From mo09@badbit.math2.nat.tu-bs.de Wed Jan 11 10:07:45 EST 1995 Article: 1126 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!hookup!kinky.eng.gtefsd.com!europa.eng.gtefsd.com!howland.reston.ans.net!Germany.EU.net!news.dfn.de!gs.dfn.de!tubsibr!news.ibr.cs.tu-bs.de!pueschel From: pueschel@math2.nat.tu-bs.de (Olaf Pueschel) Newsgroups: alt.sys.pdp8 Subject: pdp8 tad instruction Date: 10 Jan 1995 16:52:20 GMT Organization: Institut fuer Angewandte Mathematik, TU Braunschweig Lines: 29 Distribution: world Message-ID: Reply-To: mo09@badbit.math2.nat.tu-bs.de NNTP-Posting-Host: badbit.math2.nat.tu-bs.de Hi folks! Over X-mas I have been building my small own pdp8 assembler/simulator system. It is quite workable now (yet not finished) and was quit some fun. The only source of information I could usewas the pdp8-faq (posted here) and this left open a few questions: first (and most urgent): how does the TAD instruction *exactly* work with negative numbers. As I understand the -8 has negative numbers stored as 2s complement. How does this work with beeing treated as a 13 Bit unit? Further, on which condition do the several skip instruction dealing with ac negative/non-negative trigger? Is a number negative if its MSB is set? In addition if someone could provide me with the functionality of the EAE and the Memory extinsion units, on day I could impement these to :) And - in the farer future - I am planing to add a X11 based front panel emulation. I have a GIF of a pdp8 front panel, but no desciption. If anyone could provide any information, I'd be very happy :) Of course all this stuff will be made public-avaialble when It's finished. thanks to anyone providing *any* information -olf ___ Olaf P"uschel, Broitzemer Str. 10, 38118 Braunschweig, Germany, +49-531-895643 -- Olaf P"uschel Broitzemer Str. 10 38118 Braunschweig Germany +49 531 895643 ------------------------------------------------------------------------------- *** this is NOT a random-generated signature :-) From lasner@sunSITE.unc.edu Wed Jan 11 10:08:15 EST 1995 Article: 1127 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: comp.sys.dec.micro,alt.sys.pdp8 Subject: Re: DECmate diskettes (ARRGGG!!) Date: 11 Jan 1995 14:36:06 GMT Organization: University of North Carolina, Chapel Hill Lines: 138 Message-ID: <3f0qcm$175q@bigblue.oit.unc.edu> References: <3ejseo$kvt@hobbit.gandalf.ca> NNTP-Posting-Host: president.oit.unc.edu Xref: bigblue.oit.unc.edu comp.sys.dec.micro:3522 alt.sys.pdp8:1127 In article <3ejseo$kvt@hobbit.gandalf.ca>, David Huddleson wrote: >I hate when this happens... > >Out of someone's deep deep archives in the company, I have been presented >with a box of RX50 diskettes once used on a DECmate (I have books for >DECmate II as well as DECmate III, if it matters!!). > >We "unloaded" the last of our DECmates probably 2 years ago, so what >to do?? There are many different ways to get partial recovery of the data. How much you want to pay for how much of it is a factor. > >We still have a Rainbow or two around, but trying to read the diskettes >in the Rainbow failed (with both MS-DOS and CP/M). I don't know if >the Rainbow actually can read the DECmate disks anyway, but I tried! Wrong path to the wrong place. > >Can anyone suggest if or how I might be able to see what's on these diskettes, >because of course the user feels there may be important info on the old >diskettes (yeah!) > >Any help or suggestions would be appreciated. > >Thanks, > >David Huddleson Gandalf Technologies Inc., Nepean, Ontario > dhuddleson@krypton.gandalf.ca First of all, the disks clearly are *NOT* CP/M disks because all Rainbow and DECmate CP/M disks are in the same format. MS-DOS disks for the DECmate are obscure and also compatible with Rainbow MS-DOS thus the same proof applies that they aren't MS-DOS RX50 disks either. By and large WPS was what people used on DECmates. Many of our readers would think otherwise, but I am talking about the original sold purpose of the hardware, not our collective thoughts on doing it differently, etc. On the Rainbow itself there is a DEC package called WPS+. Allegedly it can convert WPS/DM disks into its format, which are at the least files that live in the Rainbow (and thus DECmate!) MS-DOS file structure with innards of a presumably similar nature. Some people have attempted to run parts of WPS+ on the DECmate MS-DOS only to find that the conversion part doesn't work there :-(. On the DECmate there are three ways to get WPS files to somewhere else using DEC-suppled utilities, and at least one additional from a non-DEC source. 1) Use DECmate MS-DOS CONVERT to move files between MS-DOS, CP/M-80 and WPS format. 2) Use DECmate CP/M-80 WPSCONV to move files between CP/M-80 and WPS format. 3) Use DECmate OS/278 WPFLOP to move files between OS/278 and WPS format. The first two can give document directory listings directly. The third can indirectly because in WPS, document #1 *IS* the directory of all of the other documents. All three methods suffer from the fact that the converted result is ASCII text sans WP-isms, etc. The non-DEC method is to run WPS80 from EBS. There is a version of it for both the Rainbow and DECmate w/CPM-80. The result is converted files that live in a CP/M file structure whose innards are presumably similar to the original WPS innards. Such CP/M diskettes can then be read wherever RX50 CP/M diskettes can be so read (22DISK on PC's, DECmate CP/M-80, Rainbow CP/M-86/80, etc.) but only WPS80 knows what to do with the file innards, etc. Moving to a PC may or may not help. While 22DISK can certainly move files from CP/M-80 RX50 per se, this does not help to deal with the actual file contents taken from any of the intermediary systems (such as WPS80). DEC sold a package called WPS+PC that presumably can convert WPS+ (Rainbow) files. The source are the WPS+ innards RX50 MS-DOS and the destination is WPS+PC MS-DOS PC files. In any case you would have to deal with the PC-DOS files in WPS+PC innards format to move them to somewhere else. An available somewhere else is to use Software Bridge for Windows. As a source it likes so-called "DX" files (what WPS+PC normally creates) and as destination it likes things like WordPerfect, etc. (Also I believe the same (different?) vendor sells a lesser product that can only read a few different input formats, while the lone output format is WordPerfect. The word "Perfect" is part of this product's name? In any case, the so-called "DX" format is the relevant input format, etc.) There also may be another way to get the files in that direction: use EBS' WPS-DOS, which reportedly can convert the original WPS-DM disks directly. Presumably, WPS-DOS is also "DX" compatible, and seems to be a competing compatible product with DEC's WPS+PC, but better in certain ways. (Questions: Does WPS-DOS and WPS+PC freely like each other's primary format directly, or do you need some additional form of file import/export conversion between them? Even if so, it would still make DX files so Software Bridge could move them to WordPerfect, etc. Does WPS-DOS like WPS+ files? Does WPS-DOS like WPS80 files? Does WPS+PC and/or WPS-DOS write any of the older formats (WPS/DM, WPS+, WPS80/CP/M) or just read them?) Finally, there is a sure-fire way to do it. For likely far more money than its worth, there exists an "anything-to-anything" convertor subsystem for PC's from Shaftstall. This consists of a hardware and software kit that adds various forms of flopy support to a PC including hard-sectored floppies and 8" drives, etc. The programs themselves are written in the hard-sectored format so forget about copying them! And of course it can't run without the supplied hardware, etc. I have seen the Shaftstall helpout Video, and it uses as the primary example the conversion of WPS/DM documents directly to Wordperfect in one step. If the person involved has a spare $15,000 to devote to this project, then it's real easy to get done :-). Additionally, there are companies in the conversion business that charge similar rates to use their Shaftstall machines. (Apparently they want about 1% of the cost of the hardware per diskette :-(.) Lastly, all WPS-type systems support communications in "DX" format (not to be confused with that notion of what the file format for WPs+PC is called!) which means in theory that the files could be sent from where they are already loaded to another site with a different WPS-type system, etc. Over the years, there have been various -11, -10/20, VAX, etc. systems that are compatible with this, as well as virtually all of the aforementioned, etc. (However, this does not necessarily represent forward progress in the conversion process! Just moving the problem around a little :-).) OR of course you could scan the printouts (if available) and then have it in yet another system, etc. :-). cjl From lasner@sunSITE.unc.edu Wed Jan 11 10:08:24 EST 1995 Article: 1128 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: alt.sys.pdp8 Subject: Re: pdp8 tad instruction Date: 11 Jan 1995 15:08:30 GMT Organization: University of North Carolina, Chapel Hill Lines: 62 Message-ID: <3f0s9e$nn9@bigblue.oit.unc.edu> References: NNTP-Posting-Host: president.oit.unc.edu In article , Olaf Pueschel wrote: >Hi folks! >Over X-mas I have been building my small own pdp8 assembler/simulator system. >It is quite workable now (yet not finished) and was quit some fun. >The only source of information I could usewas the pdp8-faq (posted here) and >this left open a few questions: >first (and most urgent): how does the TAD instruction *exactly* work with >negative numbers. As I understand the -8 has negative numbers stored as 2s >complement. How does this work with beeing treated as a 13 Bit unit? Two's complement numbers are merely added. The interpretation is what makes the numbers negative! The Link gets complemented whenever a carry occurs, making the significence of the event user-defined, although totally predictable. If the entire quantity is treated as a 13-bit unsigned number, then the user must preset the L and AC to whatever initial state makes sense for the problem at hand etc. Instructions are available to clear, complement, and test the AC and L separately and together to form any useful manipulation. If you decide that negative numbers are those with the high-order bit set, then test with SMA or SPA accordingly. The arithmetic will conform to this, but there is no hardware present forcing the issue; you have to test for it. > >Further, on which condition do the several skip instruction dealing with >ac negative/non-negative trigger? Is a number negative if its MSB is set? SMA is skip if AC[0] is set. Negativity is in the eye of the beholder :-). SPA is skip if AC[0] is not set. Same consideration otherwise. OF course the instructions are there to encourage the use of two's complement 12-bit numbers, but nothing is forced, etc. > >In addition if someone could provide me with the functionality of the EAE >and the Memory extinsion units, on day I could impement these to :) The memory extension to 32K is quite important. Emulation of a PDP-8 disk device and 03/04 console support vastly enlarges the usefulness of the project. This allows "real" PDP-8 software to run on the simulator, etc. > >And - in the farer future - I am planing to add a X11 based front panel >emulation. I have a GIF of a pdp8 front panel, but no desciption. If anyone >could provide any information, I'd be very happy :) Would be real cute, especially for beginners. But pdp-8 people need the fully-fledged simulator features more than the "show" :-).) > >Of course all this stuff will be made public-avaialble when It's finished. > >thanks to anyone providing *any* information >-olf >___ >Olaf P"uschel, Broitzemer Str. 10, 38118 Braunschweig, Germany, +49-531-895643 >-- > Olaf P"uschel Broitzemer Str. 10 38118 Braunschweig Germany +49 531 895643 >------------------------------------------------------------------------------- >*** this is NOT a random-generated signature :-) cjl From jones@pyrite.cs.uiowa.edu Wed Jan 11 15:59:04 EST 1995 Article: 1129 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!spool.mu.edu!uwm.edu!math.ohio-state.edu!hobbes.physics.uiowa.edu!news.uiowa.edu!news From: jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) Newsgroups: alt.sys.pdp8 Subject: Re: pdp8 tad instruction Date: 10 Jan 1995 17:53:29 GMT Organization: University of Iowa, Iowa City, IA, USA Lines: 67 Distribution: world Message-ID: <3euhip$54r@nexus.uiowa.edu> References: NNTP-Posting-Host: pyrite.cs.uiowa.edu >From article , by pueschel@math2.nat.tu-bs.de (Olaf Pueschel): > first (and most urgent): how does the TAD instruction *exactly* work with > negative numbers. As I understand the -8 has negative numbers stored as 2s > complement. How does this work with beeing treated as a 13 Bit unit? Negative numbers are stored as 12 bit 2's complement numbers. In adding