Article 3217 of alt.sys.pdp10: Newsgroups: alt.sys.pdp8,alt.sys.pdp10 Path: nntp1.ba.best.com!news1.best.com!news.maxwell.syr.edu!ais.net!uunet!in5.uu.net!world!world!news From: a11is0n@see.sig (Allison) Subject: Re: Could I build a PDP-8? Sender: news@world.std.com (Mr Usenet Himself) Message-ID: Date: Thu, 28 Aug 1997 03:43:53 GMT References: <19970819210601.RAA08492@ladder01.news.aol.com> <33FB770D.7DB6@InfoMation.comm> <3402AC58.CFFDEB@stoneweb.com> <3402B490.5D06@cisco.com> <5u1j0o$amv@mars.hyperk.com> Nntp-Posting-Host: world.std.com Organization: The World @ Software Tool & Die X-Newsreader: Forte Free Agent 1.0.82 Lines: 22 Xref: nntp1.ba.best.com alt.sys.pdp8:2052 alt.sys.pdp10:3217 kth@srv.net (Kevin Handy) wrote: >Isn't this the chip that was used in the DecMates? I don't think >that DEC ever made their own chip. >I've still got a couple of DecMate II bases laying around, but have no >monitors for them. Yes. The WS78 used the 6100 and the DECMate I,II,III used the 6120 which was faster and higher integration. They (the 6100 chips etal) were widely available in the late 70s early 80s and fairly cheap. It didn't take much to build a system using the 6100 and I still have mine(operational). Recently I brough up OS278 on a DMIII and its very interesting. Allison Real address is: Allisonp @ world DOT std DOT com ++++BULK Email severely not wanted+++ Article 3216 of alt.sys.pdp10: Newsgroups: alt.sys.pdp8,alt.sys.pdp10 Path: nntp1.ba.best.com!news1.best.com!nntprelay.mathworks.com!news.mathworks.com!uunet!in5.uu.net!world!world!news From: a11is0n@see.sig (Allison) Subject: Re: Could I build a PDP-8? Sender: news@world.std.com (Mr Usenet Himself) Message-ID: Date: Thu, 28 Aug 1997 03:40:27 GMT References: <19970819210601.RAA08492@ladder01.news.aol.com> <33FB770D.7DB6@InfoMation.comm> <3402AC58.CFFDEB@stoneweb.com> <3402B490.5D06@cisco.com> Nntp-Posting-Host: world.std.com Organization: The World @ Software Tool & Die X-Newsreader: Forte Free Agent 1.0.82 Lines: 19 Xref: nntp1.ba.best.com alt.sys.pdp8:2051 alt.sys.pdp10:3216 "Henry W. Miller" wrote: >Getting back to the original topic of building a PDP-8, I >just recalled that Intersil, a semiconductor company that was based >in Cupertino, CA, and which was purchased by GE in either the late >1980's or early 1990's, came out with a "PDP-8 on a chip" microprocessor >back in the 1980's, I believe. late 70s, 1976 was the introduction of the 6100 and 1981 was the introduction of the 6120 which integrated the 6100 and the 6101. I bought my first 6100 in early '79 for something like $12. Allison Real address is: Allisonp @ world DOT std DOT com ++++BULK Email severely not wanted+++ Article 3218 of alt.sys.pdp10: Newsgroups: alt.sys.pdp8,alt.sys.pdp10 Path: nntp1.ba.best.com!news1.best.com!nntprelay.mathworks.com!howland.erols.net!infeed1.internetmci.com!newsfeed.internetmci.com!192.48.96.123!in1.uu.net!uucp3.uu.net!world!news From: a11is0n@see.sig (Allison) Subject: Re: Could I build a PDP-8? Sender: news@world.std.com (Mr Usenet Himself) Message-ID: Date: Thu, 28 Aug 1997 17:08:12 GMT References: <19970819210601.RAA08492@ladder01.news.aol.com> <33FB770D.7DB6@InfoMation.comm> <3402AC58.CFFDEB@stoneweb.com> <3402B490.5D06@cisco.com> Nntp-Posting-Host: world.std.com Organization: The World @ Software Tool & Die X-Newsreader: Forte Free Agent 1.0.82 Lines: 37 Xref: nntp1.ba.best.com alt.sys.pdp8:2056 alt.sys.pdp10:3218 a11is0n@see.sig (Allison) wrote: >"Henry W. Miller" wrote: >>Getting back to the original topic of building a PDP-8, I >>just recalled that Intersil, a semiconductor company that was based >>in Cupertino, CA, and which was purchased by GE in either the late >>1980's or early 1990's, came out with a "PDP-8 on a chip" microprocessor >>back in the 1980's, I believe. >late 70s, 1976 was the introduction of the 6100 and 1981 was the >introduction of the 6120 which integrated the 6100 and the 6101. >I bought my first 6100 in early '79 for something like $12. The part wasy available and low cost. It was second sourced by harris semi as well. There are several other chips the PIE(Peripheral interface element), MEDIC (Extended memory, DMA and timer). It was widely used for low power control systems in sealed cabs where heat was a concern and the 12 bit word made it more compatable with 10 and 12bit A/D or D/A converters. Another use was to replace larger and harder to maintian PDP-8 system in use. In Kilobaud Microcomputing december 1979 there