56:15 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111214 Anne & Lynn Wheeler writes: > shift in the disk labs, and 3rd shift down at STL, and also couple > times a month supporting the operating system for the HONE complex in > palo alto). HONE was the system infrastructure that supported the people in the field, salesmen, marketing people, branch office people, etc. At one time the US HONE system in Palo Alto had grown into the largest single-system-image complex in the world. At one time, I knew it had something over 40,000 defined "userids". The US HONE system was also cloned for a number of country and/or regional centers around the world. Also, in the early '80s, the Palo Alto complex was extended with redundant centers in Dallas and Boulder for "disaster surviveability" (my wife and I later coined the terms disaster survivability and geographic survivability when we were doing HA/CMP) ... online workload was spread across the three datacenters, but if one failed the remaining two could pick up. Nearly all of the application delivery to branch & field people were written in APL ... running under CMS. One of the most important were the "configurator" applications. Starting with the 370/125 (& 115), it was no longer possible for a salesman to manual fill-out a mainframe machine order .... they all had to be done interacting with HONE configurator. random ha/cmp refs: http://www.garlic.com/~lynn/subtopic.html#hacmp -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 12 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:wl+MqYsD91whwQnPoyYp9YsuU/0= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Jul 2002 16:57:13 GMT NNTP-Posting-Host: 65.59.18.130 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025715433 65.59.18.130 (Wed, 03 Jul 2002 09:57:13 PDT) NNTP-Posting-Date: Wed, 03 Jul 2002 09:57:13 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!proxad.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111211 Anne & Lynn Wheeler writes: > basically instruction processors and I/O processors were architected > to be asyncronous .... effectively in much the same way that machines > with multiple instruction processors (SMP) are typically architected > so that the multiple instruction processors operate asyncronously. also some smp related postings: http://www.garlic.com/~lynn/subtopic.html#smp -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 28 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:faNP+7i3rM4bg/l88Ria1QGILf8= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Jul 2002 17:10:31 GMT NNTP-Posting-Host: 65.59.18.130 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025716231 65.59.18.130 (Wed, 03 Jul 2002 10:10:31 PDT) NNTP-Posting-Date: Wed, 03 Jul 2002 10:10:31 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111213 Charles Shannon Hendrix writes: > Why do I see so many notes about the 4341, but not the 4381. I programmed > a 4381 in 1986-1989 in college. > > Was the '81 much faster than the '41? I ask because the '81 I used > felt a lot faster than what I've been told about the CDC 6600, but > then again, maybe it was just I/O speed. oh yes, i did do a post referencing URL somebody's linpack table ... and extracted several of the entries for the posting (including a number of 4381 entries ... besides 6600). Not the original table at the referenced URL includes information about compiler and options used .... i scrubbed that info ... trying to reduce the size of the posting ... people wanting to see the full information should go to the reference URL the 4381 linpack entries from that posting IBM 4381 90E 1.2 IBM 4381-13 1.2 IBM 4381-22 .97 IBM 4381 MG2 .96 IBM 4381-21 .47 IBM 4381-11 .39 ref linpack posting http://www.garlic.com/~lynn/2002i.html#12 -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Followup-To: alt.folklore.computers Date: Wed, 3 Jul 2002 15:12:02 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 45 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-hog.berkeley.edu!ucberkeley!logbridge.uoregon.edu!tethys.csu.net!nntp!sn-xit-05!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111470 In article , Anne & Lynn Wheeler wrote: > Charles Shannon Hendrix writes: >> How did the two stay in sync? I mean, I have vague and fuzzy >> recollections of doing assembler on a 4381 and since the programs are >> a mix of I/O and calculation, I assume this is very important. > > basically instruction processors and I/O processors were architected > to be asyncronous .... effectively in much the same way that machines > with multiple instruction processors (SMP) are typically architected > so that the multiple instruction processors operate asyncronously. [snip] Ah, I remember that now. In some ways I regret changing track away from the IBM systems. It's just that the local colleges didn't really let you learn much about them. Everything we did on the mainframe was basically: * load up a big chunk of records into a block * process the block * repeat It always seemed to me that we should have started another I/O operation just before processing the current block, so at least in theory the next would be ready when you finished, or at least would be ready faster. Did IBM compilers like COBOL and FORTRAN do anything like this behind the scenes to increase througput? What was the proper way to interleave I/O and processing on the IBM, assuming your language would you, or you were using assembler? Most UNIX systems try to do this sort of thing automatically, although there are a lot of things you can do yourself to optimize I/O and processing to work together. The UNIX process model is good but I/O is almost totally divorced from a process. Throughput overall can be very good, but a single program with a huge I/O queue can hold up processes with small amounts of I/O for some time. That's one thing I really miss about systems like MVS/XA: the ability to control I/O at the process level when you need to. I think Solaris can do this on high-end systems now, but it's not available in all of the UNIX world yet, most especially not in the free systems. ###### From: "Eric S. Harris" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 03 Jul 2002 14:13:33 -0500 Organization: MindSpring Enterprises Lines: 86 Distribution: inet Message-ID: <3D234CDD.F33DA139@mindspring.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> Reply-To: eric_harris_76@yahoo.com NNTP-Posting-Host: 3f.d0.3f.a5 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Server-Date: 3 Jul 2002 19:16:08 GMT X-Mailer: Mozilla 4.79 [en]C-CCK-MCD (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.stueberl.de!cox.net!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!stamper.news.atl.earthlink.net!newsfeed0.news.atl.earthlink.net!news.atl.earthlink.net!news.mindspring.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111312 Walter Spector wrote: > > Ken Hunter wrote: > > > > "Eric S. Harris" wrote in message > > news:3D1F56EE.DAA8337E@mindspring.com... > That was what the hardware and the operating system supported. There were > > also other schemes for representing more than 63 (or 64)characters: 6-12 > > character sets, 8-in-12 and maybe others. (I seem to recall something about > > 7.5 characters/word, but that may just be a horrible, horrible dream.)... > > > > That 7.5 characters/word would have been when using actual 8 bit ASCII > > characters, packed as tightly into 60-bit words as possible. Probably not > > something which was frequently done. > > The 7.5 chars/word mode was quite frequently used when dealing with > 9-track 'stranger' tapes. Oh, yeah. Damn. Nightmares again. (Thank you so much ;-) > CDC definately believed in diversity. When using ASCII with NOS, you had > your choice of: > > 1.) 8-in-12 - For printers and plotters > > 2.) Packed 7.5 chars to a word - For 9-track 'stranger' tapes > > 3.) 6/12 display code - For tty I/O > > 4.) 6-bit display code - translated to ASCII in various hardware devices > > It was a mess. > > It really helps to consider the 6600 and followons as having 12-bit > bytes. Then packing two 6-bit characters per byte. (Some older > documentation even states this explicitly.) This due to the > fact that PPs had 12-bit memory, the I/O channels were 12-bits wide, > and even the central memory was composed of 12-bit memory modules > 'glued' together. When the flight simulators at McAir (McDonnell Aircraft Company) were migrating from a pair of CDC something-or-anothers (750s?) to a bunch of networked Gould SELs, back in 1985, they had a quandary: what to do about the home-brew hardware controllers? The SELs were 16-bit machines, as were the specialized I/O widgie-frammises that drove the simulators. The Cyber would do some bit-fiddling to line the 16-bit data up in 12-bit words (with padding, IIRC) and all was OK, though a little strange. A reasonable sort of strange, under the circumstances. The obvious -- to me -- approach would have been to modify the electronics to accept 16-bit data from the SELs, but that's not what was done. (Because hardware is hard, software is easy?) Instead, the SELs would input and output the 16-bit data 12 bits at a time, just like the hardware expected, with padding and aligning done in software in the SEL. Guess who wrote the memory-to-memory translation code? In Fortran, with shift and mask operations. It was horrible, yet satisfying. The code had to handle any start address, and any length of data. That meant the padding had to be accounted for differently, depending on where in in the widgie-frammis's memory the data would reside. I had a false start or two before I had code that would work correctly, was fairly clear and reasonably-sized. Much collapsing of special cases. It had a certain elegance, considering. I wish I'd kept a copy. For one thing, I don't recall just what the requirements were, much less exactly how I resolved them. Though I suppose I could "reverse engineer" it from my recollections. (Will I? Sure. Right after my javelin-catching class.) For another, it's something I'm inordinately proud of, and would like to drag out and look over, now and again. My precioussss. > If some retrocomputing person ever bothers to write a C compiler for > the CDC instruction set, a C 'char' would almost certainly be 12 bits. > > Walt > -...- > Walt Spector > (w6ws at earthlink dot net) It may just be a mis-memory, but I seem to recall something about a C compiler for CDC machines, written back then. (From a university in Texas?) Maybe I'm just thinking of Pascal, which there most definitely was. Though not from Texas. -Eric S. ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 16 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:bzen4ESUmNz5qEPsLv47eLyyd/o= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Jul 2002 19:32:22 GMT NNTP-Posting-Host: 65.59.23.53 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025724742 65.59.23.53 (Wed, 03 Jul 2002 12:32:22 PDT) NNTP-Posting-Date: Wed, 03 Jul 2002 12:32:22 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!uni-erlangen.de!news-nue1.dfn.de!news-lei1.dfn.de!newsfeed00.sul.t-online.de!newsfeed01.sul.t-online.de!t-online.de!newspeer1-gui.server.ntli.net!ntli.net!cox.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111354 Charles Shannon Hendrix writes: > Mostly I just notice that I see a lot more references to the 4341 instead > of the 4381, and I mean in general, not just in your posts. > > Was the '81 a less popular machine? > > I remember the one I used in Richmond, VA, and there was one in city > hall in Newport News, VA. But most of what I saw was 3xxx or old 370 > machines, not counting things like the baby-mainframes and the AS/400s. 4341 was possibly one of the best price/performance machines for its time. slightly related http://www.garlic.com/~lynn/2001m.html#15 departmentl servers -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 3 Jul 2002 15:54:14 -0400 Lines: 37 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d235667_3@news.iglou.com> X-Trace: news.iglou.com 1025726055 204.250.0.238 (3 Jul 2002 15:54:15 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111287 "George R. Gonzalez" wrote in message = news:afv4i0$4qe$1@laurel.tc.umn.edu... >=20 > The console was an analog stroke-vector device basically, but it had = some > smarts so it could accept a stream of 6-bit codes and automatically = draw > them. The character vector data was stored in, IIRC a not easily = changeable > rope-memory ROM. George- by any chance, do you mean one of these: http://artemis.rrze.uni-erlangen.de/iser/pictures/I0480_02.jpg > At first the console was a pretty good match, technically and = economically > for the CPU. Later on, when 6/12 and 8/12 codes became popular, and = people > might have wanted to use the console from a remote location, the = console > became a bit of an millstone around the neck. Finally by the time the > integrated circuit 17x series came out, rumor had it they whole CPU = cost > less to build than the cost of the old console! That explains why = there > was a quick move to design a newer ,smaller, cheaper console. Only = one CRT, > split/screen display, much smaller and cheaper, altho I don't know if = it > kept the vector graphics. This is possibly the CC598-PC console product, based, I think,=20 on a Zenith Z-248 PC? -dq ###### From: "George R. Gonzalez" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 3 Jul 2002 15:24:39 -0500 Organization: University of Minnesota, Twin Cities Campus Lines: 51 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> <3d235667_3@news.iglou.com> Reply-To: "George R. Gonzalez" X-Trace: laurel.tc.umn.edu 1025728166 9241 160.94.124.25 (3 Jul 2002 20:29:26 GMT) X-Complaints-To: usenet@laurel.tc.umn.edu X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!upp1.onvoy!msc1.onvoy!onvoy.com!hardy.tc.umn.edu!laurel.tc.umn.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111362 "Douglas H. Quebbeman" wrote in message news:3d235667_3@news.iglou.com... "George R. Gonzalez" wrote in message news:afv4i0$4qe$1@laurel.tc.umn.edu... > > The console was an analog stroke-vector device basically, but it had some > smarts so it could accept a stream of 6-bit codes and automatically draw > them. The character vector data was stored in, IIRC a not easily changeable > rope-memory ROM. George- by any chance, do you mean one of these: http://artemis.rrze.uni-erlangen.de/iser/pictures/I0480_02.jpg I've never seen anything like that! What is it? Rough guess: adjustable delay line? > At first the console was a pretty good match, technically and economically > for the CPU. Later on, when 6/12 and 8/12 codes became popular, and people > might have wanted to use the console from a remote location, the console > became a bit of an millstone around the neck. Finally by the time the > integrated circuit 17x series came out, rumor had it they whole CPU cost > less to build than the cost of the old console! That explains why there > was a quick move to design a newer ,smaller, cheaper console. Only one CRT, > split/screen display, much smaller and cheaper, altho I don't know if it > kept the vector graphics. This is possibly the CC598-PC console product, based, I think, on a Zenith Z-248 PC? I didnt know about that console. The one I was thinking of was still a custom made CRT console, with like a 25-inch monochrome CRT. Some of those dim pictures somebody linked to looked like it. I never thought of a PC emulating a Cyber console! Sounds a tad tricky, but doable if one had enough info. Regards, George -dq ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 34 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:TqkaBlTx5vlbZ+APYU49j2q8j1Q= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 03 Jul 2002 20:45:38 GMT NNTP-Posting-Host: 64.156.35.159 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025729138 64.156.35.159 (Wed, 03 Jul 2002 13:45:38 PDT) NNTP-Posting-Date: Wed, 03 Jul 2002 13:45:38 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!fr.usenet-edu.net!usenet-edu.net!proxad.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111371 Anne & Lynn Wheeler writes: > 4341 was possibly one of the best price/performance machines for its > time. slightly related > http://www.garlic.com/~lynn/2001m.html#15 departmentl servers the talk referenced in the above posting about 11,000+ vax machines ... was given in 1983 (spring?) note as per: http://www.garlic.com/~lynn/2002f.html#0 the total world-wide vax ships as of the end of 1982 was 14,508 and the total as of the end of 1983 was 25,070. from: http://www.isham-research.com/chrono.html 4341 announced 1/79 and fcs 11/79 4381 announced 9/83 and fcs 1q/84 workstation and PCs were starting to come on strong in the departmental server market by the time 4381s started shipping in quantity. also per: http://www.garlic.com/~lynn/2002f.html#0 while the total number of vax shipments kept climbing thru '87 ... they were micro-vax. combined 11/750 & 11/780 world wide shipments thru 1984 was 35,540 ... and then dropped to a combined total of 7,600 for 1985 and 1660 for 1986. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: 3 Jul 2002 20:53:55 GMT Organization: The National Capital FreeNet Lines: 19 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 1025729635 3594 134.117.136.30 (3 Jul 2002 20:53:55 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 3 Jul 2002 20:53:55 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!nntp.abs.net!news-peer-east1.sprintlink.net!news.sprintlink.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:111392 Charles Shannon Hendrix (shannon@news.widomaker.com) writes: > ... > Did IBM compilers like COBOL and FORTRAN do anything like this behind > the scenes to increase througput? > > What was the proper way to interleave I/O and processing on the IBM, > assuming your language would you, or you were using assembler? ... Not in Ye Olden Days, unless one used assembler and some creative I/O macros (EXCP, READ or WRITE with CHECK, and POINT if repositioning) - OR - more transparently, one used PL/I which could perform the overlap in two ways: initiate an I/O operation, then continue crunching until a WAIT; or, if the problem could be split, multitasking. These techniques had a large payoff when most of the machine could be dedicated to the job at hand, usually non-primetime end-of-{day/month/ year} type stuff. ###### Message-ID: <3D236909.87652FF8@bellsouth.net> From: "Rostyslaw J. Lewyckyj" Reply-To: urjlew@bellsouth.net X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 66 X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Wed, 03 Jul 2002 17:11:27 EDT Organization: WEBUSENET.com Date: Wed, 03 Jul 2002 17:13:45 -0400 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cyclone.bc.net!newsfeed.telusplanet.net!peer1-sjc1.usenetserver.com.MISMATCH!ps01-sjc1!news.webusenet.com!pc01.webusenet.com!e3500-atl2.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111488 Charles Shannon Hendrix wrote: > In article , Anne & Lynn Wheeler wrote: > > Charles Shannon Hendrix writes: > >> How did the two stay in sync? I mean, I have vague and fuzzy > >> recollections of doing assembler on a 4381 and since the programs are > >> a mix of I/O and calculation, I assume this is very important. > > > > basically instruction processors and I/O processors were architected > > to be asyncronous .... effectively in much the same way that machines > > with multiple instruction processors (SMP) are typically architected > > so that the multiple instruction processors operate asyncronously. > > [snip] > > Ah, I remember that now. In some ways I regret changing track away from > the IBM systems. It's just that the local colleges didn't really let > you learn much about them. > > Everything we did on the mainframe was basically: > > * load up a big chunk of records into a block > * process the block > * repeat > > It always seemed to me that we should have started another I/O operation > just before processing the current block, so at least in theory the next > would be ready when you finished, or at least would be ready faster. > > Did IBM compilers like COBOL and FORTRAN do anything like this behind > the scenes to increase througput? > > What was the proper way to interleave I/O and processing on the IBM, > assuming your language would you, or you were using assembler? > > Most UNIX systems try to do this sort of thing automatically, although > there are a lot of things you can do yourself to optimize I/O and > processing to work together. The UNIX process model is good but I/O is > almost totally divorced from a process. Throughput overall can be very > good, but a single program with a huge I/O queue can hold up processes > with small amounts of I/O for some time. > > That's one thing I really miss about systems like MVS/XA: the ability > to control I/O at the process level when you need to. I think Solaris > can do this on high-end systems now, but it's not available in all of > the UNIX world yet, most especially not in the free systems. Well for sequential processing there was a DCB parameter BUFNO, and the FORTRAN i/o libraries would automatically set up your dataset processing with BUFNO=5. I.e. the system would allocate a ring of 5 buffers each of size BLKSIZ. Your first READ would open the file fill the 5 buffers and start giving you your records. When a buffer was used up the system would start giving you records from the next buffer and asynchronously refill the freed buffer. On output you would get the same in reverse, with the last filled buffers being written out on closing the file. In case of some program terminations your buffers were properly cleaned and the files closed by the system. For other program ABENDs e.g. 322 (program time out) the buffers didn't get flushed (though they did get closed). In assembler, you of course had more flexibility, with various i/o access methods supported by macroes, and at a lower level you could write channel programs (EXCP level i/o), and at a still lower level you could if privileged program down to the metal with SIO (start i/o) and other such privileged instructions. --Rostyk ###### Message-ID: <3D2385C2.7E69ABB6@ev1.net> From: Charles Richmond Reply-To: richmond@ev1.net Organization: Cannine Computer Center X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 28 NNTP-Posting-Host: 12.237.69.162 X-Complaints-To: abuse@attbi.com X-Trace: sccrnsc03 1025731074 12.237.69.162 (Wed, 03 Jul 2002 21:17:54 GMT) NNTP-Posting-Date: Wed, 03 Jul 2002 21:17:54 GMT Date: Wed, 03 Jul 2002 21:17:54 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wn1feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!sccrnsc03.