From comp.sys.dec Tue Mar 10 11:10:20 1992 Path: sot-ecs!icdoc!ibmassc!yktnews!admin!news.ans.net!europa.asd.contel.com!darwin.sura.net!wupost!think.com!news!jpt From: jpt@telecaster.think.com (James P. Taylor) Newsgroups: vmsnet.pdp-11,comp.org.decus,comp.sys.dec,ne.org.decus Subject: Re: Definitive list of PDP processors and their differences ? Message-ID: Date: 9 Mar 92 12:19:55 GMT References: <1992Mar7.202059.6406@athena.mit.edu> <1992Mar9.133753.9169@ryn.mro4.dec.com> Organization: Thinking Machines Corporation, Cambridge MA, USA Lines: 48 Xref: sot-ecs vmsnet.pdp-11:145 comp.org.decus:689 comp.sys.dec:6166 NNTP-Posting-Host: telecaster.think.com In-reply-to: mccrohan@salthl.enet.dec.com's message of Mon, 9 Mar 1992 13:37:53 GMT In article <1992Mar9.133753.9169@ryn.mro4.dec.com> mccrohan@salthl.enet.dec.com () writes: Newsgroups: vmsnet.pdp-11,comp.org.decus,comp.sys.dec,ne.org.decus Path: think.com!mips!mips!decwrl!deccrl!news.crl.dec.com!hollie.rdg.dec.com!ryn.mro4.dec.com!salthl.enet.dec.com!mccrohan From: mccrohan@salthl.enet.dec.com () Sender: news@ryn.mro4.dec.com (USENET News System) Reply-To: mccrohan@salthl.enet.dec.com () Organization: Digital Equipment Corp. References: <1992Mar7.202059.6406@athena.mit.edu> Date: Mon, 9 Mar 1992 13:37:53 GMT Xref: think.com vmsnet.pdp-11:203 comp.org.decus:1304 comp.sys.dec:9086 ne.org.decus:51 1971 11/20 TTL SSI Core Mem Papertape reader etc 11/15 11/20 in different packaging [detailed 11 stuff deleted .....] Disclaimer: all version numbers and dates are (generally) from memory. I have probably left out a couple of machines. [detailed application stuff deleted .... ] ----------------------------------- [pdp-8 and pdp-10 stuff deleted .... ] ------------------------------------ DEC also produced the PDP-1 (36bit?), -6, -9(?), -15(18bit - half a ten) and others.... Mike -------All views, Mispellings, inaccuracies are mine.-------------------- PDP-1 was an 18 bit machine. It followed the TX0 built at MIT, by guess who?? PDP-6 was a 36 bitter [as indicated the precursor to 10] the PDP -7 , 9, and 15 [?] were all 18 bit machines built for special application areas. jim taylor Thinking Machines Corp. Cambridge MA 02142 From comp.sys.dec Tue Mar 10 11:10:21 1992 Path: sot-ecs!icdoc!uknet!mcsun!uunet!decwrl!deccrl!news.crl.dec.com!hollie.rdg.dec.com!ryn.mro4.dec.com!salthl.enet.dec.com!mccrohan From: mccrohan@salthl.enet.dec.com Newsgroups: vmsnet.pdp-11,comp.org.decus,comp.sys.dec,ne.org.decus Subject: Re: Definitive list of PDP processors and their differences ? Message-ID: <1992Mar9.133753.9169@ryn.mro4.dec.com> Date: 9 Mar 92 13:37:53 GMT References: <1992Mar7.202059.6406@athena.mit.edu> Sender: news@ryn.mro4.dec.com (USENET News System) Reply-To: mccrohan@salthl.enet.dec.com () Organization: Digital Equipment Corp. Lines: 103 Xref: sot-ecs vmsnet.pdp-11:144 comp.org.decus:688 comp.sys.dec:6155 1971 11/20 TTL SSI Core Mem Papertape reader etc 11/15 11/20 in different packaging 1972 11/40 TTL SSI and MSI. Microcode.Core and later MOS MEM. RK05 disks. 11/35 11/40 in different packaging. 