Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 6 Mar 91 05:56:15 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Wed, 6 Mar 91 05:53:14 EST Received: from sunic.sunet.se by life.ai.mit.edu (4.1/AI-4.10) id AA07376; Wed, 6 Mar 91 05:20:19 EST Received: from AIDA.CSD.UU.SE by sunic.sunet.se (5.61+IDA/KTH/LTH/1.186) id AAsunic19522; Wed, 6 Mar 91 11:19:17 +0100 Date: Wed 6 Mar 91 11:14:36 From: Johnny Billquist To: dciem!nrcaer!gandalf!mwoodwar@csri.toronto.edu Cc: pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se In-Reply-To: <9102281534.AA00855@gandalf.UUCP> Message-Id: <910306111436.30.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> Date: Wed 6 Mar 91 11:14:36 From: Johnny Billquist To: dciem!nrcaer!gandalf!mwoodwar@csri.toronto.edu Cc: pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se In-Reply-To: <9102281534.AA00855@gandalf.UUCP> As far as I know, nothing has been set up yet, I'm still hoping, though... A few different ideas has been in the air, but things been very quiet the last few months. What do all you people say? (There must be more people out there...) If we could get an archive site up and running, there is a lot of stuff that could be placed there. Anybody got any good suggestions? Johnny Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 6 Mar 91 18:54:32 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Wed, 6 Mar 91 18:51:34 EST Received: from akbar.cac.washington.edu by life.ai.mit.edu (4.1/AI-4.10) id AA00592; Wed, 6 Mar 91 18:13:45 EST Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu (5.65/UW-NDC Revision: 2.21 ) id AA27364; Wed, 6 Mar 91 15:13:06 -0800 Date: Wed, 6 Mar 1991 15:06:09 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: latest To: PDP-8 Lovers Message-Id: Date: Wed, 6 Mar 1991 15:06:09 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: latest To: PDP-8 Lovers Hi - My home menagerie received some help a few weeks ago (thank you, Messrs Lothberg and Dempster!), and I once again have a working DECSYSTEM-2020 system. I ended up replacing all of its disk drives; it now has three RM03 drives and one RM05. I have a spare RM03 and RM05; I also have four dead RM03's. On the PDP-8 front I have a "mostly working" PDP-8/f with 12K. I can key in the RIM loader, get it to load the BIN loader, and get that to load FOCAL, but when I try to run FOCAL I immediately get an error. Typing any character hangs it, but CTRL/C gives me an error and back to the prompt. I don't know if it's the interrupt system or what. Any ideas? -- Mark -- Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 6 Mar 91 20:02:01 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Wed, 6 Mar 91 19:59:09 EST Received: from gateway.sequent.com ([138.95.18.1]) by life.ai.mit.edu (4.1/AI-4.10) id AA01992; Wed, 6 Mar 91 18:57:15 EST Received: from eng3.sequent.com by gateway.sequent.com (5.61/1.34) id AA20897; Wed, 6 Mar 91 15:57:21 -0800 Received: by eng3.sequent.com (5.61/1.34) id AA22128; Wed, 6 Mar 91 15:56:08 -0800 Message-Id: <9103062356.AA22128@eng3.sequent.com> To: mrc@cac.washington.edu Cc: PDP-8 Lovers Subject: Re: latest In-Reply-To: Your message of "Wed, 06 Mar 91 15:06:09 PST." Date: Wed, 06 Mar 91 15:56:06 PST From: vandys@sequent.com To: mrc@cac.washington.edu Cc: PDP-8 Lovers Subject: Re: latest In-Reply-To: Your message of "Wed, 06 Mar 91 15:06:09 PST." Date: Wed, 06 Mar 91 15:56:06 PST From: vandys@sequent.com Cool! Hey, is your '20 on the Internet? God, would I love to have a little access to one of those guys again.... My own 8/E is still waiting for a little of my time :-(. Good job! Andy Valencia vandys@sequent.com Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 6 Mar 91 20:48:56 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Wed, 6 Mar 91 20:46:05 EST Received: from akbar.cac.washington.edu by life.ai.mit.edu (4.1/AI-4.10) id AA03294; Wed, 6 Mar 91 19:44:56 EST Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu (5.65/UW-NDC Revision: 2.21 ) id AA00944; Wed, 6 Mar 91 16:44:50 -0800 Date: Wed, 6 Mar 1991 16:27:45 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: Re: latest To: vandys@sequent.com Cc: PDP-8 Lovers In-Reply-To: <9103062356.AA22128@eng3.sequent.com> Message-Id: Date: Wed, 6 Mar 1991 16:27:45 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: Re: latest To: vandys@sequent.com Cc: PDP-8 Lovers In-Reply-To: <9103062356.AA22128@eng3.sequent.com> The 20 isn't on the Internet, but it is e-mail addressable as PANDA.PANDA.COM; that is, my e-mail address is MRC@PANDA.PANDA.COM there although MRC@PANDA.COM will also work. PANDA.COM and *.PANDA.COM are MX records that point to WSMR- SIMTEL20.ARMY.MIL which has an e-mail link with PANDA via a dial-up email protocol I invented called Cafard (this was in the days when AT&T considered the UUCP protocol proprietary). Actually, this is the second revival of PANDA. PANDA (and its e-mail connectivity) came into existance in 1985, and was shut down at the end of September 1988. In late spring 1989, living in a different state with a new job, house, and lady, I brought PANDA back up but it was seriously damaged by moving and several months' storage in an unheated damp storage locker. It was unreliable (including a head crash on my source disk) and died solidly that autumn. An unsucessful attempt was made to fix PANDA one weekend in 1990 (another head crash), but it wasn't until February 15-17 1991 that success was achieved. The repair effort involved almost totally stripping one 2020 (the former San Mateo County Office of Education system). A second 2020 (the former Stanford University Knowledge Systems Lab system) also supplied parts, but it is in an almost complete condition and may be brought up as TONTON.PANDA.COM. It also involved retiring all of PANDA's original disk drives, and replacing them with RM03 drives acquired from a surplus electronics store in Seattle and with RM05 drives donated by Boeing. The PANDA.COM menagerie is soon to get a new member; a 68040 NeXTstation running version 1.1 of the NeXT operating system. This machine will be called IKKOKU-KAN.PANDA.COM. If all goes well, it will be Internet accessible via a 9600 baud SLIP link to TOMOBIKI-CHO.CAC.WASHINGTON.EDU (my work NeXT). Work is underway to get PANDA.PANDA.COM talking TCP/IP. If successful, I will attempt to have an Ethernet link between PANDA, IKKOKU-KAN, and possibly TONTON. Although my initial TOPS-20 IP debugging will use SLIP, 9600 baud asynchronous input to a 2020 is too horrible to contemplate and I will never run the machine in that mode. Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 6 Mar 91 22:30:01 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Wed, 6 Mar 91 22:27:03 EST Received: from ama (ama.caltech.edu) by life.ai.mit.edu (4.1/AI-4.10) id AA05999; Wed, 6 Mar 91 22:06:27 EST Received: from laplace.ama.caltech.edu by ama (4.1/1.2) id AA16201; Wed, 6 Mar 91 18:04:20 PST Date: Wed, 6 Mar 91 18:04:20 PST From: ph@ama.caltech.edu (Paul Hardy) Message-Id: <9103070204.AA16201@ama> Received: by laplace.ama.caltech.edu (4.1/SMI-4.0) id AA02051; Wed, 6 Mar 91 17:40:51 PST To: vandys@sequent.com Cc: mrc@cac.washington.edu, PDP8-lovers@ai.mit.edu In-Reply-To: vandys@sequent.com's message of Wed, 06 Mar 91 15:56:06 PST <9103062356.AA22128@eng3.sequent.com> Subject: latest Date: Wed, 6 Mar 91 18:04:20 PST From: ph@ama.caltech.edu (Paul Hardy) To: vandys@sequent.com Cc: mrc@cac.washington.edu, PDP8-lovers@ai.mit.edu In-Reply-To: vandys@sequent.com's message of Wed, 06 Mar 91 15:56:06 PST <9103062356.AA22128@eng3.sequent.com> Subject: latest Gee, a PDP-8/E. This was the first PDP-8 I used. It had an ASR-33 and a (very!) small disk. I don't even remember the type, but it held something like 64 kwords. It had FOCAL on disk, as well as PAL (don't remmeber which version). The actual disk unit looked a little like the robot's head in Lost In Space (not see-through, but there was a cylinder, with a drive shaft coming down from it, and that's what I thought of when I looked at it). I don't remember what type it was -- RS08 or RF08 or something like that. It's been almost 15 years since I've used it. I've kept my programmer's manual, though. If you ever want to get rid of your 8... Does anyone have docs for a DECMate I by the way? I have a motherboard, some sort of backplane, and another board but that's it. Not even a power supply (or cabinet for that matter). I would at least like to hook up power to the thing and get a prompt from it. As for the '20, I'd settle for modem access! If there is too little interest for a PDP-20 lovers mailing list, maybe there's enough interest in a PDP-20 system to log into. Mark??? --Paul Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 6 Mar 91 23:38:58 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Wed, 6 Mar 91 23:35:56 EST Received: from watsun.cc.columbia.edu by life.ai.mit.edu (4.1/AI-4.10) id AA08074; Wed, 6 Mar 91 23:12:59 EST Received: by watsun.cc.columbia.edu (5.59/FCB) id AA13799; Wed, 6 Mar 91 23:12:35 EST Date: Wed, 6 Mar 91 23:12:34 EST From: Charles Lasner To: ph@ama.caltech.edu (Paul Hardy) Cc: vandys@sequent.com, mrc@cac.washington.edu, PDP8-lovers@ai.mit.edu Subject: Re: latest In-Reply-To: Your message of Wed, 6 Mar 91 18:04:20 PST Message-Id: Date: Wed, 6 Mar 91 23:12:34 EST From: Charles Lasner To: ph@ama.caltech.edu (Paul Hardy) Cc: vandys@sequent.com, mrc@cac.washington.edu, PDP8-lovers@ai.mit.edu Subject: Re: latest In-Reply-To: Your message of Wed, 6 Mar 91 18:04:20 PST The disk is/was the DF32 and it has only 32K words on it, but could be expanded with up to three more DS32's for a total of 128K words! The RF08 is its big brother which starts at 1 platter with 256K words and expands with RS08 expanders to 1024K words normally, but a few were bundled with the DV08 data verifier option to be capable of up to 7 expanders for a total of 2048K words. The DF32 was a negative bus peripheral, as was the RF08, but there was also the DF32D and DS32D which were the positive!negative bus versions. Most TTL -8 peripherals for the pos. bus could have negative counterparts merely by doing a cookbook buss module and cable swap. They are designated -P or -N accordingly. You are obviously describing the 4K Disk Monitor system, which is not the only or even the best software for this hardware, but was the only DEC-sold choice prior to 1970. cjl Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Thu, 7 Mar 91 00:51:05 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Thu, 7 Mar 91 00:48:13 EST Received: from akbar.cac.washington.edu by life.ai.mit.edu (4.1/AI-4.10) id AA10406; Thu, 7 Mar 91 00:28:25 EST Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu (5.65/UW-NDC Revision: 2.21 ) id AA07049; Wed, 6 Mar 91 21:28:21 -0800 Date: Wed, 6 Mar 1991 21:09:20 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: re: latest To: ph@ama.caltech.edu Cc: PDP8-lovers@ai.mit.edu In-Reply-To: <9103070204.AA16201@ama> Message-Id: Date: Wed, 6 Mar 1991 21:09:20 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: re: latest To: ph@ama.caltech.edu Cc: PDP8-lovers@ai.mit.edu In-Reply-To: <9103070204.AA16201@ama> In <9103070204.AA16201@ama>, ph@ama.caltech.edu writes: >As for the '20, I'd settle for modem access! If there is too little >interest for a PDP-20 lovers mailing list, maybe there's enough interest >in a PDP-20 system to log into. Mark??? Hmm. I'll have to think about it. There are two 1200 baud dialups on PANDA, but I'm a little worried about allowing random people to log in, particular as my tape drive is still down, hence no backups. There's enough disk space, and enough spare cycles, but giving away "idle" resources is not necessarily a no- cost operation, as I and others have found out the hard way... I may be willing to make accounts on TONTON (whose boot disk is built from PANDA's, hence no backup needed) once its disk channel is working (and preferably get the second port on its disk drive working so the drive can be accessed from both PANDA and TONTON). It would cost me approximately $1000/year (about 2KW for the CPU/disk plus extra air conditioning load, at about 5.5 cents/KWH) extra in electricity to run the second machine. Is there enough interest that would enable me to recover some or all of this cost? Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Thu, 7 Mar 91 01:34:58 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Thu, 7 Mar 91 01:32:07 EST Received: from its.rpi.edu by life.ai.mit.edu (4.1/AI-4.10) id AA11329; Thu, 7 Mar 91 01:08:18 EST Received: from sub.aix.rpi.edu (aix01srv.aix.rpi.edu) by its.rpi.edu (4.1/RPI-ITS-HUB19); id AA22879; Thu, 7 Mar 91 01:09:31 EST for pdp8-lovers@ai.mit.edu From: Matthew S Rosen Received: by sub.aix.rpi.edu (5.61-AIX-1.2/RPI-ITS-SUB10-AIX); id AA100698; Thu, 7 Mar 91 01:03:34 -0500 for pdp8-lovers@ai.mit.edu Date: Thu, 7 Mar 91 01:03:34 -0500 Message-Id: <9103070603.AA100698@sub.aix.rpi.edu> To: pdp8-lovers@ai.mit.edu Subject: Archive site! (maybe...) From: Matthew S Rosen Date: Thu, 7 Mar 91 01:03:34 -0500 To: pdp8-lovers@ai.mit.edu Subject: Archive site! (maybe...) Hi. I may have found an archive (ftp) site for PDP8 lovers here at RPI.. What I need is an idea of how much disk we would need, and how many people would find a use in it. I suppose the asiest thing to do would to email (or post to the list..) they type (and size...) of things that would be put on it as well as your own interest in the site (i.e. Would you find it useful..) As for me, I got a PDP8/f when I was 10 or so, with a TTY (110 baud) paper tape reader/punch, and 4K core. Yup. Rim loader city when the core got trashed by some misbehaving code.. How? How did I have an 8 when I was 10? Well, my dad (Arosen@ccmail.sunysb.edu) used/uses them in a Neurophysilogical research lab, doing real time data acquisation and processing (I'm sure he'd like to describe the software and hardware to anyone interested.. He spent quite some time designing and building it..) So, I got the spare 8. Programing in assembly, and 8K basic when I finally got the memory.. No onder I'm hopelessly out of date with the "modern" world of computers.. Anyway, My machine now is an 8/f with 32K MOS memory, VT220, EAE, dule RX01 Floppy, kludeged (sorry dad!) high speed paper tape reader, and misc. other stuff. Dad's got an 8/e, 32K core, Expansion box, RK05 hard disk, and duel RX02 (?) work-alikes, + some other neat stuff like a 4096 x 4096 graphics display.... (wanna talk about it, dad?) So, if anyone has any ideas about the archive site, let me know. -Matt mrosen@rpi.edu Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Thu, 7 Mar 91 08:01:34 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Thu, 7 Mar 91 07:58:44 EST Received: from watsun.cc.columbia.edu by life.ai.mit.edu (4.1/AI-4.10) id AA18681; Thu, 7 Mar 91 07:40:54 EST Received: by watsun.cc.columbia.edu (5.59/FCB) id AA17749; Thu, 7 Mar 91 06:57:38 EST Date: Thu, 7 Mar 91 6:57:37 EST