numbers to the accumulator, a carry out of the high bit of the accumulator causes the link bit to be complemented. If it was zero, it will be one; if it was one, it will be zero. Here are some examples: L/AC 0 000000000001 = 1 + 111111111111 = -1 -------------------------- 1 000000000000 = 0 L/AC 1 000000000001 = 1 + 111111111111 = -1 -------------------------- 0 000000000000 = 0 You, the programmer, are responsible for priming the link bit with an appropriate value, where the initial value matters. > Further, on which condition do the several skip instruction dealing with > ac negative/non-negative trigger? Is a number negative if its MSB is set? In the table below, I've marked the bits in L/AC that each skip instruction tests: OR GROUP 0 X00000000000 SMA - skip if AC is negative 0 XXXXXXXXXXXX SZA - skip if all bits of AC are zero X 000000000000 SNL - skip if the link is nonzero AND GROUP 0 X00000000000 SPA - skip if AC is positive 0 XXXXXXXXXXXX SNA - skip if some bit of AC is nonzero X 000000000000 SZL - skip if the link is zero > In addition if someone could provide me with the functionality of the EAE > and the Memory extinsion units, on day I could impement these to :) These are messy, and they vary depending on the PDP-8 model that interests you. You can find a short summary of some of the key EAE instructions in the pdp-8 reference card collection available by WWW from http://www.cs.uiowa.edu/~jones/pdp8/index.html Someday, I'll sift through my piles of documentation and try to put together a more complete definition in the WWW programmer's reference manual that's in the same collection. > And - in the farer future - I am planing to add a X11 based front panel > emulation. I have a GIF of a pdp8 front panel, but no desciption. If anyone > could provide any information, I'd be very happy :) You're welcome to borrow the X-11 front panel from my emulator. It has one rather substantial bug in the way it handles single-stepping, but otherwise, it works well. That emulator is indexed under the web page mentioned above. Doug Jones jones@cs.uiowa.edu http://www.cs.uiowa.edu/~jones/ From lasner@sunSITE.unc.edu Fri Jan 13 13:36:57 EST 1995 Article: 1130 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: comp.sys.dec.micro,alt.sys.pdp8 Subject: Re: DECMATE III, I've forgot what it is. Date: 13 Jan 1995 18:00:53 GMT Organization: SunSITE, University of North Carolina, Chapel Hill Lines: 28 Message-ID: <3f6f4l$142c@bigblue.oit.unc.edu> References: <3f1no6$16b@ball.datastorm.com> NNTP-Posting-Host: calzone.oit.unc.edu Xref: bigblue.oit.unc.edu comp.sys.dec.micro:3529 alt.sys.pdp8:1130 In article <3f1no6$16b@ball.datastorm.com>, John A. Maier wrote: >I just bought a DECMATE III for the large amount of free. Please help >me remember what this beast is? I want to think it's a 286 or was it a >PDP-8 word processor and or CP/M machine. There have been so may DEC >micros I get confused. Anyway, this is the one the size of a bread box >with a RX-50 drive. At the least I'll yank the RX-50 drive as a spare >in case the on in my PRO-350 goes. Hey, for those interested, I found the >guy who wrote Uniforth (for you Forth fans of old). He had (has) a >version for the PDP-11, Pro-350 and VAX. I bought a copy and am very >happy. > >thanks >jam >johnam@datastorm.com (John A. Maier) > Please be advised that there are readers of a.s.p that need spare parts desperately for various projects, including maintaining other DECmates, etc. You sound like someone who has no committment to the hardware you currently have. Please give some others a chance who have a demonstrated need, etc. Thank You cjl From jones@pyrite.cs.uiowa.edu Thu Jan 19 07:43:00 EST 1995 Article: 1131 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!math.ohio-state.edu!hobbes.physics.uiowa.edu!news.uiowa.edu!news From: jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) Newsgroups: alt.jokes.pentium,alt.sys.pdp8 Subject: Re: Valuable Offer! Date: 13 Jan 1995 19:19:40 GMT Organization: University of Iowa, Iowa City, IA, USA Lines: 21 Distribution: world Message-ID: <3f6joc$qvj@nexus.uiowa.edu> References: NNTP-Posting-Host: pyrite.cs.uiowa.edu Xref: bigblue.oit.unc.edu alt.jokes.pentium:483 alt.sys.pdp8:1131 >From article , by cjl@maestro.maestro.com (Charles Lasner): > In article <3eaq0k$lgh@paperboy.ids.net> Philip Chien writes: >>Anybody interested in a DEC PDP-8/S???? The first 'personal' (e.g. under >>$10,000) computer ever made. Makes a *GREAT* paperweight. I don't think >>it has the Pentium bug (tm) though - it could only add and subtract >>integers - never could handle those division signs! If you're serious, the machine is a museum piece, and I do have the complete maintenance manual for it. Unlike the pentium, it's an open architecture machine, albeit a bit on the very slow side, even when compared with other members of the PDP-8 family. Note, however, that it doesn't subtract. All arithmetic is done with the TAD instruction (Two's Complement Add). There is another instruction with a microcoded format that includes bits to one's complement and increment the accumulator -- if both bits are set, it does a two's complement, and that plus TAD gives you subtraction. Doug Jones jones@cs.uiowa.edu From mo09@badbit.math2.nat.tu-bs.de Thu Jan 19 07:43:20 EST 1995 Article: 1132 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!swrinde!pipex!news.oleane.net!oleane!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!scsing.switch.ch!news.dfn.de!zeus.rbi.informatik.uni-frankfurt.de!terra.wiwi.uni-frankfurt.de!news.th-darmstadt.de!zib-berlin.de!gs.dfn.de!tubsibr!news.ibr.cs.tu-bs.de!pueschel From: pueschel@math2.nat.tu-bs.de (Olaf Pueschel) Newsgroups: alt.sys.pdp8 Subject: tad instaruction / thanks! Date: 13 Jan 1995 14:27:19 GMT Organization: Institut fuer Angewandte Mathematik, TU Braunschweig Lines: 8 Distribution: world Message-ID: Reply-To: mo09@badbit.math2.nat.tu-bs.de NNTP-Posting-Host: badbit.math2.nat.tu-bs.de I would like to express many thanks to Doug Jones, Charles Lasner and Bob Supnik for the prompt reply to my question. They helped me quite a lot. tnx -olf -- Olaf P"uschel Broitzemer Str. 10 38118 Braunschweig Germany +49 531 895643 ------------------------------------------------------------------------------- *** this is NOT a random-generated signature :-) From stan@MathPro.com Thu Jan 19 07:43:42 EST 1995 Article: 1133 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!agate!howland.reston.ans.net!news.sprintlink.net!sundog.tiac.net!usenet From: Stanley Rabinowitz Newsgroups: alt.sys.pdp8 Subject: Re: pdp8 tad instruction Date: 15 Jan 1995 16:40:00 GMT Organization: The Internet Access Company Lines: 5 Message-ID: <3fbj50$nq6@sundog.tiac.net> References: <3f0s9e$nn9@bigblue.oit.unc.edu> NNTP-Posting-Host: mathpro.com So, if you implement TAD as if it were only adding unsigned integers, you won't be going wrong. - stan - From rozee@elec.canterbury.ac.nz Thu Jan 19 07:44:36 EST 1995 Article: 1134 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!taco.cc.ncsu.edu!hsdndev!yale!yale!yale.edu!noc.near.net!bigboote.WPI.EDU!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!news.starnet.net!wupost!waikato!comp.vuw.ac.nz!canterbury.ac.nz!aldebaran!rozee From: rozee@elec.canterbury.ac.nz (Robert Rozee) Newsgroups: alt.sys.pdp8 Subject: decmate III bits Date: 16 Jan 1995 02:53:16 GMT Organization: Electrical Engineering, University of Canterbury, New Zealand Lines: 32 Message-ID: <3fcn2s$c72@cantua.canterbury.ac.nz> NNTP-Posting-Host: aldebaran.canterbury.ac.nz X-Newsreader: TIN [version 1.2 PL2] hi, just been having a bit of a clean out, and as a result i have some 'bits' of decmate III to get rid of. the decmate III is labeled pc238-ab on the back. there is no keyboard or monitor. the rx-50 drive is faulty, and the case looks like it might make a fine disk enclosure (as well as being a bit too heavy to post anywhere). as a consequence the 'bits' consist of the logic board and expansion bus card. useful spares perhaps? in addition there are some disks and manuals. bl-da82a-mv dm system test disk drive 0 bl-p115e-bv cbi - intro to wps - drive 0 training bl-p116e-bv " - " " " - drive 1 " bl-n354g-bv dm/wps system diskette v 2.0 options guide, aa-dj94a-tv, list processing/sort/maths/comms/sys mgr guide. wps 'quick reference' guide. the wps full manual (if there ever was one) is _not_ there (missing). all the above is 'free to a good home'. let me know if anything is of use to you. regards, rob. -- we are so blinded by our own existence that we fail to see what is there when we are not. From cjl@maestro.maestro.com Thu Jan 19 07:44:56 EST 1995 Article: 1135 of alt.sys.pdp8 Newsgroups: alt.jokes.pentium,alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!rutgers!dziuxsolim.rutgers.edu!uunet!maestro!cjl From: cjl@maestro.maestro.com (Charles Lasner) Subject: Re: Valuable Offer! Message-ID: Organization: Maestro Technologies, Inc. References: <3e437p$4ecs@usenetp1.news.prodigy.com> <3eaq0k$lgh@paperboy.ids.net> Date: Fri, 13 Jan 1995 17:22:35 GMT Lines: 13 Xref: bigblue.oit.unc.edu alt.jokes.pentium:507 alt.sys.pdp8:1135 In article <3eaq0k$lgh@paperboy.ids.net> Philip Chien writes: > >Anybody interested in a DEC PDP-8/S???? The first 'personal' (e.g. under >$10,000) computer ever made. Makes a *GREAT* paperweight. I don't think >it has the Pentium bug (tm) though - it could only add and subtract >integers - never could handle those division signs! > >Philip Chien, Earth News - space writer and consultant PCHIEN@IDS.NET Are you serious. We at a.s.p want to know! cjl From ben@lonestar.jpl.utsa.edu Thu Jan 19 08:19:21 EST 1995 Article: 1136 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!rutgers!koriel!cs.utexas.edu!news.sprintlink.net!howland.reston.ans.net!swrinde!ringer.cs.utsa.edu!lonestar.jpl.utsa.edu!ben From: ben@lonestar.jpl.utsa.edu (Ben A. Fairbank) Newsgroups: alt.sys.pdp8 Subject: DEC-Tape Cacacity? Date: 16 Jan 1995 21:56:52 GMT Organization: University of Texas at San Antonio Lines: 11 Distribution: world Message-ID: <3feq34$ho8@ringer.cs.utsa.edu> NNTP-Posting-Host: lonestar.jpl.utsa.edu Twenty years ago, give or take a few, I worked as a PDP-8 assembly language programmer in addition to some scientific duties. My OS/8 manual is long gone, but I was recently wondering how the DEC-Tape minireels compare to our current diskettes. Can anyone tell me the capacity of one small reel? 128 K of 12-bit words seems about right, but it is a figure pulled from the air. Ben Fairbank From johnam@datastorm.com Thu Jan 19 08:24:19 EST 1995 Article: 1137 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!gatech!swrinde!hookup!news.mathworks.com!zombie.ncsc.mil!golf!news From: johnam@datastorm.com (John A. Maier) Newsgroups: comp.sys.dec.micro,alt.sys.pdp8 Subject: Re: DECMATE III, I've forgot what it is. Date: Tue, 17 Jan 95 16:28:55 PST Organization: Datastorm Technologies Inc. Lines: 22 Message-ID: <3fhfog$71a@golf.ustores.missouri.edu> References: <3f1no6$16b@ball.datastorm.com> <3f3rt7$c9d@nexus.uiowa.edu> NNTP-Posting-Host: 198.209.154.126 Mime-Version: 1.0 X-Newsreader: WinVN 0.93.0 Xref: bigblue.oit.unc.edu comp.sys.dec.micro:3545 alt.sys.pdp8:1137 In article <3f6f4l$142c@bigblue.oit.unc.edu>, lasner@sunSITE.unc.edu says... > >Please be advised that there are readers of a.s.p that need spare parts >desperately for various projects, including maintaining other DECmates, >etc. You sound like someone who has no committment to the hardware you >currently have. Please give some others a chance who have a demonstrated >need, etc. > I am just going to tinker around on it a bit just for fun. Once I have had my fill, I'll be happy to put it up on the auction block. If any of you are interested or know of an interested party, drop me a note. I'll put in a folder for future reference. BTW, I have a number of OS/278 manuals that I got from a guy a number of years ago, so if anyone is interested in these too, they will be up for grabs. I always admired the PDP-8, so this will just be one of those, just for the hell of it things, so it will be a few months. If you don't hear from me, just drop me a note in case I forget. thanks, jam johnam@datastorm.com (John A. Maier) From yankus@ITD.Sterling.COM Thu Jan 19 08:28:41 EST 1995 Article: 1138 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!agate!howland.reston.ans.net!math.ohio-state.edu!hobbes.physics.uiowa.edu!dunix.drake.edu!sparky!yankus From: yankus@ITD.Sterling.COM (Mike Yankus) Newsgroups: alt.sys.pdp8 Subject: Re: DEC-Tape Cacacity? Date: 17 Jan 1995 16:29:46 GMT Organization: Sterling Software Lines: 16 Message-ID: <3fgr9q$8h1@sparky.sterling.com> References: <3feq34$ho8@ringer.cs.utsa.edu> NNTP-Posting-Host: ssbell.itd.sterling.com In article <3feq34$ho8@ringer.cs.utsa.edu>, Ben A. Fairbank wrote: >Twenty years ago, give or take a few, I worked as a PDP-8 assembly >language programmer in addition to some scientific duties. My OS/8 >manual is long gone, but I was recently wondering how the DEC-Tape >minireels compare to our current diskettes. Can anyone tell me the >capacity of one small reel? 128 K of 12-bit words seems about right, >but it is a figure pulled from the air. > On a PDP-11 the TU56 DECtapes held 578 512-byte blocks or 289K bytes. -- Mike Yankus PRC Inc. INTERNET: yankus@ITD.Sterling.COM 1410 Wall St. Phone: (402) 291-8300 Bellevue, NE. 68005-3687 FAX: (402) 291-4362 From jones@pyrite.cs.uiowa.edu Thu Jan 19 08:29:09 EST 1995 Article: 1139 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!eff!news.umbc.edu!europa.eng.gtefsd.com!howland.reston.ans.net!math.ohio-state.edu!hobbes.physics.uiowa.edu!news.uiowa.edu!news From: jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) Newsgroups: alt.sys.pdp8 Subject: Re: DEC-Tape Cacacity? Date: 17 Jan 1995 16:29:47 GMT Organization: University of Iowa, Iowa City, IA, USA Lines: 19 Distribution: world Message-ID: <3fgr9r$qjo@nexus.uiowa.edu> References: <3feq34$ho8@ringer.cs.utsa.edu> NNTP-Posting-Host: pyrite.cs.uiowa.edu ben@lonestar.jpl.utsa.edu (Ben A. Fairbank): > Twenty years ago, give or take a few, I worked as a PDP-8 assembly > language programmer in addition to some scientific duties. My OS/8 > manual is long gone, but I was recently wondering how the DEC-Tape > minireels compare to our current diskettes. Unformatted, a standard DECtape held 849,036 lines of 3 bits each, or 2,547,108 bits. Formatted in 129 word blocks (as was standard on a PDP-8, where only 128 of the 129 words were typically used), the standard formatted tape contained 1464 blocks. 