From: "PDP-10@PaulAllen.com" Newsgroups: alt.sys.pdp10 Subject: New web and ftp site for PDP-10 Date: Mon, 5 Nov 2001 20:09:09 -0800 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 X-Complaints-To: newsabuse@supernews.com Lines: 17 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!feed2.onemain.com!feed1.onemain.com!feeder.qis.net!sn-xit-02!sn-xit-04!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7381 Announcing a new Web and ftp site for PDP-10 material - http://pdp10.paulallen.com. Paul G. Allen, co-founder of Microsoft and PDP-10 enthusiast/developer is creating a new Web site to provide easy-to-use PDP-10 files and documentation. After loading and testing various simh configurations, a DSKB image for simh with a robust configuration, a monitor that supports printing and serial logins and a DSKC file for user storage, have all been posted to the site. The TOAD-1 XKLeTen.paulallen.com will be moving soon to a location where it can again be accessed from the internet. Stay tuned for more information. For questions and comments mailto:pdp10@paulallen.com ###### From: "Zane H. Healy" Subject: Re: New web and ftp site for PDP-10 Newsgroups: alt.sys.pdp10 References: Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 20 Message-ID: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Wed, 07 Nov 2001 12:29:33 EST Date: Wed, 07 Nov 2001 17:29:33 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.tele.dk!small.news.tele.dk!205.231.236.10!newspeer.monmouth.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7404 PDP-10@PaulAllen.com wrote: > Paul G. Allen, co-founder of Microsoft and PDP-10 enthusiast/developer is > creating a new Web site to provide easy-to-use PDP-10 files and > documentation. After loading and testing various simh configurations, a DSKB > image for simh with a robust configuration, a monitor that supports printing > and serial logins and a DSKC file for user storage, have all been posted to > the site. If anyone is wanting to run TOPS-10 7.03 without having to go to the trouble of installing it. This seems to give you a *very* nice system. It looks like it's got at least the following added to it: TOPS-10 Integration Tools, V4, 10-Jan-1986 [1,2,EDT] BASIC-10 V17F [1,2,BASIC] Kermit [76,5,kermit] TOPS-10 Customer Supported CUSPs [1,2,140B] Better yet the server looks to be sitting on a nice fast connection to the net. Zane ###### From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Organization: Kilonet.net Lines: 27 Message-ID: <3BE975E8.EA0200EB@bartek.dontspamme.net> References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc) X-Accept-Language: en Date: Wed, 07 Nov 2001 17:57:53 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 1005155873 24.186.100.134 (Wed, 07 Nov 2001 12:57:53 EST) NNTP-Posting-Date: Wed, 07 Nov 2001 12:57:53 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7393 "Zane H. Healy" wrote: > > PDP-10@PaulAllen.com wrote: > > Paul G. Allen, co-founder of Microsoft and PDP-10 enthusiast/developer is > > creating a new Web site to provide easy-to-use PDP-10 files and > > documentation. After loading and testing various simh configurations, a DSKB > > image for simh with a robust configuration, a monitor that supports printing > > and serial logins and a DSKC file for user storage, have all been posted to > > the site. > > If anyone is wanting to run TOPS-10 7.03 without having to go to the trouble > of installing it. This seems to give you a *very* nice system. It looks > like it's got at least the following added to it: > TOPS-10 Integration Tools, V4, 10-Jan-1986 [1,2,EDT] > BASIC-10 V17F [1,2,BASIC] > Kermit [76,5,kermit] > TOPS-10 Customer Supported CUSPs [1,2,140B] > > Better yet the server looks to be sitting on a nice fast connection to the > net. This is a very interesting development in our world :) However, does anyone know the difference in speed between SIMH/TS10 and a REAL TOAD? aak ###### From: hsnewman@blkbox.com (HarrisNewman) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 7 Nov 2001 15:30:48 -0800 Organization: http://groups.google.com/ Lines: 31 Message-ID: References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> NNTP-Posting-Host: 206.109.122.2 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1005175848 16505 127.0.0.1 (7 Nov 2001 23:30:48 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 7 Nov 2001 23:30:48 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!paloalto-snf1.gtei.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!nntp.abs.net!feeder.qis.net!sn-xit-02!supernews.com!postnews1.google.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7424 Arthur Krewat wrote in message news:<3BE975E8.EA0200EB@bartek.dontspamme.net>... > "Zane H. Healy" wrote: > > > > PDP-10@PaulAllen.com wrote: > > > Paul G. Allen, co-founder of Microsoft and PDP-10 enthusiast/developer is > > > creating a new Web site to provide easy-to-use PDP-10 files and > > > documentation. After loading and testing various simh configurations, a DSKB > > > image for simh with a robust configuration, a monitor that supports printing > > > and serial logins and a DSKC file for user storage, have all been posted to > > > the site. > > > > If anyone is wanting to run TOPS-10 7.03 without having to go to the trouble > > of installing it. This seems to give you a *very* nice system. It looks > > like it's got at least the following added to it: > > TOPS-10 Integration Tools, V4, 10-Jan-1986 [1,2,EDT] > > BASIC-10 V17F [1,2,BASIC] > > Kermit [76,5,kermit] > > TOPS-10 Customer Supported CUSPs [1,2,140B] > > > > Better yet the server looks to be sitting on a nice fast connection to the > > net. > > This is a very interesting development in our world :) > > However, does anyone know the difference in speed between SIMH/TS10 and > a REAL TOAD? > > aak Wouldn't that depend on the system running SIMH/TS10/KLH10/KS10? -HN ###### From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Organization: Kilonet.net Lines: 19 Message-ID: <3BEA9801.C8C50583@bartek.dontspamme.net> References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc) X-Accept-Language: en Date: Thu, 08 Nov 2001 14:37:54 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 1005230274 24.186.100.134 (Thu, 08 Nov 2001 09:37:54 EST) NNTP-Posting-Date: Thu, 08 Nov 2001 09:37:54 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cyclone.bc.net!sjcppf01.usenetserver.com!usenetserver.com!news-west.rr.com!news-east.rr.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7419 HarrisNewman wrote: > > Arthur Krewat wrote in message news:<3BE975E8.EA0200EB@bartek.dontspamme.net>... > > > > This is a very interesting development in our world :) > > > > However, does anyone know the difference in speed between SIMH/TS10 and > > a REAL TOAD? > > > > aak > > Wouldn't that depend on the system running SIMH/TS10/KLH10/KS10? Yes, obviously. ;) Sarcasm aside, and following this thread, anyone have any ideas on what a TOAD-1 cycle time is, or any MIPS values? aak ###### Message-ID: <3BEAE151.20A1B2AE@jetnet.ab.ca> From: Ben Franchuk X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.19 i586) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 14 Date: Thu, 08 Nov 2001 12:47:29 -0700 NNTP-Posting-Host: 207.153.6.58 X-Trace: newsfeed.slurp.net 1005273525 207.153.6.58 (Thu, 08 Nov 2001 20:38:45 CST) NNTP-Posting-Date: Thu, 08 Nov 2001 20:38:45 CST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.slurp.net!