From: klh@panix.com (Ken Harrenstien) Newsgroups: alt.sys.pdp10 Subject: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) Date: 27 Nov 2001 02:47:54 -0500 Organization: PANIX Public Access Internet and UNIX, NYC Lines: 31 Sender: Ken Harrenstien Message-ID: <9tvgfa$h37$1@panix3.panix.com> References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> NNTP-Posting-Host: panix3.panix.com X-Trace: news.panix.com 1006847275 24880 166.84.1.3 (27 Nov 2001 07:47:55 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 27 Nov 2001 07:47:55 GMT 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!news-peer.gip.net!news.gsl.net!gip.net!panix!news.panix.com!panix3.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7822 In article <137cf56f.0111261821.25872cbd@posting.google.com>, bad bob wrote: >System build from tape: >40 minutes or so on PC164/500 >24 minutes or so on Athlon >more than an hour on 250. Um, what do you mean by "System build"? Restoring files from the installation tape (which requires user input delay), or reassembling the monitor, or...? Unfortunately I don't know of a standard benchmark for comparing the performance of real applications on TOPS-10/20, so anecdotal stuff like this is often the best we can do. There *is* one absolutely consistent benchmark, the DFKFB timing test included with the auxiliary distribution. It doesn't include any kind of I/O measurements (no disk or tape devices or images are needed to run it), just illustrates the maximum CPU performance you can get. It is a very useful tool while tweaking compiler switches to see if you can improve on the basic optimization settings of the current Makefiles. "doc/dfkfb.txt" has more. If anyone has DFKFB output from a real KL10B with cache I'd love to include it in the doc! --Ken (p.s. I'm a little surprised nobody has posted any DFKFB results yet. I only have a 900MHz Celeron so I'm not in the running :-) ###### From: tom0565@yahoo.com (tom m) Newsgroups: alt.sys.pdp10 Subject: Re: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) Date: 27 Nov 2001 12:41:27 -0800 Organization: http://groups.google.com/ Lines: 78 Message-ID: <8f576ee9.0111271241.7fb095cf@posting.google.com> References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> NNTP-Posting-Host: 64.124.70.190 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1006893687 21588 127.0.0.1 (27 Nov 2001 20:41:27 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 27 Nov 2001 20:41:27 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!feed1.newsreader.com!netnews.com!xfer02.netnews.com!feeder.qis.net!sn-xit-02!supernews.com!postnews1.google.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7847 klh@panix.com (Ken Harrenstien) wrote in message news:<9tvgfa$h37$1@panix3.panix.com>... I was running the kn10 on a 200mhz pentium and it took almost four hours to build the tops-20 monitor from sources. I upgraded to a 1.2ghz celerion just reassembled all of the monitor sources in about 15 minutes. Note, the Tops-20 load average was about 1.07 the entire time so the kn10 wasn't doing much else. > (p.s. I'm a little surprised nobody has posted any DFKFB results yet. > I only have a 900MHz Celeron so I'm not in the running :-) Here's the dfkbd results for the kn10 on a 1.2mhz celeron with 256mb memory: PDP-10 KL10 INSTRUCTION TIMING TEST (DFKFB) VERSION 0.1, SV=0.1, CPU#=1, MCV=442, MCO=660, HO=35, 60HZ SWITCHES = 400100 000000 CLK SOURCE = NORMAL, CLK RATE = FULL, AC BLK 0 MEMORY MAP = FROM TO SIZE/K 00000000 00337777 112 1 - BASIC CLOCK CYCLE IS 0 NSEC. 2 - INDEXING TAKES 3 NSEC. 3 - INDIRECT TAKES 25 NSEC. 4 - INDEXING AND INDIRECT TAKES 29 NSEC. 5 - MOVEI TAKES 41 NSEC. 6 - MOVE FROM AC TAKES 40 NSEC. 7 - MOVE FROM MEMORY TAKES 50 NSEC. 8 - HRR FROM MEMORY TAKES 60 NSEC. 9 - SETOM 0 TAKES 44 NSEC. 10 - JRST TAKES 37 NSEC. 11 - JSR TAKES 61 NSEC. 12 - PUSHJ TAKES 84 NSEC. 13 - ADD FROM MEMORY TAKES 81 NSEC. 14 - MUL (9 ADD/SUB - 18 SHIFTS) TAKES 241 NSEC. 15 - DIV TAKES 1.14 USEC. 16 - FIX A FLOATING POINT ONE TAKES 93 NSEC. 17 - FLTR AN INTERGER ONE TAKES 146 NSEC. 18 - FAD (1 RIGHT SHIFT) TAKES 188 NSEC. 19 - FAD (8 SHIFT RIGHT - 3 LEFT) TAKES 345 NSEC. 20 - FMP (7 ADD/SUB - 14 SHIFTS) TAKES 439 NSEC. 21 - FDV TAKES 800 NSEC. 22 - DMOVE FROM MEMORY TAKES 77 NSEC. 23 - DFAD (1 RIGHT SHIFT) TAKES 592 NSEC. 24 - DFAD (8 SHIFT RIGHT - 1 LEFT) TAKES 592 NSEC. 25 - DFMP (7 ADD/SUB - 32 SHIFTS) TAKES 784 NSEC. 26 - DFDV TAKES 11.26 USEC. 27 - CONO PI TAKES 77 NSEC. 28 - CONI PI TAKES 55 NSEC. 29 - DATAO APR TAKES 62 NSEC. 30 - DATAI APR TAKES 67 NSEC. 31 - MOVE TO MEMORY TAKES 