• Please review our updated Terms and Rules here

Searching for IBM S/23 Datamaster users

By necessity, more than anything, I have found the light pen test circuit.
8275 light pen circuit.png
Test of the light pen occurs at test 05 during IPL.
 
Today I have been able to trace TRAP signal (or at least most of it) and found it is triggered, with a few exceptions, by SOD. Therefore SIM instructions in the code have the responsibility to trigger this interrupt programatically.
 
Today, I have successfully had a clean IPL in the emulator!


There is still a lot of work to do, but with a bit of luck we will be able to preserve this system, even if its components fail.
 
Good to see the IBM System 23 emulation under emulation with the work continuing.
..
But as far as my self until some one gets the data for some of its external peripherals archived I will likely not be posting much here for now.
..
IBM 5247 Network Hard disk drive Photos (or fotos) of the Insides and the supporting data for its internal hardware. Ya the setup manual for this unit can be found on line, but no schematic for the network card that is required for its use has been donated online as of yet. No body has scanned and donated a copy of its internal hard drive service manual part number Sy31-9002on line either.
This unit can use the 21ED A3 (60MByte) drive.
The Data Storage Attachment Sy31-9001 could also be required if the 5247 has that card version and support hardware installed.
..
So until some one post this information on line and it gets to this forum I will likely not be posting much on this system (with my current work load) .
..
Thnk's
N. Brown
 
Good morning!

Yesterday I completed work on the keyboard (sort of). And the emulator now is capable to have its BASIC functional!


I have also made a pull request so this emulator is included in the official repo of MAME.

By the way, work on a clone has begun.

Have a nice day!
 
Brilliant work! Congrats on getting to this point, and a round of applause for all the effort you've been putting into it.

Given that BASIC can run now, and the display seems to be working as it should... would you say it's capable of demystifying the question I had brought up in that other thread, on how those "non-ROM" characters are generated?

Either way, looking forward to giving it a go!
 
Thank you very much!

Unfortunately I still can't do that. The display doesn't work completely as expected - I still haven't found the banking mechanism for the character generator ROM. While I don't find the register I won't be able to investigate that matter, but I have the feeling that the solution is near.

Again, thank you.
 
jaume-anton said:
The images we have used are fron the link above and the dir 1 command only shows the label of the disk and nothing more
Click to expand...
So I too am finding that doing a "dir 1" on Diag1 or Diag2 just returns the VOLID of the floppy (which is the IBM part number for Diag1 and Diag2) and nothing else. But doing a "dir 1" on Learn, or a number of other floppies I have will correctly show a directory listing. So I figured my Diag1 and Diag2, that I received when I bought my Datamaster, were perhaps made from the same images that people feel are suspect. But on a whim I thought I'd try to save a file to the Diag1 disk then see if I can get a directory listing. When I attempted the save operation I got back an action code/error code that according to the system messages binder means the floppy is "protected". Same for Diag2. So I'm now wondering if perhaps the abbreviated directory listing for Diag1 and Diag2 is actually correct, given they are "protected" floppies. Does one of you have any original floppies that can test that theory (do a dir 1 and see if it shows any files)? Also, can one of you with a Diag2 created from the repository images just try the "PROC START" with it in the floppy drive and see if it runs the proc procedure from the floppy? I'd do it, but I have since found that the spindle hub on my floppy was really bad and chewed up my Diag1 and Diag2 floppies, so until I can get some new ones I can't test this theory.
 
The Diag disks are very different from the other Datamaster disks. The Datamaster disk as far as I have seen always uses the standard interchange IBM format. In the case of the diagnostics floppy it just declares the files as referred in that standard but in the case of the other floppies it contains two files, SYSLABEL and SYSDATA, which contains the file system of the System/23. So, when doing a DIR on the disk it seems to ignore the files specified by track 0 of the interchange format and list exclusively the data on SYSLABEL.

So far I have found that records are 512 bytes (a sector), that offset +2 contains the type of register, offset +3 contains length of the file identifier and offset +4 contains a 17-bit EBCDIC string which is the foresaid identifier (in case of register type f1h).
 

Attachments

More progress!
I found how it marks the deleted files! When a file is deleted, the length of the file identifier at offset +3 is set to 0.

I have also been able to extract a simple BASIC program manually, I still need to identify data from SYSLABEL to do it automatically.
 