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111456 Bruce Hoult wrote: > > In article <3d215922$1@news.meer.net>, woolsey@jlw.com (Jeff Woolsey) > wrote: > > > >Other 6600 "firsts" included the use of a video display operator's console > > > > "Video" is a smidge inaccurate. It's a very glorified oscilloscope > > (Tek scopes of the day had similar calligraphic characters for > > readouts). > > At the university I went to in the early 80's there was a Tek terminal > on the VAX. It was a pain because when your text output got to the > bottom of the screen it would start again from the first line and > overprint (without erasing) whatever was there previously. So you had > to hit some sort of "clear screen" key all the time. Also, it just had > some sort of very long timeconstant phosphor, because the text (and > graphics) would fade away after a few minutes. > Sounds like a Tektronics 4010... Somewhere I have the docs for drawing points and lines on the thing. It encodes into ASCII characters the commands for graphics. Also IIRC, the xterm's in X-windows have a Tek 4010 window that you can bring up... -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ ###### From: Larry__Weiss Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 03 Jul 2002 16:23:19 -0500 Organization: Airnews.net! at Internet America Lines: 10 Message-ID: X-Orig-Message-ID: <3D236B46.3EA98CCC@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d215922$1@news.meer.net> <3D2385C2.7E69ABB6@ev1.net> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Wed Jul 3 16:34:27 2002 NNTP-Posting-Host: !aFk61k-X4E25Ti-RH8& (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!newsfeed1.uni2.dk!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wn1feed!worldnet.att.net!209.122.83.58!howland.erols.net!news.airnews.net!cabal10.airnews.net!cabal2.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111487 Charles Richmond wrote: > Sounds like a Tektronics 4010... Somewhere I have the docs for > drawing points and lines on the thing. It encodes into ASCII > characters the commands for graphics. Also IIRC, the xterm's > in X-windows have a Tek 4010 window that you can bring up... > xterm has a Tek 4014 emulation available. - LarryW ###### From: "John Homes" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 10:07:40 +1200 Organization: EDS (New Zealand) Lines: 39 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> NNTP-Posting-Host: dhcp-134-251-165-075.dhcp.nz.eds.com X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4522.1200 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-han1.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news.he.net!news-out.spamkiller.net!propagator2-maxim!propagator-maxim!news-in.spamkiller.net!news02.tsnz.net!newsfeed01.tsnz.net!news.eds.co.nz!news.nz.eds.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111381 "Heinz W. Wiggeshoff" wrote in message news:afvo93$3ga$1@freenet9.carleton.ca... > Charles Shannon Hendrix (shannon@news.widomaker.com) writes: > > > ... > > Did IBM compilers like COBOL and FORTRAN do anything like this behind > > the scenes to increase througput? > > > > What was the proper way to interleave I/O and processing on the IBM, > > assuming your language would you, or you were using assembler? > ... > > Not in Ye Olden Days, unless one used assembler and some creative > I/O macros (EXCP, READ or WRITE with CHECK, and POINT if repositioning) > - OR - more transparently, one used PL/I which could perform the overlap > in two ways: initiate an I/O operation, then continue crunching until > a WAIT; or, if the problem could be split, multitasking. > Your Olden Days must be even more Olden than me, then. For as far back as my memories go, double buffering was available in the Access Method (QSAM, which was what we used unless there were Very Good Reasons to do something else), and COBOL supported that. I think the systax has *always* been in COBOL. > These techniques had a large payoff when most of the machine could be > dedicated to the job at hand, usually non-primetime end-of-{day/month/ > year} type stuff. The payoff was almost always worth it if a. Most of the work was batch sequential. b. The machine could only cope with a small number of tasks. Both of these were usually true in my Olden Days. John Homes. ###### From: "Kent Olsen" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 3 Jul 2002 18:30:15 -0400 Organization: Sprint Advanced Network Services Lines: 57 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d200c9e.299409694@sd.znet.com> <3d21f188$1@news.ucsc.edu> <3d22ae06.92058870@sd.znet.com> NNTP-Posting-Host: 04016187.ppptlh.nettally.com X-Trace: news.utelfla.com 1025735331 16155 199.44.16.187 (3 Jul 2002 22:28:51 GMT) X-Complaints-To: usenet@news.utelfla.com NNTP-Posting-Date: Wed, 3 Jul 2002 22:28:51 +0000 (UTC) X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed-east.nntpserver.com!nntpserver.com!news.sprintnetops.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111375 "Jitze Couperus" wrote in message news:3d22ae06.92058870@sd.znet.com... > On 2 Jul 2002 11:31:36 -0800, eugene@cse.ucsc.edu (Eugene Miya) wrote: > > > > > >What was a Cyber 2000? > >Unfortunately, the FTN compiler got too old for its time. > > > > The Cyber 2000 was indeed the last, biggest, and baddest > of the Mohicans. The top end of the dual-state machines > had been the 990 (also known as Theta during its development) > but this was still a water-cooled machine. Then in the final > spasm of CDC as a manufacturer of knuckle-dragging > machines, they came out with the 2000 which was air- > cooled and had an optional vector unit - but only ran > in 180 (NOS/VE) state. The irony of this machine is > that the last few sold were not used for Fortran number > crunching at all, but filled specialised heavy-duty computing > niches in other domains completely. One was used to maintain > the mother of all databases (both very large for its day and involving > lots of very complex logic) and another site used theirs as the core > engine for the mother of all financial transaction systems. > > The overall muscle that these machines could bring to bear > on these problems (not just CPU power, but e.g. bandwidth > and address space) made both of them extrememly difficult > to replace at their sites, one of which I understand is still > trucking to this day. > > Jitze > As was typical with CDC, little was ever engineered that wasn't horrible under-engineered or massive overkill. Address space in the NOS/VE systems was one of the latter. An address space (segment) was 2^31 bytes (2BG). Each task could have up to 4095 segments. Total address space for the task 2^31 * 2^11 bytes, though some of the segments were reserved. And of course, a job step could have multiple tasks, and the ankle bone connects to the shin bone... And just to show how technology has come full circle, today's high end RS6000 and Z900 mainframes are freon cooled. They look like refrigerators that have had their liners removed. Kent ###### From: "Kent Olsen" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 3 Jul 2002 18:37:27 -0400 Organization: Sprint Advanced Network Services Lines: 22 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> NNTP-Posting-Host: 04016187.ppptlh.nettally.com X-Trace: news.utelfla.com 1025735763 16571 199.44.16.187 (3 Jul 2002 22:36:03 GMT) X-Complaints-To: usenet@news.utelfla.com NNTP-Posting-Date: Wed, 3 Jul 2002 22:36:03 +0000 (UTC) X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed-east.nntpserver.com!nntpserver.com!news.sprintnetops.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111365 <> > > When the flight simulators at McAir (McDonnell Aircraft Company) were migrating > from a pair of CDC something-or-anothers (750s?) to a bunch of networked Gould > SELs, back in 1985, they had a quandary: what to do about the home-brew hardware > controllers? Ah, McAuto. I turned down a job with their systems group back in '78 when the interview took a rather odd twist. It appeared that their offer wasn't a "job offer", but rather a "volleyball scholarship" to play in their intramural team. But I loved what they were doing there. I still wish that I could have "flown" one of the simulators. :) Kent ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> <3d235667_3@news.iglou.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 3 Jul 2002 19:23:11 -0400 Lines: 43 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d238760_1@news.iglou.com> X-Trace: news.iglou.com 1025738592 204.250.0.238 (3 Jul 2002 19:23:12 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!syros.belnet.be!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111285 "George R. Gonzalez" wrote in message = news:afvmr6$90p$1@laurel.tc.umn.edu... >=20 > "Douglas H. Quebbeman" wrote in = message > news:3d235667_3@news.iglou.com... > "George R. Gonzalez" wrote in message > news:afv4i0$4qe$1@laurel.tc.umn.edu... > > > > The console was an analog stroke-vector device basically, but it had = some > > smarts so it could accept a stream of 6-bit codes and automatically = draw > > them. The character vector data was stored in, IIRC a not easily > changeable > > rope-memory ROM. >=20 > George- by any chance, do you mean one of these: >=20 > http://artemis.rrze.uni-erlangen.de/iser/pictures/I0480_02.jpg >=20 > I've never seen anything like that! What is it? Rough guess: = adjustable > delay line? I think so, though Google's German translation called it a "memory"; technically, an acoustic delay line *is* a memory... =20 > This is possibly the CC598-PC console product, based, I think, > on a Zenith Z-248 PC? >=20 > I didnt know about that console. The one I was thinking of was still > a custom made CRT console, with like a 25-inch monochrome CRT. > Some of those dim pictures somebody linked to looked like it. >=20 > I never thought of a PC emulating a Cyber console! Sounds a tad = tricky, > but doable if one had enough info. never seen one operate, but they *do* (did?) exist... -dq ###### From: "keep-it-clean" Newsgroups: alt.folklore.computers References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Subject: Re: CDC6600 - just how powerful a machine was it? Lines: 71 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Message-ID: Date: Thu, 04 Jul 2002 00:07:44 GMT NNTP-Posting-Host: 12.89.77.52 X-Complaints-To: abuse@worldnet.att.net X-Trace: bgtnsc04-news.ops.worldnet.att.net 1025741264 12.89.77.52 (Thu, 04 Jul 2002 00:07:44 GMT) NNTP-Posting-Date: Thu, 04 Jul 2002 00:07:44 GMT Organization: AT&T Worldnet Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!newsfeed01.sul.t-online.de!t-online.de!newspeer1-gui.server.ntli.net!ntli.net!cox.net!cyclone1.gnilink.net!wn3feed!worldnet.att.net!bgtnsc04-news.ops.worldnet.att.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111355 By chance do you mean 2C39 vacuum tubes instead of 3C29 ? 2C39s could fit your description and were very popular in transmitter driver & power amplifier service....afraid I've never heard of 3C29 on the other hand. =++=++=++=++=++=++= "George R. Gonzalez" wrote in message news:afv4i0$4qe$1@laurel.tc.umn.edu... > > Some random notes on the CDC consoles: > > It was a dual-scope display. The display tubes were rather primitive > electrostatic deflection CRT's. They were very loooong, thereby > necessitating that huge deep cabinet. > > The deflection plates were driven by some really hefty tubes! Your typical > tv of that screen size would have a medium-power 6CU6, each CRT of the > console had a pair of 40-watt blast-air cooled 3C29's, one of the hottest > tubes of WWII. The repair guys were scheduled to replace these every 2000 > hours or so, thereby providing a continual source of still usable > transmitting tubes for ham radio guys. > > The console was an analog stroke-vector device basically, but it had some > smarts so it could accept a stream of 6-bit codes and automatically draw > them. The character vector data was stored in, IIRC a not easily changeable > rope-memory ROM. > Thereby the term DISPLAY CODE, and that's one of reasons the poor machines > were stuck with that unusual 6-bit code scheme instead of ASCII. I don't > think the early consoles had anyway of doing the 12 bit extended codes > either. > > The keyboard was mighty cheap looking for a $million dollar computer. It > had ugly gray styrene keys. It had to generate 6-bit display code of > course, so no lower case. > > At first the console was a pretty good match, technically and economically > for the CPU. Later on, when 6/12 and 8/12 codes became popular, and people > might have wanted to use the console from a remote location, the console > became a bit of an millstone around the neck. Finally by the time the > integrated circuit 17x series came out, rumor had it they whole CPU cost > less to build than the cost of the old console! That explains why there > was a quick move to design a newer ,smaller, cheaper console. Only one CRT, > split/screen display, much smaller and cheaper, altho I don't know if it > kept the vector graphics. > > One fella in their research lab came up with "the six pack"-- a bizarre > contraption: a gaggle of 10 PPU's, a disk, no CPU, from one to 6 consoles, > all running a simple word-processing screen editor (probably a descendant of > the "O26" console editor! IIRC the marketing guys could not figure out how > to sell such a bizarre device. > > Regards, > > > George > > > > ###### Message-ID: <3D239559.A1D80F5E@thinkage.ca> From: "Alan T. Bowler" Organization: Thinkage Ltd. X-Mailer: Mozilla 4.5 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 46 Date: Wed, 03 Jul 2002 20:22:49 -0400 NNTP-Posting-Host: 192.102.11.4 X-Trace: nnrp1.uunet.ca 1025742169 192.102.11.4 (Wed, 03 Jul 2002 20:22:49 EDT) NNTP-Posting-Date: Wed, 03 Jul 2002 20:22:49 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!snoopy.risq.qc.ca!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111290 Charles Shannon Hendrix wrote: > > It always seemed to me that we should have started another I/O operation > just before processing the current block, so at least in theory the next > would be ready when you finished, or at least would be ready faster. > > Did IBM compilers like COBOL and FORTRAN do anything like this behind > the scenes to increase througput? Not the compilers themselves, but the standard I/O libraries used by programs written in the those languages certainly did. As someone else pointed out, all you have to do was specify (or accept the default ) more then 1 buffer. It was usual to double buffer, unless you were tight on memory (and that was common). > > What was the proper way to interleave I/O and processing on the IBM, > assuming your language would you, or you were using assembler? Simply make use of the normal sequential access routines and double buffer. > > Most UNIX systems try to do this sort of thing automatically, although > there are a lot of things you can do yourself to optimize I/O and > processing to work together. The UNIX process model is good but I/O is > almost totally divorced from a process. Actually, the Unix model doesn't "do this sort of thing automatically", the usual model to the process is synchonous I/O i.e. issue request and block until data is moved to/from your address space. There is some level of async processing when you aren't using a raw device and so getting system caching. Making I/O synchronous at the program level has the advantage that it makes programming simpler, and comes out of some experience with mainframes. Namely in a multi-process environment, multiple buffering often does not increase total system throughput that much (if any). Provided that there is enough memory for sufficient other programs to be be in memory another program could almost always be dispatched while one program was blocked waiting for I/O. Double buffering and async I/O for a single program only really gives much benefit when there are only when the system is has few concurrent programs running. Unix did acquire some async I/O features later, and they are needed. Especially for networked multiple conversation type things. But they still feel like warts compared to the basic ease of the usual sync I/O paradigm. ###### From: "Rupert Pigott" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 01:26:22 +0000 (UTC) Organization: BT Openworld Lines: 27 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> <3D239559.A1D80F5E@thinkage.ca> NNTP-Posting-Host: host213-1-137-239.in-addr.btopenworld.com X-Trace: helle.btinternet.com 1025745982 1423 213.1.137.239 (4 Jul 2002 01:26:22 GMT) X-Complaints-To: news-complaints@lists.btinternet.com NNTP-Posting-Date: Thu, 4 Jul 2002 01:26:22 +0000 (UTC) X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MSMail-Priority: Normal X-Priority: 3 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!diablo.theplanet.net!news-hub.cableinet.net!blueyonder!btnet-peer!btnet-peer0!btnet-feed5!btnet!news.btopenworld.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111395 "Alan T. Bowler" wrote in message news:3D239559.A1D80F5E@thinkage.ca... > Charles Shannon Hendrix wrote: [SNIP] > Actually, the Unix model doesn't "do this sort of thing automatically", Buffer-cache (read ahead). > the usual model to the process is synchonous I/O i.e. issue > request and block until data is moved to/from your address space. > There is some level of async processing when you aren't using > a raw device and so getting system caching. Making I/O synchronous > at the program level has the advantage that it makes programming [SNIP] > Unix did acquire some async I/O features later, and they are > needed. Especially for networked multiple conversation type > things. But they still feel like warts compared to the basic > ease of the usual sync I/O paradigm. Aye, they do a little. Cheers, Rupert ###### From: "Eric S. Harris" Newsgroups: alt.folklore.computers,comp.sys.cdc,rec.arts.sf.written,alt.books.larry-niven Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 03 Jul 2002 22:17:13 -0500 Organization: MindSpring Enterprises Lines: 19 Distribution: inet Message-ID: <3D23BE39.C3BDA7FC@mindspring.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> <3d21ff60$1_1@news.iglou.com> <3d220b30$1@news.ucsc.edu> <3d22179f$1_1@news.iglou.com> Reply-To: eric_harris_76@yahoo.com NNTP-Posting-Host: a8.bf.6d.d3 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Server-Date: 4 Jul 2002 03:20:45 GMT X-Mailer: Mozilla 4.79 [en]C-CCK-MCD (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.stueberl.de!cox.net!newsfeed1.earthlink.net!newsfeed2.earthlink.net!newsfeed.earthlink.net!newsfeed0.news.atl.earthlink.net!news.atl.earthlink.net!news.mindspring.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111330 "Douglas H. Quebbeman" wrote: > > "Eugene Miya" wrote in message news:3d220b30$1@news.ucsc.edu... > > >Then, just in time, came the silicon transistor, which the manual > > >describes as "frosting on the cake". > > > > Cute: > > The golden days of the discrete component era (I can say that because I > > am more in software). > > > > Parallelism is on average at best an O(n) solution to performance problems. > > > > There are interesting technologies on the horizon. Hope some pan out. > > Quantum computers? I think the same powers that keep down fusion > research that doesn't involve tokamaks and laser confinement > systems might quash that, too. Gil "The ARM" Hamilton and the rest of the UN technology police? -Eric S. ###### From: lwinson@bbs.cpcn.com (lwin) Subject: Re: CDC6600 - just how powerful a machine was it? Newsgroups: alt.folklore.computers Organization: The PACSIBM SIG BBS Originator: root@bbs.cpcn.com Lines: 37 Message-ID: Date: Thu, 04 Jul 2002 03:25:23 GMT NNTP-Posting-Host: 207.8.186.93 X-Complaints-To: Abuse Role , We Care X-Trace: monger.newsread.com 1025753123 207.8.186.93 (Wed, 03 Jul 2002 23:25:23 EDT) NNTP-Posting-Date: Wed, 03 Jul 2002 23:25:23 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-xfer2.newshosting.com!yellow.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!POSTED.monger.newsread.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111424 x-no-archive: yes Two comments: 1) People say the old "classic" computers of the 1960s were very slow compared to today's PCs. While the CPU instruction cycle and memory access on a PC are faster today, what about the total _throughput_ capability of the classic computers? Our old S/360-40, with 192K, managed to support a batch partition and mini-CICS partition (about 3 terminals), as well as high speed printing and card reading. Could a single plain PC today support all that I/O activity? I really wonder if the bus on PCs today could match the channel control system of the mainframe. I'd think you'd have to hang an awful lot of supplementary hardware on a PC to make it work. I don't know the internals of the CDC line, but I suspect the same thing applies. 2) Our high school got access to a college CDC 6400 timesharing via a Teletype 33. It wasn't too reliable. A student had just finished typing in a long program, only to have the machine lose it. He was frustrated and typed an expletive with a period after it. Since it had a period, the system treated it as a command (and rejected it). However, all commands were shown on the system console with their origin. The operator was none too pleased, and advised the school with a printout, and the school wasn't too pleased either. We learned early on that indeed "big brother is watching you". ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1FD2A7.6690F451@attbi.com> <3d21ef6e$1@news.ucsc.edu> <3D22699B.F741E111@attbi.