11/45 TTL SSI and MSI. HiSpeed Mem. Hi oerf machine 11/50 11/45 with Bipoals mem 11/55 Some Variation of the above 11/05 Microcoded lo-cost '11. Perf = 11/20. Core and Mos. OEM 11/10 Bigger box 11/05 LSI-11 Pdp11 on 4 chips complements of Western Digital. OEM 1975 11/70 Generaltional follow on to 11/45. Faster cpu. Core and later MOS Mem. FPU, Large cache. Buses. "Mainframe Mini" LSI-11/2 Single chip '11. OEM SBC Offering 11/03 Q-bus machine with LSI-11 heart 1976 11/04 Unibus version of 11/03. All MOS memory from here on. 11/34 2board MSI follow on to 11/40. Unibus. 1977 11/60 Perf between 11/40,11/70. User definable microcode. 1979 11/34a Tidied up 11/34 11/44 (Almost) 11/70 follow on. much cheaper. Less power 1981 LSI-11/23 Single chop pdp-11. About x2 perf or 11/03. Qbus LSI-11/24 Unibus version of 11/23. F-11 Chip ?1987 11/53 J-11 chip. Faster than 11/23. Similar/same packaging Qbus 11/54? Unibus version... ?1987 11/73 Faster J-11 chip.... Perf finally Greater than 11/70. ???? c1989 11/85 Even Faster.... c1990 11/93? ...the latest..... Disclaimer: all version numbers and dates are (generally) from memory. I have probably left out a couple of machines. SW: 11/20 ran DOS/Batch which begat RSTS and later RSTS/e on the 11/40 and all later systems. Alst, RSX-11 appeared on the 11/40, 11/45 and became RSX-11M, RSX11-D, RSX11-S?. RT11 was very popular on 11/20 and many of the Qbus machines. IAS ran on the high end 1145s and 11/70s for a time. MUMPS was very popular on many large 11/70 applications. Unix, developed on a PDP-6 was migrated early to 11/45s and 11/70s.... RSTS is Resource Sharing/TimeSharing RSX is a multitasking OS. Author of RSX developed VMS and is now working with another party, involved in a major project for them. RT-11 is a Realtime OS with Foreground/Background.... MS-DOS looks a bit like it but with many of the useful bits removed. IAS is/was somewhat obscure, and I cannot remember its market or attributes. MUMPS was developed by Mass Gen Hosp and widely used for a variety of applications. MUMPS is botha a database and a language. Unix- well, I think most people have heard of UNIX. ----------------------------------- PDP8: 12 bit machine. First real minicomputer. Put DEC on the map, ready for the PDP-11 to exploit. First PDP-8 Circa 1965(?) used, I believe DTL. Followed by the PDP8-I late '60s and the PDP-8e/8m about 1970. '8e used the OMNIBUS. The OMNIBUS on the '8e and UNIBUS on the '11s were easy to interface with a myriad of third party vendors/OEMS built devices for them. This was the cornerstone of Digitals early successes. '8e was followed by 8a - similar in technology and formfactor to the 11/04. The 8/a was essentially the last PDP-8, although the DECmate-I, -II and -III were PDP-8's at heart. ------------------------------------- PDP10 was a 36bit bachine born from the lions of the PDP-6. Original KA10 built from TTL, and then the KI10 (Name changed from PDP-10 to DECsystem-10 around then) was I believe TTL/MSI. The last of the family was the KL-10 which had gobs of ECL. the '10 was a mainframe, and loved by those that used it. The '10 ran TOPS-10, a timesharing OS. A version odf the beast ran TOPS-20, and was called the DECsystem-20. A MInisized version appeared about 1979, eas called the 2020 and ran Tops-20. the '10 and '20 ran Uni- or Dual- Processors. Different subconfigurations were known as the 1060, 1090, 1099 or somesuch. ------------------------------------ DEC also produced the PDP-1 (36bit?), -6, -9(?), -15(18bit - half a ten) and others.... Mike -------All views, Mispellings, inaccuracies are mine.