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7434 Rich Alderson wrote: > Does anyone remember what benchmarks were used to give a 1 MIPS rating to the > KL-10 processor? We could always run them on the Toad-1 to get a comparable > rating for the XKL-1. Confused -- a 1 mip rating is 1,000,000 instructions per second on any CPU. Don't you mean is there a standard benchmark? Don't you have real instruction times for simple instructions for the PDP-10's and can go from there. Ben Franchuk. -- Live "Pre-historic Cpu's" -- and you thought they were extinct. www.jetnet.ab.ca/users/bfranchuk/index.html ###### From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 08 Nov 2001 21:05:44 -0500 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 21 Sender: alderson+news@panix3.panix.com Message-ID: References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> NNTP-Posting-Host: panix3.panix.com X-Trace: news.panix.com 1005271546 8460 166.84.1.3 (9 Nov 2001 02:05:46 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 9 Nov 2001 02:05:46 GMT X-Newsreader: Gnus v5.7/Emacs 20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!193.190.198.17.MISMATCH!newsfeeds.belnet.be!news.belnet.be!news.tele.dk!small.news.tele.dk!144.212.100.101!newsfeed.mathworks.com!panix!news.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7445 Arthur Krewat writes: > Sarcasm aside, and following this thread, anyone have any ideas on what a > TOAD-1 cycle time is, or any MIPS values? Some idea, yes. The basic clock on the XKL-1 CPU is 33MHz, with a syncopated instruction clock. Most instructions take 45ns, while some take 60ns, and a few take 75ns. The XKL-1 is somewhat faster than a KL-10; much greater system throughput on the Toad-1 is provided by the larger memory (minimum 32MW vs. 4MW max on the DEC-20), fast SCSI I/O, and decent Ethernet). Does anyone remember what benchmarks were used to give a 1 MIPS rating to the KL-10 processor? We could always run them on the Toad-1 to get a comparable rating for the XKL-1. -- Rich Alderson alderson+news@panix.com "You get what anybody gets. You get a lifetime." --Death, of the Endless ###### Message-ID: <3BEBFDCD.848F1BB5@srv.net> From: Kevin Handy X-Mailer: Mozilla 4.74 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> <3BEAE151.20A1B2AE@jetnet.ab.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 20 X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Fri, 09 Nov 2001 11:00:57 EST Organization: WebUseNet Corp. http://corp.webusenet.com - ReInventing the UseNet Date: Fri, 09 Nov 2001 09:01:17 -0700 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!howland.erols.net!torn!newsfeed.telusplanet.net!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7456 Ben Franchuk wrote: > > Rich Alderson wrote: > > Does anyone remember what benchmarks were used to give a 1 MIPS rating to the > > KL-10 processor? We could always run them on the Toad-1 to get a comparable > > rating for the XKL-1. > > Confused -- a 1 mip rating is 1,000,000 instructions per second on any > CPU. I believe that the 1mip rating was usually relative to the speed of a VAX 11/780, which was considered to be the standard 1 MIP machine. Due to differences in instruction sets, RISC/CISC, cache, etc. the number given was relative to the 780 in order to make it less "Meaningless Information Provided". > Don't you mean is there a standard benchmark? Don't you have > real instruction times for simple instructions for the PDP-10's and can > go > from there. Ben Franchuk. ###### From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Organization: Kilonet.net Lines: 27 Message-ID: <3BEC022B.9B6D9FF4@bartek.dontspamme.net> References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc) X-Accept-Language: en Date: Fri, 09 Nov 2001 16:22:54 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 1005322974 24.186.100.134 (Fri, 09 Nov 2001 11:22:54 EST) NNTP-Posting-Date: Fri, 09 Nov 2001 11:22:54 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!nntp-relay.ihug.net!ihug.co.nz!telocity-west!TELOCITY!gestalt.direcpc.com!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7452 Kevin Handy wrote: > > Ben Franchuk wrote: > > > > Rich Alderson wrote: > > > Does anyone remember what benchmarks were used to give a 1 MIPS rating to the > > > KL-10 processor? We could always run them on the Toad-1 to get a comparable > > > rating for the XKL-1. > > > > Confused -- a 1 mip rating is 1,000,000 instructions per second on any > > CPU. > > I believe that the 1mip rating was usually relative to the speed of a > VAX 11/780, which was considered to be the standard 1 MIP machine. > Due to differences in instruction sets, RISC/CISC, cache, etc. the > number given was relative to the 780 in order to make it less > "Meaningless Information Provided". The 780 standard is a "VUP" ... 11/780 = 1 VUP, 11/750 = ~.5 VUP etc. etc. I don't remember ever seeing VUP's compared to MIP's. I think the original poster was refering to the fact that the KL10 was supposed to be 1 MIP, and was asking what benchmark they were using to come up with that number. If that could be found, we could run it on the simulators/emulators and see what we come up with :) aak ###### From: peter@taronga.com (Peter da Silva) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 9 Nov 2001 17:03:18 GMT Organization: TSS Inc. Lines: 18 Message-ID: <9sh28m$dfr$1@citadel.in.taronga.com> References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> NNTP-Posting-Host: citadel.in.taronga.com X-Trace: citadel.in.taronga.com 1005325398 13819 10.0.0.43 (9 Nov 2001 17:03:18 GMT) X-Complaints-To: usenet@taronga.com NNTP-Posting-Date: 9 Nov 2001 17:03:18 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!64.245.249.19.MISMATCH!dfw3-feed1.news.digex.net!sfo2-feed1.news.digex.net!intermedia!newsfeed.direct.ca!look.ca!newshub2.rdc1.sfba.home.com!news.home.com!enews.sgi.com!news.spies.com!news.kjsl.com!news.usenet2.org!citadel.in.taronga.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7447 In article <3BEC022B.9B6D9FF4@bartek.dontspamme.net>, Arthur Krewat wrote: >The 780 standard is a "VUP" ... 11/780 = 1 VUP, 11/750 = ~.5 VUP >etc. etc. I don't remember ever seeing VUP's compared to MIP's. The term "VUP" is a nauseating piece of historical revisionism. The VAX was the benchmark "1 MIPS" computer for years, until someone pointed out that on better benchmarks the VAX got less than "1 MIPS", so the term "VAX MIPS" was coined. Apparently someone thought "VAX MIPS" was derogatory, so "VUP" started being used. -- Rev. Peter da Silva, ULC. WWFD? "Be conservative in what you generate, and liberal in what you accept" -- Matthew 10:16 (l.trans) ###### From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Organization: Kilonet.net Lines: 23 Message-ID: <3BEC16BA.170A0842@bartek.dontspamme.net> References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc) X-Accept-Language: en Date: Fri, 09 Nov 2001 17:47:53 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 1005328073 24.186.100.134 (Fri, 09 Nov 2001 12:47:53 EST) NNTP-Posting-Date: Fri, 09 Nov 2001 12:47:53 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!howland.erols.net!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7450 Peter da Silva wrote: > > In article <3BEC022B.9B6D9FF4@bartek.dontspamme.net>, > Arthur Krewat wrote: > >The 780 standard is a "VUP" ... 