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3d23c367$1@news.ucsc.edu> Date: 3 Jul 2002 20:39:19 -0800 X-Trace: 3 Jul 2002 20:39:19 -0800, sundance.cse.ucsc.edu Lines: 14 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:111274 In article <3D22699B.F741E111@attbi.com>, Joe Yuska wrote: >> >2. dusty decks >The main reason I was looking for the sources for the codes from the >sixties and the seventies was that there were some neat adaptations to >the 6600 architecture including hand coded algorithms that took >advantage of the functional units. In addition there were specialized >disk drivers for the 808/6638 disks that used multiple ppu's >"ping-ponging" to avoid disk interleaving. I thought they would be cool >to examine. Well in the open literature, the LFK (Livermore Fortran Kernels, aka Livermore Loops) are interesting little programs. There are some interesting apparently hardware oriented features into these codes. ###### From: "Rupert Pigott" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 06:26:49 +0000 (UTC) Organization: BT Openworld Lines: 43 Message-ID: References: NNTP-Posting-Host: host213-1-130-142.in-addr.btopenworld.com X-Trace: venus.btinternet.com 1025764009 24044 213.1.130.142 (4 Jul 2002 06:26:49 GMT) X-Complaints-To: news-complaints@lists.btinternet.com NNTP-Posting-Date: Thu, 4 Jul 2002 06:26:49 +0000 (UTC) X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MSMail-Priority: Normal X-Priority: 3 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!feedme.news.mediaways.net!fu-berlin.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!kibo.news.demon.net!demon!btnet-peer0!btnet-feed5!btnet!news.btopenworld.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111409 "lwin" wrote in message news:DePU8.622$oN.115850@monger.newsread.com... > x-no-archive: yes > > Two comments: > > 1) > People say the old "classic" computers of the 1960s were very > slow compared to today's PCs. > > While the CPU instruction cycle and memory access on a PC are > faster today, what about the total _throughput_ capability of > the classic computers? > > > Our old S/360-40, with 192K, managed to support a batch partition > and mini-CICS partition (about 3 terminals), as well as high speed > printing and card reading. Would your old S/360-40 be able to saturate 4x100Mbit Ethernet cards like anything from an old Pentium II up can ? :) I know that a VAX-11/780 could *NOT* (SBI bandwidth is too low by a factor of 4, possibly more). I suspect in most cases the raw bandwidth figures for PCI (now PCI-X) and the memory on a PC outstrip the 80s mainframes by an order of magnitude. I suspect that many mid-late 90s mainframes are also blown away by a PCI-X equipped PC server now (in terms of raw figures). > Could a single plain PC today support all that I/O activity? I > really wonder if the bus on PCs today could match the channel > control system of the mainframe. I'd think you'd have to hang an > awful lot of supplementary hardware on a PC to make it work. I think PCs are more than up to the raw megabyte bandwidth now. What they probably are not so good at is handling *lots* of devices at once. You would certainly need some fairly exotic (by PC standards) hardware to attach say 1000 terminals directly to one. :) Cheers, Rupert ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 55 Date: Thu, 04 Jul 2002 07:44:21 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news2.calgary.shaw.ca 1025768661 24.71.223.147 (Thu, 04 Jul 2002 01:44:21 MDT) NNTP-Posting-Date: Thu, 04 Jul 2002 01:44:21 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news2.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111459 On Wed, 3 Jul 2002 14:33:02 -0400, Charles Shannon Hendrix wrote: >In article , Brian Inglis wrote: > >>>Was the '81 much faster than the '41? I ask because the '81 I used >>>felt a lot faster than what I've been told about the CDC 6600, but >>>then again, maybe it was just I/O speed. >> >> IIRC about 2.0-2.5MIPS instead of about 0.5-1.0MIPS depending on >> the models, and the 4381-3 was a dual (4.5MIPS), could have up to >> 64MB memory I think, and run VM/HPO, VM/XA, or MVS realistically. >> Can't remember if it could have a second set of six channels or >> not? > >Hmmm... the execution times for a lot of the code I wrote was very quick. >Some "faster" machines executed the same code much slower. I was always >surprised to see some jobs come back and the stats gave sub-second >execution times, while locally more powerful systems were nothing like >that. If you had a local controller connection, response was a lot faster than remote, and you effectively had type ahead, as screen write CCWs were chained to zero byte reads. Typical quanta depending on load and interaction were 0.1-1.0s so could run 250K-2.5M instructions; common system path lengths were pretty short; plus had ucode assists for 10x speed up of common operations and often avoided a context switch; and the system was built to do lots of I/O fast. We typically had about 200 PROFS users banging away at any time, plus engineering, commercial, program development, and a lot of network traffic with other systems doing email, NJE work, and printing. >Of course, this system had a RAM drive for student use. It was 256MB >and I don't know what model it was. I seem to recall that it appeared >to us as a 3380, but it's been a long time. Does anyone know how this >might have been implemented? Possibly a Hitachi/Fujitsu? Solid State Disk (SSD) -- probably had a Channel-to-Channel Adapter (CTCA) interface -- which meant it dealt in units of pages at high speed. Could also have been a regular channel interface, but that would have been limited to 3MB/s plus channel overhead -- so acted liked a disk with zero seek/rotation times. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 55 Date: Thu, 04 Jul 2002 08:09:01 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news1.calgary.shaw.ca 1025770141 24.71.223.147 (Thu, 04 Jul 2002 02:09:01 MDT) NNTP-Posting-Date: Thu, 04 Jul 2002 02:09:01 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!snoopy.risq.qc.ca!cyclone.bc.net!newsfeed.telusplanet.net!peer1-sjc1.usenetserver.com.MISMATCH!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news1.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111471 On Thu, 4 Jul 2002 06:26:49 +0000 (UTC), "Rupert Pigott" wrote: >"lwin" wrote in message >news:DePU8.622$oN.115850@monger.newsread.com... >> x-no-archive: yes >> >> Two comments: >> >> 1) >> People say the old "classic" computers of the 1960s were very >> slow compared to today's PCs. >> >> While the CPU instruction cycle and memory access on a PC are >> faster today, what about the total _throughput_ capability of >> the classic computers? >> >> >> Our old S/360-40, with 192K, managed to support a batch partition >> and mini-CICS partition (about 3 terminals), as well as high speed >> printing and card reading. > >Would your old S/360-40 be able to saturate 4x100Mbit Ethernet >cards like anything from an old Pentium II up can ? :) I know >that a VAX-11/780 could *NOT* (SBI bandwidth is too low by a >factor of 4, possibly more). I suspect in most cases the raw >bandwidth figures for PCI (now PCI-X) and the memory on a PC >outstrip the 80s mainframes by an order of magnitude. I suspect >that many mid-late 90s mainframes are also blown away by a >PCI-X equipped PC server now (in terms of raw figures). > >> Could a single plain PC today support all that I/O activity? I >> really wonder if the bus on PCs today could match the channel >> control system of the mainframe. I'd think you'd have to hang an >> awful lot of supplementary hardware on a PC to make it work. > >I think PCs are more than up to the raw megabyte bandwidth now. >What they probably are not so good at is handling *lots* of >devices at once. You would certainly need some fairly exotic >(by PC standards) hardware to attach say 1000 terminals directly >to one. :) How about adding 25,000 remote terminals, and maybe 1000 disk drives out there pumping 250MB/s through the system, doing something useful with most bytes and all in 64MB? -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### From: Steve O'Hara-Smith Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 18:36:13 +0200 Organization: Wanadoo NL Lines: 33 Message-ID: <20020704183613.46af1766.steveo@eircom.net> References: NNTP-Posting-Host: i2237.vwr.wanadoo.nl Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Trace: scavenger.euro.net 1025802005 99282 194.134.216.198 (4 Jul 2002 17:00:05 GMT) X-Complaints-To: abuse@wanadoo.nl NNTP-Posting-Date: Thu, 4 Jul 2002 17:00:05 +0000 (UTC) X-Newsreader: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd4.6) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!uni-erlangen.de!news-nue1.dfn.de!news-lei1.dfn.de!newsfeed00.sul.t-online.de!t-online.de!kibo.news.demon.net!demon!news2.euro.net!news.euronet.nl!ams-gw.sohara.org!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111408 On Thu, 04 Jul 2002 08:09:01 GMT Brian Inglis wrote: BI> How about adding 25,000 remote terminals, and maybe 1000 disk BI> drives out there pumping 250MB/s through the system, doing That'd be about 20 modern drives, the 250MB/s is too much for 32 bit PCI at 33MHz but easily in range of 64 bit PCI at 66MHz so yes I think a modern (high end) PC can do that part. The 25K terminals is a bit of a problem. Are concentrator equivalents permissible ? If so there seems to be enough PCI bus bandwidth left over (just about). The memory bandwidth of a modern PC is much higher so the CPU(s) can still get at the data as it flies through the system. It's pretty close to the limit though (this week). The PC that used to take up to 5000 ftp sessions at once and one day kicked out two and a quarter terabytes of data (30MB/s average) was rather slow by todays standards. BI> something useful with most bytes and all in 64MB? That's just a matter of the right kind of software - none of which runs on a PC AFAIK. I don't see why it shouldn't be possible to write some but I also don't see anyone leaping to do it. There is one leetle problem - I don't think you can get high bandwidth PC memory that small :) -- C:>WIN | Directable Mirrors The computer obeys and wins. |A Better Way To Focus The Sun You lose and Bill collects. | licenses available - see: | http://www.sohara.org/ ###### From: "Rupert Pigott" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 16:45:38 +0000 (UTC) Organization: BT Openworld Lines: 37 Message-ID: References: NNTP-Posting-Host: host213-1-134-239.in-addr.btopenworld.com X-Trace: helle.btinternet.com 1025801138 5013 213.1.134.239 (4 Jul 2002 16:45:38 GMT) X-Complaints-To: news-complaints@lists.btinternet.com NNTP-Posting-Date: Thu, 4 Jul 2002 16:45:38 +0000 (UTC) X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MSMail-Priority: Normal X-Priority: 3 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!eusc.inter.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!diablo.theplanet.net!btnet-peer!btnet-peer0!btnet-feed5!btnet!news.btopenworld.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111400 "Brian Inglis" wrote in message news:svv7iusf51b5km2oalo792kuhniogusrao@4ax.com... [SNIP] > How about adding 25,000 remote terminals, and maybe 1000 disk > drives out there pumping 250MB/s through the system, doing > something useful with most bytes and all in 64MB? I'm fairly sure that you could rig up something close to that. 250MB/s is out of the question for PCI (unless it's 66/64bit), but well within range for PCI-X. For the terminals you'd need multiplexors, I suspect that these days you'd cheat by buying a fat pipe and a gigabit ethernet card. :) The closest thing to configurations based on PCs which I found real evidence of are the Walnut Creek site, this article is a *very* old one, but there are more recent with bigger stats. :) http://www.cnn.com/TECH/computing/9904/08/cdrom.idg/ Apparently Deja used to run on a dinky little PPro Linux setup back in the mid 90s. Sure I doubt people could all run KDE on such a system, but running mainframe forms style apps for 10k users should be well-within range for a modern PC. I suspect you'd need to do a fair bit of kernel tuning (just modding #defines & recompile) to get decent response time out of it. It's horrible I know. But PCs really are wickedly fast these days, and they are *still* gettin faster. They're also starting to use Mainframe cooling technoligies, and some of the PSUs we see now are getting into that 600W+ range that VAX-11/780's used... Sigh. I'd much rather they got smaller quieter & cooler myself. Cheers, Rupert ###### From: Larry__Weiss Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 04 Jul 2002 12:21:39 -0500 Organization: Airnews.net! at Internet America Lines: 12 Message-ID: <2A5A31D939C5383A.E688358DCFA3BF5E.5348E5B6629D9595@lp.airnews.net> X-Orig-Message-ID: <3D248423.876D9874@airmail.net> References: Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library1-aux.airnews.net NNTP-Posting-Time: Thu Jul 4 12:21:53 2002 NNTP-Posting-Host: ![/g71k-VqY`98g1-]^X (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.tele.dk!small.news.tele.dk!128.230.129.106!news.maxwell.syr.edu!news.airnews.net!cabal11.airnews.net!cabal2.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111475 Rupert Pigott wrote: > ... Sigh. I'd much rather they got smaller quieter & cooler > myself. > My Palm Pilot is plenty small/quiet/cool ... maybe something will evolve from that platform to compete with the PCs. And just to stay a little bit OT, how does the insides of a Palm Pilot compare powerwise with a CDC6600? - LarryW ###### From: jdege@jdege.visi.com (Jeffrey C. Dege) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Message-ID: User-Agent: slrn/0.9.5.4 (UNIX) Lines: 23 Date: Thu, 04 Jul 2002 17:21:44 GMT NNTP-Posting-Host: 209.98.6.59 X-Complaints-To: abuse@visi.com X-Trace: ruti.visi.com 1025803304 209.98.6.59 (Thu, 04 Jul 2002 12:21:44 CDT) NNTP-Posting-Date: Thu, 04 Jul 2002 12:21:44 CDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!eusc.inter.net!priapus.visi.com!zeus.visi.com!news-out.visi.com!ruti.visi.com!jdege Xref: chonsp.franklin.ch alt.folklore.computers:111427 On Sun, 30 Jun 2002 08:05:57 -0500, Larry__Weiss wrote: > >I just stumbled across some pics of the CDC6600: > http://www.bambi.net/computer_museum/cdc6600_and_console.jpg > http://online.sfsu.edu/~hl/mmm.html > >Just how powerful was that machine? How would it compare to modern-day >machines, especially PC's? 4096 60-bit words. 18-bit integers, 6-bit characters, ones-complement representation. I had one class in school (in 1987!) that programmed assembly on a CDC 6600 emulator running on a Cyber 76, because the instructor wouldn't upgrade his class notes. (We called it Computer Archeology). -- If ever the free institutions of America are destroyed, that event may be attributed to the omnipotence of the majority, which may at some future time urge the minorities to desperation and oblige them to have recourse to physical force. Anarchy will then be the result, but it will have been brought about by despotism. - Alexis De Toqueville ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers References: Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 13:51:18 -0400 Lines: 39 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d248b17$1_3@news.iglou.com> X-Trace: news.iglou.com 1025805079 204.250.0.238 (4 Jul 2002 13:51:19 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!nntp-out.monmouth.com!newspeer.monmouth.com!nntp.abs.net!uunet!dca.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111301 "lwin" wrote in message news:DePU8.622> 1) > People say the old "classic" computers of the 1960s were very=20 > slow compared to today's PCs. >=20 > While the CPU instruction cycle and memory access on a PC are > faster today, what about the total _throughput_ capability of > the classic computers? >=20 > Our old S/360-40, with 192K, managed to support a batch partition > and mini-CICS partition (about 3 terminals), as well as high speed > printing and card reading. >=20 > Could a single plain PC today support all that I/O activity? I > really wonder if the bus on PCs today could match the channel > control system of the mainframe. I'd think you'd have to hang an > awful lot of supplementary hardware on a PC to make it work. >=20 > I don't know the internals of the CDC line, but I suspect the > same thing applies. Actually, a CDC 6600 with 60 terminals hanging off it was *terribly* bogged down, giving crappy response time to all 60 users. I can't imagine trying to hang more _character mode_ terminals off such a system. But the comparisons often made to IBM mainframes with 1000s of terminals isn't fair; unless I'm mistaken, those 1000s of terminals were usually block-mode terminals, or ones=20 which were at least operating in block mode. Not having=20 to process an interrupt for every daggone character allows you to hang a lot more of them onto the muxes, and more muxes on the channels. CDC also provided for block-mode terminals and transaction- oriented subsystems, but I personally never saw those in use. Regards, -doug q ###### From: cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 18:56:10 +0100 Organization: Honest Chris' Sysadmin Emporium Message-ID: References: <20020704183613.46af1766.steveo@eircom.net> NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 1025805609 nnrp-01:865 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net X-Newsreader: trn 4.0-test76 (Apr 2, 2001) Originator: cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) Lines: 27 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!uni-erlangen.de!newsfeed.arcor-online.net!news.belwue.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!kibo.news.demon.net!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111486 According to Steve O'Hara-Smith : > That'd be about 20 modern drives, the 250MB/s is too much > for 32 bit PCI at 33MHz but easily in range of 64 bit PCI at 66MHz so > yes I think a modern (high end) PC can do that part. The 25K terminals > is a bit of a problem. Are concentrator equivalents permissible ? If so > there seems to be enough PCI bus bandwidth left over (just about). The > memory bandwidth of a modern PC is much higher so the CPU(s) can still > get at the data as it flies through the system. It's pretty close to the > limit though (this week). I'm curious to know how PCI performs under load... I get the impression that its sustainable bandwidth isn't that high a percentage of its theoretical peak, but that's just my perception. I suppose I just don't like the current PC architecture much: it reminds me too much of an old over-maintained program that has kludge layered on top of kludge which nobody's had the courage to redesign from scratch! At the very least I'd like to see the 16-bit support with all the hideousity that it entails dropped from new designs, but I've no doubt it would result in howls of protest from people who still want to run DOS v3 on the newest hardware... Not too sure how I'd do it if I had a blank sheet. Probably connect all the internal bits using a crossbar switch and hang all the other components (everything from mass storage to basic IO) off an array of SCSI channels. But then it'd probably be classed as a small mainframe. :) Chris. ###### From: Steve O'Hara-Smith Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 20:17:00 +0200 Organization: Wanadoo NL Lines: 14 Message-ID: <20020704201700.232b459c.steveo@eircom.net> References: NNTP-Posting-Host: rot2-p0682.dial.wanadoo.nl Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Trace: scavenger.euro.net 1025811076 48121 62.234.200.170 (4 Jul 2002 19:31:16 GMT) X-Complaints-To: abuse@wanadoo.nl NNTP-Posting-Date: Thu, 4 Jul 2002 19:31:16 +0000 (UTC) X-Newsreader: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd4.6) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!uni-erlangen.de!news-nue1.dfn.de!news-lei1.dfn.de!newsfeed.freenet.de!news2.euro.net!news.euro.net!news.euronet.nl!ams-gw.sohara.org!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111404 On Thu, 4 Jul 2002 16:45:38 +0000 (UTC) "Rupert Pigott" wrote: RP> used... Sigh. I'd much rather they got smaller quieter & cooler RP> myself. They did - they're called PDAs and pocket computers and things like that. I've even seen one of those things running a multi user OS. -- C:>WIN | Directable Mirrors The computer obeys and wins. |A Better Way To Focus The Sun You lose and Bill collects. | licenses available - see: | http://www.sohara.org/ ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 14 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:+z7Ss01dYbCh74lM2hwcsUXIoJo= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Jul 2002 18:30:15 GMT NNTP-Posting-Host: 209.244.79.161 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025807415 209.244.79.161 (Thu, 04 Jul 2002 11:30:15 PDT) NNTP-Posting-Date: Thu, 04 Jul 2002 11:30:15 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!cyclone.socal.rr.com!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111358 "Rupert Pigott" writes: >> How about adding 25,000 remote terminals, and maybe 1000 disk >> drives out there pumping 250MB/s through the system, doing >> something useful with most bytes and all in 64MB? redoing "routes" for a bit more than 25k terminal/users: http://www.garlic.com/~lynn/96.html#31 Mainframes & Unix http://www.garlic.com/~lynn/99.html#153 Uptime (was Re: Q: S/390 on PowerPC?) http://www.garlic.com/~lynn/2000.html#61 64 bit X86 ugliness (Re: Williamette trace cache (Re: First view of Willamette)) http://www.garlic.com/~lynn/2000f.html#20 Competitors to SABRE? http://www.garlic.com/~lynn/2001k.html#26 microsoft going poof [was: HP Compaq merger, here we go again.] -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <20020704183613.46af1766.steveo@eircom.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 26 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:L+j10E5+0CpYneK1P71J7jrkf+k= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Jul 2002 18:40:11 GMT NNTP-Posting-Host: 209.244.79.161 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025808011 209.244.79.161 (Thu, 04 Jul 2002 11:40:11 PDT) NNTP-Posting-Date: Thu, 04 Jul 2002 11:40:11 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-hog.berkeley.edu!ucberkeley!newshub.sdsu.edu!west.cox.net!cox.net!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111376 Steve O'Hara-Smith writes: > The PC that used to take up to 5000 ftp sessions at once and > one day kicked out two and a quarter terabytes of data (30MB/s average) > was rather slow by todays standards. remember when the netscape browser download had machines 1-19? that was somewhat alliviated when they got netscape20 machine ... a large sequent that was configured to handle 20,000 sessions at once and I believe was the first kernel that had a serious fix for the finwait problem for webservers. http wasn't so much of an issue of the number of concurrern sessions, it was the number of sessions endings per second and the length of the dangling finwait queue (high activity web servers spending 98 percent of total cpu running the finwait list). tcp finwait hadn't seen it before, even with relatively high number of concurrent sessions because tcp was somewhat presumed to be a connection protocol that lasted for some time. http1.0 is effectively a connectionless protocol being driven over a connection protocol (as a result http would drive the number of tcp session terminations per second thru the roof, as well as causing a large amount of session setup/tear-down packet chatter). -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <3d248b17$1_3@news.iglou.