Correct. And a number of DEC operating systems had card reader support. I haven?t seen them very often on DEC machines, but for example RSTS can handle at least 026 and 029 card codes, and some vague memory says there may have been a ?1401? table as well. Not well known, but there are card codes for lower case, and in fact IBM operating systems could handle that though it wasn?t commonly seen. And in fact there are card codes in the EBCDIC code table for all 256 possible 8-bit values; you?d find those in binary card decks. Again, not all that common, but 360 diagnostics would sometimes show up as binary card decks (I remember DX11 diagnostics where the IBM end was delivered in that form). And the 360 model 44 ?emulator IPL? deck (for the string instructions, which were emulated in software similar to how low end VAXes did their instruction subset) was a binary deck.

On 05/10/2015 07:58 AM, Paul Koning wrote: One *could* get an 029 with nearly any *character set* implemented. For example, CDC Display Code 029s were all over place at CDC installations. Many older machines did perform translation of punch codes into character codes in hardware (eg. the 1401, 1620, etc.) but by the mid 60s, the usual procedure was to read the card in column or row-binary and perform the translation with software. (eg. 1130 and later) --Chuck

Indeed, but I think that even in that application card readers were not all that common. The batch card model was more an IBM thing, with DEC systems tending more to terminal input. paul

On 2015-05-10 20:52, William Donzelli wrote: I don't know how common card readers were, but I don't think any DEC machines could punch cards. You normally had free standing terminals on which cards were punched, and then the cards were read in later. RSX have three types of queues, printers, batch, and card spoolers. Card spoolers were purely for punched cards, so there obviously was at some point a far amount of those around for describing jobs to be performed. I was contacted by someone (I can't remember who right now) not so long ago, about how the input card queues worked, since he couldn't get it to work on the latest version of RSX. And it turns out that that software have bit rotted. Other parts of RSX have been updated and changed to the point where the batch input card queue no longer works properly. I guess noone have tested that one in a long time. I don't remember exactly what the problem was, but I think it was either that the format of messages for the batch manager itself have changed (happened in 1993, when there was an year wrap issue coming up), or else something about named directories. Johnny

On Sun, May 10, 2015 at 1:06 PM, Johnny Billquist <bqt at update.uu.se> wrote: Almost all DEC machines up through early VAX supported card punch options. They just weren't very common other than on PDP-10. PDP-8: CP08 PDP-10: CP10 (multiple variants) and CP20 PDP-11 and VAX: CP11 PDP-15: CP15 (also earlier 18-bit machines, and maybe PDP-5, but I don't have definitive info on option numbers at hand)

Do you have any descriptions of that? I have not seen any mention of a CP11 in any peripherals handbook, nor in any OS. Was that a CSS product? Who built the punch engine? What sort of interface did it use? paul

On 5/12/2015 9:01 AM, Dave G4UGM wrote: Microdata offered a custom wiring job to punch cards with a Univac keypunch, instead of IBM. It used a standard board they had with a clocked parallel output to punch cards. Carriage control would feed cards w/o the need to send all columns to trigger card feed. There was a remote switch on the Univac Keypunch keyboard that allowed the unit to be switched to punch via the 1600. Output could be done with the same machine code as a printer driver. Interface was ascii, and logic in the keypunch custom board (wirewrapped) translated the codes. Jim

On Tue, May 12, 2015 at 9:20 AM, Paul Koning <paulkoning at comcast.net> wrote: The CP11 might have been CSS. I wouldn't have thought so since the one VAX site I knew that had one wouldn't likely have ordered CSS gear. Possibly at one time they had enough need for punching cards to justify buying CSS, but by the time I encountered it the punch was always powered down with a dust cover over it. AFAIK, DEC didn't build any card punches; they are all OEM'd with DEC-built interfaces. For the PDP-10 they OEM'd several different models, possibly from multiple vendors. I don't know the details of any of them, even though I used two different ones back in the day. IIRC, the CP20 was a Unibus device, and might or might not have been program-compatible with the CP11.