11/780 = 1 VUP, 11/750 = ~.5 VUP > >etc. etc. I don't remember ever seeing VUP's compared to MIP's. > > The term "VUP" is a nauseating piece of historical revisionism. The > VAX was the benchmark "1 MIPS" computer for years, until someone pointed > out that on better benchmarks the VAX got less than "1 MIPS", so the > term "VAX MIPS" was coined. > > Apparently someone thought "VAX MIPS" was derogatory, so "VUP" started > being used. Thanks for correcting me - in all the time I've worked with DEC equipment, I've never heard that a 780 was 1 MIP, and I started around the time the 780 came out ... I do, however, remember a lot of animosity at the time when the term "VUP" was coined... aak ###### From: "Zane H. Healy" Subject: Re: New web and ftp site for PDP-10 Newsgroups: alt.sys.pdp10 References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 7 Message-ID: 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: Fri, 09 Nov 2001 12:48:18 EST Date: Fri, 09 Nov 2001 17:48:18 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!64.245.249.19.MISMATCH!dfw3-feed1.news.digex.net!sfo2-feed1.news.digex.net!intermedia!news-out.spamkiller.net!propagator-la!news-in-la.newsfeeds.com!news-in.superfeed.net!hub1.nntpserver.com!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7457 Arthur Krewat wrote: > The 780 standard is a "VUP" ... 11/780 = 1 VUP, 11/750 = ~.5 VUP > etc. etc. I don't remember ever seeing VUP's compared to MIP's. Isn't it also 1 SpecInt92 and 1 SpecFP92? Zane ###### Newsgroups: alt.sys.pdp10 From: rivie@cougar.no.domain Subject: Re: New web and ftp site for PDP-10 References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> Reply-To: ivie@cc.usu.edu Message-ID: User-Agent: slrn/0.9.7.1 (NetBSD) Cache-Post-Path: hoth!unknown@208.186.13.136 X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) NNTP-Posting-Host: 208.186.8.4 Date: 9 Nov 2001 12:44:17 -0600 X-Trace: corp.newsgroups.com 1005331457 208.186.8.4 (9 Nov 2001 12:44:17 -0600) Lines: 19 X-Comments: This message was posted through Newsfeeds.com X-Comments2: IMPORTANT: Newsfeeds.com does not condone, nor support, spam or any illegal or copyrighted postings. X-Comments3: IMPORTANT: Under NO circumstances will postings containing illegal or copyrighted material through this service be tolerated!! X-Report: Please report illegal or inappropriate use to X-Abuse-Info: Please be sure to forward a copy of ALL headers, INCLUDING the body (DO NOT SEND ATTACHMENTS) Organization: Newsfeeds.com http://www.newsfeeds.com 80,000+ UNCENSORED Newsgroups. Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!nntp-relay.ihug.net!ihug.co.nz!out.nntp.be!propagator-SanJose!in.nntp.be!corp.newsgroups.com Xref: chonsp.franklin.ch alt.sys.pdp10:7455 In article <3BEC16BA.170A0842@bartek.dontspamme.net>, Arthur Krewat wrote: > > Thanks for correcting me - in all the time I've worked with DEC equipment, > I've never heard that a 780 was 1 MIP, and I started around the time the > 780 came out ... My dusty grey cells claim that the 780 wasn't 1 MIPS, it just performed as well on some whacky benchmark as a 1 MIPS IBM mainframe, so was widely accepted as being a 1 MIPS machine. Some time later, it became the de-facto industry standard 1 MIPS machine. -- Roger Ivie ivie@cc.usu.edu -----= Posted via Newsfeeds.Com, Uncensored Usenet News =----- http://www.newsfeeds.com - The #1 Newsgroup Service in the World! -----== Over 80,000 Newsgroups - 16 Different Servers! =----- ###### From: Tim Shoppa Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 9 Nov 2001 12:53:09 -0800 Organization: Trailing-Edge Technologies Lines: 34 Message-ID: <9shfnl04o7@drn.newsguy.com> References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> NNTP-Posting-Host: p-634.newsdawg.com X-Newsreader: Direct Read News 2.90 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!drn Xref: chonsp.franklin.ch alt.sys.pdp10:7435 In article <3BEC16BA.170A0842@bartek.dontspamme.net>, Arthur says... > >Peter da Silva wrote: >> >> In article <3BEC022B.9B6D9FF4@bartek.dontspamme.net>, >> Arthur Krewat wrote: >> >The 780 standard is a "VUP" ... 11/780 = 1 VUP, 11/750 = ~.5 VUP >> >etc. etc. I don't remember ever seeing VUP's compared to MIP's. >> >> The term "VUP" is a nauseating piece of historical revisionism. The >> VAX was the benchmark "1 MIPS" computer for years, until someone pointed >> out that on better benchmarks the VAX got less than "1 MIPS", so the >> term "VAX MIPS" was coined. >> >> Apparently someone thought "VAX MIPS" was derogatory, so "VUP" started >> being used. > >Thanks for correcting me - in all the time I've worked with DEC equipment, >I've never heard that a 780 was 1 MIP, and I started around the time the >780 came out ... > >I do, however, remember a lot of animosity at the time when the term "VUP" >was coined... If it helps any, I *do* have a copy of the Digital Review (a DEC/VAX magazine) VUP test suite somewhere around here. IIRC most of the benchmarks are in Fortran, and some may be portable enough to run on a -10's Fortran. I also have some old issues of _Creative Computing_ (from the mid-70's) with BASIC benchmarks. Most of them are on PDP-8 (EDUSYSTEM-xxx) systems, but I think at least some of them are on -10's. Tim. ###### From: "Zane H. Healy" Subject: Re: New web and ftp site for PDP-10 Newsgroups: alt.sys.pdp10 References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 13 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Fri, 09 Nov 2001 17:42:54 EST Date: Fri, 09 Nov 2001 22:42:54 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone.bc.net!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7461 Tim Shoppa wrote: > If it helps any, I *do* have a copy of the Digital Review (a DEC/VAX > magazine) VUP test suite somewhere around here. IIRC most of the benchmarks > are in Fortran, and some may be portable enough to run on a -10's Fortran. > I also have some old issues of _Creative Computing_ (from the mid-70's) > with BASIC benchmarks. Most of them are on PDP-8 (EDUSYSTEM-xxx) systems, > but I think at least some of them are on -10's. These would be interesting to see. Any chance of getting them scanned in? ane ###### From: gah@ugcs.caltech.edu (glen herrmannsfeldt) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 10 Nov 2001 01:34:35 GMT Organization: California Institute of Technology, Pasadena Lines: 21 Message-ID: <9si07b$mm4@gap.cco.caltech.edu> References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> NNTP-Posting-Host: yak.ugcs.caltech.edu X-Newsreader: NN version 6.5.0 #1 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!newsmi-eu.news.garr.it!newsmi-us.news.garr.it!NewsITBone-GARR!canoe.uoregon.edu!logbridge.uoregon.edu!nntp-server.caltech.edu!gah Xref: chonsp.franklin.ch alt.sys.pdp10:7439 Tim Shoppa writes: (snip about VAX MIPS) >If it helps any, I *do* have a copy of the Digital Review (a DEC/VAX >magazine) VUP test suite somewhere around here. IIRC most of the benchmarks >are in Fortran, and some may be portable enough to run on a -10's Fortran. >I also have some old issues of _Creative Computing_ (from the mid-70's) >with BASIC benchmarks. Most of them are on PDP-8 (EDUSYSTEM-xxx) systems, >but I think at least some of them are on -10's. Is there a C compiler to generate PDP-10 code? If so, could the usual C benchmark programs get ported, so we could compare them? I don't suppose that SPEC would be interested, though. The PDP-10 is an interesting machine to ask about a C compiler. What is sizeof(int)? What is char? -- glen ###### Message-ID: <3BED3251.EA621B89@bellatlantic.net> From: bad bob X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 36 Date: Sat, 10 Nov 2001 13:54:02 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 1005400442 138.88.75.218 (Sat, 10 Nov 2001 08:54:02 EST) NNTP-Posting-Date: Sat, 10 Nov 2001 08:54:02 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon1.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7440 Peter da Silva wrote: > > In article <3BEC022B.9B6D9FF4@bartek.dontspamme.net>, > Arthur Krewat wrote: > >The 780 standard is a "VUP" ... 