51 NSEC. 32 - LOGICAL SHIFT (35 PLACES LEFT) TAKES 62 NSEC. 33 - LOGICAL SHIFT (35 PLACES RIGHT) TAKES 60 NSEC. 34 - LOGICAL SHIFT COMBINED (71 PLACES LEFT) TAKES 143 NSEC. 35 - LOGICAL SHIFT COMBINED (71 PLACES RIGHT) TAKES 139 NSEC. 36 - INCREMENT BYTE POINTER TAKES 118 NSEC. 37 - INCREMENT AND LOAD BYTE TAKES 159 NSEC. 38 - INCREMENT AND DEPOSIT BYTE TAKES 175 NSEC. 39 - JFCL TAKES 34 NSEC. 40 - CAI TAKES 33 NSEC. 41 - JUMP TAKES 33 NSEC. 42 - CAM TAKES 36 NSEC. 43 - EQV AC TO AC TAKES 46 NSEC. 44 - EQV MEMORY TO AC TAKES 60 NSEC. 45 - SETOB TAKES 58 NSEC. 46 - AOS TO MEMORY TAKES 53 NSEC. 47 - EXCHANGE AN AC WITH AN AC TAKES 51 NSEC. 48 - EXCHANGE AN AC WITH MEMORY TAKES 63 NSEC. 49 - EXECUTE TAKES 39 NSEC. 50 - BLT MEMORY TO MEMORY TAKES 108 NSEC. 51 - BLT AC TO MEMORY TAKES 96 NSEC. 52 - DATAI TAKES 76 NSEC. 53 - DATAO TAKES 69 NSEC. ###### Message-ID: <3C040A36.379A6E6@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: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 44 Date: Tue, 27 Nov 2001 21:50:30 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 1006897830 138.88.75.218 (Tue, 27 Nov 2001 16:50:30 EST) NNTP-Posting-Date: Tue, 27 Nov 2001 16:50:30 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon1.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7823 Ken Harrenstien wrote: > > In article <137cf56f.0111261821.25872cbd@posting.google.com>, > bad bob wrote: > > >System build from tape: > >40 minutes or so on PC164/500 > >24 minutes or so on Athlon > >more than an hour on 250. > > Um, what do you mean by "System build"? Restoring files from the > installation tape (which requires user input delay), or reassembling > the monitor, or...? includes the the user delay (which in this case is consistent) involved in compiling, doing step 4, restoring from the basic tape, and rebooting !! The script that you provided in the docs, and hte .ini files makes the process faster. the notes and highlighter that I added to my hard copy made my actions faster -so the anecdotal info includes a somewhat fumble fingered bearded guy BUT who types by touch and reviewed before issuing the CR. > > Unfortunately I don't know of a standard benchmark for comparing the > performance of real applications on TOPS-10/20, so anecdotal stuff > like this is often the best we can do. > There *is* one absolutely consistent benchmark, the DFKFB timing test > included with the auxiliary distribution. It doesn't include any kind > of I/O measurements (no disk or tape devices or images are needed to > run it), just illustrates the maximum CPU performance you can get. It > is a very useful tool while tweaking compiler switches to see if you > can improve on the basic optimization settings of the current > Makefiles. "doc/dfkfb.txt" has more. > > If anyone has DFKFB output from a real KL10B with cache I'd love to > include it in the doc! > > --Ken > > (p.s. I'm a little surprised nobody has posted any DFKFB results yet. > I only have a 900MHz Celeron so I'm not in the running :-) ###### Message-ID: <3C040D09.89731F20@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: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 44 Date: Tue, 27 Nov 2001 21:58:59 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 1006898339 138.88.75.218 (Tue, 27 Nov 2001 16:58:59 EST) NNTP-Posting-Date: Tue, 27 Nov 2001 16:58:59 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon1.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7826 Ken Harrenstien wrote: > > In article <137cf56f.0111261821.25872cbd@posting.google.com>, > bad bob wrote: > > >System build from tape: > >40 minutes or so on PC164/500 > >24 minutes or so on Athlon > >more than an hour on 250. > > Um, what do you mean by "System build"? Restoring files from the > installation tape (which requires user input delay), or reassembling > the monitor, or...? includes the the user delay (which in this case is consistent) involved in compiling, doing step 4, restoring from the basic tape, and rebooting !! The script that you provided in the docs, and hte .ini files makes the process faster. the notes and highlighter that I added to my hard copy made my actions faster -so the anecdotal info includes a somewhat fumble fingered bearded guy BUT who types by touch and reviewed before issuing the CR. > > Unfortunately I don't know of a standard benchmark for comparing the > performance of real applications on TOPS-10/20, so anecdotal stuff > like this is often the best we can do. > There *is* one absolutely consistent benchmark, the DFKFB timing test > included with the auxiliary distribution. It