Searching for IBM S/23 Datamaster users

[N. Brown]

It is interesting that you could only get that disk to work. I did make some disks for someone a few months back and that was the only disk that they could get working as well. Maybe there are just a bunch of bad disk images out there.

Ya, there is a lot of bad disk images out there.
There are a lot of people who did not do the diskette images properly in the first place to begin with.
When it comes to Hard sector diskettes a lot of the disk imaging software is buggy, even knowing the authors clams elsewise.

 

[N. Brown]

I don't now much about these, but I did make someone system disks not too long ago from this repository. He reported that the all the disks worked great.

Likely you used the right hardware and software configuration to archive those diskettes, unlike some other System 23 images that are available online now days.

 

[N. Brown]

Ya, Their is a lot of bad disk Images out there, mostly done on equipment with the wrong software and hard ward configuration
to create the diskette archived image file. At work we get a lot of HardSector Diskette Images that wasn't done right in the first place to deal with.

 

[N. Brown]

As far as making diskette images with an IBM PC or older Mac computer, a lot just seam to download some software and do a bad interface cable (that has missing signal lines that most 8-inch drives need). They don't even put in any details in the disk imaging software description box/entry for the diskette they are about to archive. To lazy to even enter the original diskette type, Number of sectors (if hard sectored), Believed Sector Size. Or ever the original Diskette Label into that entry box/window, in some cases.
..
And they try to do a diskette format that there current hardware and software configuration doesn't really support.
..
So its more bad disk images we have to deal with.

 

[jaume-anton]

The only good images that @RetroAND and me are looking for are the 2 diagnostics disks. 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. If anyone knows where can we obtain them, please give us the site to download. Thank you very much.

 

[ardent-blue]

Yes, this computer is shrouded in secrecy... Model (5322/5324)

Welcome to IBM... When documenting IBM PS/2s, I foundt the Announcement Letters are a good start. Warning! You will probably end up having to read each announcement, looking for references to other Announcement Letters. The Datamaster might have been called something other than DM.

IBM Announcements

Announcement Letter Number ZG82-0243 .
IBM SYSTEM/23 5324 COMPUTER ANNOUNCED [EMEA]

Announcement Letter Number 385-021
SELECTED IBM MACHINES WITHDRAWN FROM MARKETING; PRODUCTION STATUS CHANGE FOR IBM 5324 AND IBM 5247 FROM NEW TO LIMITED NEW; DATAMASTER CUSTOMER ASSISTANCE CENTER AND INDUSTRY SUPPORT CENTER
PRODUCT SUPPORT DISCONTINUED

o IBM 5324 Computer (Datamaster)
o IBM 5247 Disk Storage Unit
o IBM 5217 Model C02 Printer
o IBM 5222 Printer All Models
o IBM 5241 Printer All Models
o IBM 5242 Printer All Models

Announcement Letter Number 283-082
IBM SYSTEM/38 BASIC LICENSED PROGRAM ANNOUNCED

IBM System/23 Datamaster MANAGEMENT SUMMARY .

 

[RetroAND]

Good evening,

By helping other users I identified causes of failure of the ROS memories where the memory is in fact fully functional:

When the memory is marked as "Not found":

1. Check continuity between pin 12 (ground) of that memory and pin 12 of another memory that is found - no matter if it is correct or not, just found.
2. Check continuity between pin 24 (5v) of that memory and the same pin of another detected memory - again, it does not matter if the second is failing.
3. Check pin 20 against the decoders that are near (does not apply for ROS 02h and 09h). They can be identified because they are DIP-14 and have jumpers near their pins. Reference is 74LS138, if there isn't continuity between any of the output pins of the 138s some trace must be broken. By restablishing that trace the memory will be detected on the next test.