11/780 = 1 VUP, 11/750 = ~.5 VUP > >etc. etc. I don't remember ever seeing VUP's compared to MIP's. > > The term "VUP" is a nauseating piece of historical revisionism. The > VAX was the benchmark "1 MIPS" computer for years, until someone pointed > out that on better benchmarks the VAX got less than "1 MIPS", so the > term "VAX MIPS" was coined. > > Apparently someone thought "VAX MIPS" was derogatory, so "VUP" started > being used. What I recall, and I may be a little rusty here, was the KL was about 1.1 VUP with the same benchmark. Some of us thought the standard should be the KL, thus the vax would be somewhat less than 1 standard unit of performance on that same benchmark. But we were a bit prejudiced. The real measure in many minds was how many concurrent users could be on the system doing meaningful work. This spawned a whole series of endeavors to create a single user test that would eat the whole machine. We did the same thing when comparing 11s to 8s, came up with a neat little basic lang matrix problem that ate rsts (hence the nick name real shitty time sharing) and tweaked things to eat vms and vax cycles .... ahh, the good days of competition!! bob > > -- > Rev. Peter da Silva, ULC. WWFD? > > "Be conservative in what you generate, and liberal in what you accept" > -- Matthew 10:16 (l.trans) ###### From: peter@taronga.com (Peter da Silva) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 10 Nov 2001 14:27:34 GMT Organization: TSS Inc. Lines: 29 Message-ID: <9sjdgm$1m35$1@citadel.in.taronga.com> References: <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BED3251.EA621B89@bellatlantic.net> NNTP-Posting-Host: citadel.in.taronga.com X-Trace: citadel.in.taronga.com 1005402454 55397 10.0.0.43 (10 Nov 2001 14:27:34 GMT) X-Complaints-To: usenet@taronga.com NNTP-Posting-Date: 10 Nov 2001 14:27:34 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news.kjsl.com!news.usenet2.org!citadel.in.taronga.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7451 In article <3BED3251.EA621B89@bellatlantic.net>, bad bob wrote: >The real measure in many minds was how >many concurrent users could be on the system doing meaningful >work. The 11/70 was about 1 VUP, and could handle twice the user load of an 11/780. Even when they were both running the same OS (Version 7 UNIX). We had a couple of RSTS boxes around. RSTS could support about four times the users as Version 7 UNIX with the same memory, but it wouldn't run the typesetting software everyone at UCB was using to write their papers... so they talked everyone into springing for more core and got rid of Really Shitty TimeSharing (yes, we used that expansion too). I was in the computer club at the time, doing a little scutwork on 4.1C BSD and helping liberal arts undergrads in exchange for a whole extra 300k quota on my computer club account. The business school refused to drop RSTS until the undergrad engineering students came up with a hack to run Basic+ under Version 7. They had a little illegal instruction trap that caught the IOT instructions RSTS used to make system calls, and emulated them via TRAP instructions which Version 7 used. -- Rev. Peter da Silva, ULC. WWFD? "Be conservative in what you generate, and liberal in what you accept" -- Matthew 10:16 (l.trans) ###### Message-ID: <3BED4485.6B52AE2A@bellatlantic.net> From: bad bob X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BED3251.EA621B89@bellatlantic.net> <9sjdgm$1m35$1@citadel.in.taronga.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 46 Date: Sat, 10 Nov 2001 15:11:40 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon2.gnilink.net 1005405100 138.88.75.218 (Sat, 10 Nov 2001 10:11:40 EST) NNTP-Posting-Date: Sat, 10 Nov 2001 10:11:40 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon2.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7441 so I don't drive Barb up the tree like I did in my last response c below Peter da Silva wrote: > > In article <3BED3251.EA621B89@bellatlantic.net>, > bad bob wrote: > >The real measure in many minds was how > >many concurrent users could be on the system doing meaningful > >work. > > The 11/70 was about 1 VUP, and could handle twice the user load of an > 11/780. Even when they were both running the same OS (Version 7 UNIX). > > We had a couple of RSTS boxes around. RSTS could support about four times > the users as Version 7 UNIX with the same memory, but it wouldn't run the > typesetting software everyone at UCB was using to write their papers... > so they talked everyone into springing for more core and got rid of Really > Shitty TimeSharing (yes, we used that expansion too). One of the things we looked at very closely was interrupt latency. Some systems had VERY long, relatively, response times. The inherent ISR (interrupt service routine - see I remmember) requirements for some OSs were huge - the hardware may have been designed for rapid context switching and quick response but the OS was not..hint DC.. > > I was in the computer club at the time, doing a little scutwork on 4.1C BSD > and helping liberal arts undergrads in exchange for a whole extra 300k > quota on my computer club account. > > The business school refused to drop RSTS until the undergrad engineering > students came up with a hack to run Basic+ under Version 7. They had a little > illegal instruction trap that caught the IOT instructions RSTS used to make > system calls, and emulated them via TRAP instructions which Version 7 used. aww man, we came up with that simple matrix inversion problem in basic for rsts that made it seem to lock up with one user...with two it went to lunch, with 3 or more it was slower than a vt78!! > > -- > Rev. Peter da Silva, ULC. WWFD? > > "Be conservative in what you generate, and liberal in what you accept" > -- Matthew 10:16 (l.trans) ###### From: Alan Greig Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: Mon, 12 Nov 2001 09:14:15 +0000 Organization: http://extra.newsguy.com Lines: 30 Message-ID: References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> NNTP-Posting-Host: p-218.newsdawg.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!logbridge.uoregon.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews1 Xref: chonsp.franklin.ch alt.sys.pdp10:7482 On Fri, 09 Nov 2001 16:22:54 GMT, Arthur Krewat wrote: >The 780 standard is a "VUP" ... 11/780 = 1 VUP, 11/750 = ~.5 VUP >etc. etc. I don't remember ever seeing VUP's compared to MIP's. > I do as I did extensive tests of a VAX 11/780 against a TOPS-20 KL-10 2060 (serial 2172). I'd put the 2060 at around 5 VUPS, That is five times as fast as the VAX 11/780 which was supposed to replace it. When it came to actual interactive system performance as experienced by the end user the 20 felt more like 10 VUPS as TOPS-10 and TOPS-20 didn't seem to have to hammer disk I/O as much to run a typical task. For example COPY FILE1.TXT FILE2.TXT would cause an image activation on VMS but not TOPS-20. Same for DIR etc. Also typically TOPS-10/20 applications did not have the RMS overhead usually incurred by VMS apps. The 750 was around .8 VUPS. It was the 730 (or was there a 725?) which was around .5 VUP. As has been pointed out you couldn't exactly compare the instruction set timings but the above figures worked for me. > >aak -- Alan ###### From: Alan Greig Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: Mon, 12 Nov 2001 09:17:52 +0000 Organization: http://extra.newsguy.com Lines: 14 Message-ID: References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BED3251.EA621B89@bellatlantic.net> NNTP-Posting-Host: p-414.newsdawg.