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 35 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:QF8taT1q/C7ZAvEuILct/DbK6Ug= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 04 Jul 2002 18:51:39 GMT NNTP-Posting-Host: 209.244.79.161 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025808699 209.244.79.161 (Thu, 04 Jul 2002 11:51:39 PDT) NNTP-Posting-Date: Thu, 04 Jul 2002 11:51:39 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp1.phx1.gblx.net!nntp.gblx.net!nntp.gblx.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111380 "Douglas H. Quebbeman" writes: > Actually, a CDC 6600 with 60 terminals hanging off it was > *terribly* bogged down, giving crappy response time to all > 60 users. I can't imagine trying to hang more _character mode_ > terminals off such a system. > > But the comparisons often made to IBM mainframes with 1000s > of terminals isn't fair; unless I'm mistaken, those 1000s > of terminals were usually block-mode terminals, or ones > which were at least operating in block mode. Not having > to process an interrupt for every daggone character allows > you to hang a lot more of them onto the muxes, and more > muxes on the channels. > > CDC also provided for block-mode terminals and transaction- > oriented subsystems, but I personally never saw those in use. in the early '90s the configuration referenced in the previous post about routes was actually a cluster of SMP mainframes .... but still having avg. peak loading of 3500 transactions per second .... 1500 or so weren't real transactions ... they were device interrupts from the ticket & boarding pass printers around the world; human operated terminals was only seeing about 2000 "transactions" a second ... about 1/4th of the 2000 or around 500/second were requests to find a route between two airports (i.e. flight segments). now, back with cp/67 on a 360/67 single processor (say maybe compareable to a large AT w/80287 co-processor) we supported 75-80 "active" 2741 terminal "mixed-mode" users with subsecond trivial response (of course this had line interrupts not character interrupts) ... mixed-mode ... apl, program development, document preperation, source editing, compilation, program debug & test, etc. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: "Rupert Pigott" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 20:26:39 +0000 (UTC) Organization: BT Openworld Lines: 31 Message-ID: References: <20020704201700.232b459c.steveo@eircom.net> NNTP-Posting-Host: host213-122-202-124.in-addr.btopenworld.com X-Trace: knossos.btinternet.com 1025814399 8723 213.122.202.124 (4 Jul 2002 20:26:39 GMT) X-Complaints-To: news-complaints@lists.btinternet.com NNTP-Posting-Date: Thu, 4 Jul 2002 20:26:39 +0000 (UTC) X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MSMail-Priority: Normal X-Priority: 3 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-xfer2.newshosting.com!btnet-peer0!btnet-feed5!btnet!news.btopenworld.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111431 "Steve O'Hara-Smith" wrote in message news:20020704201700.232b459c.steveo@eircom.net... > On Thu, 4 Jul 2002 16:45:38 +0000 (UTC) > "Rupert Pigott" wrote: > > RP> used... Sigh. I'd much rather they got smaller quieter & cooler > RP> myself. > > They did - they're called PDAs and pocket computers and things > like that. I've even seen one of those things running a multi user OS. LOL, not so very long ago I think I said that PDAs would take over from PCs in the not so distant future. I suspect that hand in hand the PDAs taking over from PCs we'll see less paperwork and more emails/IM... That means less time wasted on Word and more time wasted on web-authoring style work (aka composing HTML messages for USENET :P). To be honest I'd still like a honking big display & keyboard, just with it all hooked up to a little box the size of a cigarette packet (PSU probably about the same size). It'll come, the technology to make this workable has been around a good 3 years or so in the commodity parts bins. The software should change though, a readjustment of priorities towards power saving an economy. Cheers, Rupert ###### From: "Rupert Pigott" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 20:35:28 +0000 (UTC) Organization: BT Openworld Lines: 56 Message-ID: References: <20020704183613.46af1766.steveo@eircom.net> NNTP-Posting-Host: host213-122-202-124.in-addr.btopenworld.com X-Trace: knossos.btinternet.com 1025814928 9721 213.122.202.124 (4 Jul 2002 20:35:28 GMT) X-Complaints-To: news-complaints@lists.btinternet.com NNTP-Posting-Date: Thu, 4 Jul 2002 20:35:28 +0000 (UTC) X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MSMail-Priority: Normal X-Priority: 3 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!uni-erlangen.de!news-nue1.dfn.de!news-lei1.dfn.de!newsfeed00.sul.t-online.de!t-online.de!kibo.news.demon.net!demon!btnet-peer0!btnet-feed5!btnet!news.btopenworld.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111434 "Chris Hedley" wrote in message news:q722ga.mm.ln@teabag.cbhnet... > According to Steve O'Hara-Smith : > > That'd be about 20 modern drives, the 250MB/s is too much > > for 32 bit PCI at 33MHz but easily in range of 64 bit PCI at 66MHz so > > yes I think a modern (high end) PC can do that part. The 25K terminals > > is a bit of a problem. Are concentrator equivalents permissible ? If so > > there seems to be enough PCI bus bandwidth left over (just about). The > > memory bandwidth of a modern PC is much higher so the CPU(s) can still > > get at the data as it flies through the system. It's pretty close to the > > limit though (this week). > > I'm curious to know how PCI performs under load... I get the impression > that its sustainable bandwidth isn't that high a percentage of its > theoretical peak, but that's just my perception. I suppose I just don't IIRC you can get pretty close to peak, within 75% anyways. It depends on the workload of course. > like the current PC architecture much: it reminds me too much of an old > over-maintained program that has kludge layered on top of kludge which > nobody's had the courage to redesign from scratch! At the very least There are plenty of ways of fixing this. You can get the CPU of your choice and hang lots of controllers and stuff off it via PCI. For all that stuff which used to hang off ISA you can whack in a USB controller on your PCI bus and you've got keyboard/mouse/printer/scanner/etc connectivity for bugger all cost. :) You really don't *need* to have that ISA baggage if you don't want to, and you can still run a mainstream OS (after you've tweaked it) such as *BSD or Linux. I can almost see the light at the end of the tunnel as far as ditching the remnants of the PC ISA goes. The only thing left will probably be the partition table. :) [SNIP] > Not too sure how I'd do it if I had a blank sheet. Probably connect > all the internal bits using a crossbar switch and hang all the other > components (everything from mass storage to basic IO) off an array of > SCSI channels. But then it'd probably be classed as a small mainframe. ... Quite a few of the 64bit and 66MHz PCI 2.x slots out there are in fact point to point already ... I believe the successors to PCI are all point to point (so in effect you have an X-bar somewhere). I'd just get an embedded RISC CPU, preferably with PCI & USB on board. It would be nice to slot in an AGP display controller subsystem into that lot, but I'm guessing "it'll cost ya". AFAICT the Kyro II looks like the most power efficient of the lot right now. Cheers, Rupert ###### From: Steve O'Hara-Smith Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 23:17:59 +0200 Organization: Wanadoo NL Lines: 15 Message-ID: <20020704231759.2efd7bdc.steveo@eircom.net> References: <20020704201700.232b459c.steveo@eircom.net> NNTP-Posting-Host: rot2-p1310.dial.wanadoo.nl Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Trace: scavenger.euro.net 1025830868 48513 62.234.203.30 (5 Jul 2002 01:01:08 GMT) X-Complaints-To: abuse@wanadoo.nl NNTP-Posting-Date: Fri, 5 Jul 2002 01:01:08 +0000 (UTC) X-Newsreader: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd4.6) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!feedme.news.mediaways.net!image.surnet.ru!surnet.ru!news.euronet.nl!ams-gw.sohara.org!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111432 On Thu, 4 Jul 2002 20:26:39 +0000 (UTC) "Rupert Pigott" wrote: RP> To be honest I'd still like a honking big display & keyboard, RP> just with it all hooked up to a little box the size of a RP> cigarette packet (PSU probably about the same size). Might as well put it in the back of the honking great display, use a wireless keyboard and lose all that unsightly wire. -- C:>WIN | Directable Mirrors The computer obeys and wins. |A Better Way To Focus The Sun You lose and Bill collects. | licenses available - see: | http://www.sohara.org/ ###### From: Steve O'Hara-Smith Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 23:28:49 +0200 Organization: Wanadoo NL Lines: 15 Message-ID: <20020704232849.149e7158.steveo@eircom.net> References: <20020704183613.46af1766.steveo@eircom.net> NNTP-Posting-Host: rot2-p1310.dial.wanadoo.nl Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Trace: scavenger.euro.net 1025830869 48513 62.234.203.30 (5 Jul 2002 01:01:09 GMT) X-Complaints-To: abuse@wanadoo.nl NNTP-Posting-Date: Fri, 5 Jul 2002 01:01:09 +0000 (UTC) X-Newsreader: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd4.6) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!feedme.news.mediaways.net!news-fra1.dfn.de!gatel-ffm!newsfeed01.sul.t-online.de!t-online.de!newsrouter.chello.at!amsnews01.chello.com!newshub1.nl.home.com!news.nl.home.com!news2.euro.net!news.euro.net!news.euronet.nl!ams-gw.sohara.org!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111401 On Thu, 4 Jul 2002 20:35:28 +0000 (UTC) "Rupert Pigott" wrote: RP> You really don't *need* to have that ISA baggage if you don't want to, Everything on my current PC that probes as being on ISA is actually in one chip on the motherboard, there is no ISA slot and I don't think the ISA in the chip is much more than a figment to apease aging software. Everything real is PCI or faster. -- C:>WIN | Directable Mirrors The computer obeys and wins. |A Better Way To Focus The Sun You lose and Bill collects. | licenses available - see: | http://www.sohara.org/ ###### From: "Rupert Pigott" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Fri, 5 Jul 2002 02:23:03 +0000 (UTC) Organization: BT Openworld Lines: 22 Message-ID: References: <20020704201700.232b459c.steveo@eircom.net> <20020704231759.2efd7bdc.steveo@eircom.net> NNTP-Posting-Host: host213-1-130-161.in-addr.btopenworld.com X-Trace: knossos.btinternet.com 1025835783 9108 213.1.130.161 (5 Jul 2002 02:23:03 GMT) X-Complaints-To: news-complaints@lists.btinternet.com NNTP-Posting-Date: Fri, 5 Jul 2002 02:23:03 +0000 (UTC) X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MSMail-Priority: Normal X-Priority: 3 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!proxad.net!news-hub.cableinet.net!blueyonder!btnet-peer!btnet-peer0!btnet-feed5!btnet!news.btopenworld.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111433 "Steve O'Hara-Smith" wrote in message news:20020704231759.2efd7bdc.steveo@eircom.net... > On Thu, 4 Jul 2002 20:26:39 +0000 (UTC) > "Rupert Pigott" wrote: > > RP> To be honest I'd still like a honking big display & keyboard, > RP> just with it all hooked up to a little box the size of a > RP> cigarette packet (PSU probably about the same size). > > Might as well put it in the back of the honking great display, > use a wireless keyboard and lose all that unsightly wire. I have thought about that. But I'd like to be able to carry it around like a portable but without the extra baggage of battery, screen etc. The idea being I plug it into what's around me at the time, honking or otherwise. Cheers, Rupert ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 22:37:11 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 47 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111276 In article , Brian Inglis wrote: > If you had a local controller connection, response was a lot > faster than remote, and you effectively had type ahead, as screen > write CCWs were chained to zero byte reads. We had a 3270 controller about about 20 terminals, and a line printer. This went across the river to a DOS/VE system, which then was connected to a central 4381 in Richmond, VA. Sometimes it took hours to get results back from program runs. The program might execute in a few seconds or even a fraction of one, but the network was usually fairly slow in getting your output back. I found ways to make it happen faster by playing with priorities: they had left some holes in their security. No malice, I just wanted to get my printouts and go home... :) > We typically had about 200 PROFS users banging away at any time, > plus engineering, commercial, program development, and a lot of > network traffic with other systems doing email, NJE work, and > printing. It's amazing what good I/O hardware can do. Even a PC, if you scrap the legacy stuff and get real I/O coprocessors, can handle a lot more users than you might think. It also helps to avoid overcomplicating your code, especially the UI. The WWW is what basically amounts to a graphical, markup-based virtual 3270. > Possibly a Hitachi/Fujitsu? Solid State Disk (SSD) -- probably > had a Channel-to-Channel Adapter (CTCA) interface -- which meant > it dealt in units of pages at high speed. Could also have been a > regular channel interface, but that would have been limited to > 3MB/s plus channel overhead -- so acted liked a disk with zero > seek/rotation times. Interesting. I assume they emulated whatever drive you wanted them to. Another thing: if I remember correctly, if the machine crashed, the drive we used usually had all our files in place. I assume this was some form of battery backup system. ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 22:59:27 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fcdce_2@news.iglou.com> <3d21ee5c$1@news.ucsc.edu> <3D239559.A1D80F5E@thinkage.ca> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 77 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111277 In article <3D239559.A1D80F5E@thinkage.ca>, Alan T. Bowler wrote: > Not the compilers themselves, but the standard I/O libraries used > by programs written in the those languages certainly did. > As someone else pointed out, all you have to do was specify > (or accept the default ) more then 1 buffer. It was usual to > double buffer, unless you were tight on memory (and that was common). Was this for all languages then? >> Most UNIX systems try to do this sort of thing automatically, although >> there are a lot of things you can do yourself to optimize I/O and >> processing to work together. The UNIX process model is good but I/O is >> almost totally divorced from a process. > > Actually, the Unix model doesn't "do this sort of thing automatically", Actually, yes it does. All three of mine certainly do... :) Read-ahead on I/O requests is a fairly common speedup. Only relatively old systems don't do this to some extent. It's just that this is not really at the process level. It's more geared to system throughput, so process with small I/O demands can still get starved by processes with a lot of I/O in the queues. > the usual model to the process is synchonous I/O i.e. issue > request and block until data is moved to/from your address space. From your point of view as a programmer, yes, unless you do the double buffering yourself. But the OS will still fetch ahead to try and make sure your next request is in memory already. > There is some level of async processing when you aren't using > a raw device and so getting system caching. Right. > Making I/O synchronous at the program level has the advantage that it > makes programming simpler, and comes out of some experience with > mainframes. Namely in a multi-process environment, multiple buffering > often does not increase total system throughput that much (if any). True, but this doesn't address I/O starvation. Both types of I/O under UNIX are fairly easy, and you can do buffering without too much trouble too. I/O is optimized for overall system speed, so if you have 1000 outstanding I/O requests, most of the time your new 3 requests have a long line ahead of them. Ideally you would have a per process limit on requests, but this is obviously going to increase the complexity and optimization of system I/O queuing. There might also be other solutions as well, and if you have the sources, you can make sure applications are more friendly. But ideally, it should not matter: the system should break up I/O _and_ optimize it. > Unix did acquire some async I/O features later, and they are > needed. Especially for networked multiple conversation type > things. But they still feel like warts compared to the basic > ease of the usual sync I/O paradigm. Several systems have this, and it's nice, but doesn't solve the problem of I/O starvation when long I/O queues are present. Also, in some UNIX systems (like 4.3 BSD) this only worked for network or terminal I/O, and in SysV it only supported streams devices. A better method is to use I/O multiplexing, though it also does not solve the resource problem. Of course, most of the time UNIX systems were concerned with handling a lot of networking and terminal I/O, so that's one reason that is what was supported. ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <3d248b17$1_3@news.iglou.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 67 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:8U7RzDlfdNH21f8aMbg4y0ZZxPk= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 2002 10:00:30 GMT NNTP-Posting-Host: 209.244.76.40 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025863230 209.244.76.40 (Fri, 05 Jul 2002 03:00:30 PDT) NNTP-Posting-Date: Fri, 05 Jul 2002 03:00:30 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!news.tele.dk!small.news.tele.dk!208.49.253.98!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111373 Anne & Lynn Wheeler writes: > now, back with cp/67 on a 360/67 single processor (say maybe > compareable to a large AT w/80287 co-processor) we supported 75-80 > "active" 2741 terminal "mixed-mode" users with subsecond trivial > response (of course this had line interrupts not character > interrupts) ... mixed-mode ... apl, program development, document > preperation, source editing, compilation, program debug & test, etc. there was a lot of work to get cp/67 up to that level of performance. when I first got a copy of cp/67 at the university in jan. '68 ... it was doing good supporting 30 mixed-mode users and/or concurrent mixed-mode and a guest operating system like MFT. over the next six months I significantly reduced general pathlengths and introduced "fastpaths" (fastpath is methodology for optimized pathlength for the most common case(s) .... in contrast to just straight-forward optimized pathlength for all cases). ref to report I gave at august '68 SHARE meeting as to the result of some of the pathlength work for guest operating systems: http://www.garlic.com/~lynn/94.html#18 CP/67 and OS MFT14 Of the next 18 months at the university I also rewrote dispatch/schedule and implemted fair share scheduling, rewrote paging subsystem and implemented clock replacement algorithm, rewrote "DASD" support .... implementing ordered-seek queueing for 2311 and 2314 disks and chained scheduling for 2301 drums. The orders seek queueing reduced latency and increase thruput of both 2311 and 2314 disk drives for both paging and user i/o. The 2301 was a fixed head device that doing one i/o operations at a time was subject to average rotational latency delay for every i/o operation. In this mode, the 2301 had peak sustained paging rate of about 80 page i/os per second. With chained-scheduling, multiple requests were ordered in a single i/o ... so rotational latency typically applied to just the first operation .... given a dedicated channel the 2301 would then see peek sustained thruput of 300 page i/os per second. The other feature affecting performance that I did as an undergraduate was to implement support for being able to page selected portions of the kernel ... reducing the total fixed storage requirements ... freeing up more space user execution. Machine at the university was 768k memory ... 192 4k pages ... but that would be reduced to as little as 104 "available" 4k pages under heavy load. Some simple kernel paging could easily pick up 10 percent real storage for application execution. Eventually all of the above changes (except for kernel paging) was picked up and distributed as part of the standard release (they also distributed various other stuff that I had done like the TTY/ascii terminal support). Much of the above was carried over in the CP/67 to VM/370 port ... except for the paging and dispatch/scheduling algorithm changes. I was able to re-introduce those in the resource manager product: http://www.garlic.com/~lynn/2001e.html#45 VM/370 Resource Manager which went out initially as PRPQ (special offering) but was shortly changed into standard product status. misc. general refs: http://www.garlic.com/~lynn/subtopic.html#technology including http://www.garlic.com/~lynn/subtopic.html#fairshare http://www.garlic.com/~lynn/subtopic.html#wsclock -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 87 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:s6jt7GcTpDpyY2RP4rMKvoxFBV4= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 2002 10:32:39 GMT NNTP-Posting-Host: 209.244.76.40 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025865159 209.244.76.40 (Fri, 05 Jul 2002 03:32:39 PDT) NNTP-Posting-Date: Fri, 05 Jul 2002 03:32:39 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wn1feed!worldnet.att.net!208.49.253.98!