When the memory is marked as "Failing":

1. If all memories are failing, a short circuit between multiple address/data bus traces may be involved. As all ROS memories are connected on the same address and data busses so it is a single point of failure.
2. One or more broken traces on the address bus could cause one or many memories to be resolved as failing as floating inputs on the ROM could cause non-deterministic outputs.
3. One or more broken traces on the data bus could cause one or more ROMs to have missing bits - buffers between them and the 8085 may set them to high state and cause an incorrect lecture.

I hope this is of some help.

Regards,
Jaume

 

[RetroAND]

Hello,

Yesterday I made some advances:

1. Found what component the CPU uses to generate /IORD, /IOWR, /MRD and /MWR. From there it is only a matter of time that by following /IOWR and /IORD all ports become known.
2. By studying the code, I found the CRT controller initialization routine and with it its ports: 0x45 for commands and 0x44 for attributes. This means I could initialize the controller, but without the DMA still couldn't fetch the video data.
3. Not only found the basic commands strings in rom 02h, but also found every single rom contains its P/N within the code and a notice of copyright. Mine has notices from 1980 and 1982. The code starts after those strings and there is apparently no fixed address for them to be.
4. All strings are in EBCDIC, of course.
5. IBM used lots of illegal instructions in their code, I imagine both to have functionality and to bother any disassembly attempt.

On the development side I have three designs to deal with ROM and RAM failure. After they are fully revised, prototypes will be ordered and if they work fine this kind of Datamaster failures will be a thing of the past.

I have been able to see some 32KB DRAM boards and they seem to contain the same integrated circuits as my 64KB one. Just the half of them and no buffers. If I could document them I would be able to kill once and for all the myth of redundancy.

By the way, I imagine some may have already been asking themselves how to properly have this computer in a bench... actually to make it possible, an ATX PSU can be easily rewired to the 12 pin connector of the Datamaster board. For the video, a composite adapter is the way to go:

It can be a bit tricky to find a screen that supports it but at the end it is possible.



I am not sure at which frequencies does the Datamaster generate its video, but I am starting to suspect we could have an early version of MDA here. Making an adapter would be trivially easy knowing most of the pinout (let's be honest - it would be easier than what I have done here).

Every time I watch the 5150 and its original expansions the more I see the System/23. Some things were dropped such as the video controller, the DMA was overhauled, the CPU was replaced and the bank registers were dropped because the CPU handled it its own way. Maintained the same 12 pin connector in the prototype, the TMS4132 in the 64KB expansions and the MK36000 for the BIOS. It may be just more than an influence to the PC but the base for their first prototype where they would go replacing or removing subsystems. That would also explain how they could get it in a year instead of three.

Regards,
Jaume

 

[NeXT]

Yes, I've got a System/23 along with a complete documentation, software and training library... See below:
View attachment 1253396View attachment 1253395

At some point in the near future, I'll get started on archiving the documentation, the many software disks and the training cassettes.

-Chris

Just an update to this. This machine is now in my hands. The floppies were handed off last weekend to someone for imaging so those should appear at some point in the future and I'm working to lug the Fujitsu ADF scanner into the house to go through that mountain of manuals. Al already has the yellow binders done so I have to tackle the rest (MAPs are in the black binders), plus unbind and scan the smaller booklets. If there are photographs you would like, time permitting on the next few weeks I will be going through the system with my DSLR. I also did not verify if it was having any configuration issues but when I did the smoke test it DID come up to prompting for a disk, so I have good enough ROMs and RAM.

 

[N. Brown]

Just an update to this. This machine is now in my hands. The floppies were handed off last weekend to someone for imaging so those should appear at some point in the future and I'm working to lug the Fujitsu ADF scanner into the house to go through that mountain of manuals. Al already has the yellow binders done so I have to tackle the rest (MAPs are in the black binders), plus unbind and scan the smaller booklets. If there are photographs you would like, time permitting on the next few weeks I will be going through the system with my DSLR. I also did not verify if it was having any configuration issues but when I did the smoke test it DID come up to prompting for a disk, so I have good enough ROMs and RAM.