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews1 Xref: chonsp.franklin.ch alt.sys.pdp10:7493 On Sat, 10 Nov 2001 13:54:02 GMT, bad bob wrote: > >What I recall, and I may be a little rusty here, was the KL was about >1.1 VUP with the same benchmark. Some of us thought the standard No way! The KS was about 1.1 VUP. Perhaps an early 2040 with too little (core memory and no cache would have performed at around 1.1VUP on a benchmark requiring lots of memory. -- Alan ###### Message-ID: <3BEFF23D.1F9B8FB1@trailing-edge.com> Date: Mon, 12 Nov 2001 16:01:01 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 14 NNTP-Posting-Host: 63.73.218.130 X-Trace: 1005598866 reader1.ash.ops.us.uu.net 28429 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news2.euro.net!uunet!ash.uu.net!spool0900.news.uu.net!reader0901.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7526 Alan Greig wrote: > The 750 was around .8 VUPS. It was the 730 (or was there a 725?) > which was around .5 VUP. The 11/730 and 11/725 used the same CPU, but had different I/O systems. I think the 11/725 (which came with RC25 drives, the drives were so fragile the owners guide warned that moving them from one side of the room to the other without putting the shipping cartridge in would crash them) was ranked just a few percentage points below the 11/730 (which probably was tested with a R80 system disk - could you buy it with only RL02's?) Tim. ###### From: "Zane H. Healy" Subject: Re: New web and ftp site for PDP-10 Newsgroups: alt.sys.pdp10 References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 56 Message-ID: <5DXH7.57031$Lg.3122605@sjcpnn01.usenetserver.com> 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: Mon, 12 Nov 2001 16:59:29 EST Date: Mon, 12 Nov 2001 21:59:29 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!hub1.nntpserver.com!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7574 Alan Greig wrote: > The 750 was around .8 VUPS. It was the 730 (or was there a 725?) > which was around .5 VUP. Since I'm seeing a various numbers listed, here is a partial chart (the full chart can be found at: http://anacin.nsc.vcu.edu/~jim/mvax/vax-perf.html ). I've also added the SPECint92/SPECfp92 VAX-11/780 entry at the bottom. Zane VAX CPUs -----+---+-----+-------+--------+--------------------------------+-------------- SID | X | Id | Speed | Bus | Model Name | Nickname -----+---+-----+-------+--------+--------------------------------+-------------- ---- 700 series (1977) +--------+--------------------------------+-------------- 01 | - | 780 | 1.0 | U,M,C | VAX-11/780 | Star 01 | - | 780 | 1.8 | U,M,C | VAX-11/782 | Atlas 01 | - | 780 | 3.5 | U,M,C | VAX-11/784 | VAXimus 01 | - | 780 | 1.5 | U,M,C | VAX-11/785 | Superstar 02 | - | 750 | 0.6 | U,M,C | VAX-11/750 | Comet 03 | - | 730 | 0.3 | U | VAX-11/730, 725 | Nebula, LCN 04 | - | 790 | 4.0 | U,M,C | VAX 8600, | Venus 04 | - | 790 | 7.0 | U,M,C | VAX 8650 | Morningstar ---- 8000 series (1986)+--------+--------------------------------+-------------- 05 | - | 8SS | 0.9-2 | B,C | VAX 8200, 8300, 8250, 8350 | Scorpio 05 | - | 8SS | 0.9-2 | B,C | VAXstation 8000 | Lynx 06 | - | 8NN | 3 | B,C | VAX 8500 | Flounder 06 | - | 8NN | 4/6 | B,C | VAX 8530, 8550 | Skipjack 06 | - | 8NN | 6/12 | B,C | VAX 8700, 8800 | Nautilus ---- MicroVAX I (1984) - Decimal SID = 117440512 ----------------+-------------- 07 | - | UV1 | 0.3 | Q | MicroVAX I, VAXstation I | Seahorse ---- MicroVAX II series (1985) - Decimal SID = 134217728 --------+-------------- 08 | 1 | UV2 | 0.9 | Q | MicroVAX II,VAXstation II | Mayflower 08 | 1 | UV2 | 0.9 | Q | VAXstation II/GPX | Caylith 08 | 4 | 410 | 0.9 | none | MicroVAX 2000 | TeamMate 08 | 4 | 410 | 0.9 | none | VAXstation 2000 | VAXstar -----+---+-----+-------+--------+--------------------------------+-------------- Also: System CPU ClkMHz Cache SPECint SPECfp Info Source Name (NUMx)Type ext/in Ext+I/D base92 base92 Date Obtained ================= ========== ======= ========== ======= ======= ===== ========= DEC VAX11/780 VAX 5 8 1.0 1.0 Jan89 SPEC Ref ###### Sender: prep@k9 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <1ceG7.34203$lQ2.1381482@sjcpnn01.usenetserver.com> <3BE975E8.EA0200EB@bartek.dontspamme.net> <3BEA9801.C8C50583@bartek.dontspamme.net> <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <3BEFF23D.1F9B8FB1@trailing-edge.com> From: Paul Repacholi Message-ID: <87k7wv8mms.fsf@prep.synonet.com> Lines: 22 User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Organization: iQnet Cache-Post-Path: angelina.pe!unknown@p229.perth02.dial.usertools.net X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/) Cache-Post-Path: news.satix.net!unknown@203.132.92.2 X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/) NNTP-Posting-Host: news.satix.net X-Trace: ozemail.com.au 1005618181 203.132.96.3 (Tue, 13 Nov 2001 13:23:01 EST) NNTP-Posting-Date: Tue, 13 Nov 2001 13:23:01 EST Distribution: world Date: 13 Nov 2001 09:50:35 +0800 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!hub1.nntpserver.com!nntp-relay.ihug.net!ihug.co.nz!newsfeed.berkeley.edu!ucberkeley!sjc1.nntp.concentric.net!newsfeed.concentric.net!newsfeed.ozemail.com.au!ozemail.com.au!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7531 Tim Shoppa writes: > I think the 11/725 (which came with RC25 drives, the drives were so > fragile the owners guide warned that moving them from one side of > the room to the other without putting the shipping cartridge in > would crash them) was ranked just a few percentage points below the > 11/730 (which probably was tested with a R80 system disk - could you > buy it with only RL02's?) Sad thing was, the prototype used a UDA-50, not the Idiot Dumb C... It ran really quite nicely for its age. The real one arrived. Oh groan. The 725 was a 730 re-packaged and compactified. Pity about the RC-25s and the heads... -- Paul Repacholi 1 Crescent Rd., +61 (08) 9257-1001 Kalamunda. West Australia 6076 Raw, Cooked or Well-done, it's all half baked. EPIC, The Architecture of the future, always has been, always will be. ###### From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 12 Nov 2001 22:20:56 -0500 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 19 Sender: alderson+news@panix1.panix.com Message-ID: References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> NNTP-Posting-Host: panix1.panix.com X-Trace: news.panix.com 1005621656 20059 166.84.1.1 (13 Nov 2001 03:20:56 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 13 Nov 2001 03:20:56 GMT X-Newsreader: Gnus v5.7/Emacs 20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!nycmny1-snh1.gtei.net!news.gtei.net!panix!news.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7536 gah@ugcs.caltech.edu (glen herrmannsfeldt) writes: > Is there a C compiler to generate PDP-10 code? Yes, KCC (Kok Chen's Compiler). There is work under way on a GCC port. > If so, could the usual C benchmark programs get ported, so we could compare > them? I don't suppose that SPEC would be interested, though. The current C libraries are a crock, and artificially slow the system greatly. > The PDP-10 is an interesting machine to ask about a C compiler. What is > sizeof(int)? What is char? 36 and 8, usually. ###### From: gah@ugcs.caltech.edu (glen herrmannsfeldt) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 13 Nov 2001 04:42:43 GMT Organization: California Institute of Technology, Pasadena Lines: 20 Message-ID: <9sq8c3$guo@gap.cco.caltech.edu> References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> NNTP-Posting-Host: yak.ugcs.caltech.edu X-Newsreader: NN version 6.5.0 #1 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!193.190.198.17.MISMATCH!