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111359 Charles Shannon Hendrix writes: > The WWW is what basically amounts to a graphical, markup-based virtual > 3270. one could claim that html .... was cambridge's GML by way of all the ibm systems at CERN (aka GML was done by "G", "M", & "L" at cambridge, before becoming SGML ... and then morphing into stuff like HTML, XML, etc). 3270s did have some number of drawbacks. The 3274 controller (for 3278, 3279, etc terminals ) was slower than the original 3272 controller for 3277 terminals, in part because a lot of "head" electronics was moved out of the terminal and back into the 3274 controller (making the cost of each 3278/3279 terminal cheaper). With some local electronic tinkering it was possible to adjust the key repeat delay and key repeat speed to just about any value you wanted (on the 3277) ... improving the ability to navigate the cursor around the screen. You could also get a keystroke FIFO box for the 3277. The 3270 had a characteristic if you used to fast typing ... that if you happen to hit a key at just the same instant that something was being transferred to the screen ... the keyboard would lock and you would have to het the reset buttom. For fast typists ... that just about threw you into half-duplex mode of operation to avoid having to deal with the interruption of the keyboard lockup and having to hit reset. All of that was lost in upgrade to 3274/3278/3279 (since all the required electronics were now back in the controller and not in the terminal). It never really came back until you got ibm/pc with 3278/3279 terminal emulation. The other impact was that while local 3274s (direct channel attach) exhibited significantly better human factor response than remote 3274s controller (i.e. connected over some sort of 9.6kbit telco line ... with multiple attached terminals all sharing the controller rate) ... the 3274 command processing time on the channel was excessive. A fully configured system tended to have a lot of disk controllers and a lot of 3274 controllers all spread out on the available channels (with some disk controllers and some 3274 controllers attached to each channel). It turned out that 3274 slowness was causing high channel busy time and typically interferring with disk thruput. This wasn't immediately recognized .... however I was doing a project implementing kernel aupport for HYPERChannel Remote Device Adapter (basically channel extender over T1 telco line) for large numbers of local 3274 controllers. You would remove the 3274 controllers from all the channels and put them in a remote building. You would then attach a HYPERChannel A22x to the channel and setup a HYPERChannel network to some number of A51x boxes at the remote site. The A51x boxes emulated mainframe channels and you could attach "local" controllers (like 3274) to the A51x boxes and they would think they were talking to real channel. The result was you could put a couple hundred people and their terminals at a remote site and they typically couldn't tell the difference between being local or remote. This is in contrast to "remote" 3274s where the response and human factor degradation was significant. A side effect of this project was that it appeared that total disk thruput (and therefor total system thruput) went up by about 10-15 percent. Further analysis showed that the HYPERChannel A22x boxes that were directly attached to mainframe channel had significantly lower channel "overhead" for doing the same exact operation compared to configuration with all the local 3274s controllers attached to the real channels. This discovery resulted in some number of presentations and configuration advisories as to not mixing 3274 controllers and the same channels with anything else of significant thruput importance (like disk controllers). slightly related discussion with respect to some disk thruput & controller issues http://www.garlic.com/~lynn/2002i.html#18 AS/400 and MVS - clarification please some past HYPERChannel postings http://www.garlic.com/~lynn/subtopic.html#hsdt Slightly related HYPERChannel issue ... mentioned in some of the above postings. I had also did the standard mainframe product support for RFC1044 (aka hyperchannel) and did some tuning at Cray Research ... where we got sustained thruput between a cray machine and 4341-clone that was nearly equal to 1.5mbyte/sec hardware channel speed ... with only relatively modust cpu utilization. By comparision the "standard" base support (non-rfc1044) had trouble getting 44kbyte/sec while nearly saturating a 3090 cpu. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 27 Date: Fri, 05 Jul 2002 11:06:02 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news1.calgary.shaw.ca 1025867162 24.71.223.147 (Fri, 05 Jul 2002 05:06:02 MDT) NNTP-Posting-Date: Fri, 05 Jul 2002 05:06:02 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news1.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111467 On Fri, 05 Jul 2002 10:32:39 GMT, Anne & Lynn Wheeler wrote: >Slightly related HYPERChannel issue ... mentioned in some of the above >postings. I had also did the standard mainframe product support for >RFC1044 (aka hyperchannel) and did some tuning at Cray Research ... >where we got sustained thruput between a cray machine and 4341-clone >that was nearly equal to 1.5mbyte/sec hardware channel speed ... with >only relatively modust cpu utilization. By comparision the "standard" >base support (non-rfc1044) had trouble getting 44kbyte/sec while >nearly saturating a 3090 cpu. What kind of config was that? AT a PPOE we managed to get 30KB/s sustained TCP/IP throughput over twin 128kbps SNA links (Canada-UK) using TCP over SNA tunnelling with no significant loading, although IIRC both the TCP and SNA setups were patched up to date and tweaked according to SE instructions. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### From: cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Fri, 5 Jul 2002 12:37:25 +0100 Organization: Honest Chris' Sysadmin Emporium Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 1025870407 nnrp-10:12118 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net X-Newsreader: trn 4.0-test76 (Apr 2, 2001) Originator: cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) Lines: 16 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!eusc.inter.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!kibo.news.demon.net!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111490 According to Anne & Lynn Wheeler : > The other impact was that while local 3274s (direct channel attach) > exhibited significantly better human factor response than remote 3274s > controller (i.e. connected over some sort of 9.6kbit telco line > ... with multiple attached terminals all sharing the controller rate) > ... the 3274 command processing time on the channel was excessive. Oh yes. I remember using a 3274 (well, that and a herd of UNIX minis which emulated them and connected via an X.25 card hacked to talk SNA- style SDLC) which was connected to the remote mainframe via a piece of wet string which wobbled at 9600bps. This was over a 3-year period and it was painful, especially after a visit to the site where the mainframes lived and I got to see what the response should *really* be like (much < 1 second as opposed to several seconds) Chris. ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 52 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:LBkzzIh0ouIhOkd8jxSvH9Ad3jY= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 2002 11:49:51 GMT NNTP-Posting-Host: 209.244.76.40 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025869791 209.244.76.40 (Fri, 05 Jul 2002 04:49:51 PDT) NNTP-Posting-Date: Fri, 05 Jul 2002 04:49:51 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wn1feed!worldnet.att.net!208.49.253.98!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111342 Brian Inglis writes: > What kind of config was that? AT a PPOE we managed to get 30KB/s > sustained TCP/IP throughput over twin 128kbps SNA links > (Canada-UK) using TCP over SNA tunnelling with no significant > loading, although IIRC both the TCP and SNA setups were patched > up to date and tweaked according to SE instructions. see RFC1044 ... if nothing else find pointer to the RFC from my rfc index at: http://www.garlic.com/~lynn/rfcietff.htm Basically a channel attached hyperchannel A22x box on ibm mainframe connected to a hyperchannel network. i wrote all the code ... device driver and various optimized paths for the box in the base product. The original mainframe TCP/IP product ... in the'80s initially only supported the 8232 ... basically a pc/at with channel attach card and some number of LAN cards. The 8232 wasn't really a tcp/ip box .... but a channel to LAN gateway ... so all the TCP/IP to LAN/MAC level stuff had to be done in the mainframe (which accounted for a lot of the cpu overhead processing). The channel attached HYPERChannel box was a real TCP/IP router ... which allowed a lot of the processing for the 8232 to be bypassed. This was also the basis for what we used for the mainframe part of our internal corporate highspeed backbone. This internal corporate highspeed backbone is what the NSF audit claimed was five years ahead of all bid submissions for the NSFNET1 backbone. For the internal backbone I had done some additional stuff that didn't appear in the product (like rate-based pacing ... which the audit cliamed was included in the five years ahead .... and 15 years later it still looks to be five years ahead ... internet2 is looking at it tho). random refs: http://www.garlic.com/~lynn/internet.htm misc. 8232 refs: http://www.garlic.com/~lynn/99.html#36 why is there an "@" key? http://www.garlic.com/~lynn/2001.html#4 Sv: First video terminal? misc. stuff about hsdt ... high speed data transport: http://www.garlic.com/~lynn/subtopic.html#hsdt in the following ref: http://www.garlic.com/~lynn/94.html#33b High Speed Data Transport (HSDT) somebody in the SNA group had posted an announcement for a new newsgroup. The contrast was significant. tale slightly out of school: http://www.garlic.com/~lynn/200c.html#58 -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 42 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:qOhwPGt6n342A4C2D5/Z7XOa9RE= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 2002 12:14:36 GMT NNTP-Posting-Host: 209.244.76.40 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025871276 209.244.76.40 (Fri, 05 Jul 2002 05:14:36 PDT) NNTP-Posting-Date: Fri, 05 Jul 2002 05:14:36 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!proxad.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111345 cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) writes: > Oh yes. I remember using a 3274 (well, that and a herd of UNIX minis > which emulated them and connected via an X.25 card hacked to talk SNA- > style SDLC) which was connected to the remote mainframe via a piece > of wet string which wobbled at 9600bps. This was over a 3-year > period and it was painful, especially after a visit to the site where > the mainframes lived and I got to see what the response should *really* > be like (much < 1 second as opposed to several seconds) there were a couple level of "arguments". there were a lot of the TSO crowd claiming that sub-second response wasn't required. Some number of detailed human factor detailed studies then were done that showed that sub-second response was significant. Those studies significantly aided the case for having local 327x controllers at remote location using hyperchannel as a channel extender (as opposed to using sna-based remote 327x controllers). Then there was a east research center that was really proud of the fact that the had .22 second response under heavy load and provided one of the best time-sharing services in the world. We then pointed out that we had a .11 second response with effectively the same load and the same hardware. There was then some discussion whether less than .20 second response really had any meaning (i.e. could humans tell the difference between .11 second response and .20 second response). the sna group were less than thrilled about various of these activities (also see the recent posting mentioning tcp/ip and vtam/sna). they were really not thrilled that i was part of the four person group that created the first non-ibm controller and started the pcm controller business. another slightly related sna/hyperchannel comparison: http://www.garlic.com/~lynn/99.html#67 System/1 ? http://www.garlic.com/~lynn/2000.html#53 APPC vs TCP/IP http://www.garlic.com/~lynn/2001.html#4 Sv: First video terminal? http://www.garlic.com/~lynn/2001i.html#21 3745 and SNI http://www.garlic.com/~lynn/2001i.html#31 3745 and SNI http://www.garlic.com/~lynn/2001k.html#21 OT: almost lost LBJ tapes; Dictabelt http://www.garlic.com/~lynn/2002c.html#42 Beginning of the end for SNA? -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Fri, 5 Jul 2002 14:51:47 +0100 Organization: Honest Chris' Sysadmin Emporium Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 1025877602 nnrp-10:15685 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net X-Newsreader: trn 4.0-test76 (Apr 2, 2001) Originator: cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) Lines: 44 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!diablo.netcom.net.uk!netcom.net.uk!kibo.news.demon.net!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111495 According to Anne & Lynn Wheeler : > there were a couple level of "arguments". there were a lot of the TSO > crowd claiming that sub-second response wasn't required. Some number > of detailed human factor detailed studies then were done that showed > that sub-second response was significant. Those studies significantly > aided the case for having local 327x controllers at remote location > using hyperchannel as a channel extender (as opposed to using > sna-based remote 327x controllers). Interesting that even on our somewhat rickety setup, TSO (and TSO alone) still managed sub-second responses where a significant screen update wasn't involved. Anything which used CICS or COMIS was dreadful to use, though. Never got much of a chance to find out how the VM response was, though, in spite of trying to beg, steal or borrow a CMS or PROFS ID for the best part of a year (I eventually gave up; unfortunately account alloc and charging was a bit anal, so I couldn't get one even though I was part of the MIS group!) > Then there was a east research center that was really proud of the > fact that the had .22 second response under heavy load and provided > one of the best time-sharing services in the world. We then pointed > out that we had a .11 second response with effectively the same load > and the same hardware. There was then some discussion whether less > than .20 second response really had any meaning (i.e. could humans > tell the difference between .11 second response and .20 second > response). I'm sure I heard somewhere (may have even been from yourself?) that it's difficult to notice the difference in response times below 100ms, but any increments above that are quite obvious. My current pet hate is GUI and particularly browser response times, the likes of which would have seen heads roll in the mainframe and mini world... > the sna group were less than thrilled about various of these > activities (also see the recent posting mentioning tcp/ip and > vtam/sna). they were really not thrilled that i was part of the four > person group that created the first non-ibm controller and started the > pcm controller business. another slightly related sna/hyperchannel Good old NIH syndrome strikes again! Chris. ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 38 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:rhUFsvnow5KRo2kGSlEgv627cDw= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 2002 16:05:30 GMT NNTP-Posting-Host: 64.156.35.174 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1025885130 64.156.35.174 (Fri, 05 Jul 2002 09:05:30 PDT) NNTP-Posting-Date: Fri, 05 Jul 2002 09:05:30 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!feedme.news.mediaways.net!news-fra1.dfn.de!fu-berlin.de!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111377 cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) writes: > I'm sure I heard somewhere (may have even been from yourself?) that > it's difficult to notice the difference in response times below 100ms, > but any increments above that are quite obvious. shortly after getting hired .... i got to go to an internal technical conference held in DC at old marriott. mills gave a talk on superprogrammer. somebody from ykt gave a talk on event time perception threshold, across a wide range of people it seemd to vary between 100ms and 200ms. prior ref: http://www.garlic.com/~lynn/2001h.html#48 Whom Do Programmers Admire Now??? a later study in the early '80s found that predictability was also important .... if people could perceive the delay, they could get into a pattern anticipating it ... if the delay was longer than anticipated, it interrupted the (human) pattern and it then took the person twice as long to "recover". If the delay was about two seconds, and it went to four seconds .... a human took an additional two seconds (six seconds total) to recover. The theory was that the person's attention started to wonder when their expectation failed and the attention "recovery" time was equal to the amount of time spent "wondering". A supporting example was if the delay extended into the minutes, you might even leave your desk to do something else. other random refs: http://www.garlic.com/~lynn/93.html#31 Big I/O or Kicking the Mainframe out the Door http://www.garlic.com/~lynn/2000b.html#20 How many Megaflops and when? http://www.garlic.com/~lynn/2000b.html#24 How many Megaflops and when? http://www.garlic.com/~lynn/2000b.html#25 How many Megaflops and when? http://www.garlic.com/~lynn/2000c.html#64 Does the word "mainframe" still have a meaning? http://www.garlic.com/~lynn/2000d.html#40 360 CPU meters (was Re: Early IBM-PC sales proj.. http://www.garlic.com/~lynn/2001i.html#30 IBM OS Timeline? http://www.garlic.com/~lynn/2002e.html#8 What are some impressive page rates? -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 19 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:EuKXj0Wpa+LBJNZLoN39iWZQoFg= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 2002 16:22:15 GMT NNTP-Posting-Host: 64.156.35.174 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025886135 64.156.35.174 (Fri, 05 Jul 2002 09:22:15 PDT) NNTP-Posting-Date: Fri, 05 Jul 2002 09:22:15 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!dca6-feed2.news.algx.net!jfk3-feed1.news.algx.net!dfw3-feed1.news.algx.net!allegiance!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111366 cbh@ieya.co.REMOVE_THIS.uk (Chris Hedley) writes: > Interesting that even on our somewhat rickety setup, TSO (and TSO alone) > still managed sub-second responses where a significant screen update > wasn't involved. Anything which used CICS or COMIS was dreadful to use, > though. Never got much of a chance to find out how the VM response was, > though, in spite of trying to beg, steal or borrow a CMS or PROFS ID for > the best part of a year (I eventually gave up; unfortunately account > alloc and charging was a bit anal, so I couldn't get one even though I > was part of the MIS group!) cms subsecond .2 or less was measured even with complete screen refresh ... although that got harder to do when 3274s were first introduced since worst case 3274 controller latency could be upwards of half second. It was unusual for TSO avg. response to be subsecond. past discussion on 3274 latencies: http://www.garlic.com/~lynn/2001m.html#19 -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: stanb@dial.pipex.com (Stan Barr) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <20020704201700.232b459c.steveo@eircom.net> Organization: Metropolis Grafix Reply-To: stanb@dial.pipex.com Message-ID: X-Newsreader: slrn (0.9.5.2 UNIX) Date: 05 Jul 2002 17:31:29 GMT Lines: 15 NNTP-Posting-Host: 62-190-201-38.pdu.pipex.net X-Trace: 1025890289 news.dial.pipex.com 8509 62.190.201.38 X-Complaints-To: abuse@uk.uu.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!newsfeed01.sul.t-online.de!t-online.de!lnewspeer00.lnd.ops.eu.uu.net!lnewspost00.lnd.ops.eu.uu.net!emea.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111480 On Thu, 4 Jul 2002 20:26:39 +0000 (UTC), Rupert Pigott wrote: > >To be honest I'd still like a honking big display & keyboard, >just with it all hooked up to a little box the size of a >cigarette packet (PSU probably about the same size). Apple, of course, are putting everything in the display... iMac & eMac... -- Cheers, Stan Barr stanb@dial.pipex.com The future was never like this! ###### From: "Eric S. Harris" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Fri, 05 Jul 2002 12:55:02 -0500 Organization: MindSpring Enterprises Lines: 72 Distribution: inet Message-ID: <3D25DD76.4F5EFE4B@mindspring.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> Reply-To: eric_harris_76@yahoo.com NNTP-Posting-Host: a8.bf.6d.8e Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Server-Date: 5 Jul 2002 17:53:31 GMT X-Mailer: Mozilla 4.79 [en]C-CCK-MCD (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!proxad.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!newsfeed0.news.atl.earthlink.net!news.atl.earthlink.net!news.mindspring.