was an article about the intersil CMOS sampler package. The $49 price got you CPU, 256words ram, 1k rom with ODT, a PIE and UART. For $39 more an etched board it could be put on. I have one and I still play with it as it's extended to 3kw ram and a bunch of parallel ports. Allison Real address is: Allisonp @ world DOT std DOT com ++++BULK Email severely not wanted+++ Article 3419 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!howland.erols.net!nntp.abs.net!news.abs.net!not-for-mail Message-ID: <3A84D116.5642636C@mail.bcpl.net> From: Ken McMonigal X-Mailer: Mozilla 4.61 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: possible pdp-8 good idea for projects here (posted from afc) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 47 Date: Sat, 10 Feb 2001 00:26:46 -0500 NNTP-Posting-Host: 208.242.126.28 X-Complaints-To: abuse@bcpl.net X-Trace: news.abs.net 981782622 208.242.126.28 (Sat, 10 Feb 2001 00:23:42 EST) NNTP-Posting-Date: Sat, 10 Feb 2001 00:23:42 EST Xref: nntp1.ba.best.com alt.sys.pdp10:3419 > Subject: > Re: Oldest operating computer on the internet? > Date: > Sat, 10 Feb 2001 02:18:47 GMT > From: > djg@drs-esg.com (David Gesswein) > Newsgroups: > alt.folklore.computers > References: > 1 , 2 > > > > > In article <3A8392C3.294E9354@ev1.net>, > Charles Richmond writes: > > rhn@nicholson.com wrote: > >> > >> What are some candidates for the oldest operating computer CPU running > >> and available (maybe only periodically) on the public internet (via > >> tcp/ip)? Are there any '70's vintage minicomputers that will respond > >> to a ping? > >> > > Someone had a PDP-8 that you could use via the internet...I think you had > > to telnet...do *not* know about the "ping". > > > That was me, http://www.pdp8.net. The machine itself doesn't talk directly > to the internet, I have a 486 Linux machine that does that. The PDP-8 > has its serial ports connected to the 486 which are access via telnet. > The front panel of the machine is controlled from the 486 parallel port > by a custom board I made. A java interface is provided to allow you to > see the lights and play with the switches. > > The 8/E's were manufactured from end 1970 to 1978, I am not sure when mine > was made, probably towards the latter part of the period. At some point > I will be putting the 8/I next to it online. The 8/I's were made from > 1968 to 1971. > > The PDP-8 never had TCP/IP support that I know of. They did have DECnet. > The 8/E I have is fully configured with 32k 12 bit words but even > that doesn't leave much room for OS, applications, and TCP/IP. They > did fit a multiuser operating system in that. My 8/I have only 8k 12 > bit words. > > David Gesswein > http://www.pdp8.net/ -- Run an old computer with blinkenlights > Article 6469 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nycmny1-snh1.gtei.net!lsanca1-snf1!news.gtei.net!newsfeed2.earthlink.net!newsfeed.earthlink.net!news.mindspring.net!not-for-mail From: djg@pdp8.net (David Gesswein) Newsgroups: alt.sys.pdp10,alt.sys.pdp8 Subject: Re: Local Connection Between one PDP-8 and one PDP-10 and TSS/8 Date: 29 Oct 2001 00:15:49 GMT Organization: www.pdp8.net Lines: 60 Message-ID: <9ri73l$55t$1@slb2.atl.mindspring.net> References: <9rbjpb$1p5$2@bob.news.rcn.net> NNTP-Posting-Host: 42.95.66.5e X-Newsreader: trn 4.0-test76 (Apr 2, 2001) Originator: djg@pdp8.net (David Gesswein) Xref: nntp1.ba.best.com alt.sys.pdp10:6469 alt.sys.pdp8:1000 In article , Sergio Pedraja wrote: > >4) In the TSS/8 manual appears a reference to PT08's. That is the interface > for the ASR33 teletype. Right ? > Yes. It also had a RS-232 option. >5) The manual says that can support up to 32 "teletype controls" (PT08's > or 680/I). It speak too about support for one DA10 Interface, and one > Bit Synchronous Communication Unit (Type 637). About the 680/I, > the manual says it's divided in one DL8/I Data Line Interface and > one Serial Line Multiplexer Unit 685. > This is the description of the Hardware that could be used by the > TSS/8. With independence of the additional emulation possibilities > that it offers, here comes my question itself. > From my small computer handbook. The 685 is simply a switch which allows the 681 to be connected to any one of 64 teletype lines. It has the IOT's for it, a little long to type right now, email if you need it. >6) In case of PT08's... How works this board ? How much boards > could be plugged in one PDP/8 ? What manner ? Must the boards plugged > in one additional cabinet with nothing special ? Are they only used > to manage the ASR33's ? > Where can be located documentation about these units ? > Seems to have same IOT as standard console port. Normally they were Line IOT 1 40/41 2 42/43 3 44/45 4 46/47 5 11/12 Special model for 8/S that went inside an ASR-33 (PT08A) used 03/04 (normal console IOT). I have print set but it does not have programming information other than table above. It uses a W706/W707 which I assume is functionally the same as the M706/M707 in an 8/I but negative bus with RTL IC's. They were about a half row double high in a 19" rack mount. Attached to the external negative bus on original 8, 8/S and 8/I. I have one but have not tried it yet. How many you can have is the external bus limit on number of devices/cable length. If you think the print set will be of use I can put it on the scan pile. >9) FInally, it's the Sync.Comm.Unit 637. What do exactly this unit ? How > treat it the TSS/8 ? Where can be located documentation ? > Didn't find it but the books aren't big on indexes. David Gesswein http://www.pdp8.net/ -- Run an old computer with blinkenlights Have any PDP-8 stuff you're willing to part with? Article: 114107 of alt.folklore.computers Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newspeer1.nwr.nac.net!news.astraweb.com!news-small.astraweb.com!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!not-for-mail From: John Everett Newsgroups: alt.folklore.computers Subject: Re: TSS/8 and RTS8 ? Message-ID: References: X-Newsreader: Forte Agent 1.9/32.560 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 66 Date: Thu, 27 Feb 2003 15:24:58 GMT NNTP-Posting-Host: 209.86.101.145 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1046359498 209.86.101.145 (Thu, 27 Feb 2003 07:24:58 PST) NNTP-Posting-Date: Thu, 27 Feb 2003 07:24:58 PST Organization: EarthLink Inc. -- http://www.EarthLink.net Xref: dfw-artgen alt.folklore.computers:114107 On Wed, 26 Feb 2003 09:15:51 -0800, "George R. Gonzalez" wrote: >Way back, DEC came out with the "time share" option for the PDP-8. >This was an amazingly RISCy way to virtualize the PDP-8. > >Basically if you enabled "time sharing" mode, all the dangerous >instructions, >like HLT, read switch register, and all the I/O instructions, would >interrupt to the monitor, which would do the proper steps to make it look >like the instruction executed. Since all the memory-bank switching >instructions had been shoehowned into the I/O instruction space, this also >allowed the monitor to do paging and virtual memory! > >Just like today, the basic DOS (OS/8) didnt do anything with this feature, >but TSS/8 and RTS8 did! > >Looking thru the RTS8 source code, there was a big section that looked like >it may have been bodily copied from TSS/8. > >I wonder if the cognoscenti in this NG can elaborate on what was the >relationship between TSS/8 and RTS8. One might guess that they were pretty >close cousins. This cognoscente can't elaborate on the relationship between TSS/8 and RTS8, but as co-author of TSS/8 I can shed some light on the origins of that system and its supporting hardware. During the late '60s Gordon Bell had taken a "leave" from DEC and was a professor of computer science at Carnegie-Mellon. He had a Dutch grad student named Adrian Van Der Goor (sp?). Somehow between the two of them they came up with the idea that with some small hardware mods a time-sharing system could be built for the PDP-8. Gordon and Adrian visited Maynard sometime in 1967 or 1968 (I'm guessing) and made a presentation to some execs and also some of us in the (tiny) PDP-8 programming group. Gordon sort of just oversaw the meeting; Adrian made the presentation. The proposal was to create a "time-sharing option" which would create "user mode" and "exec mode" run states. In user mode all I/O operations, HLT, switch register reads, etc. would cause an interrupt. Of course at interrupt the system would switch into exec mode. The proposed time-sharing monitor would reside in page 0, while user images and swappable parts of the monitor would reside in extended memory. It was proposed to give each user a 4k virtual (basic) PDP-8. The design of the monitor (as we then called operating systems) would be based upon a research system originally proposed (and perhaps actually built) for a PDP-4 or PDP-7. I can't now recall where the design came from, but it was described in detail in a computer science text