It will be cool once some of the IBM system 23 missing manuals are scanned. As anybody how is working with a systems by now knows the archive at Bitsavers and Archive.org are incomplete. There is a lot of missing data for this system and its good to see other people bring the machine back to life.

 

[N. Brown]

Hello,

Yesterday I made some advances:

1. Found what component the CPU uses to generate /IORD, /IOWR, /MRD and /MWR. From there it is only a matter of time that by following /IOWR and /IORD all ports become known.
2. By studying the code, I found the CRT controller initialization routine and with it its ports: 0x45 for commands and 0x44 for attributes. This means I could initialize the controller, but without the DMA still couldn't fetch the video data.
3. Not only found the basic commands strings in rom 02h, but also found every single rom contains its P/N within the code and a notice of copyright. Mine has notices from 1980 and 1982. The code starts after those strings and there is apparently no fixed address for them to be.
4. All strings are in EBCDIC, of course.
5. IBM used lots of illegal instructions in their code, I imagine both to have functionality and to bother any disassembly attempt.

On the development side I have three designs to deal with ROM and RAM failure. After they are fully revised, prototypes will be ordered and if they work fine this kind of Datamaster failures will be a thing of the past.

I have been able to see some 32KB DRAM boards and they seem to contain the same integrated circuits as my 64KB one. Just the half of them and no buffers. If I could document them I would be able to kill once and for all the myth of redundancy.

By the way, I imagine some may have already been asking themselves how to properly have this computer in a bench... actually to make it possible, an ATX PSU can be easily rewired to the 12 pin connector of the Datamaster board. For the video, a composite adapter is the way to go:

It can be a bit tricky to find a screen that supports it but at the end it is possible.

View attachment 1276544

I am not sure at which frequencies does the Datamaster generate its video, but I am starting to suspect we could have an early version of MDA here. Making an adapter would be trivially easy knowing most of the pinout (let's be honest - it would be easier than what I have done here).

Every time I watch the 5150 and its original expansions the more I see the System/23. Some things were dropped such as the video controller, the DMA was overhauled, the CPU was replaced and the bank registers were dropped because the CPU handled it its own way. Maintained the same 12 pin connector in the prototype, the TMS4132 in the 64KB expansions and the MK36000 for the BIOS. It may be just more than an influence to the PC but the base for their first prototype where they would go replacing or removing subsystems. That would also explain how they could get it in a year instead of three.

Regards,
Jaume

If you have a frequency counter you could check the systems Hor and Ver signals frequencies and yes they are very close to the IBM MDA standard.
In fact we use to adapt other IBM monitors for this system, Mostly from systems like IBM displaywriter (6580) and or other similar grayscale monitors .
Some times you must manually tweak the monitors Hor and Ver Hold signal pot controls due to the different frame rates for their Hor and Ver Signals.
If this helps any ?

 

[N. Brown]

Hello,

Yesterday I made some advances:

1. Found what component the CPU uses to generate /IORD, /IOWR, /MRD and /MWR. From there it is only a matter of time that by following /IOWR and /IORD all ports become known.
2. By studying the code, I found the CRT controller initialization routine and with it its ports: 0x45 for commands and 0x44 for attributes. This means I could initialize the controller, but without the DMA still couldn't fetch the video data.
3. Not only found the basic commands strings in rom 02h, but also found every single rom contains its P/N within the code and a notice of copyright. Mine has notices from 1980 and 1982. The code starts after those strings and there is apparently no fixed address for them to be.
4. All strings are in EBCDIC, of course.
5. IBM used lots of illegal instructions in their code, I imagine both to have functionality and to bother any disassembly attempt.

On the development side I have three designs to deal with ROM and RAM failure. After they are fully revised, prototypes will be ordered and if they work fine this kind of Datamaster failures will be a thing of the past.

I have been able to see some 32KB DRAM boards and they seem to contain the same integrated circuits as my 64KB one. Just the half of them and no buffers. If I could document them I would be able to kill once and for all the myth of redundancy.