doesn't include any kind > of I/O measurements (no disk or tape devices or images are needed to > run it), just illustrates the maximum CPU performance you can get. It > is a very useful tool while tweaking compiler switches to see if you > can improve on the basic optimization settings of the current > Makefiles. "doc/dfkfb.txt" has more. > > If anyone has DFKFB output from a real KL10B with cache I'd love to > include it in the doc! > > --Ken > > (p.s. I'm a little surprised nobody has posted any DFKFB results yet. > I only have a 900MHz Celeron so I'm not in the running :-) ###### From: "Zane H. Healy" Subject: Re: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) Newsgroups: alt.sys.pdp10 References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 9 Message-ID: <%aUM7.585401$Lg.23468114@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: Tue, 27 Nov 2001 17:10:03 EST Date: Tue, 27 Nov 2001 22:10:03 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!howland.erols.net!bloom-beacon.mit.edu!hub1.nntpserver.com!sjcppf01!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7852 tom m wrote: > I was running the kn10 on a 200mhz pentium and it took almost four > hours to build the tops-20 monitor from sources. I upgraded to a > 1.2ghz celerion just reassembled all of the monitor sources in about You know, I get the impression that these PDP-10 emulators are causing a lot of us to get new systems :^) Zane ###### Message-ID: <3C041A70.A491545D@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: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> <%aUM7.585401$Lg.23468114@sjcpnn01.usenetserver.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 15 Date: Tue, 27 Nov 2001 22:53:19 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 1006901599 138.88.75.218 (Tue, 27 Nov 2001 17:53:19 EST) NNTP-Posting-Date: Tue, 27 Nov 2001 17:53:19 EST 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!atl1-feed1.news.digex.net!dca6-feed2.news.digex.net!intermedia!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon1.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7828 "Zane H. Healy" wrote: > > tom m wrote: > > I was running the kn10 on a 200mhz pentium and it took almost four > > hours to build the tops-20 monitor from sources. I upgraded to a > > 1.2ghz celerion just reassembled all of the monitor sources in about > > You know, I get the impression that these PDP-10 emulators are causing a lot > of us to get new systems :^) > Yep, and work the electrons os older alphas too!! 8) > Zane ###### Message-ID: <3C041A77.936B5B6F@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: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> <%aUM7.585401$Lg.23468114@sjcpnn01.usenetserver.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 15 Date: Tue, 27 Nov 2001 22:53:25 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 1006901605 138.88.75.218 (Tue, 27 Nov 2001 17:53:25 EST) NNTP-Posting-Date: Tue, 27 Nov 2001 17:53:25 EST 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!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.pdp10:7834 "Zane H. Healy" wrote: > > tom m wrote: > > I was running the kn10 on a 200mhz pentium and it took almost four > > hours to build the tops-20 monitor from sources. I upgraded to a > > 1.2ghz celerion just reassembled all of the monitor sources in about > > You know, I get the impression that these PDP-10 emulators are causing a lot > of us to get new systems :^) > Yep, and work the electrons os older alphas too!! 8) > Zane ###### From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: Re: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) Date: Wed, 28 Nov 2001 00:17:00 -0000 Message-ID: Sender: Timothy Stark References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> <3C042354.A2F68872@Empire.Net> User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 28 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!skynet.be!skynet.be!sn-xit-03!