From: Charles Lasner To: Matthew S Rosen Cc: pdp8-lovers@ai.mit.edu Subject: Re: Archive site! (maybe...) In-Reply-To: Your message of Thu, 7 Mar 91 01:03:34 -0500 Message-Id: Date: Thu, 7 Mar 91 6:57:37 EST From: Charles Lasner To: Matthew S Rosen Cc: pdp8-lovers@ai.mit.edu Subject: Re: Archive site! (maybe...) In-Reply-To: Your message of Thu, 7 Mar 91 01:03:34 -0500 Ok, here is a rough guess: I would like to see all OS/8 family sources on-line including all known variants such as release-dependent versions and user-created versions, etc. This includes the original DEC sources as well as user add-ons, etc. In addition, I would like to see all TSS8 and DM sources, as well as all known machine readable manuals for all of the above. My operating system, P?S/8 is proprietary, so it will NOT be posted here; interested parties can contact me off the net for this software. Certain diagnostics have their sources and/or writeups available as files as well. I would also encourage the posting of the LINC family source files as well, since they can be converted to an ASCII format PAL language as used on LAP6-DIAL, P?S/8 PAL, and DIAL10 for TOPS10/20. This would get us all of the LINC-8 and PDP-12 files that were available. The DECmate family also hopefully contributes to the collection as well. We would welcome the sources for GTE, Master Menu, the hard disk formatter, System Test Diskette, DEC-X8 diagnostics, etc., and of course any PDP-8 specific games like adventure-8, spacewar, and editors like TECO-8 and the DECmate Edit program, CPODT, etc. Assuming all of this were to be submitted, I can't imagine it taking more than 50 Megabytes compressed.Z format. BTW, KERMIT-12 need not be placed in this archive, as watsun.cc.columbia.edu serves as the home of all kermits, not just mine. cjl (lasner@watsun.cc.columbia.edu and author of KERMIT-12) ps: I can contribute a fair amount of the above and I know others that would also. Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Thu, 7 Mar 91 09:16:35 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Thu, 7 Mar 91 09:13:44 EST Received: from mcsun.EU.net by life.ai.mit.edu (4.1/AI-4.10) id AA19543; Thu, 7 Mar 91 08:43:48 EST Received: from kestrel.ukc.ac.uk by mcsun.EU.net with SMTP; id AA28350 (5.65a/CWI-2.76); Thu, 7 Mar 91 14:43:37 +0100 Received: from axion.bt.co.uk by kestrel.Ukc.AC.UK via PSS (UKC CAMEL FTP) id aa22705; 7 Mar 91 12:37 GMT Received: from kbss by zaphod.axion.bt.co.uk via SMTP channel id aa21178; 7 Mar 91 6:36 GMT Received: from topaz.kbss.bt.co.uk by jasper.kbss.bt.co.uk; Thu, 7 Mar 91 12:34:44 GMT From: Chris Green Date: Thu, 7 Mar 91 12:34:41 GMT Message-Id: <991.9103071234@topaz.kbss.bt.co.uk> To: PDP8-lovers@ai.mit.edu, to@kbss.bt.co.uk Subject: Re: latest, and also PDP-12's From: Chris Green Date: Thu, 7 Mar 91 12:34:41 GMT To: PDP8-lovers@ai.mit.edu, to@kbss.bt.co.uk Subject: Re: latest, and also PDP-12's Quite a burst of activity recently, it awoke my memories. 1974 to 1977 I worked on PDP-12's, anyone else out there know that lovely beast? The one I spent most time on was in Saudi Arabia and had 16k of memory (a big one!) and a 256k RF08 disk plus lots of other peripherals like a fast lineprinter and so on. The PDP-12 was great fun, it had a PDP-8/E (I think it was a /E) in it plus a LINC machine, you could switch from one instruction set to the other 'on the fly'. The LINC was the only machine I have ever worked on which did 1's complement arithmetic. Chris Green (chris@kbss.bt.co.uk or cgreen@essex.ac.uk) Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Thu, 7 Mar 91 15:48:01 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Thu, 7 Mar 91 15:45:09 EST Received: from watsun.cc.columbia.edu by life.ai.mit.edu (4.1/AI-4.10) id AA00289; Thu, 7 Mar 91 14:58:14 EST Received: by watsun.cc.columbia.edu (5.59/FCB) id AA21035; Thu, 7 Mar 91 14:57:56 EST Date: Thu, 7 Mar 91 14:57:55 EST From: Charles Lasner To: Chris Green Cc: PDP8-lovers@ai.mit.edu, to@kbss.bt.co.uk Subject: Re: latest, and also PDP-12's In-Reply-To: Your message of Thu, 7 Mar 91 12:34:41 GMT Message-Id: Date: Thu, 7 Mar 91 14:57:55 EST From: Charles Lasner To: Chris Green Cc: PDP8-lovers@ai.mit.edu, to@kbss.bt.co.uk Subject: Re: latest, and also PDP-12's In-Reply-To: Your message of Thu, 7 Mar 91 12:34:41 GMT Not quite: The PDP-12 is an 8/i with a LINC "grafted" on. It was originally to be called LINC-8/i as some of the modules actually say! There is no separate machine, rather the processor is integrated as both with mode changing instructions. There is no peripheral implementation as implied or coprocessor concept either. Unlike its predecessor, the LINC-8, there is only one set of registers, and yes, while in LINC mode, the arithmetic instructions of the LINC prevail which are all but one a set of one's complement arithmetic instructions. The instructions are totally different, and the similarity ends at the fact that they both have 12 bits. Addressing schemes are thoroughly incompatible as well, and sometimes "clash" in terms of OS conventions, interrupts, etc. Late model PDP-12s also have hardware Automatic Priority Interrupt (API) which never made it into the newer -8s. Further, the FPP-12 had a special lockout mode when it was enabled and connected only to a -12 which speeds up FPP Fortran. (This was reimplemented in the FPP-8/e,a for the newer machines though.) EAE for these machines is still an option, not built-in even though so much of the hardware is required to be a LINC anyway. I believe two cards gives you EAE here. Let's not forget the PDP-12 console: virtually every register is displayed simultaneously in individual bit lamps (and the bulbs plug in, not soldered, and are replaceable today with LEDs). There are two switch registers each 12 switches, 6 sense switches, 8 knobs, 8 a-d stereo phone jacks (optionally up to 16 more) all available on the front panel as input devices. There are also 12 TTL inputs available (not on the front panel though) and 6 SPDT relays as well. And let's not forget the 12" 'scope display which is 512x512 complete with character generator hardware (well, 1/2 char anyway :-), and a formica table top to lean on in front of the machine. The switches are horizontally mounted rocker switches. The console allows fetch/exec stop, automatic restart delay (for slow-motion operation) variable over four decades of speed. While there is no hardware boot per se, there is the ability to execute an instruction directly off of the switches, not even in memory, so booting the machine generally consists of executing a tape read instruction and manual startup. The console functions include filling or examining memory via hardware autoincrement, so you can just watch the lights until you see something "important" should that matter, without even having to hit a step key (which is also available if you want), and a distinctive feature: the ability to examine a location and change its contents without having to reestablish the address. Further, the address is specified on one set of switches, and the contents being stored there on the other set which allows multiple addresses to be patched with the same value with minimal effort. Since this machine has true fetch/exec stop, it is possible to debug interrupt-driven code effectively without modifying it. Should a desired instruction have been left out, just execute it on the switches and keep going. We often forget to raise the TTY output flag, so