-------------------- S From comp.sys.dec Tue Mar 10 11:10:21 1992 Path: sot-ecs!icdoc!ibmassc!yktnews!admin!news.ans.net!europa.asd.contel.com!uunet!sun-barr!west.West.Sun.COM!cronkite.Central.Sun.COM!texsun!exucom.exu.ericsson.se!ericom!seunet!sunic!kth.se!mattis From: mattis@elixir.lne.kth.se (Mattis Andersson) Newsgroups: vmsnet.pdp-11,comp.org.decus,comp.sys.dec,ne.org.decus Subject: Re: Definitive list of PDP processors and their differences ? Message-ID: <1992Mar9.170831.2481@kth.se> Date: 9 Mar 92 17:08:31 GMT References: <1992Mar7.202059.6406@athena.mit.edu> <1992Mar9.133753.9169@ryn.mro4.dec.com> Sender: usenet@kth.se (Usenet) Organization: Dept. of EE, Royal Institute of Technology, Stockholm, Sweden Lines: 33 Xref: sot-ecs vmsnet.pdp-11:146 comp.org.decus:691 comp.sys.dec:6173 Nntp-Posting-Host: elixir.lne.kth.se Actually the LSI-11 (the original, quad-size board), and the LSI-11/2 (dual-size board) is using the same chipset. I also think that PDP-11/03 is some kind of market name for a system based on either of these two. The PDP-11/04 is a completley different CPU bulit by PLAs and ordinary TTL-logic. The CPU is one hex-size board. PDP-1 was a 18 bit computer. PDP-4 was also 18-bit but slower than PDP-1, and as I remember incompatible with the PDP-1. PDP-5 was a 12 bit computer. PDP-6 and DEC-10/20 was 36-bit computers PDP-7, PDP-9, PDP-15 is 18 bit computers compatble with the PDP-4 PDP-12 is a specialized laboratory computer, successor to LINC and LINC-8. If you are interested in more information concerning old DEC computers I recommend "Computer Engineering, A DEC view of hardware systems design" by Gordon Bell, J. Graig Mudge, John E. McNamara. Digital Press. Mattis Andersson mattis@elixir.lne.kth.se From comp.sys.dec Tue Mar 10 11:10:21 1992 Path: sot-ecs!icdoc!ibmassc!yktnews!admin!news.ans.net!europa.asd.contel.com!uunet!olivea!bu2.bu.edu!bu.edu!world!iecc!johnl From: johnl@iecc.cambridge.ma.us (John R. Levine) Newsgroups: vmsnet.pdp-11,comp.org.decus,comp.sys.dec,ne.org.decus Subject: Re: Definitive list of PDP processors and their differences ? Message-ID: <1992Mar9.194033.21005@iecc.cambridge.ma.us> Date: 9 Mar 92 19:40:33 GMT References: <1992Mar7.202059.6406@athena.mit.edu> <1992Mar9.133753.9169@ryn.mro4.dec.com> Organization: I.E.C.C. Lines: 54 Xref: sot-ecs vmsnet.pdp-11:147 comp.org.decus:692 comp.sys.dec:6174 In article <1992Mar9.133753.9169@ryn.mro4.dec.com> mccrohan@salthl.enet.dec.com () writes: >1971 11/20 TTL SSI Core Mem Papertape reader etc > 11/15 11/20 in different packaging > > >1972 11/40 TTL SSI and MSI. Microcode.Core and later MOS MEM. RK05 disks. > 11/35 11/40 in different packaging. > > 11/45 TTL SSI and MSI. HiSpeed Mem. Hi oerf machine > 11/50 11/45 with Bipoals mem > 11/55 Some Variation of the above > > 11/05 Microcoded lo-cost '11. Perf = 11/20. Core and Mos. OEM > 11/10 Bigger box 11/05 > LSI-11 Pdp11 on 4 chips complements of Western Digital. OEM > >1975 11/70 Generaltional follow on to 11/45. Faster cpu. Core and > later MOS Mem. FPU, Large cache. Buses. "Mainframe Mini" The 11/70 had the same CPU as the 11/45 but with a cache and a larger physical (not logical) address space. Phsical addressing only: 11/03 /04 /05 /10 /15 /20 16 bit virtual: 11/23 /24 /34 /35 /40 /60 /53? /73 virtual