1464 x 128 = 187,392 words per tape, 12 bits each. Converting to 8 bit bytes, this is 281,088 bytes. Someday, I've got to get my DECtape drives working! Doug Jones jones@cs.uiowa.edu From johnam@datastorm.com Thu Jan 19 08:29:28 EST 1995 Article: 1140 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!rutgers!utcsri!newsflash.concordia.ca!news.mcgill.ca!mcrcim.mcgill.edu!bloom-beacon.mit.edu!satisfied.elf.com!news.mathworks.com!zombie.ncsc.mil!golf!news From: johnam@datastorm.com (John A. Maier) Newsgroups: comp.sys.dec.micro,alt.sys.pdp8 Subject: Re: DECMATE III, I've forgot what it is. Date: Tue, 17 Jan 95 16:32:25 PST Organization: Datastorm Technologies Inc. Lines: 29 Message-ID: <3fhfv0$71a@golf.ustores.missouri.edu> References: <3f1no6$16b@ball.datastorm.com> <3f3rt7$c9d@nexus.uiowa.edu> <3fhfog$71a@golf.ustores.missouri.edu> NNTP-Posting-Host: 198.209.154.126 Mime-Version: 1.0 X-Newsreader: WinVN 0.93.0 Xref: bigblue.oit.unc.edu comp.sys.dec.micro:3547 alt.sys.pdp8:1140 In article <3fhfog$71a@golf.ustores.missouri.edu>, johnam@datastorm.com says... >I am just going to tinker around on it a bit just for fun. Once I have >had my fill, I'll be happy to put it up on the auction block. If any >of you are interested or know of an interested party, drop me a note. I'll >put in a folder for future reference. BTW, I have a number of OS/278 >manuals that I got from a guy a number of years ago, so if anyone is >interested in these too, they will be up for grabs. I always admired the >PDP-8, so this will just be one of those, just for the hell of it things, >so it will be a few months. If you don't hear from me, just drop me a note >in case I forget. > >thanks, >jam >johnam@datastorm.com (John A. Maier) > Oh yes, if any one has a good TK-50 around, I might be pursuaded to drop the whole lot into a box and ship it all off in exchange...hint, hint. Yea, I have my price. jam johnam@datastorm.com (John A. Maier) From mo09@badbit.math2.nat.tu-bs.de Thu Jan 19 08:38:54 EST 1995 Article: 1141 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!godot.cc.duq.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!Germany.EU.net!news.dfn.de!zeus.rbi.informatik.uni-frankfurt.de!terra.wiwi.uni-frankfurt.de!news.th-darmstadt.de!fauern!gs.dfn.de!tubsibr!news.ibr.cs.tu-bs.de!pueschel From: pueschel@math2.nat.tu-bs.de (Olaf Pueschel) Newsgroups: alt.sys.pdp8 Subject: -8 memory managment question Date: 17 Jan 1995 17:24:07 GMT Organization: Institut fuer Angewandte Mathematik, TU Braunschweig Lines: 20 Distribution: world Message-ID: Reply-To: mo09@badbit.math2.nat.tu-bs.de NNTP-Posting-Host: badbit.math2.nat.tu-bs.de Hi folks! When reading some of the documentation located on Doug Jones *great* www pages, I ran into the following question: How are the instruction/data field values handled to form a physical address? are they a) used to select one of 8 mutually excusive memory banks b) prepended to the mem addr to form some sort of segment register (yes I worked on 80x86 machines :( ) c) used in som other magical way? -olf PS: are there any other germans out here? --- Olaf P"uschel, Broitzemer Str. 10, 38118 Braunschweig, Germany, +49-531-895643 -- Olaf P"uschel Broitzemer Str. 10 38118 Braunschweig Germany +49 531 895643 ------------------------------------------------------------------------------- *** this is NOT a random-generated signature :-) From lasner@sunSITE.unc.edu Thu Jan 19 08:39:16 EST 1995 Article: 1142 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: alt.sys.pdp8 Subject: Re: DEC-Tape Cacacity? Date: 19 Jan 1995 13:26:02 GMT Organization: SunSITE, University of North Carolina, Chapel Hill Lines: 97 Message-ID: <3flp9a$mbj@bigblue.oit.unc.edu> References: <3feq34$ho8@ringer.cs.utsa.edu> NNTP-Posting-Host: calzone.oit.unc.edu In article <3feq34$ho8@ringer.cs.utsa.edu>, Ben A. Fairbank wrote: >Twenty years ago, give or take a few, I worked as a PDP-8 assembly >language programmer in addition to some scientific duties. My OS/8 >manual is long gone, but I was recently wondering how the DEC-Tape >minireels compare to our current diskettes. Can anyone tell me the >capacity of one small reel? 128 K of 12-bit words seems about right, >but it is a figure pulled from the air. > >Ben Fairbank > > > A *standard* PDP-8 DECtape *officially* holds 2702 octal blocks of 129 decimal words each. Other than the Disk Monitor and related software, most PDP-8 software ignores the 129th word. There is a hardware restriction that since the mark-track design is a multiple of 6 info bits over 3 data bits each, the tape block length must be a multiple of 18 bits. However, the only two official formats that DEC ever official endorsed are the PDP-8 format (derived from an arbitrary formula that is quite conservative) and the PDP-10/20/7/9/15/11 format which is 384 decimal words each block and a total of 1102 blocks. Note that this format gives a whole lot more data, but is only considered an archival format on the PDP-8 (except for OS8 (no slash there!), see below). The DECtape formatting program uses MARK384 as a special exception case that "violates" the formula to allow the extra data. MARK129 just plugs into the formula producing the standard result, etc. The extra capacity is there because there are far fewer inter-block zones, and additionally all of the tape is used. Thus, while the tape was sold as a *nominal* 260 feet, you really need 10-20 feet more to properly support the MARK384 format. Unfortunately, later DECtape sales were for shortened tape (sometimes to less than 260 feet; people measured it :-( !), and caused tapes to be relegated to MARK129 format strictly, or worse, when the worn end was cut off, it was then too short and was trashed :-(. The older tapes that could format MARK384 with ease were often set aside for use as "extra-long" PDP-8 DECtapes formatted using a patched version of the format program that made MARK129 perform a format of 3000 blocks. P?S/8's version of the formatter has option switches to set this mode more easily, etc. When the same media is used as LINCtape, even more blocks can be had. The reasons are as follows: Other than for use on the PDP-12 version of the Disk Monitor, there is no sane reason to have 129 words/block, since LINC hardware only demands multiples of 12-bits, etc. (Note: P?S/8 and OS/8 LINCtape PDP-12 handlers all check for the possible presence of the nuisance 129th word. The potentially destroyed 129th word is stored before the DMA is started, and restored afterwards!) Thus, using 128 words/block gets more blocks. LINCtape has less overhead in the inter-block zone. As a consequence, it cannot transfer backwards (but can search backwards; most PDP-8 software transfers forwards-only anyway.) But this also contributes to higher tape block count. LINCtape writes at a marginally higher density than DECtape. As a defense against the ever-shortening tape lengths, some users cranked up the mark clock on their DECtape controllers (especially TC08) to get proper tape lengths anyway. This could lead to trouble on TD8E-based systems. As an alternative to tweaking the clock, "laying on of hands" (usually a thumb) is often used to get a little more tape usage on marginal length tapes. Again, the TD8E consequences could be bad. The best possible tape length achieved is 4000 octal blocks of 128 decimal 12-bit words on a LINC-8. However, there is an anomaly there: The LINCtape format doesn't use end-zone sensing to turn the tape around! Instead, dummy blocks -1, -2, -3, -4 etc. are written so the arithmetic search works the turn-around instead. Additionally, the inner end should have blocks higher than the largest desired for the same reason. However, on the PDP-12, the end-zone will tend to make the tape reverse, so this isn't a problem region, etc. But on the LINC-8, if a block 4000 is searched forward after a transfer of block 3777, the search algorithm is blown, i.e., it will tend to make the tape turn *further* into the inner end, only to have the end-zone sensing reverse the tape direction. Thus, the tape can get trapped in a search loop unless manually overridden. Good tape handlers can avoid this assuming the next block transferred is a lower-numbered block,e tc. So, the most optimistic PDP-8 capacity for this media is 4000 blocks or 2048x128x12 bits. I think that works out to 384 KBytes for those who can't count higher than 8. cjl From lasner@sunSITE.unc.edu Thu Jan 19 08:39:24 EST 1995 Article: 1143 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: comp.sys.dec.micro,alt.sys.pdp8 Subject: Re: DECMATE III, I've forgot what it is. Date: 19 Jan 1995 13:31:08 GMT Organization: SunSITE, University of North Carolina, Chapel Hill Lines: 18 Message-ID: <3flpis$1ekn@bigblue.oit.unc.edu> References: <3f1no6$16b@ball.datastorm.com> <3f3rt7$c9d@nexus.uiowa.edu> <3fhfog$71a@golf.ustores.missouri.edu> NNTP-Posting-Host: calzone.oit.unc.edu Xref: bigblue.oit.unc.edu comp.sys.dec.micro:3550 alt.sys.pdp8:1143 In article <3fhfog$71a@golf.ustores.missouri.edu>, John A. Maier wrote: References: <3feq34$ho8@ringer.cs.utsa.edu> <3fgr9q$8h1@sparky.sterling.com> NNTP-Posting-Host: calzone.oit.unc.edu In article <3fgr9q$8h1@sparky.sterling.com>, Mike Yankus wrote: >On a PDP-11 the TU56 DECtapes held 578 512-byte blocks or 289K bytes. Nope. There are 578 blocks, but each holds 256 18-bit words. The PDP-11 just chooses to waste some of the bits for most storage purposes. Thus, your figure should be inflated by 1/8th. (Unless you are referring to 9-bit bytes :-).) cjl From lasner@sunSITE.unc.edu Thu Jan 19 08:39:51 EST 1995 Article: 1145 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: alt.sys.pdp8 Subject: Re: -8 memory managment question Date: 19 Jan 1995 13:45:44 GMT Organization: SunSITE, University of North Carolina, Chapel Hill Lines: 53 Message-ID: <3flqe8$nhq@bigblue.oit.unc.edu> References: NNTP-Posting-Host: calzone.oit.unc.edu In article , Olaf Pueschel wrote: >Hi folks! >When reading some of the documentation located on Doug Jones *great* >www pages, I ran into the following question: > >How are the instruction/data field values handled to form a physical address? > >are they a) used to select one of 8 mutually excusive memory banks On a 32K machine model this is the case. > b) prepended to the mem addr to form some sort of segment register > (yes I worked on 80x86 machines :( ) > c) used in som other magical way? There are two different forms of memory beyond 32K. One is supported by two different (incompatible) memory controllers; the second is only supported by one of them. The first is from DEC. Memory is decoded further into "banks" each uniquely containing 32K of 8 fields as in the smaller model. By careful redefinition, an additional addressing bit was inserted into the CIF and CDF instructions, and the additional support was added in other instructions. The further additional bit to raise the addressing space to 128K is obtained by taking away all IOT's in the range of 630x-637x. For compatibility, this second part can be disabled, limiting the addressing to 64K. The second is from CESI. There are bank registers you have to load with additional instructions. 