a routine would hang. Just stop the machine, put 6046 on the switches, and press DO, and then continue and you're on your way. The PDP-12 was clearly the "ergonomic" top-of-the-line, as well as having little subtle touches as well, such as nylon washers on the door hinge points to prevent squeaking. A PDP-8 "cadillac" if there ever was one. cjl From N.M.Troop@loughborough.ac.uk Thu Mar 7 20:30:33 1991 Return-Path: Received: from cunixf.cc.columbia.edu by watsun.cc.columbia.edu (5.59/FCB) id AA23828; Thu, 7 Mar 91 20:30:05 EST Received: from life.ai.mit.edu by cunixf.cc.columbia.edu (5.59/FCB) id AA12857; Thu, 7 Mar 91 20:29:43 EST Received: from sun2.nsfnet-relay.ac.uk by life.ai.mit.edu (4.1/AI-4.10) id AA09519; Thu, 7 Mar 91 20:04:11 EST Received: from loughborough.ac.uk by sun2.nsfnet-relay.ac.uk via JANET with NIFTP id <12041-0@sun2.nsfnet-relay.ac.uk>; Thu, 7 Mar 1991 12:39:08 +0000 Message-Id: <697.9103071241@hpc.lut.ac.uk> From: Nick Troop Subject: reuest To: pdp8-lovers@ai.mit.edu Date: Thu, 7 Mar 91 12:41:27 GMT X-Mailer: ELM [version 2.3 PL0 (LUT)] help Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Fri, 8 Mar 91 07:05:42 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Fri, 8 Mar 91 07:02:52 EST Received: from mcsun.EU.net by life.ai.mit.edu (4.1/AI-4.10) id AA18938; Fri, 8 Mar 91 04:30:55 EST Received: from kestrel.ukc.ac.uk by mcsun.EU.net with SMTP; id AA13453 (5.65a/CWI-2.76); Fri, 8 Mar 91 10:30:44 +0100 Received: from axion.bt.co.uk by kestrel.Ukc.AC.UK via PSS (UKC CAMEL FTP) id aa22261; 8 Mar 91 9:25 GMT Received: from kbss by zaphod.axion.bt.co.uk via SMTP channel id aa28666; 8 Mar 91 3:25 GMT Received: from topaz.kbss.bt.co.uk by jasper.kbss.bt.co.uk; Fri, 8 Mar 91 09:23:35 GMT From: Chris Green Date: Fri, 8 Mar 91 09:23:34 GMT Message-Id: <7456.9103080923@topaz.kbss.bt.co.uk> To: PDP8-lovers@ai.mit.edu Subject: PDP-12 continued From: Chris Green Date: Fri, 8 Mar 91 09:23:34 GMT To: PDP8-lovers@ai.mit.edu Subject: PDP-12 continued Re cjl's long message on the PDP-12, thanks for the clarification and extra description - it's a long time since I worked on the PDP-12! Yes, of course, it's a PDP8/i plus a LINC machine. And don't I remember the horrendous problems brought on by being able to switch between the two, as you say incompatible, architectures. In particular the wierd and wonderful things that had to be done to handle interrupts, so that you could switch to the right mode to execute the interrupt handler and then when finished return to the right mode and address. And yes, the PDP-12 console was wonderful, always gave one the image of a mad professor with his machine, all those flashing lights. I wish they had been LED's when I used it though as it was a weekly service job to replace all the blown bulbs. Users of other computers at the time use to envy the simple boot sequence on the PDP-12 because of being able to put a tape read instruction in the switch registers and execute it, 0701 7310 wasn't it, followed by loading the start address and off you go - something like that anyway. The 'scope display was also something special as the programmers editor on the PDP-12 used it as a VDU (though one still had to use the KSR-35 as a keyboard) and it was actually quite a civilised programming env- ironment, especially with single stepping, break points and so on easily available from the switches on the console. The systems I worked on were actually 'Clinilab-12' hospital laboratory systems, I think these were about the last PDP-12's that DEC actively sold. Certainly when I was involved DEC was withdrawing support for the Clinilab-12 and my company (G.D.Searle) attempted to take over the market by continuing to support them, unsuccessfully I fear! I spent quite some time at DEC in Maynard picking the brains of the few remaining PDP-12 experts at that time. I was soon completely on my own though and even more so when I went out to Saudi Arabia, all the other hospital computers there were PDP-11's (new fangled 16 bit machines!!). Chris Green Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Fri, 8 Mar 91 18:48:06 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Fri, 8 Mar 91 18:45:09 EST Received: from watsun.cc.columbia.edu by life.ai.mit.edu (4.1/AI-4.10) id AA04542; Fri, 8 Mar 91 18:07:31 EST Received: by watsun.cc.columbia.edu (5.59/FCB) id AA03773; Fri, 8 Mar 91 18:06:54 EST Date: Fri, 8 Mar 91 18:06:53 EST From: Charles Lasner To: Chris Green Cc: PDP8-lovers@ai.mit.edu Subject: Re: PDP-12 continued In-Reply-To: Your message of Fri, 8 Mar 91 09:23:34 GMT Message-Id: Date: Fri, 8 Mar 91 18:06:53 EST From: Charles Lasner To: Chris Green Cc: PDP8-lovers@ai.mit.edu Subject: Re: PDP-12 continued In-Reply-To: Your message of Fri, 8 Mar 91 09:23:34 GMT 0701 7310 was the tape operation to reconfigure and boot LAP6-DIAL's (potentially) disk-based DIAL-MS system. The typical boot is 0701 7300 which is a relic from the LINC-8 days. With DIAL-MS, it means that the tape is loaded, and then re-boots to the current configuration. 0701 7310 does a different thing: it guessworks the best available configuration that DIAL can run from, which might be merely the tapes if that's all that's available, but will auto-configure to 1 or 2 DF32s, RF08, or RK08 if available, rewrite the configuration on the tape, and then do the equivalent of 0701 7300 to boot it up. It is curious that a tape is ALWAYS involved in booting a disk-oriented system. Btw, you used Clinical-Lab-12 systems? So, your machines were blue! Most of us have green PDP-12s in "normal" grey-side cabinets, not white cabinets and PDP-15-color rocker switches! Both of mine are green, but I know another guy with machines of both colors, as he was also a Clinical-Lab-12 person. Dialco and others sell a plug-in replacement LED with limiting resistor built-in to use instead of the bulbs which really looks like an incandescent bulb. DEC uses them disks like RK05 to replace the same lamp there, but the Dialco LED is brighter; there is also an alternate vendor, but the name escapes me; may have it on file though. Regarding interrupts: Much of our PDP-8 code for the -12 has various instruction sequences in them to prevent the interrupt problem. While not widely documented, there are a few tricks that work: IOF {whatever} ION This is the classic method to prevent interrupts. The whatever part might include the sequence: LINC {whatever in LINC code} PDP It is vital that the code after the LINC and before the PDP not be interrupted. One problem with this is that you have to have the ION and IOF instructions so the code is longer, slower, etc., and unnecessarily prevents interrupts during the sequence. A better way on PDP-8's with extended memory is to use: CIF 00 /or whatever the current field is LINC {whatever} PDP {whatever without a JMP or JMS} JMP WHEREVER /or JMS WHEREVER This takes care of the interrupt by inhibiting it after the CIF until the next JMP or JMS, and doesn't need an explicit IOF. The problem is on 4K machines where CIF is not implemented because there is no extended memory hardware. If a program must be viable on a 4K machine, you can't use this trick. Well, it turns out that even 4K PDP-12's *DO* implement the memory extension control, just no extended memory. The reason is that the PDP-12 extended memory control understands the LINC portion of the machine as a special case, and maintains LINC status. All PDP-12's are extended memory as far is the LINC is