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7840 John Sauter wrote: > My memory isn't precise, but I believe the DFKFB numbers > that Tom posted are about 200 times faster than the > KA-10. The computer he ran it on is about 1000 times > faster than a KA-10, so we are losing about a factor > of 5 to the emulator software. That's about right, > in my experience. > Even though the numbers are reasonable, they take my > breath away. There is something astonishing about > looking at the familiar DFKFB listing and seeing > such incredible speed. I wonder how fast computers > will be 30 years from now. > John Sauter (J_Sauter@Empire.Net) Yeah! PDP10 emulators will be Cray-10 emulators in 30 years later. :-) I had a very old copy of dhrystone. I tried to run it on my Pentium III 800 Mhz system and it complaint that my procesor is too fast to measure! I increased number of loops and it resulted 1 million dhrystones! I wrote small assembly language program to count 1 billion and it only took one second to complete. Soon, I will get 2 Ghz Pentium 4 CPU because I now have new P4T-E motherboard and see what happened in my emulators. In 10 years, there will be 50+ ghz processors... -- Tim Stark ###### Message-ID: <3C042F4F.87D1F612@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: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> <3C042354.A2F68872@Empire.Net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 42 Date: Wed, 28 Nov 2001 00:22:22 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon2.gnilink.net 1006906942 138.88.75.218 (Tue, 27 Nov 2001 19:22:22 EST) NNTP-Posting-Date: Tue, 27 Nov 2001 19:22:22 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon2.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7832 sword7@speakeasy.org wrote: > > John Sauter wrote: > > My memory isn't precise, but I believe the DFKFB numbers > > that Tom posted are about 200 times faster than the > > KA-10. The computer he ran it on is about 1000 times > > faster than a KA-10, so we are losing about a factor > > of 5 to the emulator software. That's about right, > > in my experience. > Does anyone know what the Toad1 numbers are? Now, how do these numbers compare to a 2060? hmmm > > Even though the numbers are reasonable, they take my > > breath away. There is something astonishing about > > looking at the familiar DFKFB listing and seeing > > such incredible speed. I wonder how fast computers > > will be 30 years from now. > > John Sauter (J_Sauter@Empire.Net) > > Yeah! PDP10 emulators will be Cray-10 emulators in 30 years later. :-) > I had a very old copy of dhrystone. I tried to run it on my Pentium III > 800 Mhz system and it complaint that my procesor is too fast to measure! > I increased number of loops and it resulted 1 million dhrystones! I > wrote small assembly language program to count 1 billion and it only took > one second to complete. Yeah, and you are the guy who made me go buy this Athlon! Thank you. Sitting there with the freebsd environment, I guess I am a bsd geek after all - happily, I was able to get thru my problems quicker and get thru klh. NOW on to TS10 and SIMH!! I have room to put them all on each box, run them, get some numbers, and see what they look like on 3 OS implementations!! And I still want to go back to NetBSD on the 3000s. > > Soon, I will get 2 Ghz Pentium 4 CPU because I now have new P4T-E > motherboard and see what happened in my emulators. > > In 10 years, there will be 50+ ghz processors... > > -- Tim Stark ###### Message-ID: <3C042FA5.4479F40F@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: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> <3C042354.A2F68872@Empire.Net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 42 Date: Wed, 28 Nov 2001 00:23:47 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon2.gnilink.net 1006907027 138.88.75.218 (Tue, 27 Nov 2001 19:23:47 EST) NNTP-Posting-Date: Tue, 27 Nov 2001 19:23:47 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon2.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7833 sword7@speakeasy.org wrote: > > John Sauter wrote: > > My memory isn't precise, but I believe the DFKFB numbers > > that Tom posted are about 200 times faster than the > > KA-10. The computer he ran it on is about 1000 times > > faster than a KA-10, so we are losing about a factor > > of 5 to the emulator software. That's about right, > > in my experience. > Does anyone know what the Toad1 numbers are? Now, how do these numbers compare to a 2060? hmmm > > Even though the numbers are reasonable, they take my > > breath away. There is something astonishing about > > looking at the familiar DFKFB listing and seeing > > such incredible speed. I wonder how fast computers > > will be 30 years from now. > > John Sauter (J_Sauter@Empire.Net) > > Yeah! PDP10 emulators will be Cray-10 emulators in 30 years later. :-) > I had a very old copy of dhrystone. I tried to run it on my Pentium III > 800 Mhz system and it complaint that my procesor is too fast to measure! > I increased number of loops and it resulted 1 million dhrystones! I > wrote small assembly language program to count 1 billion and it only took > one second to complete. Yeah, and you are