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!logbridge.uoregon.edu!nntp-server.caltech.edu!gah Xref: chonsp.franklin.ch alt.sys.pdp10:7525 Rich Alderson writes: >gah@ugcs.caltech.edu (glen herrmannsfeldt) writes: >> Is there a C compiler to generate PDP-10 code? >Yes, KCC (Kok Chen's Compiler). There is work under way on a GCC port. (snip) >> The PDP-10 is an interesting machine to ask about a C compiler. What is >> sizeof(int)? What is char? >36 and 8, usually. sizeof(int) is how many char's fit into one int. I think it has to be an integer. That would leave 1, 2, 3, or 4 as choices, for 36 bit, 18 bit, 12 bit, or 9 bit char's. I suppose it doesn't matter much as long as they are converted when writing a text file. Being word addressable, char pointers would be pretty strange. -- glen ###### From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 13 Nov 2001 12:54:32 -0500 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 27 Sender: alderson+news@panix1.panix.com Message-ID: References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sq8c3$guo@gap.cco.caltech.edu> NNTP-Posting-Host: panix1.panix.com X-Trace: news.panix.com 1005674072 3590 166.84.1.1 (13 Nov 2001 17:54:32 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 13 Nov 2001 17:54:32 GMT X-Newsreader: Gnus v5.7/Emacs 20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!panix!news.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7529 gah@ugcs.caltech.edu (glen herrmannsfeldt) writes: > Rich Alderson writes: >>> The PDP-10 is an interesting machine to ask about a C compiler. What is >>> sizeof(int)? What is char? >> 36 and 8, usually. > sizeof(int) is how many char's fit into one int. I think it has to be an > integer. That would leave 1, 2, 3, or 4 as choices, for 36 bit, 18 bit, 12 > bit, or 9 bit char's. I suppose it doesn't matter much as long as they are > converted when writing a text file. You can tell that I've never been a C programmer, can't you? I should probably learn the language one of these days. :-> As far as I can tell, the internal representation of characters is as 8-bit bytes in the 36-bit word, so sizeof(int) would then be 4. > Being word addressable, char pointers would be pretty strange. Right, they're not equivalent to array indices. -- Rich Alderson alderson+news@panix.com "You get what anybody gets. You get a lifetime." --Death, of the Endless ###### Sender: prep@k9 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sq8c3$guo@gap.cco.caltech.edu> From: Paul Repacholi Message-ID: <874rnytu14.fsf@prep.synonet.com> Lines: 21 User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Organization: iQnet Cache-Post-Path: angelina.pe!unknown@p041.perth01.dial.usertools.net X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/) Cache-Post-Path: news.satix.net!unknown@203.132.92.2 X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/) NNTP-Posting-Host: news.satix.net X-Trace: ozemail.com.au 1005679498 203.132.96.3 (Wed, 14 Nov 2001 06:24:58 EST) NNTP-Posting-Date: Wed, 14 Nov 2001 06:24:58 EST Distribution: world Date: 14 Nov 2001 02:17:27 +0800 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newshub2.rdc1.sfba.home.com!news.home.com!sjc1.nntp.concentric.net!newsfeed.concentric.net!newsfeed.ozemail.com.au!ozemail.com.au!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7528 gah@ugcs.caltech.edu (glen herrmannsfeldt) writes: > sizeof(int) is how many char's fit into one int. Um, size in address units of an int I think. 1, for all non-MS values of 1. > I think it has to be an integer. That would leave 1, 2, 3, or 4 as > choices, for 36 bit, 18 bit, 12 bit, or 9 bit char's. I suppose it > doesn't matter much as long as they are converted when writing a > text file. > Being word addressable, char pointers would be pretty strange. Sizeof and pointers take one a whole new world of meaning. How portable is your C code ;) -- Paul Repacholi 1 Crescent Rd., +61 (08) 9257-1001 Kalamunda. West Australia 6076 Raw, Cooked or Well-done, it's all half baked. EPIC, The Architecture of the future, always has been, always will be. ###### Message-ID: <3BF199DD.6F1C8017@trailing-edge.com> Date: Tue, 13 Nov 2001 22:08:29 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sq8c3$guo@gap.cco.caltech.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 40 NNTP-Posting-Host: 63.73.218.130 X-Trace: 1005707304 reader2.ash.ops.us.uu.net 11569 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!newsfeeds.belnet.be!news.belnet.be!newsgate.cistron.nl!news2.euro.net!uunet!ash.uu.net!spool0900.news.uu.net!reader0902.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7524 Rich Alderson wrote: > As far as I can tell, the internal representation of characters is as 8-bit > bytes in the 36-bit word, so sizeof(int) would then be 4. This is less than perfectly satisfying, as a 36-bit int has more information in it than 4 8-bit bytes. (Never mind that a pointer to char would be completely incompatible with a pointer to int so that it is silly to make such a comparison anyway!) The ANSI C Rationale (see http://www.lysator.liu.se/c/rat/a.html#1-6 ) drives my beliefs, saying on this subject The definitions of object, bit, byte, and alignment reflect a strong consensus, reached after considerable discussion, about the fundamental nature of the memory organization of a C environment: * All objects in C must be representable as a contiguous sequence of bytes, each of which is at least 8 bits wide. * A char (or signed char or unsigned char) occupies exactly one byte. (Thus, for instance, on a machine with 36-bit words, a byte can be defined to consist of 9, 12, 18, or 36 bits, these numbers being all the exact divisors of 36 which are not less than 8.) A comment of my own: I'm not perfectly satisfied with the idea that even pointers must be representable as a contiguous sequence of bytes. So I don't fully agree with the rationale myself. A question of my own: do any machines have word sizes that are prime numbers? Tim. ###### From: klh@panix.com (Ken Harrenstien) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 14 Nov 2001 03:33:19 -0500 Organization: PANIX Public Access Internet and UNIX, NYC Lines: 55 Sender: Ken Harrenstien Message-ID: <9sta8f$fo9$1@panix2.panix.com> References: <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> NNTP-Posting-Host: panix2.panix.com X-Trace: news.panix.com 1005726800 20366 166.84.1.2 (14 Nov 2001 08:33:20 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 14 Nov 2001 08:33:20 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!colt.net!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!nycmny1-snh1.gtei.net!news.gtei.net!panix!news.panix.com!panix2.