concerned, since the LINC is a 1K-oriented architecture with an additional 1K data-only segment. So even a lowly 4K PDP-12 has to maintain two bits of every register, so the instructions for memory control ALWAYS work! So, for example, the P?S/8 system handler for the PDP-12 with 4K or more allows interrupts during the I/O call. The few LINC operations during the handler are interrupt-protected with a leading CIF instruction. Interrupts can freely happen during all of the rest of the operation of the handler; LINC mode operation lasts only about 15 microseconds. The handler waits for the tape to finish in -8 mode in a totally interruptible manner. cjl (Charles Lasner, lasner@watsun.cc.columbia.edu) ps: It is most notable that the -8 family KERMIT is known as KERMIT-12. KERMIT-12 runs on 12-bit PDP-8 family machines and the prompt is related to the cpu hardware. Running KERMIT-12 on an 8/i gives a prompt of KERMIT-8/e> while running it on the PDP-12 gives a prompt of KERMIT-12> so I guess that's its "true" home. :-) In any case, I wrote the code that way. :-) Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Fri, 8 Mar 91 21:25:09 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Fri, 8 Mar 91 21:22:16 EST Received: from watsun.cc.columbia.edu by life.ai.mit.edu (4.1/AI-4.10) id AA11431; Fri, 8 Mar 91 20:57:48 EST Received: by watsun.cc.columbia.edu (5.59/FCB) id AA04667; Fri, 8 Mar 91 20:57:16 EST Date: Fri, 8 Mar 91 20:57:15 EST From: Charles Lasner Cc: Chris Green , PDP8-lovers@ai.mit.edu Subject: Re: PDP-12 continued In-Reply-To: Your message of Fri, 8 Mar 91 18:06:53 EST Message-Id: Date: Fri, 8 Mar 91 20:57:15 EST From: Charles Lasner Cc: Chris Green , PDP8-lovers@ai.mit.edu Subject: Re: PDP-12 continued In-Reply-To: Your message of Fri, 8 Mar 91 18:06:53 EST a typo: KERMIT-12 says the prompt KERMIT-12> on a -12, but it says KERMIT-8/i> on an 8/i, and KERMIT-8/e> on an 8/e, sorry for the confusion. cjl Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Sat, 9 Mar 91 15:59:20 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Sat, 9 Mar 91 15:56:29 EST Received: from rice-chex (rice-chex.ai.mit.edu) by life.ai.mit.edu (4.1/AI-4.10) id AA05754; Sat, 9 Mar 91 15:37:19 EST From: rs@ai.mit.edu (Robert E. Seastrom) Received: by rice-chex (4.1/AI-4.10) id AA25797; Sat, 9 Mar 91 15:37:14 EST Date: Sat, 9 Mar 91 15:37:14 EST Message-Id: <9103092037.AA25797@rice-chex> To: lasner@watsun.cc.columbia.edu Cc: chris@kbss.bt.co.uk, PDP8-lovers@ai.mit.edu In-Reply-To: Charles Lasner's message of Fri, 8 Mar 91 20:57:15 EST Subject: PDP-12 continued From: rs@ai.mit.edu (Robert E. Seastrom) Date: Sat, 9 Mar 91 15:37:14 EST To: lasner@watsun.cc.columbia.edu Cc: chris@kbss.bt.co.uk, PDP8-lovers@ai.mit.edu In-Reply-To: Charles Lasner's message of Fri, 8 Mar 91 20:57:15 EST Subject: PDP-12 continued Date: Fri, 8 Mar 91 20:57:15 EST From: Charles Lasner a typo: KERMIT-12 says the prompt KERMIT-12> on a -12, but it says KERMIT-8/i> on an 8/i, and KERMIT-8/e> on an 8/e, sorry for the confusion. cjl While you're at it, could you post an explanation of how FOCAL (and presumably kermit) could tell what kind of machine they were running on? ---Rob Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Sat, 9 Mar 91 18:30:46 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Sat, 9 Mar 91 18:27:54 EST Received: from watsun.cc.columbia.edu by life.ai.mit.edu (4.1/AI-4.10) id AA07579; Sat, 9 Mar 91 18:07:43 EST Received: by watsun.cc.columbia.edu (5.59/FCB) id AA12043; Sat, 9 Mar 91 18:07:29 EST Date: Sat, 9 Mar 91 18:07:28 EST From: Charles Lasner To: rs@ai.mit.edu (Robert E. Seastrom) Cc: chris@kbss.bt.co.uk, PDP8-lovers@ai.mit.edu Subject: Re: PDP-12 continued In-Reply-To: Your message of Sat, 9 Mar 91 15:37:14 EST Message-Id: Date: Sat, 9 Mar 91 18:07:28 EST From: Charles Lasner To: rs@ai.mit.edu (Robert E. Seastrom) Cc: chris@kbss.bt.co.uk, PDP8-lovers@ai.mit.edu Subject: Re: PDP-12 continued In-Reply-To: Your message of Sat, 9 Mar 91 15:37:14 EST How to identify what the machine is on any PDP-8 family cpu and other sordid subjects. This subject started because FOCAL, 1969 tells you (sometimes erroneously) what the machine is. So, here is how it goes: The 8/i-type cpu can do some strange operate combinations. When you rotate right and left at the same time, the bits on the end get lopped off. So the combination CLA CMA RAL RAR makes 3776 in the accumulator. So, the first test is whether this is an 8/i-kinda cpu. If this fails, then it's either a newer type or a straight-8. Newer types have things the -8 lacks, so you can test for stuff like BSW. Focal, 1969 has an error in reporting an 8/i versus an 8/l. They assume two things you should NEVER do: a) All 8/i's are faster than all 8/l's. This is often just not true since these are tunable asynchronous memory machines. b) The console is 110 baud. Very risky. The FOCAL, 1969 check is to output a TTY: character and execute a lot of 6000 instructions. If the flag is up, then the machine is slow (an 8/l) else an 8/i. This is foolish and I discovered an improved method (later). The PDP-5 is easy: since the program counter is in location zero, just store an address-1 there and you will jump to the address of an "I am a PDP-5" routine. This is necessary for FOCAL, 1969 because PDP-5 interrupts are to location 2, not 1, and it must be changed. The PDP-8/s is done by noticing that CMA and RAL can't be done at the same time here. So, CLA CMA CLL RAL fails to achieve 1 7776 on the 8/s. Note that the test for -5 and -8/s are only used in FOCAL, 1969, since all PDP-8 operation systems that "matter" don't run on these retarded machines. Rumor has it that there were people modifying both CPUs to be as smart as a straight -8 by adding on outboard logic, etc. The newer machines represent a different challenge: one of packaging. The KK-8/A CPU can be differentiated from the KK-8/F CPU because while both have BSW, they handle the RAL RAR problem differently. On the KK-8/F, CLA CLL RAL RAR loads the current page address + 16 into the accumulator. On the KK-8/A, the updated 12-bit PC address is loaded into the accumulator. A handler was once written for an omnibus peripheral that depended on this; the instruction was located at 15 past the start of the page to ensure compatibility either way. So, it is easy to know what the CPU is, but what box is it in? KK-8/F is the cpu for any 8/e, f, m, or 8A-500, 600, or theoretically 400 box. The KK-8/A is the CPU for an 8A-400, 500, 600 box also, depending on configuration. My own machine is an 8/e in a 40-slot box with an 8/a 20-slot hex box on the back (or is that the other way around?) with both front panels present. (This is totally legal; read the configuration guide.) So, what is MY machine to be called? The micro-8/e-type machines are a different problem: the 6100 machines are to be known as -78 or -6100 and can be discovered because the RAL RAR combination does nothing at all, yet they also have BSW. (It's a straight -8 if you can't do NL7776 and also don't have BSW.). It also has a strange quirk that's almost a feature: Instructions like TAD I 10 are autoindexers. The octal is 1410, which means that the operand 10 is on page zero. If you move the instruction itself to page zero somewhere, then the bit for current page is effectively a NOP, thus, on page zero 1410 and 1610 do the same thing, right? Wrong on the 6100! The current page bit sets to PREVENT the auto-indexing! The random instruction