By the way, I imagine some may have already been asking themselves how to properly have this computer in a bench... actually to make it possible, an ATX PSU can be easily rewired to the 12 pin connector of the Datamaster board. For the video, a composite adapter is the way to go:

It can be a bit tricky to find a screen that supports it but at the end it is possible.

View attachment 1276544

I am not sure at which frequencies does the Datamaster generate its video, but I am starting to suspect we could have an early version of MDA here. Making an adapter would be trivially easy knowing most of the pinout (let's be honest - it would be easier than what I have done here).

Every time I watch the 5150 and its original expansions the more I see the System/23. Some things were dropped such as the video controller, the DMA was overhauled, the CPU was replaced and the bank registers were dropped because the CPU handled it its own way. Maintained the same 12 pin connector in the prototype, the TMS4132 in the 64KB expansions and the MK36000 for the BIOS. It may be just more than an influence to the PC but the base for their first prototype where they would go replacing or removing subsystems. That would also explain how they could get it in a year instead of three.

Regards,
Jaume

Cool, anyways about your number 5
IBM used lots of illegal instructions in their code both to have functionality and to bother any disassembly attempt. More or less this is true.
The system 23 uses a area in memory to emulate the systems 34 processors registers and other hardware, the rest is handled by a Custom preprogramed MCU on certain devices I/O card. The system is likely also paging ROM and RAM memory banks depending on the system 34 processor instruction under execution. They are doing invalid instruction redirection to a overlay in memory that has a routine to emulated the required function.
..
Yes every single rom contains its P/N within the code and a notice of copyright and normally some free space. I have commented about this in another message. With the Datamaster (system 23) you can key software so it will only run on selected systems depend what is in the ROS Coding. This was done with some system software from the Fed. Sys. Div. for these systems. (Aka. Mil. like Applications).

 

[Chuck(G)]

Heckuva setup for an 8085 system.

 

[RetroAND]

Cool, anyways about your number 5
IBM used lots of illegal instructions in their code both to have functionality and to bother any disassembly attempt. More or less this is true.
The system 23 uses a area in memory to emulate the systems 34 processors registers and other hardware, the rest is handled by a Custom preprogramed MCU on certain devices I/O card. The system is likely also paging ROM and RAM memory banks depending on the system 34 processor instruction under execution. They are doing invalid instruction redirection to a overlay in memory that has a routine to emulated the required function.
..
Yes every single rom contains its P/N within the code and a notice of copyright and normally some free space. I have commented about this in another message. With the Datamaster (system 23) you can key software so it will only run on selected systems depend what is in the ROS Coding. This was done with some system software from the Fed. Sys. Div. for these systems. (Aka. Mil. like Applications).

As always, your posts are both welcome and also very interesting.
I guess then that I should spend some time researching the System/34 in order to know its architecture. It's pretty amazing they emulated the whole S/34 there.

I have a question maybe you already know the answer: The way the diagnostics are done seems more suitable to a System/36, System/34, etc. with an indicator panel than just the bare screen diagnostics like it is actually. Knowing there is a connector for diagnostics latches, do you think the construction of such part would be possible? I think they would be useful in case the necessary tests prior to enabling the screen fail.

Just an update to this. This machine is now in my hands. The floppies were handed off last weekend to someone for imaging so those should appear at some point in the future and I'm working to lug the Fujitsu ADF scanner into the house to go through that mountain of manuals. Al already has the yellow binders done so I have to tackle the rest (MAPs are in the black binders), plus unbind and scan the smaller booklets. If there are photographs you would like, time permitting on the next few weeks I will be going through the system with my DSLR. I also did not verify if it was having any configuration issues but when I did the smoke test it DID come up to prompting for a disk, so I have good enough ROMs and RAM.

Congratulations! And thank you very much for your commitment. Every single bit is welcome... and to be fair I am pretty eager to take a look to those manuals.
It seems after all this will be a very good year for this system.