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111328 A reply regarding Flight Sim and characters from a former coworker, below. Some corrections to what I wrote. Expected; it's why I bcc-ed him. He worked there for years; I was there for just 12 months. "Eric S. Harris" wrote: > > Walter Spector wrote: [snippage] > When the flight simulators at McAir (McDonnell Aircraft Company) were migrating > from a pair of CDC something-or-anothers (750s?) to a bunch of networked Gould > SELs, back in 1985, they had a quandary: what to do about the home-brew hardware > controllers? > > The SELs were 16-bit machines, as were the specialized I/O widgie-frammises that > drove the simulators. The Cyber would do some bit-fiddling to line the 16-bit > data up in 12-bit words (with padding, IIRC) and all was OK, though a little > strange. A reasonable sort of strange, under the circumstances. > > The obvious -- to me -- approach would have been to modify the electronics to > accept 16-bit data from the SELs, but that's not what was done. (Because > hardware is hard, software is easy?) Instead, the SELs would input and output > the 16-bit data 12 bits at a time, just like the hardware expected, with padding > and aligning done in software in the SEL. > > Guess who wrote the memory-to-memory translation code? In Fortran, with shift > and mask operations. > > It was horrible, yet satisfying. The code had to handle any start address, and > any length of data. That meant the padding had to be accounted for differently, > depending on where in in the widgie-frammis's memory the data would reside. I > had a false start or two before I had code that would work correctly, was fairly > clear and reasonably-sized. Much collapsing of special cases. It had a certain > elegance, considering. > > I wish I'd kept a copy. For one thing, I don't recall just what the > requirements were, much less exactly how I resolved them. Though I suppose I > could "reverse engineer" it from my recollections. (Will I? Sure. Right after > my javelin-catching class.) > > For another, it's something I'm inordinately proud of, and would like to drag > out and look over, now and again. My precioussss. Corrections and comments from Steve. -Eric S. Not all of the Flight Hardware was built to take 12 bit data, but all of the "Channel" interfaces were. All of the flight graphics equipment for instance was built for 16 bit so they were front ended with 12/16 bit converters. As far as performing the 12 to 16 bit packing/unpacking in the CPU... You had it easy! I had to do this in the PP! (1- 18 bit register, 12 bit words) But you are correct! it was fun, exciting and still find myself today pulling/using information that I learned when I was deep inside NOS and the H/W. As to tape, don't forget the CDC Internal tape format. (LABEL,TAPE,F=I,PO=R,M=GE) The controller took advantage of the fact that the OS used 6 bit characters and we were using 9 track tapes. So unlike IBM and the rest of the world that mostly wasted one 1 per frame (the 8th bit is almost always zero in ASCII), we put our 6 bit character + the next 2 bits of the next character in a single frame, giving us a net of 3 bits saved per frame over the competitors... -Steve "Shift to the Left, Shift to the Right, Pop-up, Push-down, Byte! Byte! Byte! ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 41 Date: Fri, 05 Jul 2002 19:56:17 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news2.calgary.shaw.ca 1025898977 24.71.223.147 (Fri, 05 Jul 2002 13:56:17 MDT) NNTP-Posting-Date: Fri, 05 Jul 2002 13:56:17 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-han1.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone.bc.net!newsfeed.telusplanet.net!peer1-sjc1.usenetserver.com.MISMATCH!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news2.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111510 On Fri, 05 Jul 2002 12:14:36 GMT, Anne & Lynn Wheeler wrote: >Then there was a east research center that was really proud of the >fact that the had .22 second response under heavy load and provided >one of the best time-sharing services in the world. We then pointed >out that we had a .11 second response with effectively the same load >and the same hardware. There was then some discussion whether less >than .20 second response really had any meaning (i.e. could humans >tell the difference between .11 second response and .20 second >response). Yes, if you had used Xedit under CMS under VM then asked to do some work using SPF Edit under TSO under MVS. You've got to have experienced it on local attached 3270s to notice the difference: the MVS programmers on remote SNA 3270s never knew what they were missing. My VM HPO systems always gave less than .1 second trivial interactive response -- even if it meant spending a weekend moving hot and cold spots around between packs. Page down does not seem quite as fast on a local Windows PC editor app as those systems. OTOH most Unix xterm and editor windows seem to be fast enough, even running with a Windows X server. >the sna group were less than thrilled about various of these >activities (also see the recent posting mentioning tcp/ip and >vtam/sna). VM/VTAM seemed to be snappier than the MVS original when compared outside heavy use hours. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 41 Date: Fri, 05 Jul 2002 19:56:17 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news2.calgary.shaw.ca 1025898977 24.71.223.147 (Fri, 05 Jul 2002 13:56:17 MDT) NNTP-Posting-Date: Fri, 05 Jul 2002 13:56:17 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-han1.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone.bc.net!newsfeed.telusplanet.net!peer1-sjc1.usenetserver.com.MISMATCH!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news2.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111510 On Fri, 05 Jul 2002 12:14:36 GMT, Anne & Lynn Wheeler wrote: >Then there was a east research center that was really proud of the >fact that the had .22 second response under heavy load and provided >one of the best time-sharing services in the world. We then pointed >out that we had a .11 second response with effectively the same load >and the same hardware. There was then some discussion whether less >than .20 second response really had any meaning (i.e. could humans >tell the difference between .11 second response and .20 second >response). Yes, if you had used Xedit under CMS under VM then asked to do some work using SPF Edit under TSO under MVS. You've got to have experienced it on local attached 3270s to notice the difference: the MVS programmers on remote SNA 3270s never knew what they were missing. My VM HPO systems always gave less than .1 second trivial interactive response -- even if it meant spending a weekend moving hot and cold spots around between packs. Page down does not seem quite as fast on a local Windows PC editor app as those systems. OTOH most Unix xterm and editor windows seem to be fast enough, even running with a Windows X server. >the sna group were less than thrilled about various of these >activities (also see the recent posting mentioning tcp/ip and >vtam/sna). VM/VTAM seemed to be snappier than the MVS original when compared outside heavy use hours. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### X-Trace-PostClient-IP: 24.67.16.79 From: Brian Inglis Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D25DD76.4F5EFE4B@mindspring.com> X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 32 Date: Fri, 05 Jul 2002 20:06:18 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: news3.calgary.shaw.ca 1025899578 24.71.223.147 (Fri, 05 Jul 2002 14:06:18 MDT) NNTP-Posting-Date: Fri, 05 Jul 2002 14:06:18 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!newsfeed.telusplanet.net!peer1-sjc1.usenetserver.com.MISMATCH!ps01-sjc1!news.webusenet.com!pd2nf1so.cg.shawcable.net!residential.shaw.ca!news3.calgary.shaw.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111502 On Fri, 05 Jul 2002 12:55:02 -0500, "Eric S. Harris" wrote: >A reply regarding Flight Sim and characters from a former coworker, below. > >Some corrections to what I wrote. Expected; it's why I bcc-ed him. He worked >there for years; I was there for just 12 months. >Corrections and comments from Steve. -Eric S. > As to tape, don't forget the CDC Internal tape format. > (LABEL,TAPE,F=I,PO=R,M=GE) > > The controller took advantage of the fact that the OS used 6 bit characters >and we were using 9 track tapes. So unlike IBM and the rest of >the world that mostly wasted one 1 per frame (the 8th bit is almost always zero >in ASCII), we put our 6 bit character + the next 2 bits of the >next character in a single frame, giving us a net of 3 bits saved per frame over >the competitors... Hoy! IBM 3x0 didn't use ASCII and needed the full eight bits -- letters and digits (0xF0-0xF9) are at the top end of the EBCDIC character set. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply tosspam@aol.com abuse@aol.com abuse@yahoo.com abuse@hotmail.com abuse@msn.com abuse@sprint.com abuse@earthlink.com abuse@cadvision.com abuse@ibsystems.com uce@ftc.gov spam traps ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 33 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:OvSfHTaBTNA1OkU2Q5f9d7S7c84= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 2002 21:42:23 GMT NNTP-Posting-Host: 64.156.39.83 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025905343 64.156.39.83 (Fri, 05 Jul 2002 14:42:23 PDT) NNTP-Posting-Date: Fri, 05 Jul 2002 14:42:23 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!cyclone.socal.rr.com!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111507 Brian Inglis writes: > My VM HPO systems always gave less than .1 second trivial > interactive response -- even if it meant spending a weekend > moving hot and cold spots around between packs. > Page down does not seem quite as fast on a local Windows PC > editor app as those systems. > OTOH most Unix xterm and editor windows seem to be fast enough, > even running with a Windows X server. exchange from last on slightly related topic http://www.garlic.com/~lynn/2001f.html#57 one of the "new" releases saw a 10-15 percent degradation compared to a previous release .... which raised a lot of attention as to performance in general. One of my things that had been laying around had to do with the internal processing of full-screen i/o transactions. If you relied on the standard product internal response numbers it was based on elapsed time between certain physical events ... actually 3 different response events .... with the avg. time for each one going into the calculation. In the above reference, I had collapsed the three seperate events into a single event .... resulting in higher aggregate thruput because of better optimization processing ... but the internal avg. response calculation might look worse (since I had a single event that was three times longer than previous three single events each one-third as long). Until my changes were merged into the standard product ... a system with my changes in this area with .1 second resposne was actually three times better than an unmodified system that produced a calculation of .1 second response. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: Alan Clifford Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Fri, 5 Jul 2002 21:44:16 +0000 (UTC) Organization: Lots of data but no information. Lines: 22 Message-ID: References: <20020704201700.232b459c.steveo@eircom.net> <20020704231759.2efd7bdc.steveo@eircom.net> NNTP-Posting-Host: host213-122-58-134.in-addr.btopenworld.com Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: venus.btinternet.com 1025905456 7799 213.122.58.134 (5 Jul 2002 21:44:16 GMT) X-Complaints-To: news-complaints@lists.btinternet.com NNTP-Posting-Date: Fri, 5 Jul 2002 21:44:16 +0000 (UTC) In-Reply-To: X-X-Sender: alan@mundungus.clifford.ac X-Message-Flag: Linux Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!feedme.news.mediaways.net!news-fra1.dfn.de!eusc.inter.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!kibo.news.demon.net!demon!btnet-peer0!btnet-feed5!btnet!news.btopenworld.com!mundungus.clifford.ac!alan Xref: chonsp.franklin.ch alt.folklore.computers:111505 On Fri, 5 Jul 2002, Rupert Pigott wrote: RP> > RP> > Might as well put it in the back of the honking great display, RP> > use a wireless keyboard and lose all that unsightly wire. RP> RP> I have thought about that. But I'd like to be able to carry it RP> around like a portable but without the extra baggage of battery, RP> screen etc. The idea being I plug it into what's around me at RP> the time, honking or otherwise. RP> You need a virtual keyboard. One was demonstrated on "Tomorrows World". An infrared device that displayed a keyboard on any flat surface and detected which "key" you pressed. Alan ( If replying by mail, please change scatology to AC2 and expect a password autoresponder. Mail to scatology is not read. ) ###### From: "Charlie Gibbs" Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: 05 Jul 02 13:06:54 -0800 Organization: http://extra.newsguy.com Lines: 25 Message-ID: <1220.951T2744T7866341@sky.bus.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> NNTP-Posting-Host: p-418.newsdawg.com X-Newsreader: THOR 2.5a (Amiga;TCP/IP) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews4 Xref: chonsp.franklin.ch alt.folklore.computers:111513 In article lynn@garlic.com (Anne & Lynn Wheeler) writes: >a later study in the early '80s found that predictability was also >important .... if people could perceive the delay, they could get >into a pattern anticipating it ... if the delay was longer than >anticipated, it interrupted the (human) pattern and it then took the >person twice as long to "recover". If the delay was about two seconds, >and it went to four seconds .... a human took an additional two >seconds (six seconds total) to recover. The theory was that the >person's attention started to wonder when their expectation failed and >the attention "recovery" time was equal to the amount of time spent >"wondering". A supporting example was if the delay extended into the >minutes, you might even leave your desk to do something else. I once heard a rumour that a transaction monitor on Sperry->Unisys 1100 systems would shoot for consistency by delaying responses if they were ready too fast. -- cgibbs@sky.bus.com (Charlie Gibbs) Remove the first period after the "at" sign to reply. I don't read top-posted messages. If you want me to see your reply, appropriately trim the quoted text and put your reply below it. ###### From: aek@spies.com (Al Kossow) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Fri, 05 Jul 2002 16:56:36 -0700 Organization: Apple Computer, Inc. Lines: 8 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <1220.951T2744T7866341@sky.bus.com> NNTP-Posting-Host: il0502a-dhcp53.apple.com X-Trace: news.apple.com 1025913394 9440 17.205.24.53 (5 Jul 2002 23:56:34 GMT) X-Complaints-To: abuse@news.apple.com NNTP-Posting-Date: Fri, 5 Jul 2002 23:56:34 +0000 (UTC) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-lei1.dfn.de!newsfeed00.sul.t-online.de!t-online.de!newsfeed.wirehub.nl!news.tiscali.nl!news-out.nuthinbutnews.com!propagator-sterling!news-in.nuthinbutnews.com!news.kjsl.com!news.spies.com!forum.apple.com!news.apple.com!il0502a-dhcp53.apple.com!user Xref: chonsp.franklin.ch alt.folklore.computers:111517 > I once heard a rumour that a transaction monitor on Sperry->Unisys > 1100 systems would shoot for consistency by delaying responses if > they were ready too fast. > WAIT LAST INPUT IGNORED Demand under Exec-8 was a PIA ###### From: "Ken Hunter" Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Thu, 4 Jul 2002 12:12:37 -0600 Organization: CompuServe Interactive Services Lines: 32 Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3d23020b_1@news.iglou.com> NNTP-Posting-Host: lax9-tgn-lvt-vty4.as.wcom.net X-Trace: nntp-m01.news.aol.com 1025806365 12681 216.193.23.4 (4 Jul 2002 18:12:45 GMT) X-Complaints-To: newsmaster@compuserve.com NNTP-Posting-Date: Thu, 4 Jul 2002 18:12:45 +0000 (UTC) X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!netnews.com!xfer02.netnews.com!feed1.newsreader.com!ngpeer.news.aol.com!news.compuserve.com!news-master.compuserve.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111660 "Douglas H. Quebbeman" wrote in message news:3d23020b_1@news.iglou.com... NAM and IAF (with NOS 2) support a transparent mode, which delivers the raw ASCII characters input from a terminal. I believe the input characters are provided to the job (and output characters expected from the job) are in an 8-in-12 format, not 7.5-in-60. The control codes recognized by IAF are all of the form 00xxB, so 4000B is not one of them. I don't know if this is related to the TELEX 8-in-12 format. Full Screen Editor uses the IAF/NAM transparent mode, and I remember seeing a problem where following certain aborts (notably Operator Drop), FSE will drop out leaving IAF in transparent mode. IAF will cheerfully enter 8-in-12 ASCII characters into the command input buffer. Which 1AJ will declare to be an invalid command, no matter what you type, including "LOGOUT." The only way I found to get control of the job back was to force it into central memory, then manually enter "FSE." in Display Code into the command buffer from the console. Then going back to the terminal and exiting FSE normally would shift IAF back to Display Code (or possibly ASCII 6/12) mode, allowing normal entry of commands. If CDSI ever fixed this, I don't think it was till the final release of NOS 2. Ken Hunter ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Sat, 6 Jul 2002 12:45:16 -0400 Organization: Too Much Message-ID: References: <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news-out.nuthinbutnews.com!propagator-sterling!news-in.nuthinbutnews.com!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111535 In article , Anne & Lynn Wheeler wrote: > backbone I had done some additional stuff that didn't appear in the > product (like rate-based pacing ... which the audit cliamed was > included in the five years ahead .... and 15 years later it still > looks to be five years ahead ... internet2 is looking at it tho). What is rate-based pacing? This sounds a lot like rate-limiting, where either the application or the TCP/IP stack limits throughput. ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Sat, 6 Jul 2002 13:44:52 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 29 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!newsfeed.it.ip-plus.net!news.it.ip-plus.net!news.it.colt.net!itgate.net!nntp1.phx1.gblx.net!nntp.gblx.net!nntp.gblx.net!newsfeed.news2me.com!sn-xit-05!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111534 In article , Anne & Lynn Wheeler wrote: > Then there was a east research center that was really proud of the > fact that the had .22 second response under heavy load and provided > one of the best time-sharing services in the world. We then pointed > out that we had a .11 second response with effectively the same load > and the same hardware. There was then some discussion whether less > than .20 second response really had any meaning (i.e. could humans > tell the difference between .11 second response and .20 second > response). I have to laugh when people suggest it's not important. Of course it is, if only because you are probably lowering CPU usage if you cut the response time down (assuming you aren't burning CPU solely to get better response time). > the sna group were less than thrilled about various of these > activities (also see the recent posting mentioning tcp/ip and > vtam/sna). they were really not thrilled that i was part of the four > person group that created the first non-ibm controller and started the > pcm controller business. another slightly related sna/hyperchannel SNA seems more like an umbrella of support from IBM to make it all work than a communications protocol. When I last worked with it, every "SNA" connection we had as a different medium and protocol. ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <3D1F56EE.DAA8337E@mindspring.com> <65OT8.211813$_j6.10841149@bin3.nnrp.aus1.giganews.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 68 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:FssIrBRzMDBplxIyLJBFpemkP1w= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 06 Jul 2002 19:10:26 GMT NNTP-Posting-Host: 64.156.32.38 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025982626 64.156.32.38 (Sat, 06 Jul 2002 12:10:26 PDT) NNTP-Posting-Date: Sat, 06 Jul 2002 12:10:26 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!newsfeed01.sul.t-online.de!t-online.de!newspeer1-gui.server.ntli.net!ntli.net!cox.net!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111632 Charles Shannon Hendrix writes: > What is rate-based pacing? > > This sounds a lot like rate-limiting, where either the application or > the TCP/IP stack limits throughput. rate-based pacing can be used for rate-limiting. one of the reasons for rate-based pacing is that window-based pacing is non-stable as is slow-start. in fact window-based pacing (in part what slow-start uses as a base implementation) is a misuse of a paradigm. window-based pacing was originally developed for direct end-to-end connections and overruning buffers at the end-point. Window-based pacing is possibly 2-3 levels of indirection from having anything to do with network congestion. I've conjectured that possibly part of the reason for trying to leverage window-based pacing to addressing congestion issues is that many of the machines & systems used in the '80s had eather very poor hardware and/or software support for time-based facitlities. rate-based pacing was one of the features for HSP (high speed protocol), standards activity in the late '80s. However, part of the problem was that HSP was being worked on in ANSI (x3s3.3) targeted for ISO ... and ISO had this strong OSI bent .... aka if it violated OSI model it wouldn't get very far. Much of HSP had to do with taking the level 4 (transport) interface directly to the LAN interface ... approximately the middle of level 3 (network). ISO sort of tried to squint real hard and ignore that LANs had collapsed into a single layer ... the bottom half of layer 3, and all of layer 2 and 1 (aka not only did LANs collapse several layers into one ... but the LAN boundary didn't correspond to any defined OSI boundry ... being in the middle of layer 3). HSP was to cut directly from layer 4 interface directly to LAN interface. Again the ISO forces had to squint a lot when objecting to HSP .... it bypassed the level 3 boundary interface and talked directly to the LAN boundary interface ... but at least it had a boundary interface that corresponded to defined OSI layer (level 4) ... which LANs didn't. random rate-based pacing, slow-start, congestion, and hsp postings. http://www.garlic.com/~lynn/94.html#22 CP spooling & programming technology http://www.garlic.com/~lynn/99.html#0 Early tcp development? http://www.garlic.com/~lynn/99.html#114 What is the use of OSI Reference Model? http://www.garlic.com/~lynn/99.html#115 What is the use of OSI Reference Model? http://www.garlic.com/~lynn/2000b.html#1 "Mainframe" Usage http://www.garlic.com/~lynn/2000b.html#5 "Mainframe" Usage http://www.garlic.com/~lynn/2000b.html#9 "Mainframe" Usage http://www.garlic.com/~lynn/2000b.html#11 "Mainframe" Usage http://www.garlic.com/~lynn/2000b.html#59 7 layers to a program http://www.garlic.com/~lynn/2000e.html#19 Is Al Gore The Father of the Internet?