I came here to the forums to clear misconceptions and misunderstandings about this computer.
I own a unit, a model 5322-124 and a second unit, only logics, has to arrive soon. I have studied my specimen for about three years, so I am confident enough to attempt repairs and also wrote the MAME driver, which is yet to be accepted into the main repo.
Except for some questions about the software, and info about expansion cards I don’t have, I can provide data for the motherboard and a bit for the FDC.
Please, don’t hesitate to ask your doubts. I will answer them as much as I can.
Actually, I’m on a personal crusade. Having been so secretive and at the same time unsuccessful has propagater a lot of rumours that have been accepted as a facts, when in reality is incorrect information. As part of my goal of clearing the information about this computer, I am contacting every single place I find with various search engines and contact the owners and expose them my information in order to help them correct their texts.
This is one of many sites where I think I can help with my experience with the system.
Are you interested in anything in particular about this computer? Not everything I know is published, although I try to publish as much as I can.
I confess I have no special interest, although I do have a general interest as it’s an intriguing part of computer history.
It would be interesting here to have your personal story of how you got interested in the machine and your experience with it.
As for your campaign to correct misapprehensions, perhaps a good way is to write a Q&A which sets out the mistaken ideas, and your best understanding of the reality. Of course you already have your blog and your own wiki.
Meanwhile, I found various links which might be of interest here:
Okay, I’ve got some questions or maybe just thoughts that might be misconceptions.
If I understand correctly, this used a CPU with 64K address space, right?
And this caused frustration with software development, leading eventually to the IBM PC leaning toward something with a bigger address space. Is that right?
But I’m also reminded of the 6509 used in some Commodore computers, which had a 64K address space but a bad form of bank switching. Software development was a nightmare because you had to be excessively clever when dealing with bank switches. Could this have been the real problem with System/23 Datamaster software development?
It makes me wonder how the IBM PC might have gone if this system used a better RAM banking system that would make software development more pleasant. For example, you could only bank a 16K window, which leaves 48K for code that you don’t have to worry about bank switching for. A simple byte specifying the 16K RAM bank would provide up to 1024K of banked memory.
Right, the 8085 has an address space of 64KB, so it can only read/write at a range of 64KB.
But with two banks, one at RAM and another at ROM they made it possible to access a theoretical maximum of 272KB of ROM and 128KB of RAM.
Also right. The bank switching was something so painful they wanted a device with a more extense range for the next device.
At the same time they wanted to reuse as much as they could and also simplify it to reduce costs. That’s why for example the Intel 8203 was dropped from the PC design.