tests for the 8/e don't work on the VT-78 because of this. Also all code generated by any assembler can't exploit this feature unless you express it in octal. The 6120 machines (all DECMATES) have an additional feature. They also do nothing with RAL RAR, and have BSW, but they also have R3L which rotates the AC left three spaces WITHOUT involving the link. So you can make instructions like CLA IAC R3L to load the AC with 0010 without affecting the link. It's a little hard to pin down which DECmate because they are all quite similar; a DECmate three looks an awful lot like a DECmate II with floppies only, and a DECmate III+ looks an awful lot like a DECmate II with a hard disk. The DECmate I has different I/O ports which can be/are detected within KERMIT-12. The LINC machines are no problem at all. A LINC-8 is an -8 with a LINC section where the -8 can access the LINC registers. The -12 is an 8/i with the LINC CPU mode. Just go into LINC mode with the LINC instruction and do a few innocuous instructions like COM to complement the AC and then do a PDP to come back. If the AC complemented, then it's a -12, not an 8/i. If it is an 8/i, all the instructions look like AND instructions and are innocuous. FOCAL, 1969 also checks for a LAB-8 configuration, meaning an AX08 is hooked up. This is so the FADC and FDIS functions can use the right hardware. There is a bit settable in one AX08 register that is readable in another. This isn't relevant to KERMIT-12 operation. Further, FOCAL, 1969 fails to distinguish LAB-8 from LAB-8/i from LAB-8/l, etc. Since FOCAL, 1969 was written in 1969, newer stuff always defaults to being claimed as PDP-8 only. There is also some "defensive business practices" code in FOCAL-8, the deficient newer patched version of FOCAL, 1969. This version contains some constants and instructions in different places cleverly hidden. The source is also XLISTed at that point so the listing they would sell you (NOT the source they would NOT!) shows code that gets overlaid by the REAL instructions which do the following: If a strange combination of operates produces an unusual result attributed to the DCC-112 and DCC-112H machines made by Digital Computer Controls (later a division of Data General), FOCAL-8 deliberately subtlely self-destructs by pointing the stack at the code, etc. This is to make the crash seem like unreliable hardware, etc., and to cover the tracks of the code that did the dastardly deed! I have somewhere a file which is the source of how to patch the binary of FOCAL, 1969 so it matches the binary of FOCAL-8 which was obtained by dis-assembling the differences between the two versions. Other than this trap, the only fix in FOCAL-8 is taking advantage of the 8/e superset hardware to avoid TTY: overrun of paper-tape instead of a proper software fix, and fixing a one-line missing bug in the TYPE/ASK code that was already fixed in FOCAL-12. Their implementation was therefore one-line longer than the original and involved axing some adjacent code. I invented an alternate patch myself which is used in P?S/8 FOCAL which is based on FOCAL, 1969 which is the same length as the original, but involves optimizing out a word and moving down a couple dozen words to make room for the extra word, but the net result is a clean patch over couple dozen words+1, not DEC's rewrite of the code. The only other change in FOCAL-8 is removing the code to care about all of those other machines, since it now won't work on any of them. They also charged extra for a paper-tape which merely restores the former code so it works properly again on pre-8/e machines. So, other than the one-word TYPE/ASK fix, you have effectively come right back to FOCAL, 1969 except it can't be used on the -5 or 8/s at all since it doesn't check and self-modify. Better to implement my patch on FOCAL, 1969. The method of PROPERLY determining an 8/i versus an 8/l is something I stumbled on: the normal operate instruction CLA actually has three flavors. 7200 is group one CLA, 7600 is group two CLA, and 7601 is group three (EAE) CLA. The 8/l has inhibitor logic for ANY EAE instruction, so 7601 will NOT clear the AC! Further, any 8/i or -12 for that matter always decodes all CLA variants as CLA since this is independent of the rest of the optional EAE hardware. So, just load a number in the AC and execute 7601. If the AC is unchanged, this is an 8/l. So, summarizing a list of models and where applicable KERMIT-12 prompt and info lines: CPU or model KERMIT-12 prompt info method of discovery PDP-5 N. A. N. A. (IAM5-1) > PC at 00000 PDP-8 KERMIT-8> PDP-8 no RAL RAR no BSW LINC-8 KERMIT-LINC-8> LINC-8 no RAL RAR no BSW LINC regs PDP-8/s N. A. N. A. no CMA RAR PDP-8/i KERMIT-8/i> PDP-8/i NL3776 7601 clear PDP-8/l KERMIT-8/l> PDP-8/l NL3776 7601 non-clear PDP-12 KERMIT-12> PDP-12 NL3776 LINC mode PDP-8/e KERMIT-8/e> PDP-8/e BSW page+16 RAL RAR PDP-8/f KERMIT-8/e> PDP-8/e BSW page+16 RAL RAR PDP-8/M KERMIT-8/e> PDP-8/e BSW page+16 RAL RAR PDP-8/A with KK-8/A KERMIT-8/a> PDP-8/a BSW absolute RAL RAR PDP-8/A with KK-8/F KERMIT-8/e> PDP-8/e BSW page+16 RAL RAR VT-78 KERMIT-8/e> 6100 BSW RAL RAR NOP non-XR VT-278 (DECmate 1) KERMIT-278> 6120 BSW NL0010 DM I ports PC-278 (DECmate 2) KERMIT-278> 6120 BSW NL0010 DM II ports PC-238 (DECmate 3) KERMIT-278> 6120 BSW NL0010 DM II ports PC-239 (DECmate 3+) KERMIT-278> 6120 BSW NL0010 DM II ports The info line is used as [6100] and the prompt is KERMIT-78>. There is also a connect mode line such as "type C to get back to your VT-78" or whatever. The only problems here are that "your {whatever}" will be PDP-8/e or PDP-8/a according to CPU type, and DECmate II, III, III+ all indicate DM II as a consequence of CPU identification, not box type where applicable. Further, there are a few maverick machines out there that are based on 6100, etc., so a box would be called VT-78 instead of PCM-12, etc., so you can't handle it all perfectly. Better than saying "type ^D C to get back to your {computer that has a 6100 type CPU in it}" for all of the actual VT-78s, etc. cjl Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Sun, 10 Mar 91 05:29:47 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Sun, 10 Mar 91 05:26:49 EST Received: from isy.liu.se ([130.236.1.3]) by life.ai.mit.edu (4.1/AI-4.10) id AA17015; Sun, 10 Mar 91 05:05:39 EST Received: from rainier.isy.liu.se by isy.liu.se with SMTP (5.61-bind 1.2+ida/IDA-1.2.8.2/LTH/LiTH) id AA00247; Sun, 10 Mar 91 11:05:45 +0100 Received: by rainier.isy.liu.se; Fri, 8 Mar 91 22:02:11 +0100 Date: Fri, 8 Mar 91 22:02:11 +0100 From: P{r Emanuelsson Message-Id: <9103082102.AA10978@rainier.isy.liu.se> To: ph@ama.caltech.edu Subject: Re: latest Cc: PDP8-lovers@ai.mit.edu Date: Fri, 8 Mar 91 22:02:11 +0100 From: P{r Emanuelsson To: ph@ama.caltech.edu Subject: Re: latest Cc: PDP8-lovers@ai.mit.edu We have a 2020 "on the internet". Telnet to 130.236.254.13 port 3001 and you're there (hopefully). What happens is that you will be connected to a terminal line into the 2020. At present there is only one line, but if I get time I might wire up a couple more. Also, we have no "guest" account, but might perhaps create one if there is interest... Also, I'm having trouble with the disks at the moment so the 2020 may be frequently inaccessible indefinitely... :-) /Pell Received: from BULLDOG.