Possible architecture game?

For those of us who are interested in computer architecture, there are a lot of interesting architectures from the past that we don't see anymore: the DEC PDP-8 with its 12-bit words and minimalist instruction set, the IBM 7090 with its 36 bit words, character oriented machines like the IBM 1401, architectures like many of the early IBM machines that, rather than general registers, had the accumulator, the multiplier-quotient (MQ), and some index registers (IBM 1130/1800, IBM 7090, many other machines).

I am thinking it would be interesting to have a "game", if you want to call it that, that would provide emulators for interesting old architectures. Reserved pointers in low storage would point to routines for reading, printing, and talking to the console, so people would not have to learn an operating system at the same time they are learning the architecture. There are open-source emulators available for many old architectures, and architecture manuals are available on bitsavers.org.

This would clearly be a niche game, and I would not expecting it to be a moneymaker. I was thinking about it because I think exploring old architectures is fun, but when I read, for example, about people getting an IBM 7090 emulator running, I think "That would be fun, but then I would have to learn IBSYS, the 7090 operating system, which would be a big time investment, and I couldn't do any real work on it."

What do you think?
 
Slightly related (and windows only) but I love the idea behind Jason, a GUI for the Hercules mainframe emulator. It makes it a little more like a game, setting up all the machines visually (scroll down for some screenshots).

(Written by the same guy behind OllyDbg (a windows DRM cracking cult classic!)).
 
For those of us who are interested in computer architecture, there are a lot of interesting architectures from the past that we don't see anymore: the DEC PDP-8 with its 12-bit words and minimalist instruction set, the IBM 7090 with its 36 bit words, character oriented machines like the IBM 1401, architectures like many of the early IBM machines that, rather than general registers, had the accumulator, the multiplier-quotient (MQ), and some index registers (IBM 1130/1800, IBM 7090, many other machines).

I am thinking it would be interesting to have a "game", if you want to call it that, that would provide emulators for interesting old architectures. Reserved pointers in low storage would point to routines for reading, printing, and talking to the console, so people would not have to learn an operating system at the same time they are learning the architecture. There are open-source emulators available for many old architectures, and architecture manuals are available on bitsavers.org.

This would clearly be a niche game, and I would not expecting it to be a moneymaker. I was thinking about it because I think exploring old architectures is fun, but when I read, for example, about people getting an IBM 7090 emulator running, I think "That would be fun, but then I would have to learn IBSYS, the 7090 operating system, which would be a big time investment, and I couldn't do any real work on it."

What do you think?
I would not call it a game.
 
game (n.)
c. 1200, from Old English gamen "joy, fun; game, amusement," common Germanic (cognates: Old Frisian game "joy, glee," Old Norse gaman "game, sport; pleasure, amusement," Old Saxon gaman, Old High German gaman "sport, merriment," Danish gamen, Swedish gamman "merriment")

Everything is a game given the right state of mind!
 
I am thinking it would be interesting to have a "game", if you want to call it that, that would provide emulators for interesting old architectures.
The B5500 would be interesting to reincarnate. MULTICS has a significantly active support effort.

But, there are far more MPU oddities that have just "disappeared". iAPX432, anyone? 8x300? 99105?

Personally, I would be more interested in actual hardware to emulate these devices; given how much faster technology has become, I suspect many larger FPGAs could be called on for the task.
 
MULTICS would be cool. I never studied its hardware architecture, but I did some work on a MULTICS system, and the operating system was nice.
 
Maybe things like MULTICS will achive the same status as flint arrowtips have now. We have more people making them as a hobby as there were people making them 'back then' for a life.
 
MULTICS would be cool. I never studied its hardware architecture, but I did some work on a MULTICS system, and the operating system was nice.
MULTICS at least tried to be "highly available" -- something that really isn't the case with modern hardware OR software. You could disassemble a MULTICS system and still maintain functionality. Try pulling a CPU out of your FOSS box while its running. Or, try running diagnostics on the hardware while it is still running applications...

The MULTICS goal was to have computing as a "service" -- much like telecommunications, electric power, etc.
 
Back
Top