32K remains compatible, as long as you zero out the bank extensions. Again, this gets you to 128K and it is hardware compatible with DEC's collections of up-to 128K memory boards. However, it also supports a different buss decode for up to 512K using boards DEC never made. The programming is essentially the same, only "wider" thus occupying more "bank" bits, etc. (The KT8A buss was defined as a 1-of-4 select; the CESI 512K mode makes it a 4-bit encode, etc.) > >-olf > >PS: are there any other germans out here? >--- >Olaf P"uschel, Broitzemer Str. 10, 38118 Braunschweig, Germany, +49-531-895643 >-- > Olaf P"uschel Broitzemer Str. 10 38118 Braunschweig Germany +49 531 895643 >------------------------------------------------------------------------------- >*** this is NOT a random-generated signature :-) cjl From kgg@mv.mv.com Tue Jan 24 11:48:36 EST 1995 Article: 1146 of alt.sys.pdp8 Newsgroups: alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!udel!news.sprintlink.net!mv!mv.mv.com!kgg From: kgg@mv.mv.com (Kenn Goutal) Subject: Re: DEC-Tape Cacacity? Message-ID: Nntp-Posting-Host: mv.mv.com Sender: usenet@mv.mv.com (System Administrator) Organization: MV Communications, Inc. Date: Wed, 18 Jan 1995 05:51:31 GMT References: <3feq34$ho8@ringer.cs.utsa.edu> <3fgr9r$qjo@nexus.uiowa.edu> Lines: 15 In article <3fgr9r$qjo@nexus.uiowa.edu>, Douglas W. Jones,201H MLH,3193350740,3193382879 wrote: >Someday, I've got to get my DECtape drives working! When you do, I've got some files that need converting! Actually, most of mine are LINCtape, though I also have a couple DECtapes. I remember being impressed when, after my box of tapes had sat in the back of my VW through 3 or 4 freezing winters and blazing summers, I plopped some of them onto a drive, and after a few "warm-up" spins, they read in just fine! -- Kenn Goutal kenn@mv.mv.com http://www.mv.com/users/kgg/kenn.html From ben@lonestar.jpl.utsa.edu Tue Jan 24 11:49:01 EST 1995 Article: 1147 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!agate!howland.reston.ans.net!swrinde!ringer.cs.utsa.edu!lonestar.jpl.utsa.edu!ben From: ben@lonestar.jpl.utsa.edu (Ben A. Fairbank) Newsgroups: alt.sys.pdp8 Subject: Was Re: DEC-Tape Cacacity? Now gravitational interaction. Date: 18 Jan 1995 14:02:41 GMT Organization: University of Texas at San Antonio Lines: 19 Message-ID: <3fj721$71b@ringer.cs.utsa.edu> References: <3feq34$ho8@ringer.cs.utsa.edu> <3fgr9r$qjo@nexus.uiowa.edu> NNTP-Posting-Host: lonestar.jpl.utsa.edu In article kgg@mv.mv.com (Kenn Goutal) writes: snip >I remember being impressed when, after my box of tapes had sat in the back >of my VW through 3 or 4 freezing winters and blazing summers, I plopped >some of them onto a drive, and after a few "warm-up" spins, they read in >just fine! > >-- Kenn Goutal >kenn@mv.mv.com Which reminds me that in '73 or so our lab had retreaded a clerk/administrator into a PDP-8 programmer (and a damn good one). Her husband was our electonics technician and had explained to her that the magnetic dots on the tape tended to accumulate at the edges of the tape unless they were regularly rotated. For months we, not knowing what he had told her, watched her every Monday picking up the tapes and turning them over. Finally someone asked her what she was doing... From rla@rahul.net Tue Jan 24 11:49:11 EST 1995 Article: 1148 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!gatech!howland.reston.ans.net!swrinde!sgiblab!rahul.net!a2i!rla.a2i!rla From: Bob Armstrong Newsgroups: comp.os.vms,alt.sys.pdp11,alt.sys.pdp8,vmsnet.misc Subject: RX01/2 drive needed! Followup-To: poster Date: 19 Jan 1995 21:09:02 GMT Organization: a2i network Lines: 7 Distribution: usa Message-ID: <3fmkde$prt@hustle.rahul.net> NNTP-Posting-Host: foxtrot.rahul.net NNTP-Posting-User: rla Keywords: RX01 RX02 Xref: bigblue.oit.unc.edu comp.os.vms:84462 alt.sys.pdp8:1148 vmsnet.misc:3190 I need a RX01 or RX02 drive unit. I already have the controller. Any offers or suggestions would be appreciated! Thanks, -- Bob Armstrong From hassinger@eisner.decus.org Tue Jan 24 11:49:36 EST 1995 Article: 1149 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!rutgers!koriel!lll-winken.llnl.gov!uwm.edu!news.alpha.net!news.mathworks.com!noc.near.net!eisner!hassinger From: hassinger@eisner.decus.org (Bob Hassinger) Newsgroups: alt.sys.pdp8 Subject: Re: DEC-Tape Cacacity? Message-ID: <1995Jan19.000834.8760@eisner> Date: 19 Jan 95 00:08:33 -0500 References: <3feq34$ho8@ringer.cs.utsa.edu> <3fgr9r$qjo@nexus.uiowa.edu> Distribution: world Organization: DECUServe Lines: 23 In article <3fgr9r$qjo@nexus.uiowa.edu>, jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) writes: > ben@lonestar.jpl.utsa.edu (Ben A. Fairbank): > >> Twenty years ago, give or take a few, I worked as a PDP-8 assembly >> language programmer in addition to some scientific duties. My OS/8 >> manual is long gone, but I was recently wondering how the DEC-Tape >> minireels compare to our current diskettes. > > Unformatted, a standard DECtape held 849,036 lines of 3 bits each, or > 2,547,108 bits. Formatted in 129 word blocks (as was standard on a > PDP-8, where only 128 of the 129 words were typically used), the > standard formatted tape contained 1464 blocks. > > 1464 x 128 = 187,392 words per tape, 12 bits each. Converting to 8 bit > bytes, this is 281,088 bytes. Not as well know where schemes to get more than the standard on a tape. I recall Jim VanZee telling me about using schemes as extreme as pressing on the edge of the reel while the tape was being formatted to slow it down. That let him get more blocks on the tape. Bob Hassinger hassinger@merlin.lmig.com From bqt@Krille.Update.UU.SE Tue Jan 24 11:50:03 EST 1995 Article: 1150 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!godot.cc.duq.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!Germany.EU.net!EU.net!sunic!columba.udac.uu.se!Krille.Update.UU.SE!Krille.Update.UU.SE!not-for-mail From: bqt@Krille.Update.UU.SE (Johnny Billquist) Newsgroups: alt.sys.pdp8 Subject: Re: -8 memory managment question Date: 18 Jan 1995 14:40:28 +0100 Organization: Update Computer Club Lines: 26 Message-ID: <3fj5oq$jcq@Krille.Update.UU.SE> References: NNTP-Posting-Host: krille.update.uu.se In pueschel@math2.nat.tu-bs.de (Olaf Pueschel) writes: >Hi folks! >When reading some of the documentation located on Doug Jones *great* >www pages, I ran into the following question: >How are the instruction/data field values handled to form a physical address? >are they a) used to select one of 8 mutually excusive memory banks > b) prepended to the mem addr to form some sort of segment register > (yes I worked on 80x86 machines :( ) > c) used in som other magical way? Answer is a and b. They are really the same thing. 80x86 machines are a third variety, where the segment is shifted, and *added* to the virtual address to get the physical address. (The MMU on the pdp-11 works this way too, as do most modern machines, the difference being that modern machines have the same size for virtual and physical addresses). Johnny -- Johnny Billquist || "I'm on a bus CS student at Uppsala University || on a psychedelic trip email: bqt@minsk.docs.uu.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol From billh@haygood.com Tue Jan 24 11:51:12 EST 1995 Article: 1151 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!godot.cc.duq.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!news.sprintlink.net!nwnexus!krel.iea.com!comtch!billh From: billh@comtch.iea.com (Bill Haygood) Newsgroups: alt.sys.pdp8 Subject: Re: -8 memory managment question Date: 20 Jan 1995 00:27:00 GMT Organization: home Lines: 38 Distribution: world Message-ID: <3fn00k$cj9@krel.iea.com> References: Reply-To: billh@haygood.com NNTP-Posting-Host: comtch.iea.com X-Newsreader: TIN [version 1.2 PL1] Olaf Pueschel (pueschel@math2.nat.tu-bs.de) wrote: : How are the instruction/data field values handled to form a physical address? : are they a) used to select one of 8 mutually excusive memory banks On *indirect* AND, TAD, ISZ and DCA instructions only, the PDP-8 locates the field relative address (low order 12 bits) of the operand by fetching it either from the current instruction field page or page zero of the current instruction field. The machine then prepends the DF to this 12-bit address and the operand gets fetched from that address. For all other *indirect* (and *all* direct instructions), the DF does not come into play. : b) prepended to the mem addr to form some sort of segment register : (yes I worked on 80x86 machines :( ) In the PDP-8, since the DF always gets "prepended" rather than shifted a few bits (as in the 80x86 uprocs), I personally find the PDP-8 scheme much easier to understand. For example, in the PDP-8, one specifies an absolute address as a 5-digit octal number: 41600 (field 4, relative address 1600). We generally don't use any other way to specify an absolute address. In the Intel 80x86 uprocs, OTOH, you could specify absolute address 45678 as segment 4000, offset 5678 or as segment 4001, offset 5668, or as segment 4002, offset 5658, or as segment 4100, offset 4678, ... (this could really get tedious after a while!). (I once has to write a subroutine to return a true or false as to whether two 80x86 addresses matched or not [given segment,offset as input]!) : c) used in som other magical way? To many of us, the PDP-8 == pure magic! After all, some consider the -8 as the very first RISC machine! If either Bob Supnik at DEC and/or cjl at sunsite and/or anyone else read(s) this post, please feel free to critique my understanding of the DF register as outlined above. Bill From jones@pyrite.cs.uiowa.edu Tue Jan 24 11:52:28 EST 1995 Article: 1152 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!godot.cc.duq.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!math.ohio-state.edu!hobbes.physics.uiowa.edu!news.uiowa.edu!news From: jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) Newsgroups: alt.sys.pdp8 Subject: Re: DEC-Tape Cacacity? Date: 18 Jan 1995 19:46:15 GMT Organization: University of Iowa, Iowa City, IA, USA Lines: 46 Distribution: world Message-ID: <3fjr67$qes@nexus.uiowa.edu> References: NNTP-Posting-Host: pyrite.cs.uiowa.edu >From article , by kgg@mv.mv.com (Kenn Goutal): > > Actually, most of mine are LINCtape, though I also have a couple DECtapes. > I remember being impressed when, after my box of tapes had sat in the back > of my VW through 3 or 4 freezing winters and blazing summers, I plopped > some of them onto a drive, and after a few "warm-up" spins, they read in > just fine! DECtape is an archivist's dream. The sandwich mylar construction of the tape makes it far less sensitive than normal mag-tape to such environmental considerations such as humidity, and the massive redundancy of the recording format (all bits recorded twice) makes it resistant to the random flaking and dropout problems that plague those who try to read ancient conventional tapes. You can take a hole punch to a DECtape (the kind you'd use to punc a ticket or make a hole in a page you wanted to put in a binder), and there's a good chance you won't damage any data. You lose out, though, if you're unlucky enough to punch a hole opposite a bad spot in the tape that wiped out the other copy of the bits you punched out. I know some wonderful DECtape stories, the best being from a friend of mine who loaded a DECtape on his PDP-10 upside down, then hit the load switch, at which point, the whole reel of tape was dumped on the floor at high speed. He must have lost a bit too much sleep, but anyway, his reaction was to pull the reels, pop them on right side up, and hit the load switch again, at which point the tape was snapped back onto the reels at high speed, complete with many folds, dirt from the floor, and various crumples. His solution to the folds and crumples was to rock the tape quickly back and forth over its whole length. That got out most of the wrinkles, but it set the others even harder. Then, he pressed the tape down over the head with his finger and wound the tape back and forth again to get it all flat. When he tried to read the tape, there were plenty of bad bits, but he figured that they were just places where the wrinkles were holding the tape away from the head, so he pressed his finger against the head while he read the tape. The friction was too much, but by hand cranking the tape reels with a pencil while he held it against the head with a finger, he successfully read the tape, without errors. That was a good thing, it was the only tape he had with that data on it. Doug Jones jones@cs.uiowa.edu From alexs@cyberquest.com Tue Jan 24 11:53:08 EST 1995 Article: 1153 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!agate!howland.reston.ans.net!swiss.ans.net!news.cyberquest.com!usenet From: alexs@cyberquest.com Newsgroups: alt.sys.pdp8 Subject: PDP-8 spare parts/systems. Date: 20 Jan 1995 06:34:35 GMT Organization: CyberQuest BBS Lines: 15 Message-ID: <3fnlhr$jn8@starbug.cyberquest.com> NNTP-Posting-Host: godfatherttyp8.cyberquest.com X-Newsreader: AIR News 3.X (SPRY, Inc.) I was in the last class year at the University of Michigan Engin School to have logic lab/assembly programming class on an old LINC/8 machine. They switched to Z80s the next semester. I also did work study programming on various PDP-8s at the Cooley Laboratory on the U of M's North Campus. I recall now that they had lots of those pesky plug-in circuit cards, just enough to keep their systems in repair. You might look up Ted Birdsall or Kurt Metzger at the U of M to see whatever became of their PDP-8 systems. They might be long surplused, but it's worth a try. -Alexander Stoll, U of M class of 1980. From wilsonj@alum01.its.rpi.edu Tue Jan 24 11:53:31 EST 1995 Article: 1154 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!news2.near.net!howland.reston.ans.net!news.sprintlink.net!pipex!sunic!psinntp!psinntp!psinntp!rebecca!rpi!wilsonj From: wilsonj@alum01.its.rpi.edu (John Wilson) Newsgroups: alt.sys.pdp8 Subject: Re: DEC-Tape Cacacity? Date: 17 Jan 1995 04:25:28 GMT Organization: Rensselaer Polytechnic Institute, Troy NY Lines: 16 Message-ID: <3ffgro$rhj@usenet.rpi.edu> References: <3feq34$ho8@ringer.cs.utsa.edu> NNTP-Posting-Host: alum01.its.rpi.edu In article <3feq34$ho8@ringer.cs.utsa.edu>, Ben A. Fairbank wrote: >Can anyone tell me the >capacity of one small reel? 128 K of 12-bit words seems about right, >but it is a figure pulled from the air. Right ballpark, it's actually a little more. The "standard" format was 1474. blocks (somehow that's easier to remember in octal) of 129. words each. I think TOG-8 would let you mess around with weird formats but at your own risk, the hardware requires a block to be a multiple of 36 bits but otherwise you could pretty much do your own thing and it would work, with your own driver. Even with stock format the tape would spin off the end of the reel while formatting the end zone unless you had only threaded it just enough to barely stay on the reel at the beginning. John Wilson From supnik@human.enet.dec.com Tue Jan 24 11:53:50 EST 1995 Article: 1155 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!taco.cc.ncsu.edu!lll-winken.llnl.gov!decwrl!pa.dec.com!sousa.tay1.dec.com!human.enet.dec.com!supnik From: supnik@human.enet.dec.com (Bob Supnik) Newsgroups: alt.sys.pdp8 Subject: Re: pdp8 tad instruction Date: 12 JAN 95 14:18:24 Organization: Digital Equipment Corporation Lines: 13 Distribution: world Message-ID: <3f3vck$9te@sousa.tay1.dec.com> References: NNTP-Posting-Host: human.enet.dec.com In article , pueschel@math2.nat.tu-bs.de (Olaf Pueschel) writes... > [text deleted] >In addition if someone could provide me with the functionality of the EAE >and the Memory extinsion units, on day I could impement these to :) > I realize it takes the fun out, but check one of the published simulators (mine or Bill Haygood's) for the memory extension and EAE. I wrote mine from the maintenance manual and code checked Bill's for compliance. Bob Supnik >Supnik@human.enet.dec.com >All opinions expressed are those of a hardline microcoder >and do not reflect those of Digital Equipment Corporation From Pchien@ids.net Tue Jan 24 11:54:17 EST 1995 Article: 1156 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!taco.cc.ncsu.edu!lll-winken.llnl.gov!sol.ctr.columbia.edu!howland.reston.ans.net!pipex!uunet!paperboy.ids.net!pslip016.ksc-fl.ids.net!Pchien From: Philip Chien Newsgroups: alt.jokes.pentium,alt.sys.pdp8 Subject: Re: Valuable Offer! Date: 16 Jan 1995 20:21:06 GMT Organization: Earth News Lines: 28 Distribution: world Message-ID: <3fekfi$2i8@paperboy.ids.net> References: NNTP-Posting-Host: pslip016.ksc-fl.ids.net X-Newsreader: Nuntius Version 1.2 X-XXMessage-ID: X-XXDate: Mon, 16 Jan 1995 20:21:47 GMT Xref: bigblue.oit.unc.edu alt.jokes.pentium:559 alt.sys.pdp8:1156 In article Charles Lasner, cjl@maestro.maestro.com writes: >In article <3eaq0k$lgh@paperboy.ids.net> Philip Chien writes: >> >>Anybody interested in a DEC PDP-8/S???? The first 'personal' (e.g. under >>$10,000) computer ever made. Makes a *GREAT* paperweight. I don't think >>it has the Pentium bug (tm) though - it could only add and subtract >>integers - never could handle those division signs! >> >>Philip Chien, Earth News - space writer and consultant PCHIEN@IDS.NET > >Are you serious. We at a.s.p want to know! Actually, considering that I just had to carry the 200 lb. monster 20 miles to my new home, yes I am serious - I've got a DEC PDP-8/S here. Makes a great paperweight ... Philip Chien, Earth News - space writer and consultant PCHIEN@IDS.NET __ __^__ __________ | \ +---/ \---+ (========= |____\___________ +---\_____/---+ // >____)| | \__ \ \______//___ >/ |________| \ [ _____\ >|____________________\ \_______/ Roger, go at throttle up CHR$(32) the final frontier From rtschalaer@cvax.psi.ch Tue Jan 24 11:54:38 EST 1995 Article: 1157 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!news.alpha.net!uwm.edu!math.ohio-state.edu!howland.reston.ans.net!pipex!news.oleane.net!oleane!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!scsing.switch.ch!pss100.psi.ch!psiclu.psi.ch!rtschalaer From: rtschalaer@cvax.psi.ch (Life is hard... and then you die!) Newsgroups: alt.sys.pdp8 Subject: circuit diagrams for the VT78 Date: 23 Jan 95 16:26:07 +0200 Organization: Paul Scherrer Institute Lines: 16 Message-ID: <1995Jan23.162607.1@psiclu.psi.ch> NNTP-Posting-Host: psiclb.psi.ch Hi, Does anybody happen to have circuit diagrams for the VT78? The power supply is bust, and I'm having trouble fixing it because many of the parts are labeled in an unconventional (for me) way. Since the power supply is switchable for 110V and 230V, I assume they used the same one for models sold in the states. If anybody has access to these diagrams and could send me photocopies (postage and packing will be paid) or if anybody could give me some pointers as to where these could be attained (DEC here wasn't very helpful) I'd be grateful. TIA, Ronald From Owie@ix.netcom.com Sat Jan 28 12:44:05 EST 1995 Article: 1158 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!news.alpha.net!uwm.edu!cs.utexas.edu!howland.reston.ans.net!ix.netcom.com!netnews From: Owie@ix.netcom.com (Owen Fisk) Newsgroups: alt.sys.pdp8 Subject: Re: circuit diagrams for the VT78 Date: 24 Jan 1995 01:03:44 GMT Organization: Netcom Lines: 28 Distribution: world Message-ID: <3g1jlg$4h3@ixnews2.ix.netcom.com> References: <1995Jan23.162607.1@psiclu.psi.ch> NNTP-Posting-Host: ix-sac1-25.ix.netcom.com In <1995Jan23.162607.1@psiclu.psi.ch> rtschalaer@cvax.psi.ch (Life is hard... and then you die!) writes: > > > Hi, > >Does anybody happen to have circuit diagrams for the VT78? The power supply >is bust, and I'm having trouble fixing it because many of the parts are >labeled in an unconventional (for me) way. Since the power supply is >switchable for 110V and 230V, I assume they used the same one for models >sold in the states. If anybody has access to these diagrams and could send >me photocopies (postage and packing will be paid) or if anybody could give >me some pointers as to where these could be attained (DEC here wasn't very >helpful) I'd be grateful. > > TIA, > > Ronald > > Try Jerrasic Park - this thing is a real dinasaur From russ@cs.indiana.edu Sat Jan 28 14:16:55 EST 1995 Article: 1159 of alt.sys.pdp8 Newsgroups: alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!news.alpha.net!uwm.edu!spool.mu.edu!news.cs.indiana.edu!russ@cs.indiana.edu From: "J Russ" Subject: Re: PDP-8 spare parts/systems. Message-ID: <1995Jan23.194547.28781@news.cs.indiana.edu> Organization: Computer Science, Indiana University References: <3fnlhr$jn8@starbug.cyberquest.com> Date: Mon, 23 Jan 1995 19:45:43 -0500 Lines: 29 In article <3fnlhr$jn8@starbug.cyberquest.com>, wrote: >I also did work study programming on various PDP-8s at the >Cooley Laboratory on the U of M's North Campus. I recall >now that they had lots of those pesky plug-in circuit cards, >just enough to keep their systems in repair. You might look >up Ted Birdsall or Kurt Metzger at the U of M to see whatever >became of their PDP-8 systems. They might be long surplused, >but it's worth a try. I bought all the PDP8/LINC8 stuff at the Cooley Laboratory back in 1987. It was an interesting collection of stuff. One of the machines was an old LINC8 down in the basement buried away in a back room. Near the machine was a copy of a radical newspaper from the late 1960's so the machine may have been there for a long time. The LINC8 was missing the core memory and various odds & ends but eventually I was able to rebuild the machine and run the SUDSY diagnostics and demo programs. Another interesting artifact was a box of smoke damaged A series modules that were out of a LINC8 in Florida that had caught on fire. I also got a couple of the cloned DECTAPE drives (I can't remember the name of the manufacturer) but haven't tried firing them up yet. U of M and Ann Arbor are interesting places to visit. If you're ever up there stop in at the university surplus building and check out all of the interesting junk they have for sale. When I was there in November I picked up a couple of junked Apollo workstations for use as parts in my Apollo workstation collection. I didn't see any PDP8 stuff but it may pass through occassionly. Before you head out of town be sure to check out the new and used book stores in downtown Ann Arbor. Jeff Russ Bloomington, IN From frank@rover.uchicago.edu Sat Jan 28 15:17:46 EST 1995 Article: 1160 of alt.sys.pdp8 Newsgroups: alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!usenet.eel.ufl.edu!usenet.cis.ufl.edu!caen!msunews!uchinews!rover.uchicago.edu!frank From: frank@rover.uchicago.edu (Frank R. Borger) Subject: Re: DEC-Tape Cacacity? Message-ID: Lines: 23 Sender: news@uchinews.uchicago.edu (News System) Organization: Radiation Therapy, Michael Reese Hospital References: <3feq34$ho8@ringer.cs.utsa.edu> <3ffgro$rhj@usenet.rpi.edu> Date: Wed, 25 Jan 1995 00:24:29 GMT In article <3ffgro$rhj@usenet.rpi.edu> wilsonj@alum01.its.rpi.edu (John Wilson) writes: >... Even with stock format the tape would spin >off the end of the reel while formatting the end zone unless you had only >threaded it just enough to barely stay on the reel at the beginning. That was a commony symptom of a mal-adjsted formatter. The interface used an adjustable oscillator to determine how OFTEN you wrote formatting data, with the tape moving at constant reel speed. If the frequency of the oscillator was a little low, the format marks were too far apart and you ended up running off the end of the reel. I remember an article that actually made it into DECUScope that described how one software specialist re-wrote the formatter to only do 6' instead of 10' leader/trailer sections. Frank R. Borger - Physicist ___ What's the difference between: Michael Reese - U of Chicago |___ (1) having a baby and Center for Radiation Therapy | |_) _ (2) doing an RSX11M sysgen? net: Frank@rover.uchicago.edu | \|_) ph: 312-791-8075 fa: 791-2517 |_) Having a baby only takes 9 months! From rla@rahul.net Sat Jan 28 15:18:03 EST 1995 Article: 1161 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!gatech!howland.reston.ans.net!news.sprintlink.net!news.clark.net!rahul.net!a2i!rla.a2i!rla From: Bob Armstrong Newsgroups: comp.os.vms,alt.sys.pdp11,alt.sys.pdp8 Subject: DEC part number for RL02 unit number plugs ? Followup-To: poster Date: 26 Jan 1995 20:10:31 GMT Organization: a2i network Lines: 12 Distribution: usa Message-ID: <3g8vjn$rbu@hustle.rahul.net> NNTP-Posting-Host: bolero.rahul.net NNTP-Posting-User: rla Xref: bigblue.oit.unc.edu comp.os.vms:84972 alt.sys.pdp8:1161 Can someone tell me the part number for the unit select plugs on the front of a RL02 ? I have an RL02 that would work if I just had the right numbered plug for it! I need a #2 if it makes a difference. DEC part number assistance has not been able to figure this out - the best they could do was to come up with the part number for the entire front panel! Thanks, -- Bob Armstrong From prep@yarrow.wt.uwa.edu.au Sat Jan 28 15:19:34 EST 1995 Article: 1162 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!taco.cc.ncsu.edu!lll-winken.llnl.gov!noc.near.net!bigboote.WPI.EDU!news.mathworks.com!zombie.ncsc.mil!news.duke.edu!news-feed-1.peachnet.edu!gatech!howland.reston.ans.net!swrinde!ihnp4.ucsd.edu!munnari.oz.au!news.uwa.edu.au!news!prep From: prep@yarrow.wt.uwa.edu.au (Paul Repacholi) Newsgroups: alt.sys.pdp8 Subject: Re: circuit diagrams for the VT78 Date: 24 Jan 1995 22:53:20 GMT Organization: Winthrop Technology Lines: 9 Distribution: world Message-ID: References: <1995Jan23.162607.1@psiclu.psi.ch> <3g1jlg$4h3@ixnews2.ix.netcom.com> NNTP-Posting-Host: yarrow.wt.uwa.edu.au In-reply-to: Owie@ix.netcom.com's message of 24 Jan 1995 01:03:44 GMT Or try a vt100 print set, or vt5x... -- ~Paul +61 (09) 257-1001 prep@yarrow.wt.uwa.edu.au ( preferred ) 1 Crescent Rd, zrepachol@cc.curtin.edu.au Kalamunda, West Aust 6076 From seymour@u.washington.edu Sat Jan 28 15:20:04 EST 1995 Article: 1163 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!gatech!howland.reston.ans.net!news.sprintlink.net!pipex!uunet!news.u.washington.edu!seymour From: seymour@u.washington.edu (Richard Seymour) Newsgroups: comp.os.vms,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: DEC part number for RL02 unit number plugs ? Date: 27 Jan 1995 23:03:58 GMT Organization: University of Washington Lines: 18 Distribution: usa Message-ID: <3gbu4u$bui@nntp1.u.washington.edu> References: <3g8vjn$rbu@hustle.rahul.net> NNTP-Posting-Host: saul2.u.washington.edu Xref: bigblue.oit.unc.edu comp.os.vms:85072 alt.sys.pdp8:1163 In article <3g8vjn$rbu@hustle.rahul.net>, Bob Armstrong wrote: > Can someone tell me the part number for the unit select plugs on >the front of a RL02 ? I have an RL02 that would work if I just had >the right numbered plug for it! I need a #2 if it makes a difference. > > DEC part number assistance has not been able to figure this out - >the best they could do was to come up with the part number for the >entire front panel! it's 12-12691-00 (for unit zero... break pins for other numbers) ??? can't DEC use IPB fiche anymore?? --dick (from the land of obsolete parts...) hmmm... i just looked it up in a 1993 catalog... the "-00" denotes unit zero, -01 unit one, etc. but the prices are interesting: zero and one are $2 each. two and three are $4 each. five and six are $1. From ivie@cc.usu.edu Sat Jan 28 15:20:30 EST 1995 Article: 1164 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!rutgers!rochester!galileo.cc.rochester.edu!ub!news.kei.com!news.mathworks.com!udel!news.sprintlink.net!cs.utexas.edu!news.cs.utah.edu!cc.usu.edu!ivie From: ivie@cc.usu.edu Newsgroups: alt.sys.pdp8 Subject: Re: circuit diagrams for the VT78 Message-ID: <1995Jan25.095613.39182@cc.usu.edu> Date: 25 Jan 95 09:56:13 MDT References: <1995Jan23.162607.1@psiclu.psi.ch> <3g1jlg$4h3@ixnews2.ix.netcom.com> Distribution: world Organization: Utah State University Lines: 14 In article , prep@yarrow.wt.uwa.edu.au (Paul Repacholi) writes: > Or try a vt100 print set, or vt5x... > The power supply on a VT78 is different from the power supply on the VT5x. The VT52 has a large transformer in the "printer" area which, presumably, had to be moved to put the PDP-8 in there. The VT78 has a power supply module in the back of the terminal. I have a print set for the VT52, but it certainly did _me_ no good when my VT78's power supply decided to become a paperweight. -- ----------------+------------------------------------------------------ Roger Ivie | Don't think of it as a 'new' computer, think of it as ivie@cc.usu.edu | 'obsolete-ready' From lasner@sunSITE.unc.edu Sat Jan 28 15:21:07 EST 1995 Article: 1165 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: alt.sys.pdp8 Subject: Re: PDP-8 spare parts/systems. Date: 28 Jan 1995 19:15:59 GMT Organization: University of North Carolina, Chapel Hill Lines: 175 Message-ID: <3ge55f$lke@bigblue.oit.unc.edu> References: <3fnlhr$jn8@starbug.cyberquest.com> <1995Jan23.194547.28781@news.cs.indiana.edu> NNTP-Posting-Host: president.oit.unc.edu In article <1995Jan23.194547.28781@news.cs.indiana.edu>, J Russ wrote: >I bought all the PDP8/LINC8 stuff at the Cooley Laboratory back in 1987. It >was an interesting collection of stuff. One of the machines was an old LINC8 >down in the basement buried away in a back room. Near the machine was a copy >of a radical newspaper from the late 1960's so the machine may have been >there for a long time. The LINC8 was missing the core memory and various odds >& ends but eventually I was able to rebuild the machine and run the SUDSY >diagnostics and demo programs. Another interesting artifact was a box of smoke >damaged A series modules that were out of a LINC8 in Florida that had caught on >fire. I also got a couple of the cloned DECTAPE drives (I can't remember the >name of the manufacturer) but haven't tried firing them up yet. The Cooley Labs LINC-8 collection is the original source of quite a few LINC8-specific things. The unmodified 4-K LINC-8 cannot support running P?S/8 due to the complexity of the LINCtape control. (Similarly, you cannot run OS/8 on an unmodified 8K LINC-8.) By making a few hardware changes, it's possible to run a kludgy P?S/8 and/or (if you have 8K!) OS/8 system with some limitations: 1) Inadequate error recovery (sorely needed on a tape system) 2) Always searches forwards first, even though statistically, 99.99% of the time backwards search is best. (Quite pessimized :-).) (Note: there is the possibility of a P?S/8 handler for an 8K *un-modified* machine and an OS/8 system handler for a 12K unmodified machine. Also, one LINC-8 was definitely used with a DF32 using the standard system handler for the DF32 and two-page non-system handlers for the LINCtapes, etc. An item being sought towards this and related ends is the modifications to the LINC-8 so that you could have DMA devices past 4K. The stock LINC-8 doesn't support this! Yet, apparently DEC came up with ECO's to the machine to support it for at least this very site. (LINCtape isn't DMA, and the buss is taken over for DMA purposes by the LINC CPU, etc. so there isn't available a DMA channel for extended memory, etc.) Anyone with any info about these hardware mod's? I suspect the machine was installed at WVU at one point, etc.) Several individuals at Syracuse University and I participated in an "upgrade" to the Cooley-defined changes. As a requirement of our spec, we incorporate other changes from Cooley that allow DECtape to be read/written on the LINC-8 (as well as NOVA LINCtapes. BTW, NOVA LINCtapes also use the strange drives Jeff mentioned; they were made by Computer Operations in Beltsville, MD.) The DECtape changes are useful, but are somewhat unreliable for the purpose. We expanded on both sets of changes to add additional features: i) Faster access to the registers that mattered, such as a direct skip-on-flag-up set instead of the longer ICON interface that multiplexes several registers with sense bits, etc. ii) One instruction to force tape searching mode directly. iii) Coupled with ii, a means of forcing an initial tape search direction "guess" where it self-clears when a transfer is initiated thus all subsequent transfers are forwards, but the search ls (likely) backwards. The DECtape-oriented changes involve adding a device selector to get three IOT's. We added 4 extra instructions using full decoding to get all of the above improvements. Thus, the net result is compatible with the un-modified machine, the slightly modified machine with or without the DECtape-oriented enhancements, and of course itself. Using all of the improvements, the following have been successfully written: 1) A 4K P?S/8 system handler which I use on my 4K LINC-8. It can handle up to 8 drives (if you implement them! I have heard of machines with drives 4,5 added, but never the other 4 drives!), and can search with an intelligent guess ala the P?S/8 DECtape TC01/08 system handler. Additionally, no compromise is needed regarding DMA devices such as DF32. (Note: when the minimal change version was used, we compromised the handler such that if the DF32 were also present, it would destroy the tape handler! Both P?S/8 and OS/8 share a common rule that had been broken in this case: Always use locations 07750-07755 as temporaries only since these could be DMA locations for TC01/08 or TC58 or DF32/RF08. That older handler stuck code and/or constants into this area and thus was in jeopardy, etc. This newer handler that depends on all of the hardware mods now obeys this rule, etc.) 2) A similarly equipped OS/8 handler for 8K machines. The mods have been applied to both my 4K LINC-8 and also an 8K LINC-8 now residing in my basement. 3) one-page non-system handlers without compromise are now possible for both systems. (Note: even without hardware mods, it's possible to produce a slightly compromised one-page handler, and no problem in two-pages.) 4) Much more reliable DECtape <-> LINCtape transfers. I wrote a block-oriented utility (a fore-runner of the more generic P?S/8 utility BLKCPY) to move DECtape blocks around on the modified LINC-8. The following commands are typical of usage: (Note: the "*" is the prompt char!) .R BLKCPY *DTA1:0-2701>LTA0:0 {transfer an entire DECtape to a LINCtape} *LTA0:0-2701" imply a read then write operation to perform a copy. The direction is "pointed" to by the arrow key used. Either an input or output range must be specified, or alternatively an explicit block count using "=". A command line option exists to specify whether to verify the write with a readback-compare after the write, etc. The LINC-8 version of this program can run on the 4K LINC-8. The LINCtape handlers can be one page if the hardware mods are used, else are two pages. The DECtape handler is 6 pages including lookup tables for the data conversion. (Note: DECtape bit directions are "sideways" compared to LINCtape conventions! Invariably, this means that some sort of lookup table is required.) All P?S/8 non-system handlers are page and field relocatable, and can be up to an entire field long unlke the OS/8 two-page max restriction, etc. As a practical consideration, since the hardware mods are required for the DECtape conversion, the modified LINC-8 LINCtape handler appies as a likely system device to run the program from, etc. (But it could also be something else such as a DF32, RF08, or an IPB such as on my home machine. In that case, the "boot device" is actually a disk or real TC01/08 DECtape on the "other" side of the IPB on my -8/e, etc. This allows both drives of the LINC-8 to be available to the program without changing tapes, etc.) Cooley Labs is also responsible for CPS and CPS-4K. These are systems that run on LINCtape as usually found on the LINC-8. However, the LINC-8 tape hardware is actually a PDP-8 peripheral! (PROGOFOP is a trap simulator that helps to "fool" LINC programs into believing they are running on complete LINC hardware, which is actually partially simulated, especially the tape portion, etc.) The actual mechanism for the tape is to appropriate the LINC CPU registers as tape registers during the time the LINC CPU is shut down (during normal PDP-8 operations, which in LINC mode are only true during interrupts, such as interrupts generated from trapping LINC instructions, etc.). Kurt Metzger and company defined an interface for the Posibus (on an -8/l) that essentially is the LINC-8 register sequence for tape only. Thus, LINc-8 software for the PDP-8 such as the Library System, CPS (-4K), P?S/8 for LINC-8, OS/8 for LINC-8, etc. can run on LINCtapes on PDP-8's. (Of course you cannot run systems for LINC such as LAP6, since these run primarily in LINC mode, although you can write file conversion utilities for these machines, etc.) These "transplanted" drives were actually Computer Operations tape drives with a Cooley-designed interface. Undoubtedly, this is what Jeff has recovered, etc. Note to Jeff: I have sent you e-mail months ago but you never responded regarding archive site "synchronization" etc. If you are reading this, please e-mail me at lasner@sunsite.unc.edu etc. Further note to Jeff: The smoked LINC-8 in Florida: was it at the Miami-based Mt Sinai Medical Center? (They also had a research division, but I don't remember their title, etc.) I have the "other" machine from there, etc. Please contact me about that as well, etc. Sorry to use the a.s.p group for private correspondance, but there doesn't seem any other way (why?) etc. cjl From lasner@sunSITE.unc.edu Sat Jan 28 15:21:29 EST 1995 Article: 1166 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: alt.sys.pdp8 Subject: Re: DEC-Tape Cacacity? Date: 28 Jan 1995 20:18:28 GMT Organization: University of North Carolina, Chapel Hill Lines: 112 Message-ID: <3ge8qk$ob2@bigblue.oit.unc.edu> References: <3feq34$ho8@ringer.cs.utsa.edu> <3ffgro$rhj@usenet.rpi.edu> NNTP-Posting-Host: president.oit.unc.edu In article , Frank R. Borger wrote: >In article <3ffgro$rhj@usenet.rpi.edu> wilsonj@alum01.its.rpi.edu (John Wilson) writes: > >>... Even with stock format the tape would spin >>off the end of the reel while formatting the end zone unless you had only >>threaded it just enough to barely stay on the reel at the beginning. > > That was a commony symptom of a mal-adjsted formatter. Not necessarily. > > The interface used an adjustable oscillator to determine how > OFTEN you wrote formatting data, with the tape moving at constant > reel speed. If the frequency of the oscillator was a little low, > the format marks were too far apart and you ended up running off > the end of the reel. True. But the problem is that DEC mis-specified the tape length so that later when 3M sold *correctly-specified* tapes, they were too short for