the guy who made me go buy this Athlon! Thank you. Sitting there with the freebsd environment, I guess I am a bsd geek after all - happily, I was able to get thru my problems quicker and get thru klh. NOW on to TS10 and SIMH!! I have room to put them all on each box, run them, get some numbers, and see what they look like on 3 OS implementations!! And I still want to go back to NetBSD on the 3000s. > > Soon, I will get 2 Ghz Pentium 4 CPU because I now have new P4T-E > motherboard and see what happened in my emulators. > > In 10 years, there will be 50+ ghz processors... > > -- Tim Stark ###### Message-ID: <3C042FB3.C4AE6AC7@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: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> <3C042354.A2F68872@Empire.Net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 42 Date: Wed, 28 Nov 2001 00:24:08 GMT NNTP-Posting-Host: 138.88.75.218 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 1006907048 138.88.75.218 (Tue, 27 Nov 2001 19:24:08 EST) NNTP-Posting-Date: Tue, 27 Nov 2001 19:24:08 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon1.gnilink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7836 sword7@speakeasy.org wrote: > > John Sauter wrote: > > My memory isn't precise, but I believe the DFKFB numbers > > that Tom posted are about 200 times faster than the > > KA-10. The computer he ran it on is about 1000 times > > faster than a KA-10, so we are losing about a factor > > of 5 to the emulator software. That's about right, > > in my experience. > Does anyone know what the Toad1 numbers are? Now, how do these numbers compare to a 2060? hmmm > > Even though the numbers are reasonable, they take my > > breath away. There is something astonishing about > > looking at the familiar DFKFB listing and seeing > > such incredible speed. I wonder how fast computers > > will be 30 years from now. > > John Sauter (J_Sauter@Empire.Net) > > Yeah! PDP10 emulators will be Cray-10 emulators in 30 years later. :-) > I had a very old copy of dhrystone. I tried to run it on my Pentium III > 800 Mhz system and it complaint that my procesor is too fast to measure! > I increased number of loops and it resulted 1 million dhrystones! I > wrote small assembly language program to count 1 billion and it only took > one second to complete. Yeah, and you are the guy who made me go buy this Athlon! Thank you. Sitting there with the freebsd environment, I guess I am a bsd geek after all - happily, I was able to get thru my problems quicker and get thru klh. NOW on to TS10 and SIMH!! I have room to put them all on each box, run them, get some numbers, and see what they look like on 3 OS implementations!! And I still want to go back to NetBSD on the 3000s. > > Soon, I will get 2 Ghz Pentium 4 CPU because I now have new P4T-E > motherboard and see what happened in my emulators. > > In 10 years, there will be 50+ ghz processors... > > -- Tim Stark ###### From: Ben Franchuk Newsgroups: alt.sys.pdp10 Subject: Re: DFKFB (was Re: SUCCESS!! Re: klh10 question - on a new box) Date: Tue, 27 Nov 2001 17:31:47 -0700 Organization: Posted via Supernews, http://www.supernews.com Message-ID: <3C043073.97F445AD@jetnet.ab.ca> X-Mailer: Mozilla 4.79 [en] (Win95; U) X-Accept-Language: en MIME-Version: 1.0 References: <3C019003.F5B8B9C5@bellatlantic.net> <3C01B0BA.AAA30D48@bellatlantic.net> <137cf56f.0111261821.25872cbd@posting.google.com> <9tvgfa$h37$1@panix3.panix.com> <8f576ee9.0111271241.7fb095cf@posting.google.com> <3C042354.A2F68872@Empire.Net> <3C043000.5D64F3C6@jetnet.ab.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 25 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!netnews.com!xfer02.netnews.com!feeder.qis.net!sn-xit-02!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:7816 Ben Franchuk wrote: > > sword7@speakeasy.org wrote: > > > Soon, I will get 2 Ghz Pentium 4 CPU because I now have new P4T-E > > motherboard and see what happened in my emulators. > > Still running a P166 here. That was a big jump in the last few years. > But I doub't if the next OS will run on anything slower than a 3 > GHZ machine with 1 Gig of internal ram. While the old machines > were painful to use they in some ways where more efficent in resources. > > > In 10 years, there will be 50+ ghz processors... > > > > -- Tim Stark > Still remember the line "All computers wait at the same speed." > Bigger and faster may not be better. > -- > Ben Franchuk --- Pre-historic Cpu's -- > www.jetnet.ab.ca/users/bfranchuk/index.html -- Ben Franchuk --- Pre-historic Cpu's -- www.jetnet.ab.ca/users/bfranchuk/index.html