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7513 In article , Rich Alderson wrote: >gah@ugcs.caltech.edu (glen herrmannsfeldt) writes: > >> Is there a C compiler to generate PDP-10 code? > >Yes, KCC (Kok Chen's Compiler). There is work under way on a GCC port. That was the initial derivation of KCC, yes, but I want to point out that KCC as it currently exists has virtually nothing to do with Kok Chen. Various people at Stanford and Columbia worked on it before I rewrote it completely into an ANSI C compiler and (with Ian Macky) did both a C and Unix syscall emulation library. The math library I think is derived from a portable C package from Fred Fish. It runs on both TOPS-10 and TOPS-20, extended or non-extended, producing either extended or non-extended code for any of the PDP-10 models. I don't know whether anyone has really maintained it since I left SRI other than XKL and CompuServe. Lars Brinkhoff is using KCC in some form to help with his GCC port. >> If so, could the usual C benchmark programs get ported, so we could compare >> them? I don't suppose that SPEC would be interested, though. > >The current C libraries are a crock, and artificially slow the system greatly. Well, gee, thanks. I agree they are not for performance, though. They are in C for portability (KCC compiles itself) and the point was to be able to import software, not run it fast; even at the time KCC was written, Sun workstations could do many things faster than a KL. But by the same token, it *should* be possible to port the benchmarks relatively easily, just for amusement value. >> The PDP-10 is an interesting machine to ask about a C compiler. What is >> sizeof(int)? What is char? > >36 and 8, usually. For KCC, sizeof(int) is 4, sizeof(char) is 1. Those correspond to 36 and 9 bits respectively. It does have extensions that allow you to use special types of 6, 7, 8, and 18 bits. "short" is 18 bits, "long" is 36. "float" and "double" are 36 and 72. As you might suspect, C code must follow the Standard's pointer conversion rules quite strictly in order to run without incident on the PDP-10, and anything that (unportably) assumes a char is always 8 bits will fail. If there is interest, I could pull my KCC directories back off tape, or post the KCC doc file; the two companies I mentioned should be canvassed for bug fixes or updates though. Or wait for Lars, although it looks like GCC will only run on machines (real or virtual) with extended addressing. --Ken ###### From: Erno Palonheimo Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 14 Nov 2001 17:41:17 +0200 Organization: Helsinki University of Technology, Finland Lines: 15 Message-ID: References: <3BEAE151.20A1B2AE@jetnet.ab.ca> <3BEBFDCD.848F1BB5@srv.net> <3BEC022B.9B6D9FF4@bartek.dontspamme.net> <9sh28m$dfr$1@citadel.in.taronga.com> <3BEC16BA.170A0842@bartek.dontspamme.net> <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sq8c3$guo@gap.cco.caltech.edu> <3BF199DD.6F1C8017@trailing-edge.com> NNTP-Posting-Host: kosh.hut.fi X-Trace: nntp.hut.fi 1005752477 29269 130.233.228.10 (14 Nov 2001 15:41:17 GMT) X-Complaints-To: usenet@nntp.hut.fi NNTP-Posting-Date: 14 Nov 2001 15:41:17 GMT X-Newsreader: Gnus v5.7/Emacs 20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!193.190.198.17.MISMATCH!newsfeeds.belnet.be!news.belnet.be!newsfeed00.sul.t-online.de!t-online.de!nntp-relay.ihug.net!ihug.co.nz!newsfeed.media.kyoto-u.ac.jp!news.cc.tut.fi!news.helsinki.fi!newsfeed3.funet.fi!newsfeeds.funet.fi!nntp.hut.fi!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7517 Tim Shoppa writes: > A question of my own: do any machines have word sizes that are prime > numbers? Russian-made MIR-2 computer from early '70s had a word size of 13 bits. There aren't too many of them around, though. AFAIK the only unit outside the former Soviet Union is now located at Helsinki University of Technology and they're most likely going to scrap it in near future because of lack of interest in preserving it. It was taken out of service in early '80s. -- Erno Palonheimo Tel: +358 50 560 4765 Unix Systems Administrator Email: esp@iki.fi ###### From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 14 Nov 2001 19:06:32 -0500 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 63 Sender: alderson+news@panix1.panix.com Message-ID: References: <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sta8f$fo9$1@panix2.panix.com> NNTP-Posting-Host: panix1.panix.com X-Trace: news.panix.com 1005782793 13785 166.84.1.1 (15 Nov 2001 00:06:33 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 15 Nov 2001 00:06:33 GMT X-Newsreader: Gnus v5.7/Emacs 20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.mathworks.com!panix!news.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7608 klh@panix.com (Ken Harrenstien) writes: > In article , > Rich Alderson wrote: >> Yes, KCC (Kok Chen's Compiler). There is work under way on a GCC port. > That was the initial derivation of KCC, yes, but I want to point out that KCC > as it currently exists has virtually nothing to do with Kok Chen. Various > people at Stanford and Columbia worked on it before I rewrote it completely > into an ANSI C compiler and (with Ian Macky) did both a C and Unix syscall > emulation library. The math library I think is derived from a portable C > package from Fred Fish. I didn't know the later history of KCC in detail; I only knew the name derived from the original author's. >> The current C libraries are a crock, and artificially slow the system >> greatly. > Well, gee, thanks. I'm sorry if I've given offense, but performance is the measure by which I have to judge compilers and their run-time libraries. > I agree they are not for performance, though. They are in C for portability > (KCC compiles itself) and the point was to be able to import software, not > run it fast; even at the time KCC was written, Sun workstations could do many > things faster than a KL. But by the same token, it *should* be possible to > port the benchmarks relatively easily, just for amusement value. I was concerned about the effect of the non-optimized performance of library code on the benchmarks, but for basic instruction timing benchmarks there are probably not that many library calls. Should probably have kept my mouth shut. >>> The PDP-10 is an interesting machine to ask about a C compiler. What is >>> sizeof(int)? What is char? >> 36 and 8, usually. > For KCC, sizeof(int) is 4, sizeof(char) is 1. Those correspond to 36 and 9 > bits respectively. It does have extensions that allow you to use special > types of 6, 7, 8, and 18 bits. "short" is 18 bits, "long" is 36. "float" > and "double" are 36 and 72. As you might suspect, C code must follow the > Standard's pointer conversion rules quite strictly in order to run without > incident on the PDP-10, and anything that (unportably) assumes a char is > always 8 bits will fail. My experience with KCC-built apps is here at XKL, where 8-bit compatibility was the issue: X11R4/5, GNU Emacs 18.59, and Perl 4.036. Perhaps if they had been made 9-bit-friendly instead of Unix-friendly they'd be faster; in any case, my assumption about the 8-bit-ness of KCC derives from this usage. > If there is interest, I could pull my KCC directories back off tape, or post > the KCC doc file; the two companies I mentioned should be canvassed for bug > fixes or updates though. Or wait for Lars, although it looks like GCC will > only run on machines (real or virtual) with extended addressing. I'm sure there'd be a great deal of interest. -- Rich Alderson alderson+news@panix.com "You get what anybody gets. You get a lifetime." --Death, of the Endless ###### From: gah@ugcs.caltech.edu (glen herrmannsfeldt) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 15 Nov 2001 06:01:32 GMT Organization: California Institute of Technology, Pasadena Lines: 37 Message-ID: <9svlns$4jt@gap.cco.caltech.edu> References: <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sta8f$fo9$1@panix2.panix.com> NNTP-Posting-Host: yak.ugcs.caltech.edu X-Newsreader: NN version 6.5.0 #1 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!nntp-relay.ihug.net!ihug.co.nz!logbridge.uoregon.edu!nntp-server.caltech.edu!gah Xref: chonsp.franklin.ch alt.sys.pdp10:7607 Rich Alderson writes: >klh@panix.com (Ken Harrenstien) writes: >> In article , >> Rich Alderson wrote: >>> Yes, KCC (Kok Chen's Compiler). There