Thank you very much,
Jaume

 

[RetroAND]

Today the unknown component under reference 4481775 has been successfully identified as an Intel 8202; this leaves only two integrated circuits not present on any list as still to be identified. Some progress has been done near the CPU to find the port decoding logic.

 

[N. Brown]

To be real they only emulated enough of the systems 34 to fully support its basic and programs compiled by the systems 34 Assembly language compiler (converted to system 23 level code).
This Including any systems call back to the hardware level from basic.
..
Trying to find the standalone RPG II programming language disk for the Sys23 is impossible today. The remote sys 23 RPG II generator for the System 34 is also impossible to get now days.
The compiler for the System 23 to run System 34 ASM is impossible to find a copy now days, the same is true for the system 23 Basic Assembler and others as well.
..
As far as learning the System 34 architecture, the data available on that is also limited. It is only really needed if you plan to run System 34 applications with out instruction code conversion
and or without the recompiling of those applications. This conversion of course can be done in firmware on a system 23, but IBM only originally did the enough to get the system 34 basic
and its related extended processor function calls to work. IBM did reserve address space up to 1M on the system 23 but most of it is un implemented on models below the 500 numbers.
..
If you plan to add the system 34 multiuser runtime unit, expand the memory over 1M byte via bank switched, Use system 34 peripherals. Then yes the knowledge of the system 34 architecture
is needed to do so. Some 3Rd parity did make memory boards for the system 34 that exceeded IBM systems memory limitations, but that is another subjected to be noted on.
..
A diagnostics indicator panel for the system 23.
To do this simply I would use a Modified ISA-Bus Diagnostic unit
(One made for the 8086 that has the extra extended function connector would be the best to use). This would be the simplest way to go about it.
In theory some one could design a new diagnostic unit that used the System 23 diagnostics connector, but the time and resources to do so is another question.
Yes, the way the diagnostics are done seems more suitable to a System/36, System/34 so the service technical could attach the correct cables into a existing diagnostic unit from one of those systems..
..
There was a reason why the System 23 launch date was delayed by a year.

 

[N. Brown]

Right on, some one else has noted IBM semiconductor device 4481775 as identified to an Intel 8202.
Hopefully some others will list IBM 4481102 and 4178626 chips.
..
The 4481102 chip in the Word processor video extension/expansion card has a different version number, and yes there is a difference then the one on system 23 motherboard.
So some versions of this chip when removed from the slave Video card and used as a replacement to the one on the motherboard don't work properly. (So I note-it in this message)
..
Yes you will notice another CRTC controller on the master board, but no Firmware is on ether one. This board also adds some extra video attributes to the display for word processing
(,more or less). There is a lot I could re type (re-enter) on it, but we leave it empty/out for done.
Some 3Rd party graphic programs for the system 23 will need this board installed and yes the 16K of on board Dram can be dedicated to the systems display file.
..
In order to do more them 80 columns on this system, they do a trick, they basically break the screen up into two sections of 80 columns, but this subject is more or less a future item.
On the IBM PC they use a more advanced Video controller that is capable of higher resolutions, but then again it still has the same limited video memory issues. (memory was quite
expensive back then).

 

[N. Brown]

As far as CPM-80 and GSX they where likely done by DR. DR hade some legal issues with IBM with their GSX module for the System23, as well as the displaywriter. (IBM 6580)
Both these products are listed in the IBM Fed. Sys. Catalog but they are not listed in there general software catalog, the same with COBAL and a version of APL. (Cobol and APL could of been by a 3Rd Parity, as well).