the intended purpose(s) many of the time, etc. As a bandaid, many people upped the density either with fingers on the hubs or by increasing the mark clock on the various controllers, since the clock speed was quite conservative. Thus, you often could get away with a little marginal reliability (other than on the TD8E!) to get more density and thus fit what you wanted on less tape, etc. > > I remember an article that actually made it into DECUScope that > described how one software specialist re-wrote the formatter to > only do 6' instead of 10' leader/trailer sections. No, that happened earlier. The standard PDP-8 DECtape formatter has two modes of operation: 1) Anything other than MARK384 uses a stated conservative formula found in some early DEC manuals that will calculate out to 2702 blocks octal of 129 words with 10 foot leader/trailers and will completely succeed if the tapes really are a bit longer than the nominal 260 foot claimed (instead of really being about 10-15 feet longer as they usually were at the beginning!). Other selected values will give corresponding variations that use the same amount of tape. As the tapes started being made shorter (some even being *less* than thenominal 260 feet!), more and more tapes couldn't be formatted correctly. Note that this effect is compounded by too-slow mark clock adjustments, and the tendancy of some ignorant lusers who erroneously assumed the program needed "help" so they wound the tape on some number of turns before starting the formatting operation (which is dead wrong - the program writes its own leader. An additional variant: Lusers would cut off the leader because it was "worn out" or something perceived but incorrect. The truth is that the tape format requires nothing of the leader other than to exist; the writing occurs further down the tape. All that could happen is that a tape could move to the inner-most portion of the end-zone, and be turned around back towards the data. During the PDP-8 formatting process, you turn on the mark clock, and do a first pass over the tape. The tape could fall off, but is usually not fatal, although many lusers didn't realize that, etc. You turn off the mark clock, remount the tape if necessary, then enter the next phase (hit a CR to continue). The tape is first rewound about 20-30 feet, then turned around to write forward the pattern "BLOCK 2701" an indefinite number of times (resulting in an anomalous tape so far!) until the end zone is hit at its inner end. At this point, the last-written block *IS* block 2701. Previous block patterns can then be written in descending order from 2700 down to 0000 at which point another end zone error should occur at the tape beginning, hitting the inner end of the front end zone, etc. Thus, as long as the tape can successfully not fall off a second time, the tape is valid, albeit possibly with a short inner end zone, etc. 2) When the specific command MARK384 is given, the formula is bypassed and constants plugged in to create the variant PDP-10 format tape. The end zones are written as 6 feet instead of 10 feet to accomodate getting 1102 blocks of 600 words (600=384 decimal). Were the standard formula used, far less blocks would be written, etc. Thus, from the beginning, the notion of 6 ft end zones was standard. Since the standard PDP-8 format (from the formula) didn't require as much tape, 10 ft end zones were written. The P?S/8 version of the formatter has an option to run such that MARK129 is the command that uses the plug-in values. The number of blocks is taken from the =xxxx user-specified and the end-zone is shorteded to 6 ft. As a result, "extra-long" DECtapes can be created as long as the tapes are truly full-length, etc. A common semi-standard was to achieve 3300 blocks instead of 2702. When some late-production tapes were sold that were too short, the P?S/8 method was the only way to achieve 2702 blocks! The MARK384 command invariably failed. The tape falling off only once became the standard as never falling off ceased to exist, and falling off twice (and flunking the formatter!) become all too common :-(. Some people then purchased tapes directly from 3M. If they stated "DIGL WIND" then they are DECtapes right out of the box. Else they are wound as LINCtapes and need to be reversed, etc. Unfortunately, most of these tapes were accurately 260 ft long, thus guaranteeing no further too-short tape problems, but also preventing any more extra-long tapes either. > >Frank R. Borger - Physicist ___ What's the difference between: >Michael Reese - U of Chicago |___ (1) having a baby and >Center for Radiation Therapy | |_) _ (2) doing an RSX11M sysgen? >net: Frank@rover.uchicago.edu | \|_) >ph: 312-791-8075 fa: 791-2517 |_) Having a baby only takes 9 months! cjl From wilsonj@alum01.its.rpi.edu Mon Jan 30 18:00:11 EST 1995 Article: 1167 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!gatech!udel!news.mathworks.com!solaris.cc.vt.edu!swiss.ans.net!emi.com!pauling.wadsworth.org!rebecca!rpi!wilsonj From: wilsonj@alum01.its.rpi.edu (John Wilson) Newsgroups: comp.os.vms,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: DEC part number for RL02 unit number plugs ? Date: 28 Jan 1995 16:25:14 GMT Organization: Rensselaer Polytechnic Institute, Troy NY Lines: 13 Distribution: usa Message-ID: <3gdr5b$npu@usenet.rpi.edu> References: <3g8vjn$rbu@hustle.rahul.net> <3gbu4u$bui@nntp1.u.washington.edu> NNTP-Posting-Host: alum01.its.rpi.edu Xref: bigblue.oit.unc.edu comp.os.vms:85126 alt.sys.pdp8:1167 In article <3gbu4u$bui@nntp1.u.washington.edu>, Richard Seymour wrote: >hmmm... i just looked it up in a 1993 catalog... the "-00" denotes > unit zero, -01 unit one, etc. but the prices are interesting: > zero and one are $2 each. two and three are $4 each. five and six are $1. The pricing weirdness is probably due to demand or rarity or something. Note that RL controllers support only drives 0-3, the numbers >=4 are for RK06/RK07 drives, which use the same plugs, but are less common. John From m.bischoff@tu-bs.de Mon Jan 30 18:00:46 EST 1995 Article: 1168 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!agate!howland.reston.ans.net!Germany.EU.net!netmbx.de!CNB.CompuNet.DE!zrz.TU-Berlin.DE!math.fu-berlin.de!zib-berlin.de!gs.dfn.de!tubsibr!news.ibr.cs.tu-bs.de!mbi From: mbi@mo.math.nat.tu-bs.de (Michael Bischoff) Newsgroups: alt.sys.pdp8 Subject: page wrap for PDP-8? Date: 25 Jan 1995 13:44:42 GMT Organization: Institut fuer Angewandte Mathematik, TU Braunschweig Lines: 11 Distribution: world Message-ID: Reply-To: m.bischoff@tu-bs.de NNTP-Posting-Host: mo.math.nat.tu-bs.de Hi, to which page does a 'TAD addr' instruction refer, if the opcode is at memory address 128*n+127 (decimal)? Is it the page of the instruction, or the next? Michael -- Michael Bischoff, Abt. Mathematische Optimierung, Pockelsstrasse 14 TU Braunschweig, 38106 Braunschweig, Germany. e-mail: m.bischoff@tu-bs.de Tel: +49-531-391-7555, Fax: +49-531-391-7559 From jones@pyrite.cs.uiowa.edu Mon Jan 30 18:01:21 EST 1995 Article: 1169 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!udel!news.sprintlink.net!howland.reston.ans.net!math.ohio-state.edu!hobbes.physics.uiowa.edu!news.uiowa.edu!news From: jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) Newsgroups: alt.sys.pdp8 Subject: Re: page wrap for PDP-8? Date: 27 Jan 1995 17:50:31 GMT Organization: University of Iowa, Iowa City, IA, USA Lines: 33 Distribution: world Message-ID: <3gbbp7$b02@nexus.uiowa.edu> References: NNTP-Posting-Host: pyrite.cs.uiowa.edu >From article , by mbi@mo.math.nat.tu-bs.de (Michael Bischoff): > to which page does a 'TAD addr' instruction refer, if > the opcode is at memory address 128*n+127 (decimal)? If you examine the flowcharts of the CPU logic for any PDP-8 model, you'll find something like this: Fetch: CPMA = PC PC = PC + 1 MB = MEMORY[CPMA] IR = MB CPMA = MB if MB = 0 then CPMA = 0 -- page zero mode if MB = 1 then begin MB = MEMORY[CPMA] CPMA = MB end There, PC is the program counter, CPMA is the central processor memory address register, MB is the memory buffer, and IR is the instruction register. As a result of the above, current page mode refers to the page from which the instruction was fetched. Doug Jones jones@cs.uiowa.edu From bqt@Krille.Update.UU.SE Mon Jan 30 18:01:37 EST 1995 Article: 1170 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!agate!howland.reston.ans.net!pipex!sunic!columba.udac.uu.se!Krille.Update.UU.SE!Krille.Update.UU.SE!not-for-mail From: bqt@Krille.Update.UU.SE (Johnny Billquist) Newsgroups: alt.sys.pdp8 Subject: Re: page wrap for PDP-8? Date: 28 Jan 1995 12:16:31 +0100 Organization: Update Computer Club Lines: 21 Message-ID: <3gd92t$qcd@Krille.Update.UU.SE> References: NNTP-Posting-Host: krille.update.uu.se In mbi@mo.math.nat.tu-bs.de (Michael Bischoff) writes: >Hi, >to which page does a 'TAD addr' instruction refer, if >the opcode is at memory address 128*n+127 (decimal)? >Is it the page of the instruction, or the next? It refers to an address on the same page as the instruction. I think the PC, as all other auto-incrementing stuff is incremented before it is used. (Well, okay, the only other auto-incrementing stuff I can remember off hand is the index registers.) Johnny -- Johnny Billquist || "I'm on a bus CS student at Uppsala University || on a psychedelic trip email: bqt@minsk.docs.uu.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol From aek@wiretap.spies.com Mon Jan 30 18:02:06 EST 1995 Article: 1171 of alt.sys.pdp8 Newsgroups: alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!godot.cc.duq.edu!hudson.lm.com!netline-fddi.jpl.nasa.gov!wiretap.spies.com!aek From: aek@wiretap.spies.com (Al Kossow) Subject: whereabouts of Gary Coleman Message-ID: Organization: The Internet Wiretap Date: Sun, 29 Jan 1995 00:58:05 GMT Lines: 6 Has anyone heard from Gary Coleman? I thought if there was anyone who might know if Gary Coleman of Allen Bradley in Cleveland was, it would be here. He was a BIG 8 collector in the early 80's, had a pdp12 in his basement. From ivie@cc.usu.edu Mon Jan 30 18:02:46 EST 1995 Article: 1172 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!news.mathworks.com!udel!news.sprintlink.net!cs.utexas.edu!news.cs.utah.edu!cc.usu.edu!ivie From: ivie@cc.usu.edu Newsgroups: alt.sys.pdp8 Subject: Re: page wrap for PDP-8? Message-ID: <1995Jan28.193148.39659@cc.usu.edu> Date: 28 Jan 95 19:31:48 MDT References: <3gd92t$qcd@Krille.Update.UU.SE> Organization: Utah State University Lines: 18 In article <3gd92t$qcd@Krille.Update.UU.SE>, bqt@Krille.Update.UU.SE (Johnny Billquist) writes: > I think the PC, as all other auto-incrementing stuff is incremented > before it is used. If that were the case, you would have to enter the address of the instruction before the instruction you wanted to start into the switches. I thought about that sort of thing because on a PDP-5-like processor, if you extended the autoincrement to include Z 0, you could have immediate constants. However, it wound up as a special case because of loading addresses into the PC through the console. Not that I've done more than think about my PDP-5-in-a-Xilinx mind you (I've estimated it would take about 300 CLBs). -- ----------------+------------------------------------------------------ Roger Ivie | Don't think of it as a 'new' computer, think of it as ivie@cc.usu.edu | 'obsolete-ready' From ard@siva.bris.ac.uk Mon Jan 30 18:03:16 EST 1995 Article: 1173 of alt.sys.pdp8 Newsgroups: comp.os.vms,alt.sys.pdp11,alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!gatech!howland.reston.ans.net!pipex!warwick!bsmail!siva.bris.ac.uk!ard From: ard@siva.bris.ac.uk (PDP11 Hacker .....) Subject: Re: DEC part number for RL02 unit number plugs ? Message-ID: <30JAN199518581687@siva.bris.ac.uk> News-Software: VAX/VMS VNEWS 1.41 Sender: usenet@info.bris.ac.uk (Usenet news owner) Nntp-Posting-Host: siva.bris.ac.uk Organization: University of Bristol Physics Department References: <3g8vjn$rbu@hustle.rahul.net> <3gbu4u$bui@nntp1.u.washington.edu> Distribution: usa Date: Mon, 30 Jan 1995 17:58:00 GMT Lines: 25 Xref: bigblue.oit.unc.edu comp.os.vms:85254 alt.sys.pdp8:1173 In article <3gbu4u$bui@nntp1.u.washington.edu>, seymour@u.washington.edu (Richard Seymour) writes... > it's 12-12691-00 (for unit zero... break pins for other numbers) Not on an RL-drive. There are 2 pins that encode what number it is, and the shape of the top and bottom edges of these pins is what matters. There is no easy way of converting one plug into another. Mind you, it's possible to cut/jumper the switch board in the RL-drive in an emergency >hmmm... i just looked it up in a 1993 catalog... the "-00" denotes > unit zero, -01 unit one, etc. but the prices are interesting: > zero and one are $2 each. two and three are $4 each. five and six are $1. RL's are only 0-3 - there are only 2 drive select lines in the hardware. The same plugs fit RK06's/07's, and also some