with split I/D: 11/45 /70 /85 > Unix, developed on a PDP-6 was migrated early to 11/45s and > 11/70s.... Close, Unix started on a PDP-7, a much smaller machine than the 6. >DEC also produced the PDP-1 (36bit?), -6, -9(?), -15(18bit - half a ten) PDP-1: 18 bit, based on MIT designs PDP-4: 18 bit, not particularly compatible with -1 PDP-5: cut down PDP-4, first DEC 12 bit micro PDP-6: 36 bit quasi-mainframe (quasi because of mediocre I/O) PDP-7: improved PDP-4 PDP-8: improved PDP-5, wildly successful LINC-8: PDP-8 with Lincoln Labs CPU bolt-on, used for lab work PDP-9: improved PDP-7 PDP-10: improved PDP-6, fairly successful, greatly beloved PDP-11: all new 16 bit, wildly successful PDP-12: improved LINC-8 with single integrated CPU, really cool machine PDP-14: strange little process control box PDP-16: stranger set of do it yourself modules DEC-20: improved PDP-10 See Bell, Mudge, and McNamara, "Computer Engineering" for true facts. -- John R. Levine, IECC, POB 349, Cambridge MA 02238, +1 617 492 3869 johnl@iecc.cambridge.ma.us, {ima|spdcc|world}!iecc!johnl Adm. Grace Hopper: 1906-1992 From comp.sys.dec Tue Mar 10 11:10:21 1992 Path: sot-ecs!icdoc!ibmassc!yktnews!admin!news.ans.net!europa.asd.contel.com!darwin.sura.net!wupost!think.com!mips!mips!munnari.oz.au!network.ucsd.edu!dog.ee.lbl.gov!hellgate.utah.edu!cc.usu.edu!ivie From: ivie@cc.usu.edu (CP/M lives!) Newsgroups: comp.org.decus,comp.sys.dec Subject: Re: Definitive list of PDP processors and their differences ? Message-ID: <1992Mar9.141342.53550@cc.usu.edu> Date: 9 Mar 92 20:13:42 GMT References: <1992Mar7.202059.6406@athena.mit.edu> <1992Mar9.133753.9169@ryn.mro4.dec.com> Organization: Utah State University Lines: 29 Xref: sot-ecs comp.org.decus:690 comp.sys.dec:6171 In article <1992Mar9.133753.9169@ryn.mro4.dec.com>, mccrohan@salthl.enet.dec.com () writes: > > PDP8: > 12 bit machine. First real minicomputer. Put DEC on the map, ready for > the PDP-11 to exploit. First PDP-8 Circa 1965(?) used, I believe DTL. The PDP-8 actually started with the PDP-5 introduced at Wescon in August of 1963 (I recently wandered through some old computer mags). The PDP-5 ran the same instruction set as the -8, although there were some limitations on the combinations of operate microinstructions that you could use. It originally sold for $27,000 and sold pretty much like hotcakes; about the only things that seemed to be doing better than the -5 were some NCR machines, according to the monthly computer census in whatever magazine I was looking through ("Computers And Control" or "Control And Automation", something like that). The -8 was announced by a two-page spread in the magazine in May of '64 or '65 (I forget what year I was wandering through when I saw the ad), and it outsold in a few months what the -5 had sold in the year or two since its announcement. It originally cost $18,000 and was an order of magnitude faster than the PDP-5. The odd thing about the ad for the -8 was that it claimed that it was smaller than the -5. I've seen both a -5 and an original -8, and it just ain't true. The difference was in the packaging. The -5's circuits were bolted across a 19-inch rack while the -8 had two back-to-back backplanes that made good use of the depth of the rack. This let the -8 be sold as a tabletop machine. Roger Ivie ivie@cc.usu.edu