of the period. The Time-Sharing Option was designed by a junior member of PDP-8 engineering who's name I've long since forgotten. Don Witcraft, who had worked on the PDP-10 monitor (later to be called TOPS-10) and I wrote TSS/8 based upon the suggestions made by Gordon and Adrian. BTW, the system was initially to be called TS-8, a name which persisted during most of its development. We changed it to TSS/8 when IBM announced TSS/360. It just seemed so right! One of the catch phrases around DEC at the time, "Man is going to the moon, and time-sharing on the PDP-8." jeverett3earthlinknet http://home.earthlink.net/~jeverett3 Article: 19462 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!nntp-relay.ihug.net!ihug.co.nz!news.linkpendium.com!xuxa.iecc.com!not-for-mail From: johnl@iecc.com (John R. Levine) Newsgroups: alt.folklore.computers,alt.sys.pdp10 Subject: Re: Speed of APL on 360s, was Any DEC 340 Display System Doco ? Date: 10 Apr 2003 22:22:21 -0400 Organization: I.E.C.C., Trumansburg NY USA Lines: 63 Message-ID: References: <3e8ae086.45754328@news.m.iinet.net.au> NNTP-Posting-Host: xuxa.iecc.com X-Trace: xuxa.iecc.com 1050027741 29695 208.31.42.42 (11 Apr 2003 02:22:21 GMT) X-Complaints-To: abuse@iecc.com NNTP-Posting-Date: 11 Apr 2003 02:22:21 GMT Xref: dfw-artgen alt.folklore.computers:119151 alt.sys.pdp10:19462 >I think though that APL\360 itself was slow. You didn't have all that >large a workspace -- I think 32K -- yet even memory primitives took >noticable time. It was pretty slow, since it just cranked away one operator at a time creating and discardit array temporaries. The reason APL\360 gained popularity was that it was well designed for time-sharing with very efficient context switching and monitor that let you run dozens of users on a 360/50 with snappy response. In later years people started doing compilation with interesting optimizations like type and shape inference, but that wasn't until the mid 1970s. I wrote an APL subset to Basic compiler as a class project in 1972 (written in Trac, believe it or not) that built up Basic expressions representing arbitrary elements of APL expressions and only generated loops when it needed to store an array or print a value. I later realized it was considerably ahead of the then state of the art in APL compilers. > On the other hand, I don't know how fast a 360/67 was. It was > supposed to have been one of the faster models of 360s (this was > early 1970s) but I recall that a 360/30 was *very* slow. I'm SO GLAD you asked, since I happen to have the Functional Characteristics manuals for the 360/67 and 360/30 here. (I bought them from IBM's web site a year or two ago.) The /67 was a pretty fast machine for the mid 1960s with 32 bit integer datapaths, 200ns cycle time, and a 60 bit adder for floating point. Register to register integer load or add was 650 ns, memory-to-register load was 1.2us, add 1.4us. Floating add from memory 2.43us single, 2.45us double. Floating multiply 4.4us short, 7.6us long. Floating divide 7.3us short, 14.1us long. Floating compare 1.98us, conditional branch 800ns not taken, 1.1us taken. The dual processor machine was slightly slower, 690 ns load, to 14.33us floating divide. The /30, on the other hand, was really slow. It was byte serial, 1us cycle time and 2us memory cycle on early machines, later sped up to 750ns and 1.5us, and microcode did everything including simulating channel I/O. Register to register integer load was 17us, add was 22us, memory-to-register load was 24us, add 29us. Floating add from memory 52us single, 69us double. Floating multiply 208us short, 472us long. Floating divide 301us short, 1665us long. Floating compare 65us, conditional branch 21us not taken, 22us taken. > The stupid prime number calculation program (check all odd numbers > from 3 to the square root to see if any are a factor) was faster in > interpreted FOCAL on a PDP-8/e than in compiled Fortran on the > 360/30. I'm not surprised. The 8/e was a 12 bit machine with 1.2us memory and hard-wired instructions that ran the memory at full speed, giving it a head start over the /30's 1.5us memory and microcoded instruction set. Also, the floating point package around which Focal was built was quite well written. -- John R. Levine, IECC, POB 727, Trumansburg NY 14886 +1 607 387 6869 johnl@iecc.com, Village Trustee and Sewer Commissioner, http://iecc.com/johnl, Member, Provisional board, Coalition Against Unsolicited Commercial E-mail