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 13 Mar 91 06:26:33 EST Received: from life.ai.mit.edu by BULLDOG.CS.YALE.EDU; Wed, 13 Mar 91 06:23:52 EST Received: from sunic.sunet.se by life.ai.mit.edu (4.1/AI-4.10) id AA16192; Wed, 13 Mar 91 05:39:13 EST Received: from AIDA.CSD.UU.SE by sunic.sunet.se (5.61+IDA/KTH/LTH/1.186) id AAsunic22059; Wed, 13 Mar 91 11:38:55 +0100 Date: Wed 13 Mar 91 11:38:53 From: Johnny Billquist Subject: Re: Archive site! (maybe...) To: mrosen@rpi.edu Cc: pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se In-Reply-To: <9103070603.AA100698@sub.aix.rpi.edu> Message-Id: <910313113853.17.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> Date: Wed 13 Mar 91 11:38:53 From: Johnny Billquist Subject: Re: Archive site! (maybe...) To: mrosen@rpi.edu Cc: pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se In-Reply-To: <9103070603.AA100698@sub.aix.rpi.edu> To give you a starter (I'm sure others will add to this) I have about 4 meg of pdp8 software which would probably be of interest to put on the archive. Much of this is common stuff for sure, but about 1 meg is homecooked software, some of it very useful. To mention a few bits: A fancy KERMIT, which requires omnibus pdp8. A heavily modified FORTRAN IV compiler. A lot of small CUSPs. And soon (hopefully) DUNGEON (the DECUS version) rehacked to run under OS/8 FORTRAN IV... Received: from eli.cs.yale.edu by BUGS.SYSTEMSY.CS.YALE.EDU; Wed, 13 Mar 91 14:30:27 EST Received: from life.ai.mit.edu by eli.cs.yale.edu; Wed, 13 Mar 91 14:26:48 EST Received: from watsun.cc.columbia.edu by life.ai.mit.edu (4.1/AI-4.10) id AA02478; Wed, 13 Mar 91 13:58:46 EST Received: by watsun.cc.columbia.edu (5.59/FCB) id AA19244; Wed, 13 Mar 91 13:57:37 EST Date: Wed, 13 Mar 91 13:57:36 EST From: Charles Lasner To: Johnny Billquist Cc: mrosen@rpi.edu, pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se Subject: Re: Archive site! (maybe...) In-Reply-To: Your message of Wed 13 Mar 91 11:38:53 Message-Id: Date: Wed, 13 Mar 91 13:57:36 EST From: Charles Lasner To: Johnny Billquist Cc: mrosen@rpi.edu, pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se Subject: Re: Archive site! (maybe...) In-Reply-To: Your message of Wed 13 Mar 91 11:38:53 From: Charles Lasner To: Johnny Billquist Subject: Re: Archive site! (maybe...) In-Reply-To: Your message of Wed 13 Mar 91 11:38:53 Sorry about the test, but mail system is flakey at the moment. I am curious about your mention of dungeon. Who is working on that? The history of dungeon (aka zork) is a checkered one. First it was written at MIT in MDL, an mit list-processing language. Apparently, MDL is very powerful, and presumably suitable for writing zork in. MDL only runs under TOPS10,20,ITS, etc., since no one ever ported it to any other architecture. The game evolved and there became a "standard" version played at many universities, etc. Several less than ethical events occurred. First, someone who is either a DEC employee or one level removed, stole a copy of the zork files, or at least the "cave" description portion. This eventually became the hacked-up DECUS zork for the -11/vax world. Essentially they hacked on adventure to make it do the cave stuff for zork. This means that the game is somewhat different in scope. The original game was meant to be used with a smart parser and a somewhat critical path strategy. Due to the stupid parser, several command simplicifcations bordering on hints have to be used, and the overall criticality is relaxed. I also stumbled on some data errors in the cave description which affects the overall "map", although no change to the outcome (proceed sw, not se to get to the same place as in -20 zork, etc.). The other event was that a company called infocom "bought off" some of the MIT now ex-students who took with them the entire game file set. This borders on theft, as MIT has copyrighted ZORK. Infocom had them modify the particulars of the game so it is now incompatible enough that maps to the original are useless, and some of the new features are somewhat juvenile. I get the impression that the programmers brought into Infocom were not the prose writers who created the bulk of the original game. In any case, Infocom had them write a new implementation language which they call ZIL or Zork Information Language. Presumably, this language, which incidentally was developed on a -20, is fairly close to MDL so they could port the game to it. Object code was generated to run on Z80-based CP/M systems. One problem is sheer size. The game got split up into three arbitrary sections, which means much of the original spirit of the game is lost, since now they have decided a major amount of the strategy for you. In actuality, the original game consisted of small mini-adventures whose order was mostly immaterial, or at least consisted of groups which had to be time-ordered, but whose members were not, etc. Now they arbitrate how much you have seen/not seen, and you can't even go back to where you have been. Further, the flow of the game is disrupted because you onll "win" part one or two or three by getting to their arbitrary "end" points. Eventually, they ported the object code to the 8088 world, using paranoically protected MS-DOS-type disks which had to boot standalone on 360K PC disk drives. Copying disks was only minimally allowed, and a copy couldn't be copied. You could not run these from MS-DOS on a hard disk either. Worse, the original restrictive segmentation arising from CP/M-80 still applies; they are still known as the zork trilogy. Someone unlocked these disks so they could be run as normal MS-DOS files, including from a hard disk, but it's still as bad as CP/M-80 version, and a far cry from the classic game it rips off. On a related subject, PDP-8 adventure is a modification of the rt-11 version of adventure taken from the PDP-10 adventure. I understand that a major change in the -11 version stems from the inability to define 36-bit storage elements, so they reduced the commands to 4 characters to fit in 32 bits to accomodate the -11's limitations. Well, PDP-8 fortran DOES 36 bit data! SO, it would appear that the -8 version should be repaired by restoring the PDP-10 original code, not the -11 code which was used by m. Subnik for RT-11, etc. Perhaps, in this way, the PDP-8 can have a more authentic version of adventure. Btw, the only redeeming feature of the Infocom version is that the parser was maintained; they even claimed that it was somewhat enhanced, and playing the game, even in the CP/M/MS-DOS variant does allow for the smarter commands that the DECUS version lacks. cjl Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Mon, 18 Mar 91 17:17:41 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Mon, 18 Mar 91 17:14:23 EST Received: from sunic.sunet.se by life.ai.mit.edu (4.1/AI-4.10) id AA29315; Mon, 18 Mar 91 16:20:44 EST Received: from AIDA.CSD.UU.SE by sunic.sunet.se (5.61+IDA/KTH/LTH/1.186) id AAsunic24924; Mon, 18 Mar 91 22:20:31 +0100 Date: Mon 18 Mar 91 22:20:06 From: Johnny Billquist Subject: [Johnny Billquist : Re: Archive site! (maybe...)] To: pdp8-lovers@ai.mit.edu Message-Id: <910318222006.21.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> Date: Mon 18 Mar 91 22:20:06 From: Johnny Billquist Subject: [Johnny Billquist : Re: Archive site! (maybe...)] To: pdp8-lovers@ai.mit.edu Mail-From: D89.JOHNNY-BILLQUIST created at 18-Mar-91 22:05:42 Date: Mon 18 Mar 91 22:05:42 From: Johnny Billquist Subject: Re: Archive site! (maybe...) To: lasner@watsun.cc.columbia.edu cc: D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE In-Reply-To: Message-ID: <910318220542.21.