Attachments

RetroAND said:
The Diag disks are very different from the other Datamaster disks. The Datamaster disk as far as I have seen always uses the standard interchange IBM format. In the case of the diagnostics floppy it just declares the files as referred in that standard but in the case of the other floppies it contains two files, SYSLABEL and SYSDATA, which contains the file system of the System/23. So, when doing a DIR on the disk it seems to ignore the files specified by track 0 of the interchange format and list exclusively the data on SYSLABEL.
Click to expand...
OK, so doing a "dir" on Diag1 or Diag2 had having no files listed is actually correct behavior? So everybody who got their Diag1 and Diag2 built from the image who thinks it's bad because they do a "dir" and not see files are actually good. If you use Diag2 (built from image) and do a "link prepare" or "proc start" does it run the program off Diag2 correctly?
 
My father found that to execute diagnostics disks you need to pres a combination of letters instead of the normal procedure. I will ask him.
 
It's the PREPARE (for formatting floppies) and START (startup sequence, date, etc) that are on Diag2, a.k.a., VOL002, that I'm interested in. When the system boots it gives you the "PROC START" prompt, wanting you to press Enter in order to run the START procedure off of Diag2 floppy. And for PREPARE, instructions say to insert Diag2 (VOL002) and type "LINK PREPARE".

I know what you're saying about the key strokes for diagnostics, but that's now what I'm interested in. Thanks!
 
Yesterday evening I was able to extract programmatically four BASIC source files using the information from SYSLABEL to access SYSDATA. I am having trouble with non-BASIC files, however. Still, it is better than nothing and I imagine I will be able to extract those files too very soon.
 

Attachments

I have wanted one of those machines for a long time. I worked for IBM (as an intern) and before I got involved with it, I was swept over to the IBM PC project. It was the first desktop computer from IBM with the beautiful green phosphor like the 5251 terminals. I also recall that it had word processor capabilities possibly built-in - I don't think you had to load them from disk, but I could be wrong. I was moved to the top secret PC project over in Del Ray Beach. I'm blown away that RetroAND can reverse engineer the electronics. Now if I ever find one in reasonable condition not too far from me (mid-Atlantic), I might consider grabbing it.
 
gkaplan said:
I have wanted one of those machines for a long time. I worked for IBM (as an intern) and before I got involved with it, I was swept over to the IBM PC project. It was the first desktop computer from IBM with the beautiful green phosphor like the 5251 terminals. I also recall that it had word processor capabilities possibly built-in - I don't think you had to load them from disk, but I could be wrong. I was moved to the top secret PC project over in Del Ray Beach. I'm blown away that RetroAND can reverse engineer the electronics. Now if I ever find one in reasonable condition not too far from me (mid-Atlantic), I might consider grabbing it.
Click to expand...
To be fair, I have only identified a couple of components myself. The real heroes here are an unknown number of persons that made a series of cross-references list. This is basically what enabled me to do my part. What really intrigues me is that I am only a newbie in electronics and until I started with the Datamaster, I wasn't that into IBM stuff. What I mean is that I am really surprised that having people around with more electronics knowledge than me and also into IBM stuff that they didn't do this before me. I am glad, however, that we seem to be reversing the situation with the Datamasters and maybe we can avoid them a rotting death.

If you find one, contact me. I am trying to build a small community of these machines. But get advised, they are pretty cryptic!
 
I don't know if anyone was aware that IBM made a special machine, based on the Datamaster, just for State Farm insurance. It included a built-in printer. Both the Datamaster and the built-in printer were developed at Boca Raton. We had one in the printer lab, but I never had to work with it.
 
I am repairing @Lulu 's machine, and so far I got some success. From doing nothing, the machine has started to display the diagnostics codes. Now I am stuck at test 5, which deals with the Intel 8275 interrupt first and the light pen later. His original CRTC was dead and has been replaced with a stock 8275. According to the datasheet of such component, enabling interrupts before initializing the component is a legal move, and that's how the Datamaster operates. However, with the new 8275 I can't get any interrupt out and it seems like I have to initialize the CRTC first. How could I check this? Is there any solution other than check various 8275 or modifying the code?

Maybe the long term solution would be to reprogram ROM 02 by disabling the light pen test code and making a proper initialization of the component before enabling interrupts. That way the 8276 could also be used with this computer as a replacement of dead CRTCs, don't you think?
 
You must log in or register to reply here.
Back
Top