is work under way on a GCC port. (snip) >>>> The PDP-10 is an interesting machine to ask about a C compiler. What is >>>> sizeof(int)? What is char? >>> 36 and 8, usually. >> For KCC, sizeof(int) is 4, sizeof(char) is 1. Those correspond to 36 and 9 >> bits respectively. It does have extensions that allow you to use special >> types of 6, 7, 8, and 18 bits. "short" is 18 bits, "long" is 36. "float" >> and "double" are 36 and 72. As you might suspect, C code must follow the >> Standard's pointer conversion rules quite strictly in order to run without >> incident on the PDP-10, and anything that (unportably) assumes a char is >> always 8 bits will fail. >My experience with KCC-built apps is here at XKL, where 8-bit compatibility was >the issue: X11R4/5, GNU Emacs 18.59, and Perl 4.036. Perhaps if they had been >made 9-bit-friendly instead of Unix-friendly they'd be faster; in any case, my >assumption about the 8-bit-ness of KCC derives from this usage. Where does the 8 bit problem come in? Programs converting int to char and expecting it to be mod 256 instead of 512? >> If there is interest, I could pull my KCC directories back off tape, or post >> the KCC doc file; the two companies I mentioned should be canvassed for bug >> fixes or updates though. Or wait for Lars, although it looks like GCC will >> only run on machines (real or virtual) with extended addressing. Or run as a cross compiler. gcc is usually pretty good at that. -- glen ###### From: klh@panix.com (Ken Harrenstien) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 15 Nov 2001 03:42:20 -0500 Organization: PANIX Public Access Internet and UNIX, NYC Lines: 34 Sender: Ken Harrenstien Message-ID: <9svv5c$bhu$1@panix2.panix.com> References: <9sta8f$fo9$1@panix2.panix.com> NNTP-Posting-Host: panix2.panix.com X-Trace: news.panix.com 1005813740 22663 166.84.1.2 (15 Nov 2001 08:42:20 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 15 Nov 2001 08:42:20 GMT 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!fu-berlin.de!feeder.qis.net!yellow.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!panix!news.panix.com!panix2.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7606 In article , Rich Alderson wrote: >klh@panix.com (Ken Harrenstien) writes: >My experience with KCC-built apps is here at XKL, where 8-bit compatibility was >the issue: X11R4/5, GNU Emacs 18.59, and Perl 4.036. Perhaps if they had been >made 9-bit-friendly instead of Unix-friendly they'd be faster; in any case, my >assumption about the 8-bit-ness of KCC derives from this usage. X11? I vaguely recall hearing that XKL asked Ian Macky to port X11 to TOPS-20, but the project was abandoned when it was realized just how many places in the X11 code (esp the protocol handling) uncompromisingly assumed 8-bit chars. But it would have been trivial to configure KCC to use 8-bit chars by default, so I wonder if that's what XKL tried later (apparently with success, if I understand you correctly)? (Aside: another neat thing about PDP-10 byte pointers is that you can easily generate function code that works for arguments of any plausible byte size!) >> If there is interest, I could pull my KCC directories back off tape, or post >> the KCC doc file; the two companies I mentioned should be canvassed for bug >> fixes or updates though. Or wait for Lars, although it looks like GCC will >> only run on machines (real or virtual) with extended addressing. > >I'm sure there'd be a great deal of interest. OK, I've already found my tape and have asked REG for the XKL mods. Should know soon whether it makes sense to distribute those, or re-integrate, or do something else. --Ken ###### From: klh@panix.com (Ken Harrenstien) Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 15 Nov 2001 04:07:23 -0500 Organization: PANIX Public Access Internet and UNIX, NYC Lines: 16 Sender: Ken Harrenstien Message-ID: <9t00kb$d0h$1@panix2.panix.com> References: <9sta8f$fo9$1@panix2.panix.com> <9svlns$4jt@gap.cco.caltech.edu> NNTP-Posting-Host: panix2.panix.com X-Trace: news.panix.com 1005815244 23053 166.84.1.2 (15 Nov 2001 09:07:24 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 15 Nov 2001 09:07:24 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!skynet.be!skynet.be!europa.netcrusader.net!usenetserver.com!198.138.0.5!newshub.northeast.verio.net!verio!newsfeed.mathworks.com!panix!news.panix.com!panix2.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7598 In article <9svlns$4jt@gap.cco.caltech.edu>, glen herrmannsfeldt wrote: >Or run as a cross compiler. gcc is usually pretty good at that. Only problem there would be the desire to do constants and constant expression evaluation (both in the preprocessor and code generator) and getting the same answers as if you were compiling on the target itself -- eg using 36/72-bit integer or floating-point arithmetic. This idea was actually the initial seed for KLH10; I wrote a PDP-10 arithmetic package that was intended to allow KCC to run on 32-bit Unixboxen and cross-compile code for PDP-10s. I left SRI before I could finish and distribute that variant of KCC, but the package still survives as KLH10's "kn10ops.c". --Ken ###### From: Ben Franchuk Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: Thu, 15 Nov 2001 09:18:55 -0700 Organization: Posted via Supernews, http://www.supernews.com Message-ID: <3BF3EAEF.384E0E15@jetnet.ab.ca> X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.19 i586) X-Accept-Language: en MIME-Version: 1.0 References: <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sta8f$fo9$1@panix2.panix.com> <9svlns$4jt@gap.cco.caltech.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 15 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!cleanfeed.casema.net!leda.casema.net!isdnet!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7591 glen herrmannsfeldt wrote: > Where does the 8 bit problem come in? Programs converting int to char > and expecting it to be mod 256 instead of 512? Since you can pack 5 7 bit chars in a word data >= 128 could be real problem with programs like zip and unzip that expect 8 bit data. Byte to integer conversions in structures are a pain. union { int_32 foo ; char_8 foobar[4]; } or other fancy tricks like char_8 array ... int_ptr = & array[n]; * int_ptr = foo-16; Ben Franchuk. -- Live "Pre-historic Cpu's" -- and you thought they were extinct. www.jetnet.ab.ca/users/bfranchuk/index.html ###### From: Lars Brinkhoff Newsgroups: alt.sys.pdp10 Subject: Re: New web and ftp site for PDP-10 Date: 24 Nov 2001 17:09:39 +0100 Organization: nocrew Lines: 11 Message-ID: <85elmoywu4.fsf@junk.nocrew.org> References: <9shfnl04o7@drn.newsguy.com> <9si07b$mm4@gap.cco.caltech.edu> <9sta8f$fo9$1@panix2.panix.com> NNTP-Posting-Host: junk.nocrew.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: yggdrasil.utfors.se 1006618007 21701 212.73.17.42 (24 Nov 2001 16:06:47 GMT) X-Complaints-To: abuse@utfors.se NNTP-Posting-Date: 24 Nov 2001 16:06:47 GMT User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.tele.dk!small.news.tele.dk!195.158.233.21!news1.ebone.net!news.ebone.net!newsfeed.sunet.se!news01.sunet.se!news.utfors.se!unknown!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7785 > klh@panix.com (Ken Harrenstien) writes: > > it looks like GCC will only run on machines (real or virtual) with > > extended addressing. Yes. Although I (try to) support code generation for all machines back to the PDP-6, GCC most certainly needs an address space larger than 256K to run. -- Lars Brinkhoff http://lars.nocrew.org/ Linux, GCC, PDP-10 Brinkhoff Consulting http://www.brinkhoff.se/ programming