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> As to who is doing the port of DUNGEON to the pdp8: I am. I have done about 2/3 of it. It has required some RALF hacking as well as rewriting of the parser and I/O subroutines. The DECUS version is copyrighted Infocom 1980. As I am running on a TOPS-20 machine, I have tested the MDL version, and I have not seen any discrepancies(?) yet. The two things I know the DECUS version do not handle, which I wished it did is AGAIN and ALL BUT... Also, one room (or if it was two) is not in there, but they play no real part in the game. I think the MDL version has a room halfway down the chute. The latest, and last, version from DECUS contains the endgame, as well as the royal puzzle. I can also add that "my" version of DUNGEON will require a 32k pdp8, with my own version of OS/8 FORTRAN IV. Reasons: boolean operators need to work on integers, dynamic open, close and create files. One difference from the pdp-11 version will be that everything is in uppercase, since OS/8 FORTRAN IV uses sixbit. Other than that, everything will probably look nice enough. Also, I can add that I don't think the FORTRAN DUNGEON is based on ADVENTURE codewise. The parser in DUNGEON is much more versatile than ADVENTURE. It was a long time since I looked at the code for ADVENTURE, but I remember that there is no similarities in the data structures for the parser. One more requirement: You cannot use RX01 floppies, since the text data file is almost 500 blocks, and the "raw" data file is about 300 blocks. Both is assumed to be on DSK:, also the savefile will be put on DSK: I don't know how fast (or slow) it will be, but I suspect that a real FPP would be nice to have. Is there anybody who has a spare FPP-8A? I WANT ONE! Johnny Received: from BULLDOG.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Mon, 18 Mar 91 20:45:11 EST Received: from life.ai.mit.edu by BULLDOG.CS.YALE.EDU; Mon, 18 Mar 91 20:42:42 EST Received: from akbar.cac.washington.edu by life.ai.mit.edu (4.1/AI-4.10) id AA05606; Mon, 18 Mar 91 20:10:16 EST Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu (5.65/UW-NDC Revision: 2.21 ) id AA29509; Mon, 18 Mar 91 17:10:13 -0800 Date: Mon, 18 Mar 1991 17:05:20 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: RM03/5 for 36-bit machines To: PDP-8 Lovers Message-Id: Date: Mon, 18 Mar 1991 17:05:20 -0800 (PST) From: Mark Crispin Sender: Mark Crispin Subject: RM03/5 for 36-bit machines To: PDP-8 Lovers The question of how to modify the drive control logic on an RM03 or RM05 so it can be used with a 2020 has come up a few times in this forum. Here's the answer: The way to fix an RM03/RM05 to work with a 36-bit CPU is to install a jumper on the MBA backplane between E06E1 and E09C2. If the jumper is not installed, nothing works; you can't format the drive in 18-bit mode or even recognize an 18-bit format. The jumper grounds signal BP144 ENB M on page D56 of the MBA printset. Received: from BULLDOG.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Mon, 18 Mar 91 21:34:13 EST Received: from life.ai.mit.edu by BULLDOG.CS.YALE.EDU; Mon, 18 Mar 91 21:31:37 EST Received: from sunic.sunet.se by life.ai.mit.edu (4.1/AI-4.10) id AA06557; Mon, 18 Mar 91 20:56:58 EST Received: from AIDA.CSD.UU.SE by sunic.sunet.se (5.61+IDA/KTH/LTH/1.186) id AAsunic09975; Tue, 19 Mar 91 02:56:38 +0100 Date: Tue 19 Mar 91 02:55:31 From: Johnny Billquist Subject: MDL-ZORK vs. DECUS-DUNGEON. To: lasner@watsun.cc.columbia.edu, pdp8-lovers@ai.mit.edu Message-Id: <910319025531.21.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> Date: Tue 19 Mar 91 02:55:31 From: Johnny Billquist Subject: MDL-ZORK vs. DECUS-DUNGEON. To: lasner@watsun.cc.columbia.edu, pdp8-lovers@ai.mit.edu After a fast check in MDL-ZORK, I stand (probably much) corrected. Lasner has obviously spent some time delving into the secrets of ZORK. The MDL version has a max score of 616, while the DECUS version has max 585, more stuff than I know of must be missing. Johnny Summary-line: 15-Feb John_Wilson@mts.rpi.edu #PDP-10, PDP-11 Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Fri, 15 Feb 91 14:49:27 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Fri, 15 Feb 91 14:46:44 EST Received: from rpi.edu by life.ai.mit.edu (4.1/AI-4.10) id AA00137; Fri, 15 Feb 91 14:16:02 EST Received: from MTS.RPI.EDU by rpi.edu (4.1/RPI-ITS-SM55); id AA19580; Fri, 15 Feb 91 14:12:29 EST for PDP8-LOVERS@ai.mit.edu Date: Fri, 15 Feb 91 14:12:16 EST From: John_Wilson@mts.rpi.edu To: PDP8-LOVERS@ai.mit.edu Message-Id: <2165828@MTS.RPI.EDU> Subject: PDP-10, PDP-11 Date: Fri, 15 Feb 91 14:12:16 EST From: John_Wilson@mts.rpi.edu To: PDP8-LOVERS@ai.mit.edu Subject: PDP-10, PDP-11 Does anyone know of any PDP-10- or PDP-11-specific mailing lists? John Summary-line: 25-Mar ph@ama.caltech.edu #MDL-ZORK vs. DECUS-DUNGEON. Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Mon, 25 Mar 91 15:17:13 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Mon, 25 Mar 91 15:14:03 EST Received: from ama (ama.caltech.edu) by life.ai.mit.edu (4.1/AI-4.10) id AA24157; Mon, 25 Mar 91 14:32:36 EST Received: from laplace.ama.caltech.edu by ama (4.1/1.2) id AA02935; Mon, 25 Mar 91 11:58:31 PST Date: Mon, 25 Mar 91 11:58:31 PST From: ph@ama.caltech.edu (Paul Hardy) Message-Id: <9103251958.AA02935@ama> Received: by laplace.ama.caltech.edu (4.1/SMI-4.0) id AA00723; Mon, 25 Mar 91 11:56:17 PST To: D89.JOHNNY-BILLQUIST@aida.csd.uu.se Cc: lasner@watsun.cc.columbia.edu, pdp8-lovers@ai.mit.edu In-Reply-To: Johnny Billquist's message of Tue 19 Mar 91 02:55:31 <910319025531.21.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> Subject: MDL-ZORK vs. DECUS-DUNGEON. Date: Mon, 25 Mar 91 11:58:31 PST From: ph@ama.caltech.edu (Paul Hardy) To: D89.JOHNNY-BILLQUIST@aida.csd.uu.se Cc: lasner@watsun.cc.columbia.edu, pdp8-lovers@ai.mit.edu In-Reply-To: Johnny Billquist's message of Tue 19 Mar 91 02:55:31 <910319025531.21.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> Subject: MDL-ZORK vs. DECUS-DUNGEON. MDL-ZORK?! I would definitely like a copy of this. I first played this in '78, and didn't snarf the sources because I assumed it would live on as free hacker software forever (oh, to be young again!) --Paul Summary-line: 25-Mar -BILLQUIST@aida.csd.uu.se #Re: MDL-ZORK vs. DECUS-DUNGEON. Received: from ELI.CS.YALE.EDU by BUGS.SYSTEMSY.CS.YALE.EDU; Mon, 25 Mar 91 16:17:57 EST Received: from life.ai.mit.edu by ELI.CS.YALE.EDU; Mon, 25 Mar 91 16:14:52 EST Received: from sunic.sunet.se by life.ai.mit.edu (4.1/AI-4.10) id AA25619; Mon, 25 Mar 91 15:21:11 EST Received: from AIDA.CSD.UU.SE by sunic.sunet.se (5.61+IDA/KTH/LTH/1.186) id AAsunic14387; Mon, 25 Mar 91 21:21:02 +0100 Date: Mon 25 Mar 91 21:19:32 From: Johnny Billquist Subject: Re: MDL-ZORK vs. DECUS-DUNGEON. To: ph@ama.caltech.edu Cc: lasner@watsun.cc.columbia.edu, pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se In-Reply-To: <9103251958.AA02935@ama> Message-Id: <910325211932.19.D89.JOHNNY-BILLQUIST@AIDA.CSD.UU.SE> Date: Mon 25 Mar 91 21:19:32 From: Johnny Billquist Subject: Re: MDL-ZORK vs. DECUS-DUNGEON. To: ph@ama.caltech.edu Cc: lasner@watsun.cc.columbia.edu, pdp8-lovers@ai.mit.edu, D89.JOHNNY-BILLQUIST@aida.csd.uu.se In-Reply-To: <9103251958.AA02935@ama> Sorry, don't have the sources... I do like it, though... Actually I don't know in what form the game is stored. But I suspect it is compiled. Johnny (Yes, I use a TOPS-20 machine)