CDC drives, like the 9448 RK06's/7's have numbers 0-7 > -tony Bristol University takes no responsibility for the views expressed in this posting. They are the personal views of the user concerned. From lasner@sunSITE.unc.edu Mon Jan 30 18:10:12 EST 1995 Article: 1174 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: alt.sys.pdp8 Subject: Re: whereabouts of Gary Coleman Date: 30 Jan 1995 23:10:54 GMT Organization: University of North Carolina, Chapel Hill Lines: 19 Message-ID: <3gjrlu$19fj@bigblue.oit.unc.edu> References: NNTP-Posting-Host: president.oit.unc.edu In article , Al Kossow wrote: >Has anyone heard from Gary Coleman? >I thought if there was anyone who might know if Gary Coleman of >Allen Bradley in Cleveland was, it would be here. He was a BIG >8 collector in the early 80's, had a pdp12 in his basement. > I haven't heard of Gary Coleman in years. However, I and several people I know have far larger collections. cj "What, you only have *one* PDP-12 in your basement?" l ps: Jeff Russ, who still hasn't contacted me, might know his whereabouts. If someone can get Jeff to contact me, it would be appreciated. (Gary too :-).) From ivie@cc.usu.edu Tue Jan 31 08:01:41 EST 1995 Article: 1175 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!news-server.ncren.net!news.duke.edu!agate!dog.ee.lbl.gov!news.cs.utah.edu!cc.usu.edu!ivie From: ivie@cc.usu.edu (Roger Ivie) Newsgroups: alt.sys.pdp8 Subject: DM2 formatter for RD53? Message-ID: <1995Jan29.211618.39752@cc.usu.edu> Date: 29 Jan 95 21:16:18 MDT Organization: Utah State University Lines: 15 Greetings! I suddenly and unexpectedly find myself with a surplus of large MFM hard disks, including a couple of RD53 70 MB drives. This, of course, has me dreaming of stashing them in my DECmate IIs. Does anyone have a formatter for the DM2 capable of doing RD53s that they are willing to send my way, or do I have to write one? Also, once I have them formatted, what do I do with all the space? I know I can suck up 32MB for CP/M (four 8MB partitions). Can OS/278 handle more than two partitions connected at once? -- ----------------+------------------------------------------------------ Roger Ivie | Don't think of it as a 'new' computer, think of it as ivie@cc.usu.edu | 'obsolete-ready' From lasner@sunSITE.unc.edu Tue Jan 31 09:15:56 EST 1995 Article: 1176 of alt.sys.pdp8 Path: bigblue.oit.unc.edu!sunSITE!lasner From: lasner@sunSITE.unc.edu (Charles Lasner) Newsgroups: alt.sys.pdp8 Subject: Re: DM2 formatter for RD53? Date: 31 Jan 1995 14:16:22 GMT Organization: University of North Carolina, Chapel Hill Lines: 168 Message-ID: <3glgnm$17rr@bigblue.oit.unc.edu> References: <1995Jan29.211618.39752@cc.usu.edu> NNTP-Posting-Host: president.oit.unc.edu In article <1995Jan29.211618.39752@cc.usu.edu>, Roger Ivie wrote: >Greetings! > >I suddenly and unexpectedly find myself with a surplus of large MFM hard disks, >including a couple of RD53 70 MB drives. This, of course, has me dreaming of >stashing them in my DECmate IIs. Does anyone have a formatter for the DM2 >capable of doing RD53s that they are willing to send my way, or do I have to >write one? I have the so-called "clandestine" formatter in binary executable form. It's apparently standalone, thus can be wrapped around any form of OS/(27)8 system or could even be transferred to COS or P?S/8 accordingly, etc. It's "distributed" on an OS/278 RX50 without documentation, but all you need to give it (manually on the keyboard) are the # of cylinders and heads. It does (most of) the rest. There are several problems: 1) It doesn't apply the slushware, but that's easily done from the System Test Diskette. (Note: the STD has two functions: one formats the disk and applies slushware automatically, the other *just* applies the slushware. The catch with the STD is that it will *ONLY* format 5,10,20 MB disks to one head shy of their respective sizes, and will do no other sizes at all. These sizes are in turn "sacred" (see below).) 2) It cannot make any "sacred" disk sizes which leads to a problem with Master Menu. Master Menu pragmatically checks for the three sacred sizes of one head shy of 5 or 10 or 20 MB, the three normally used disk sizes on DM II/III+. Anything else is claimed to be "corrupted" when you attempt to use the "X" option to compress the disk. Thus, the use of these disks demands manual volume placement, which means a whole lot of backup/restore, etc. :-(. (My present situation: I use an ST-4096 wasting a head to get exactly 64 MB of formatted space. And a whole lot of RX50's of volume backups, although backing up per se isn't that bad an idea :-).) (Note: obtaining the sources to Master Menu will likely allow fixing this. Older prerelease versions of MM actually didn't mind alternate sizes, but cannot be used due to other internal problems. All are warned not to attempt to use old MM copies to get around this - it will permanently corrupt your entire disk!!!) Other than that, arbitrary disk sizes can be used. The maximum number of heads is 8, even though newer disks allow up to 16. It is rumored that the controller can handle more than the Rainbow limit of 1024 cylinders; it would be nice if an RD54 was attempted to see if you can get to 1224 cylinders (even if wasting all of those other heads!). If that's true, anyone have any leads where to find larger disks by cylinders, but with less heads (Microscience?). 3) There is no facility to lockout bad blocks. Even Master Menu's facility is a non-workable kludge. What's needed is a routine to read test the disk and find and mark bad blocks/replacements. (Better yet, a read/write test used just after the clandestine formatter even before systems are applied, but after the slushware is in place so there even is a bad block table to mark!) I so far have been lucky in that I've found disks with no apparent problems. Perhaps it's because DEC writes lower-density disks than the rest of the industry? (Remember, DEC formats for 16 512 byte sectors/track, not 17 as in PC's, etc.) > >Also, once I have them formatted, what do I do with all the space? I know I >can suck up 32MB for CP/M (four 8MB partitions). Can OS/278 handle more than >two partitions connected at once? There are several problems for each system: 1) MS-DOS for some reason cannot boot a volume larger than about 10.5 MB. However, non-system volumes of up to 32 MB each work fine for drive F:, G:, H:. 2) I cannot confirm how large CP/M-80 volumes can be, but again I would check if the non-system size is different max than the bootable system size, or if there's a problem referencing different sizes from a bootable floppy system. However, there is a specific problem with CP/M-80: You cannot directly create a volume that starts past the first 10-20 MB of disk address. The install diskette gets confused by the extra set high-order bits. Thus, the workaround is to place a CP/M-80 system on a near-empty disk or smaller hard disk system and then use backup/restore to place the now-existent system where you want it. (Note: see how this couples well with the MM limitations :-( :-).) While both CP/M-80 and MS-DOS allow the same configuration of A:, B: RX50 and E:,F:,G:,H: hard disk volumes (booted to E: on the hard disk version), a bug in the CP/M configuration prevents C: and D: from being accessed even if you have the second RX50. (MS-DOS does it correctly.) Only up-to-8 all-floppy CP/M systems can access the second drive pair. A side note for CP/M-80 users: The install diskette is for any APU or XPU configuration, yet the systems it creates are for APU or XPU only, and can never be "crossed". This even applies to bootable volumes! Additionally, the installation of DM II-oriented CP/M-80 must be performed on the DM III+ by first modifying the machine to have two diskette drives. Afterwards, all will work except the HD utility's backup/restore feature since that only addresses the B: drive. (Unless you make the second drive permanent!) 3) COS volumes can be created by Master Menu, but are created corrupted. Apparently there are COS commands to correct this after the fact. However, COS is distributed with a second diskette that contains a "stub" version of Master Menu that redundantly includes a facility to delete and add volumes in COS format. Volumes created this way are acceptable to COS directly. I haven't tried any of this on a non-standard disk size though (used a 10 MB machine standardly formatted at the time). The diskette itself is actually a bootable pre-release OS/278 diskette that auto-executes the COS-oriented stub derivative from MM. I would not trust either the overall worthiness of the underlying OS/278 system compared to the released system, nor would I attempt to move the executable to OS/278 V2. (This inadvertant "lock-in" is characteristic of a lot of early install software; apparently these programs were written for the "OS/278 system of the week" back then, etc. Once OS/278 stabilized as more like OS/8 (as it should be!) these programs with their dependent kludges became as orphaned as the systems they were released with, etc.) 4) OS/278 as distributed only specifically addresses RD50: and RD51:. While CP/M and MS-DOS are limited to 4 logical volumes, WPS and OS/278 are able to access all 8 logical volumes, at least in theory. OS/278 needs to be reconfigured with more handlers using the SET HANDLER utility. (Note: BUILD is not compatible with OS/278 specifically, but this can eventually be overcome!) Distributed with OS/278 is the handler for RD52: and RD53:. This can be applied directly as a replacement for the RX02 handlers (RX20: and RX21: for example, since hard disks and RX01/02 never meet :-).) The designated way to create additional OS/278 volumes is through Master Menu directly. The "A" command will allocate any of the other systems as advertised on the screen. However, you can also indicate "OS278" as an undocumented choice! There are a few problems acquiring access to the other four volumes: The OS/278 MOUNT program as distributed is limited to accessing logical drives 1-3 only (0-3 from diskette-based systems; the improvement makes sense; you don't want to dismount the system boot drive from a hard disk system!). However, Master Menu allows you to pre-mount all 8 volumes (btw, this is required for WPS normally), thus this would at least allow a half-fixed and half remountable system, etc. However, as distributed, there are no handlers for RD54: through RD57:. >From the sources, it should be fairly easy to modify for the extended logical volumes. I haven't ever bothered merely because I just use a lot of MOUNT commands on the four available logical volumes. This allows me to designate RD53: as the WPSVOL1 normally associated with WPS documents, and still have a system volume and two additional data volumes. WPFLOP can convert document files in either direction to/from OS/278, thus allowing text files to go between OS/278 and WPS format. Since both CP/M-80 and MS-DOS allow local conversions to/from those systems and WPS, this allows WPS format used as an intermediary to be the means of file conversion between all of these systems, etc. without resorting to floppies. There are higher priority issues to be dealt with in improving OS/278, so we'll all have to "suffer along" with "only" four volumes for now :-). >-- >----------------+------------------------------------------------------ >Roger Ivie | Don't think of it as a 'new' computer, think of it as >ivie@cc.usu.edu | 'obsolete-ready' cjl From jlothian@castle.ed.ac.uk Tue Jan 31 23:11:17 EST 1995 Article: 1177 of alt.sys.pdp8 Newsgroups: comp.os.vms,alt.sys.pdp11,alt.sys.pdp8 Path: bigblue.oit.unc.edu!concert!gatech!swrinde!howland.reston.ans.net!pipex!exnet2!exnet!dcs.ed.ac.uk!festival!jlothian From: jlothian@castle.ed.ac.uk (J Lothian) Subject: Re: DEC part number for RL02 unit number plugs ? X-Newsreader: TIN [version 1.1 PL8] References: <30JAN199518581687@siva.bris.ac.uk> Message-ID: Sender: news@festival.ed.ac.uk (remote news read deamon) Followup-To: comp.os.vms,alt.sys.pdp11,alt.sys.pdp8 Organization: Edinburgh University Date: Tue, 31 Jan 1995 16:43:47 GMT Lines: 20 Xref: bigblue.oit.unc.edu comp.os.vms:85288 alt.sys.pdp8:1177 PDP11 Hacker ..... (ard@siva.bris.ac.uk) wrote: : In article <3gbu4u$bui@nntp1.u.washington.edu>, seymour@u.washington.edu (Richard Seymour) writes... : > it's 12-12691-00 (for unit zero... break pins for other numbers) : Not on an RL-drive. There are 2 pins that encode what number it is, and the : shape of the top and bottom edges of these pins is what matters. There is no : easy way of converting one plug into another. A nail-file, he said darkly, works wonders... James -- ------------------------------------------------------- James Lothian | "It's life, Jim, james@uk.ac.ed.caad | but not as we know it" ------------------------------------------------------- These opinions have nothing to do with Edinburgh University.