 

[RetroAND]

Right on, some one else has noted IBM semiconductor device 4481775 as identified to an Intel 8202.
Hopefully some others will list IBM 4481102 and 4178626 chips.
..
The 4481102 chip in the Word processor video extension/expansion card has a different version number, and yes there is a difference then the one on system 23 motherboard.
So some versions of this chip when removed from the slave Video card and used as a replacement to the one on the motherboard don't work properly. (So I note-it in this message)
..
Yes you will notice another CRTC controller on the master board, but no Firmware is on ether one. This board also adds some extra video attributes to the display for word processing
(,more or less). There is a lot I could re type (re-enter) on it, but we leave it empty/out for done.
Some 3Rd party graphic programs for the system 23 will need this board installed and yes the 16K of on board Dram can be dedicated to the systems display file.
..
In order to do more them 80 columns on this system, they do a trick, they basically break the screen up into two sections of 80 columns, but this subject is more or less a future item.
On the IBM PC they use a more advanced Video controller that is capable of higher resolutions, but then again it still has the same limited video memory issues. (memory was quite
expensive back then).

I checked among other pins 35 and 36 por in case I made a mistake and was a 8203 instead. However, to work accordingly with the memories it has and to work with an external clock both pins should have been tied to +5V, which is not the case. I am sure it is what I said... On the other hand could it be that the controller in the slave card was a 8203 configured to work in 16k mode? How could I have some kind of confirmation on that?

I found a direct connection from the 74LS670 on the edge of the board to the ROS memories. If two registers were for the CPU memory map and the other two were for DMA use, It would only amount to 128KB per bank as the registers are only four bits. Something is missing from my scheme, it's just I still don't know what it is.

By the way, your boards look gorgeous. I am glad they did survive... My unit only has the floppy drive controller board so it will be difficult for me to ever come close to one of those.

To be real they only emulated enough of the systems 34 to fully support its basic and programs compiled by the systems 34 Assembly language compiler (converted to system 23 level code).
This Including any systems call back to the hardware level from basic.
..
Trying to find the standalone RPG II programming language disk for the Sys23 is impossible today. The remote sys 23 RPG II generator for the System 34 is also impossible to get now days.
The compiler for the System 23 to run System 34 ASM is impossible to find a copy now days, the same is true for the system 23 Basic Assembler and others as well.
..
As far as learning the System 34 architecture, the data available on that is also limited. It is only really needed if you plan to run System 34 applications with out instruction code conversion
and or without the recompiling of those applications. This conversion of course can be done in firmware on a system 23, but IBM only originally did the enough to get the system 34 basic
and its related extended processor function calls to work. IBM did reserve address space up to 1M on the system 23 but most of it is un implemented on models below the 500 numbers.
..
If you plan to add the system 34 multiuser runtime unit, expand the memory over 1M byte via bank switched, Use system 34 peripherals. Then yes the knowledge of the system 34 architecture
is needed to do so. Some 3Rd parity did make memory boards for the system 34 that exceeded IBM systems memory limitations, but that is another subjected to be noted on.
..
A diagnostics indicator panel for the system 23.
To do this simply I would use a Modified ISA-Bus Diagnostic unit
(One made for the 8086 that has the extra extended function connector would be the best to use). This would be the simplest way to go about it.
In theory some one could design a new diagnostic unit that used the System 23 diagnostics connector, but the time and resources to do so is another question.
Yes, the way the diagnostics are done seems more suitable to a System/36, System/34 so the service technical could attach the correct cables into a existing diagnostic unit from one of those systems..
..
There was a reason why the System 23 launch date was delayed by a year.

I knew they released it late because they wanted BASIC compatibility with the S/34, but other than that I ignore what is the degree they implemented. Please, have in mind I started with this system a year ago and work on it had been halted most of the time - my experience is very limited. I wish I knew half the things you know - that's why I am asking.

 

[RetroAND]

Hello!

My work towards RAM replacement will, with a bit of luck, end soon. A prototype designed with available and inexpensive parts has been ordered. It's not perfect but I think it will do the trick. With it, unrepairable memory cards could be replaced or, in case the originals are working fine, serve as an internal upgrade. It contains 64KB (8 bits plus parity) of static memories and does not require anything else than 5V. It should arrive in about two weeks.



At the same time, a prototype to replace ROMs 02 and 09 have also been ordered. The one for the rest of ROS memories is still being worked on, and don't know when will it be ordered. With a bit of luck, in about a month, I could have something decent.

Regards,
Jaume