^ http://www.garlic.com/~lynn/2000f.html#38 Ethernet efficiency (was Re: Ms employees begging for food) http://www.garlic.com/~lynn/2001b.html#57 I am fed up! http://www.garlic.com/~lynn/2001e.html#24 Pre ARPAnet email? http://www.garlic.com/~lynn/2001e.html#25 Pre ARPAnet email? http://www.garlic.com/~lynn/2001h.html#44 Wired News :The Grid: The Next-Gen Internet? http://www.garlic.com/~lynn/2001k.html#62 SMP idea for the future http://www.garlic.com/~lynn/2001m.html#15 departmental servers http://www.garlic.com/~lynn/2001n.html#15 Replace SNA communication to host with something else http://www.garlic.com/~lynn/2001n.html#27 Unpacking my 15-year old office boxes generates memory refreshes http://www.garlic.com/~lynn/2002.html#38 Buffer overflow http://www.garlic.com/~lynn/2002b.html#4 Microcode? (& index searching) http://www.garlic.com/~lynn/2002c.html#54 Swapper was Re: History of Login Names http://www.garlic.com/~lynn/2002g.html#19 Why did OSI fail compared with TCP-IP? http://www.garlic.com/~lynn/2002g.html#26 Why did OSI fail compared with TCP-IP? http://www.garlic.com/~lynn/2002g.html#46 Why did OSI fail compared with TCP-IP? http://www.garlic.com/~lynn/2002g.html#49 Why did OSI fail compared with TCP-IP? http://www.garlic.com/~lynn/2002g.html#50 Why did OSI fail compared with TCP-IP? -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 24 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:K9/pz3wf1fWB6JenCkHwmJxc6+8= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 06 Jul 2002 19:21:01 GMT NNTP-Posting-Host: 64.156.32.38 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1025983261 64.156.32.38 (Sat, 06 Jul 2002 12:21:01 PDT) NNTP-Posting-Date: Sat, 06 Jul 2002 12:21:01 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!newsfeed01.sul.t-online.de!t-online.de!fr.clara.net!heighliner.fr.clara.net!proxad.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111622 Charles Shannon Hendrix writes: > SNA seems more like an umbrella of support from IBM to make it all work > than a communications protocol. > > When I last worked with it, every "SNA" connection we had as a different > medium and protocol. well SNA was primarily an infrastructure for managing communication with large terminal populations ... mainframe application talking to upwards of 60,000 to 100,000 terminals (maybe more). in that sense it included a communication protocol (but not a networking protocol). Another example might be the data processing infrastructure that talks to the head-end of majority of the cable tv systems using LU0 for addressing every set-top box. sna had a mainframe api (like lu6.2), vtam (sscp, or pu5) and 37xx (ncp, or pu4). there was some conjecture that driving factor in the pu5/pu4 evoluation/definotion was largely because of a project that i worked on as undergraduate that created PCM controller business. misc. refs: http://www.garlic.com/~lynn/subtopic.html#360pcm -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Message-ID: <3D27D4EC.5EBAD9BA@earthlink.net> From: Walter Spector Organization: Not much. X-Mailer: Mozilla 4.79 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 25 Date: Sun, 07 Jul 2002 05:43:39 GMT NNTP-Posting-Host: 158.252.220.196 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1026020619 158.252.220.196 (Sat, 06 Jul 2002 22:43:39 PDT) NNTP-Posting-Date: Sat, 06 Jul 2002 22:43:39 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!news.imp.ch!nntp-out.monmouth.com!newspeer.monmouth.com!news-out.visi.com!hermes.visi.com!newsfeed1.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111580 "Eric S. Harris" wrote: > > Walter Spector wrote: > > ... > > If some retrocomputing person ever bothers to write a C compiler for > > the CDC instruction set, a C 'char' would almost certainly be 12 bits. > > It may just be a mis-memory, but I seem to recall something about a C compiler > for CDC machines, written back then. (From a university in Texas?) Yes, there was one from UT. We had a copy of it. I held the tape in my hands. However our fellow in charge of installing it could never get it to install. Apparently we were not alone, because at the last VIM conference I attended (spring 1984), there was a session devoted to it. A sizable lynch mob showed up, but the speaker did not. It then turned into a bull session on how to get CDC to wake up to modern programming languages. I'd really enjoy seeing a copy of the UT C compiler to see some of the design decisions that were made. But fear it has gone to the Great Bit Bucket in the Sky. (Someone please prove me wrong...) Walt -...- Walt Spector (w6ws at earthlink dot net) ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Sun, 07 Jul 2002 17:59:24 -0500 Organization: Airnews.net! at Internet America Lines: 27 Message-ID: X-Orig-Message-ID: <3D28C7CC.4DBF266C@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library1-aux.airnews.net NNTP-Posting-Time: Sun Jul 7 18:06:51 2002 NNTP-Posting-Host: !cNK41k-WsrmtI9/i#$Z (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!news.airnews.net!cabal10.airnews.net!cabal1.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111555 Walter Spector wrote: > "Eric S. Harris" wrote: > > Walter Spector wrote: > > > ... > > > If some retrocomputing person ever bothers to write a C compiler for > > > the CDC instruction set, a C 'char' would almost certainly be 12 bits. > > It may just be a mis-memory, but I seem to recall something about a C > > compiler for CDC machines, written back then. > > (From a university in Texas?) > > Yes, there was one from UT. We had a copy of it. I held the tape in my > hands. However our fellow in charge of installing it could never get it to > install. Apparently we were not alone, because at the last VIM conference > I attended (spring 1984), there was a session devoted to it. A sizable > lynch mob showed up, but the speaker did not. It then turned into a bull > session on how to get CDC to wake up to modern programming languages. > I'd really enjoy seeing a copy of the UT C compiler to see some of the > design decisions that were made. But fear it has gone to the Great Bit > Bucket in the Sky. (Someone please prove me wrong...) > I remember more of a Pascal movement at UT back then. Don't know what was home-grown there or what was imported. The UT Computation Center would just make languages publically available without much comment as to their origins, IIRC. - LarryW ###### Message-ID: <3D292AF7.1040402@beagle-ears.com> Date: Sun, 07 Jul 2002 23:02:31 -0700 From: Lars Poulsen User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:0.9.4.1) Gecko/20020314 Netscape6/6.2.2 X-Accept-Language: da,en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Lines: 26 NNTP-Posting-Host: 207.154.106.6 X-Trace: azure.impulse.net 1026108151 187 207.154.106.6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!opentransit.net!kinglear.mobilixnet.dk!sienna.impulse.net!azure.impulse.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111566 Rupert Pigott wrote: > It's horrible I know. But PCs really are wickedly fast these > days, and they are *still* gettin faster. They're also starting > to use Mainframe cooling technoligies, and some of the PSUs we > see now are getting into that 600W+ range that VAX-11/780's > used... Sigh. I'd much rather they got smaller quieter & cooler > myself. Sure, that just as readily available. Get a TransMeta-powered laptop. Or a NetWinder 3100 appliance. Something like 25W total power draw. I just bought a pair of Micro-ATX Mainboards and stuffed them with AMD K6/2-500 cpus. Underclocked them at 400 MHz. Should be very happy and robust little utility horses in the lab. Cost about $110 each. Only games need more horsepower. These MBs include sound chipset, 3D-accellerated video. Just add chassis, power, ethernet card, hard drive and DRAM. I'm using them to fix a pair of broken e-machines, which supplied all those missing parts. -- / Lars Poulsen +1-805-569-5277 http://www.beagle-ears.com/lars/ 125 South Ontare Rd, Santa Barbara, CA 93105 USA lars@beagle-ears.com ###### Newsgroups: alt.folklore.computers,comp.sys.cdc From: "Douglas A. Gwyn" Subject: Re: CDC6600 - just how powerful a machine was it? X-Nntp-Posting-Host: swctd.arl.army.mil Content-Type: text/plain; charset=us-ascii Message-ID: <3D29A4C0.62EFE32@null.net> Sender: news@arl.army.mil (News Administration ) Content-Transfer-Encoding: 7bit Organization: U.S. Army Research Laboratory X-Accept-Language: en References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> Mime-Version: 1.0 Date: Mon, 8 Jul 2002 14:42:08 GMT X-Mailer: Mozilla 4.79 [en] (X11; U; SunOS 5.8 sun4u) Lines: 31 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!news.imp.ch!newsfeed.stueberl.de!kibo.news.demon.net!demon!newsfeed.icl.net!newsfeed.fjserv.net!skynet.be!skynet.be!oanews!arlnews!blaze!news Xref: chonsp.franklin.ch alt.folklore.computers:111743 Walter Spector wrote: > > It may just be a mis-memory, but I seem to recall something about a C compiler > > for CDC machines, written back then. (From a university in Texas?) > Yes, there was one from UT. We had a copy of it. I held the tape in my > hands. However our fellow in charge of installing it could never get it to > install. Apparently we were not alone, because at the last VIM conference > I attended (spring 1984), there was a session devoted to it. A sizable lynch > mob showed up, but the speaker did not. It then turned into a bull session on > how to get CDC to wake up to modern programming languages. At BRL (Aberdeen Proving Ground) we obtained a copy of UT's C compiler, which as I recall was based on PCC and thus available only with proof of license. Our computing activities were divided between a central mainframe (CDC, later Denelcor HEP) "job shop" and a distributed network of (super)minicomputers, with attendant friction between camps. The hope for the Cyber C compiler was to port some significant supermini apps to the mainframe. I don't recall the details of that effort, but it wasn't long before the mainframes were replaced by supercomputers that came with vendor-supported C compilers, running UniCOS or some other flavor of Unix. > I'd really enjoy seeing a copy of the UT C compiler to see some of the design > decisions that were made. But fear it has gone to the Great Bit Bucket in the Sky. There is a *small* chance that somebody at the former BRL still has the CDC C compiler on 9-track 1/2" magtape. I have recently recovered disk images from our PDP-11 Unix systems, and there is a 300MB pack still not processed from our primary VAX Unix system; it is possible if not likely that somewhere among those files is a copy of Cyber C. I'll put it on my list of things to look for as I rummage through the files in an attempt to reconstruct some of those old systems. ###### From: tmm@spamfilter.asns.tr.unisys.com (Tim McCaffrey) Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 8 Jul 2002 18:43:31 +0000 (UTC) Organization: A series networking Lines: 20 Message-ID: References: <3d248b17$1_3@news.iglou.com> NNTP-Posting-Host: 192.63.212.94 X-Trace: trsvr.tr.unisys.com 1026153811 3886 192.63.212.94 (8 Jul 2002 18:43:31 GMT) X-Complaints-To: usenet@tr.unisys.com NNTP-Posting-Date: Mon, 8 Jul 2002 18:43:31 +0000 (UTC) X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!fu-berlin.de!cox.net!priapus.visi.com!news-out.visi.com!hermes.visi.com!uunet!ash.uu.net!bbnews1.unisys.com!trsvr.tr.unisys.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111753 In article <3d248b17$1_3@news.iglou.com>, dquebbeman@ixnayamspayacm.org says... > >Actually, a CDC 6600 with 60 terminals hanging off it was >*terribly* bogged down, giving crappy response time to all >60 users. I can't imagine trying to hang more _character mode_ >terminals off such a system. > Well, you didn't have it setup correctly :). MSU had a 6500 running upto 100 terminals (response was good at 30, tolerable at 60, and very bad at 100). Of course, most of the interactive traffic load was handled by a Interdata 7/32 (or 8/32?) and the Argus PP. Most terminal responses were handled by Manager, a single process on the 6500 (so, no swapping needed). MSU spent a lot of effort customizing SCOPE to handle interactive usage. It paid off when the Cyber 170/750 showed. That box could handle 200 users easily. - Tim ###### From: Eric Sosman Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Mon, 08 Jul 2002 15:51:56 -0400 Organization: Sun Microsystems Lines: 22 Message-ID: <3D29ED5C.1F3862C0@sun.com> References: <20020704201700.232b459c.steveo@eircom.net> <20020704231759.2efd7bdc.steveo@eircom.net> NNTP-Posting-Host: tardis.east.sun.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsmi-eu.news.garr.it!NewsITBone-GARR!feed.news.nacamar.de!newsfeed01.sul.t-online.de!t-online.de!newsrouter.chello.at!amsnews01.chello.com!news-hub.cableinet.net!blueyonder!btnet-peer!btnet-peer0!btnet-feed5!btnet!carbon.eu.sun.com!new-usenet.uk.sun.com!eastnews1.East.Sun.COM!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111756 Alan Clifford wrote: > > On Fri, 5 Jul 2002, Rupert Pigott wrote: > > RP> > > RP> > Might as well put it in the back of the honking great display, > RP> > use a wireless keyboard and lose all that unsightly wire. > RP> > RP> I have thought about that. But I'd like to be able to carry it > RP> around like a portable but without the extra baggage of battery, > RP> screen etc. The idea being I plug it into what's around me at > RP> the time, honking or otherwise. > RP> > > You need a virtual keyboard. One was demonstrated on "Tomorrows World". > An infrared device that displayed a keyboard on any flat surface and > detected which "key" you pressed. Gives a new meaning to the term "fat-finger" ... -- Eric.Sosman@sun.com ###### From: Charles Shannon Hendrix Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 9 Jul 2002 01:11:00 -0400 Organization: Too Much Message-ID: References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> Reply-To: shannon@nospam.widomaker.com User-Agent: slrn/0.9.7.3 (Linux) X-Complaints-To: newsabuse@supernews.com Lines: 44 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!newsfeed01.sul.t-online.de!t-online.de!newspeer1-gui.server.ntli.net!ntli.net!sn-xit-05!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!escape!nobody Xref: chonsp.franklin.ch alt.folklore.computers:111823 In article , Anne & Lynn Wheeler wrote: > Charles Shannon Hendrix writes: >> SNA seems more like an umbrella of support from IBM to make it all work >> than a communications protocol. >> >> When I last worked with it, every "SNA" connection we had as a different >> medium and protocol. > > well SNA was primarily an infrastructure for managing communication > with large terminal populations ... That's not what the ads and the salesmen's spiel said... :) > mainframe application talking to upwards of 60,000 to 100,000 > terminals (maybe more). in that sense it included a communication > protocol (but not a networking protocol). Another example might be > the data processing infrastructure that talks to the head-end of > majority of the cable tv systems using LU0 for addressing every > set-top box. Everything we had was file transfer between systems, except cases where we had to use IBM EasyMail (odd system). Of course, doing the file transfers with the UNIX and VMS systems was trivial. We could specify jobs to be executed on the mainframe after each transfer, and in fact a minimal job was mandatory if I remember correctly. What we did not get was any protocol to tell us if the transfer was correct. An IBMer's suggestion was to specify a remote job that would verify the file and send an OK back to our system. This was after being told, "SNA, in any sensible terms, doesn't really exist as a ``thing''". The problem was that queuing up files on 9600bps links meant hours or days before we knew if the transfer worked. In college, SNA was only covered very briefly, and I jumped out of the program before learning anything but trivial stuff. ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 9 Jul 2002 12:44:22 -0400 Lines: 26 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d2b12e8_1@news.iglou.com> X-Trace: news.iglou.com 1026233064 204.250.0.238 (9 Jul 2002 12:44:24 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed-east.nntpserver.com!nntpserver.com!news-out.visi.com!hermes.visi.com!uunet!ash.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111763 "Douglas A. Gwyn" wrote in message = news:3D29A4C0.62EFE32@null.net... > > I'd really enjoy seeing a copy of the UT C compiler to see some of = the design > > decisions that were made. But fear it has gone to the Great Bit = Bucket in the Sky. >=20 > There is a *small* chance that somebody at the former BRL still has = the > CDC C compiler on 9-track 1/2" magtape. I have recently recovered = disk > images from our PDP-11 Unix systems, and there is a 300MB pack still > not processed from our primary VAX Unix system; it is possible if not > likely that somewhere among those files is a copy of Cyber C. I'll = put > it on my list of things to look for as I rummage through the files in = an > attempt to reconstruct some of those old systems. Andy Mickel has a stack of Pascal Compiler distro tapes; the one he sent me is readable... I didn't think to ask him what other stuff he might still have! -dq ###### Newsgroups: alt.folklore.computers,comp.sys.cdc From: "Douglas A. Gwyn" Subject: Re: CDC6600 - just how powerful a machine was it? X-Nntp-Posting-Host: swctd.arl.army.mil Content-Type: text/plain; charset=us-ascii Message-ID: <3D2B1A9B.920A6E14@null.net> Sender: news@arl.army.mil (News Administration ) Content-Transfer-Encoding: 7bit Organization: U.S. Army Research Laboratory X-Accept-Language: en References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> Mime-Version: 1.0 Date: Tue, 9 Jul 2002 17:17:15 GMT X-Mailer: Mozilla 4.79 [en] (X11; U; SunOS 5.8 sun4u) Lines: 2 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!kibo.news.demon.net!demon!transit.news.xs4all.nl!skynet.be!skynet.be!oanews!arlnews!blaze!news Xref: chonsp.franklin.ch alt.folklore.computers:111820 So far, I have found some software we used to "RJE" access our Cyber mainframe from a PDP-11 UNIX system. ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Tue, 09 Jul 2002 13:01:21 -0500 Organization: Airnews.net! at Internet America Lines: 16 Message-ID: <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> X-Orig-Message-ID: <3D2B24F1.3DE9C128@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library1-aux.airnews.net NNTP-Posting-Time: Tue Jul 9 13:05:39 2002 NNTP-Posting-Host: !d>Y11k-WV>Yu%l/i#co (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!news.airnews.net!cabal10.airnews.net!cabal2.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111768 "Douglas A. Gwyn" wrote: > So far, I have found some software we used to "RJE" access > our Cyber mainframe from a PDP-11 UNIX system. > That recalls in my mind the actual RJE terminals that were used with UT Austin's CDC6600. They had a pretty good cardreader and lineprinter attached. Does anyone have a picture of one of them (I think they were CDC products). IIRC, the carddeck "job" structure you used at the remote site was identical to what you'd use at the main Computation Center. I think one difference was that it would not steal your second card (containing your account's password). - LarryW ###### Newsgroups: alt.folklore.computers Subject: Re: CDC6600 - just how powerful a machine was it? References: <3d248b17$1_3@news.iglou.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 21 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:HDWX5YCgoVdpjFUmX02utxrFg8U= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 09 Jul 2002 19:06:54 GMT NNTP-Posting-Host: 65.59.21.36 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1026241614 65.59.21.36 (Tue, 09 Jul 2002 12:06:54 PDT) NNTP-Posting-Date: Tue, 09 Jul 2002 12:06:54 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!feedme.news.mediaways.net!newsfeed.media.kyoto-u.ac.jp!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111793 tmm@spamfilter.asns.tr.unisys.com (Tim McCaffrey) writes: > Well, you didn't have it setup correctly :). MSU had a 6500 running > upto 100 terminals (response was good at 30, tolerable at 60, and > very bad at 100). Of course, most of the interactive traffic load > was handled by a Interdata 7/32 (or 8/32?) and the Argus PP. Most > terminal responses were handled by Manager, a single process on the > 6500 (so, no swapping needed). MSU spent a lot of effort customizing > SCOPE to handle interactive usage. It paid off when the Cyber 170/750 > showed. That box could handle 200 users easily. i worked on the original terminal controller that was built with the interdata/3 ... later enhanced to interdata/4 using multiple interdata/3s as linescanners. That along with the ibm channel interface started the PCM controller business. Later perkin/elmer bought them up. Not more than a couple years ago, i ran into perkin/elmer terminal controller doing some pretty heavy duty work in a large transaction shop. http://www.garlic.com/~lynn/subtopic.html#360pcm -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: couperusNOSPAM@znet.com (Jitze Couperus) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 10 Jul 2002 02:58:15 GMT Organization: Posted via Supernews, http://www.supernews.com Message-ID: <3d2ba070.526367777@sd.znet.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> X-Newsreader: Forte Free Agent 1.21/32.243 X-Complaints-To: newsabuse@supernews.com Lines: 28 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!newsfeed.stueberl.de!cox.net!sn-xit-02!sn-xit-06!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111841 On Tue, 09 Jul 2002 13:01:21 -0500, Larry__Weiss wrote: > >That recalls in my mind the actual RJE terminals that were >used with UT Austin's CDC6600. They had a pretty good >cardreader and lineprinter attached. Does anyone have >a picture of one of them (I think they were CDC products). > You referring to the "200 User Terminal" or UT200? I'm pretty sure that was a CDC product - there is a reference to 22681-18029 1500 UTERM-CDC 200 USER TERMINAL SIMULATOR in the page at http://www.spies.com/~aek/pdf/hp/HP_SoftwarePNs.txt and another reference to an emulator for this box running on an HP-2100 - last reference on the page at http://bluehawk.monmouth.edu/~msoklic/Publications.html but that's about all I can find - alas no pictures. Jitze ###### From: Larry__Weiss Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 10 Jul 2002 04:04:30 -0500 Organization: Airnews.net! at Internet America Lines: 7 Message-ID: <9C888DBAB9501543.F0BF43020EBBCDBB.C106558F92566D17@lp.airnews.net> X-Orig-Message-ID: <3D2BF89E.F76C7891@airmail.net> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library1-aux.airnews.net NNTP-Posting-Time: Wed Jul 10 04:06:27 2002 NNTP-Posting-Host: !c1g]1k-WLrVg(r1-[&b (Encoded at Airnews!) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!fr.usenet-edu.net!usenet-edu.net!fr.clara.net!heighliner.fr.clara.net!isdnet!howland.erols.net!news.airnews.net!cabal10.airnews.net!cabal2.airnews.net!news-f.iadfw.net!usenet Xref: chonsp.franklin.ch alt.folklore.computers:111842 Jitze Couperus wrote: > but that's about all I can find - alas no pictures. > Same here. No pictures, just memories. Good memories! - LarryW ###### Reply-To: "Gerard J van der Grinten" From: "Gerard J van der Grinten" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 10 Jul 2002 12:15:58 +0200 Organization: Philips Medical Systems X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Lines: 43 Message-ID: <3d2c0988$0$227$4d4ebb8e@read-nat.news.nl.uu.net> NNTP-Posting-Host: gw-ehv05.pnl.philips.com X-Trace: 1026296201 read-nat.news.nl.uu.net 227 212.153.190.4 X-Complaints-To: abuse@nl.uu.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!skynet.be!skynet.be!newsfeed.online.be!bnewspeer00.bru.ops.eu.uu.net!bnewspost00.bru.ops.eu.uu.net!emea.uu.net!read-nat.news.nl.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111895 A picture of the "memory" of the 200UT is at: HTTP://artemis.rrze.uni-erlangen.de/servlet/Anzeigen45?inventarnummer-10304 It is a 9.98 ms delay line (don't hit the cabinet type) The bits traveled from the transducer to the pickup. When cards where read the line was used for card buffer. It hold 8 cards (as the display was 8 by 80) Regards, Gerard. "Jitze Couperus" wrote in message news:3d2ba070.526367777@sd.znet.com... > On Tue, 09 Jul 2002 13:01:21 -0500, Larry__Weiss > wrote: > > > > > >That recalls in my mind the actual RJE terminals that were > >used with UT Austin's CDC6600. They had a pretty good > >cardreader and lineprinter attached. Does anyone have > >a picture of one of them (I think they were CDC products). > > > > You referring to the "200 User Terminal" or UT200? I'm pretty > sure that was a CDC product - there is a reference to > > 22681-18029 1500 UTERM-CDC 200 USER TERMINAL SIMULATOR > > in the page at > > http://www.spies.com/~aek/pdf/hp/HP_SoftwarePNs.txt > > and another reference to an emulator for this box running on > an HP-2100 - last reference on the page at > > http://bluehawk.monmouth.edu/~msoklic/Publications.html > > but that's about all I can find - alas no pictures. > > Jitze > ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <9C888DBAB9501543.F0BF43020EBBCDBB.C106558F92566D17@lp.airnews.net> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 10 Jul 2002 07:17:35 -0400 Lines: 15 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d2c17d0_2@news.iglou.com> X-Trace: news.iglou.com 1026299856 204.250.0.238 (10 Jul 2002 07:17:36 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newspump.monmouth.com!newspeer.monmouth.com!jfk3-feed1.news.algx.net!allegiance!news-out.visi.com!hermes.visi.com!uunet!ash.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111838 "Larry__Weiss" wrote in message = news:9C888DBAB9501543.F0BF43020EBBCDBB.C106558F92566D17@lp.airnews.net...= > Jitze Couperus wrote: > > but that's about all I can find - alas no pictures. > >=20 >=20 > Same here. No pictures, just memories. Good memories! There are some line art pictures in the KRONOS 2.1 EXPORT IMPORT REFERENCE MANUAL; I'll see if I can scan those and put them up somewhere... -dq ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <3d2c0988$0$227$4d4ebb8e@read-nat.news.nl.uu.net> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Wed, 10 Jul 2002 07:23:13 -0400 Lines: 25 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d2c1922$1_3@news.iglou.com> X-Trace: news.iglou.com 1026300194 204.250.0.238 (10 Jul 2002 07:23:14 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!opentransit.net!fr.clara.net!heighliner.fr.clara.net!news-hub.siol.net!zur.uu.net!dca.uu.net!ash.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111839 "Gerard J van der Grinten" = wrote in message news:3d2c0988$0$227$4d4ebb8e@read-nat.news.nl.uu.net... > A picture of the "memory" of the 200UT is at: > = HTTP://artemis.rrze.uni-erlangen.de/servlet/Anzeigen45?inventarnummer-103= 04 >=20 > It is a 9.98 ms delay line (don't hit the cabinet type) > The bits traveled from the transducer to the pickup. When cards where = read > the line was used for card buffer. It hold 8 cards (as the display was = 8 by > 80) That url didn't work, but this one does, and is partly translated, but watch out for the link wrap: http://216.239.35.120/translate_c?hl=3Den&u=3Dhttp://artemis.rrze.uni-erl= angen.de/servlet/Anzeigen45%3Finventarnummer%3DI0304&prev=3D/search%3Fq%3= DISER-Katalog%26hl%3Den%26safe%3Doff%26sa%3DG It appears this delay line was used for multiple purposes by CDC. -dq ###### Message-ID: <3D2C81B0.FED3F952@attbi.com> From: Joe Yuska X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 25 NNTP-Posting-Host: 24.91.67.33 X-Complaints-To: abuse@attbi.com X-Trace: sccrnsc03 1026326594 24.91.67.33 (Wed, 10 Jul 2002 18:43:14 GMT) NNTP-Posting-Date: Wed, 10 Jul 2002 18:43:14 GMT Organization: AT&T Broadband Date: Wed, 10 Jul 2002 18:43:14 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!extra.newsguy.com!lotsanews.com!nf3.bellglobal.com!wn1feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!sccrnsc03.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111908 Jitze Couperus wrote: > > On Tue, 09 Jul 2002 13:01:21 -0500, Larry__Weiss > wrote: > > > > >That recalls in my mind the actual RJE terminals that were > >used with UT Austin's CDC6600. They had a pretty good > >cardreader and lineprinter attached. Does anyone have > >a picture of one of them (I think they were CDC products). > > > > You referring to the "200 User Terminal" or UT200? I'm pretty > sure that was a CDC product - there is a reference to The UT200 referred both to the RJE terminal and to the synchronous protocol used to drive it. IIRC, it consisted of a small table-sized box o' logic with a CDC 211 type terminal on it, a 250 cpm reader and a 300 lpm printer. Emulators were written for several minicomputers including some HP's, PDP-11's, as well as CDC 1700's and CYBER 18's. Later versions of NOS (and the CDC mini's ) also provided a HASP protocol for RJE work. Joe Yuska ###### From: ctill@nc.rr.com (Chuck Till) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Message-ID: <3d2e21f7.91178227@news-server.nc.rr.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <3D2C81B0.FED3F952@attbi.com> X-Newsreader: Forte Free Agent 1.21/32.243 Lines: 13 Date: Fri, 12 Jul 2002 00:31:58 GMT NNTP-Posting-Host: 66.57.4.71 X-Complaints-To: abuse@rr.com X-Trace: twister.southeast.rr.com 1026433918 66.57.4.71 (Thu, 11 Jul 2002 20:31:58 EDT) NNTP-Posting-Date: Thu, 11 Jul 2002 20:31:58 EDT Organization: Road Runner - NC Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!cyclone.tampabay.rr.com!news-post.tampabay.rr.com!twister.southeast.rr.com.POSTED!53ab2750!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:112026 >UT200... emulators were written for several minicomputers >including some HP's, PDP-11's, as well as CDC 1700's and >CYBER 18's. I imagine an emulator was written for nearly every popular mini of the 1970s. Georgia Tech and the University of Georgia used mainly Interdata 7/xx and TI 990, but I know there were many more. >Later versions of NOS (and the CDC mini's ) also provided a HASP >protocol for RJE work. Also the Tieline package, available from CDC as a pro services product, that let a PP and a 6671 port emulate a 3780. ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <3d2c0988$0$227$4d4ebb8e@read-nat.news.nl.uu.net> <3D2DC9F7.92FC7BE5@bellsouth.net> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Fri, 12 Jul 2002 10:03:52 -0400 Lines: 30 MIME-Version: 1.0 Content-Type: text/plain; charset="koi8-r" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d2ee1c9_1@news.iglou.com> X-Trace: news.iglou.com 1026482633 204.250.0.238 (12 Jul 2002 10:03:53 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.esat.net!news-out.visi.com!hermes.visi.com!uunet!ash.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:111997 "Rostyslaw J. Lewyckyj" wrote in message = news:3D2DC9F7.92FC7BE5@bellsouth.net... > Gerard J van der Grinten wrote: >=20 > > A picture of the "memory" of the 200UT is at: > > = HTTP://artemis.rrze.uni-erlangen.de/servlet/Anzeigen45?inventarnummer-103= 04 > > >=20 > The URL doesn't seem to load anything except the header HTML . I am = using > Netscape V 4.79. Backing up to just //artemis.rrze.uni-erlanden.de = puts up a=20 > page with a link to ISER which also doesn't work. I get a page with = bad buttons,=20 > and undispayable pictures. Trying the link on the first button gets a = 404 page=20 > not found error. What gives??? Either try the link I posted in my reply to his message, or try this = fixed version of what he typed above: http://artemis.rrze.uni-erlangen.de/servlet/Anzeigen45?inventarnummer=3DI= 0304 Regards, -doug quebbeman ###### From: ctill@nc.rr.com (Chuck Till) Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? Message-ID: <3d317cb9.132067956@news-server.nc.rr.com> References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <3D2C81B0.FED3F952@attbi.com> <3d2e21f7.91178227@news-server.nc.rr.com> <3D2F2812.4583905C@attbi.com> X-Newsreader: Forte Free Agent 1.21/32.243 Lines: 14 Date: Sun, 14 Jul 2002 13:31:29 GMT NNTP-Posting-Host: 24.74.139.167 X-Complaints-To: abuse@rr.com X-Trace: twister.southeast.rr.com 1026653489 24.74.139.167 (Sun, 14 Jul 2002 09:31:29 EDT) NNTP-Posting-Date: Sun, 14 Jul 2002 09:31:29 EDT Organization: Road Runner - NC Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed1.cidera.com!Cidera!cyclone.tampabay.rr.com!news-post.tampabay.rr.com!twister.southeast.rr.com.POSTED!53ab2750!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:112192 >> Also the Tieline package, available from CDC as a pro services >> product, that let a PP and a 6671 port emulate a 3780. > >Now, that's waking up a few memory cells but not enough to clear the >fog. Was this an outgrowth of the CDC data centers' interconnections >with SBC? This would be after they picked up the SBC stuff in the >lawsuit. The story I heard was that some university paid CDC to write it, and CDC got the marketing rights. Might be accurate, might not. I never saw much evidence that CDC did anything with SBC, not even when CDC was selling IBM 360-compatible memory, disks, and eventually entire 360-compatible systems. ###### Message-ID: <3D31B1EB.DD8D2F2B@attbi.com> From: Joe Yuska X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <3D2C81B0.FED3F952@attbi.com> <3d2e21f7.91178227@news-server.nc.rr.com> <3D2F2812.4583905C@attbi.com> <3d317cb9.132067956@news-server.nc.rr.com> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 32 NNTP-Posting-Host: 24.91.67.33 X-Complaints-To: abuse@attbi.com X-Trace: sccrnsc03 1026666791 24.91.67.33 (Sun, 14 Jul 2002 17:13:11 GMT) NNTP-Posting-Date: Sun, 14 Jul 2002 17:13:11 GMT Organization: AT&T Broadband Date: Sun, 14 Jul 2002 17:13:11 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!arclight.uoregon.edu!wn4feed!worldnet.att.net!204.127.198.204!attbi_feed4!attbi_feed3!attbi.com!sccrnsc03.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:112157 Chuck Till wrote: > > >> Also the Tieline package, available from CDC as a pro services > >> product, that let a PP and a 6671 port emulate a 3780. > > > >Now, that's waking up a few memory cells but not enough to clear the > >fog. Was this an outgrowth of the CDC data centers' interconnections > >with SBC? This would be after they picked up the SBC stuff in the > >lawsuit. > > The story I heard was that some university paid CDC to write it, and > CDC got the marketing rights. Might be accurate, might not. I never > saw much evidence that CDC did anything with SBC, not even when CDC > was selling IBM 360-compatible memory, disks, and eventually entire > 360-compatible systems. They owned SBC. By 1969, the data processing service provided by Control Data had expanded to include timesharing for smaller companies and was large enough to spin off into its own area called CYBERNET. An interesting result of Control Data’s 1968 lawsuit against IBM was that as part of the settlement, IBM transferred its subsidiary, Service Bureau Corporation, to Control Data for an insignificant amount. The acquisition of this company more than doubled Control Data’s data services business and greatly broadened its business and customer base. http://www.digitalcentury.com/encyclo/update/cdc.html Joe Yuska ###### Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <3D2C81B0.FED3F952@attbi.com> <3d2e21f7.91178227@news-server.nc.rr.com> <3D2F2812.4583905C@attbi.com> <3d317cb9.132067956@news-server.nc.rr.com> Subject: Re: CDC6600 - just how powerful a machine was it? Date: Sun, 14 Jul 2002 19:26:25 -0400 Lines: 25 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d3208a2_2@news.iglou.com> X-Trace: news.iglou.com 1026689186 204.250.0.238 (14 Jul 2002 19:26:26 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!ash.uu.net!news.iglou.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:112211 "Chuck Till" wrote in message = news:3d317cb9.132067956@news-server.nc.rr.com... > >> Also the Tieline package, available from CDC as a pro services > >> product, that let a PP and a 6671 port emulate a 3780. > > > >Now, that's waking up a few memory cells but not enough to clear the > >fog. Was this an outgrowth of the CDC data centers' interconnections > >with SBC? This would be after they picked up the SBC stuff in the > >lawsuit. >=20 > The story I heard was that some university paid CDC to write it, and > CDC got the marketing rights. Might be accurate, might not. I never > saw much evidence that CDC did anything with SBC, not even when CDC > was selling IBM 360-compatible memory, disks, and eventually entire > 360-compatible systems. We were one of many sites who became so customized, that we got code from wherever we could. We got code from a Naval site for the 20 PP support, we got TIELINE from University of Georgia (I think), and we got Multi-Mainframe (MMF) from Manchester or Cambridge. It was a crazy-quilt Kronos... -dq ###### Newsgroups: alt.folklore.computers,comp.sys.cdc Subject: Re: CDC6600 - just how powerful a machine was it? References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3D226048.9012D949@earthlink.net> <3D234CDD.F33DA139@mindspring.com> <3D27D4EC.5EBAD9BA@earthlink.net> <3D29A4C0.62EFE32@null.net> <3d2b12e8_1@news.iglou.com> <3D2B1A9B.920A6E14@null.net> <360B7FCC53A00712.E7C5EC7922AFD020.AB85F76EA306470D@lp.airnews.net> <3d2ba070.526367777@sd.znet.com> <3D2C81B0.FED3F952@attbi.com> <3d2e21f7.91178227@news-server.nc.rr.com> <3D2F2812.4583905C@attbi.com> <3d317cb9.132067956@news-server.nc.rr.com> <3D31B1EB.DD8D2F2B@attbi.com> <3d340b2d.141222942@news-server.nc.rr.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 26 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.2 (i386-msvc-nt4.0.1381) Cancel-Lock: sha1:yRkJclrvAwz6eNnDkLoNe0hsALw= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 16 Jul 2002 13:36:12 GMT NNTP-Posting-Host: 64.156.37.233 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1026826572 64.156.37.233 (Tue, 16 Jul 2002 06:36:12 PDT) NNTP-Posting-Date: Tue, 16 Jul 2002 06:36:12 PDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp1.phx1.gblx.net!nntp.gblx.net!nntp.gblx.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!03b822ca!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:112318 ctill@nc.rr.com (Chuck Till) writes: > Sounds like PR fluff in retrospect. It didn't take much to > double CDC's data service business, and I wonder how many > SBC customers in 1969 were still using CDC/SBC a few years > later. i don't know the circumstances ... but ibm real estate retained the SBC building out in burlington mall ... and the vm/370 group moved out there when the outgrew the space on 3rd floor 545 tech. sq. the cp/67 group had split off from CSC and sometime in the time-frame of starting the morph of cp/67 to vm/370 they started rapid growth ... including essentially absorbing the (ibm) boston programming center on the 3rd floor (bpc had maybe 3/4ths of the 3rd floor, csc was about 2/3rds of the 4th floor with the csc machine room taking about 1/2 the 2nd floor). the boston programming center was responsible for cps (conversational programming system ... interpreted pl/i ran under os/360 ... and there was special microcode available for 360/50 that made it run faster). Jean Sammet, Nat Rochester, couple others had their offices in the boston programming center (so the vm/370 group sort of absorbed them also). When vm/370 group moved out to sbc bldg. in burlington mall they stayed in 545 tech. sq. -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ####### From: "ed thelen" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3d1fef08_1@news.iglou.com> <3D209F50.A6CAA51E@attbi.com> <3d2154fe$1@news.meer.net> Subject: Re: CDC6600 - just how powerful a machine was it? Lines: 80 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Message-ID: NNTP-Posting-Host: 12.235.26.140 X-Complaints-To: abuse@attbi.com X-Trace: rwcrnsc52.ops.asp.att.net 1027124386 12.235.26.140 (Sat, 20 Jul 2002 00:19:46 GMT) NNTP-Posting-Date: Sat, 20 Jul 2002 00:19:46 GMT Organization: AT&T Broadband Date: Sat, 20 Jul 2002 00:19:46 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wn1feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!rwcrnsc52.ops.asp.att.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:112524 Oh how fun - and unusual - a difference of opinion :-))) Maybe we are talking different eras. I worked in CDC Special Systems from 1965 to 1970 - Our group shipped about 5 6xxx machines, into the hybrid computing and data acquisition/on-line-analysis market. something like 1 6600 3 6500 oh heck - I can't remember - and NONE had the (?rare?) 20 PP option. We figured it was a sales gimick - we did improve PP usage - better optimize seeks based on rotational position - better optimize tape read/writes - PP did not go back into the pool after every read/write - and some other stuff. Maybe different uses and different eras have different stories - Cheers Ed Thelen OH - I have posted some documents on line http://ed-thelen.org/comp-hist/CDC-6600-R-M.html (6600 Reference Manual) http://ed-thelen.org/comp-hist/CRAY-1-HardRefMan/CRAY-1-HRM.html 1st three chapters of Cray-1 hardware reference manual and Tom Uban posted a PDF copy of Jim Thornton's book http://www.spies.com/~aek/pdf/cdc/DesignOfAComputer_CDC6600.pdf "Jeff Woolsey" wrote in message news:3d2154fe$1@news.meer.net... > In article <3D209F50.A6CAA51E@attbi.com>, Joe Yuska wrote: > > >Kent Olsen wrote: > > >> > There were Peripheral Processors which did the I/O. There were 10 or 20 of > >> > them, each of which had its own memory (12 bits of addressing?) and > >full access > >> > to the central processor's memory. > >> > >> Most 6000s were configured with a full complement of 20 PPs. Certain > >> pricing options allowed for the purchase of machines with less, but > >> throughput suffered pretty badly. Many were field upgraded to 20 PPs. > > > >I'll take serious issue with this. Virtually all of the 6000's and Cyber 70's I > >worked on during a 17-year period had only 10 ppu. The extra 10-ppu option was > >expensive and required a second cabinet. Most of the 20 ppu > >configurations I was > >aware of were connected to exotic type peripherals some of which you > >could be killed > >for knowing about. > > And I take issue with that. > > Of the dozen or so systems I worked on, only the 6400 and one Cyber 73 only > had 10 PPs. All of the rest were either ordered with more, or upgraded > shortly after delivery. The other Cyber 73 has 17 PPs, the Cyber 74 had > 14 and then 17, and all the rest were 170s or later, all with at > least 14 PPs. None had the full complement, though. > > >> The PP was a 12 bit processor. It had 4095.5 words. (Inside joke. > >> It had 4096 words, but word 7777 was rarely used as it wasn't directly > >> addressable.) > > It's addressible via instructions which do not involve the 12-bit > incrementer... Limits uesfulness a bit. > -- > Jeff Woolsey {woolsey,jlw}@{jlw,jxh}.com > "I didn't get a 'Harrumph!' out of _that_ guy." -Gov Le Petomaine > "Delete! Delete! OK!" -Dr. Bronner on disk space management > "A toy robot!!!!" -unlucky Japanese scientist ###### From: "ed thelen" Newsgroups: alt.folklore.computers,comp.sys.cdc References: <50F723405B6BFCB7.9620EB809FCFA439.F2B1D34FA3F21C6C@lp.airnews.net> <3D1F56EE.DAA8337E@mindspring.com> <3d1fef08_1@news.iglou.com> <3D209F50.A6CAA51E@attbi.com> <3d220101_2@news.iglou.com> Subject: Re: CDC6600 - just how powerful a machine was it? Lines: 31 X-Newsreader: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Message-ID: NNTP-Posting-Host: 12.235.26.140 X-Complaints-To: abuse@attbi.com X-Trace: sccrnsc03 1027139180 12.235.26.140 (Sat, 20 Jul 2002 04:26:20 GMT) NNTP-Posting-Date: Sat, 20 Jul 2002 04:26:20 GMT Organization: AT&T Broadband Date: Sat, 20 Jul 2002 04:26:24 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wn1feed!worldnet.att.net!204.127.198.204!attbi_feed4!attbi_feed3!attbi.com!sccrnsc03.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:112475 > "Douglas H. Quebbeman" wrote in message news:3d220101_2@news.iglou.com... > "Tim McCaffrey" wrote in message news:afso22> > > > I would be surprised if there was a 10 PPU add on option, at least for the > > 6500. Michigan State would have probably bought it, if available. When the > > Cyber 170/750 came in we could only use 10 of the 20 PP's, because our > > customized OS needed to be updated. When the additional 10 PPs were usable, > > it made a significant difference. One reason may have been because the OS > > dedicated (I think) 4 PPs to certain tasks (monitor, Argus (terminal support), > > Console, and something else). > > You *would* post this on a day I'm not carrying the 6000 Series > Hardware refman with me, so I can't quote the option number, but > yeah, it was a "standard option". take a peek at http://ed-thelen.org/comp-hist/CDC-6600-R-M.html#PA-00 for "Augmented I/O Buffer and Control"