Emulators of Classic Computers

From: Tom Jennings <tomj_at_wps.com>
Date: Tue Jan 20 14:36:43 2004

Well my equally crazy (but smarter) friend David Forbes and I talked
about making Williams memory, and the problems (modern) thereof. Well it
turns out he makes oscilloscope clocks (quadrature character generators
and nice things) and we talked about mocking one up -- using the 68HC08
on board, simulating a 1, 4, 8, ... bit Williams memory cell, what with
all the circle/dot/slash/focus/defocus modes, we could tightly
characterize readaround issues, very repidly reinvent those wheels of
the past. He buys CRTs from the manufacturer and knows a lot about this
stuff. http://www.cathodecorner.com.

The pedant in me says a CPU bigger than all combined planteary computing
resources of 1955 driving a single memory element might be cheating.
Those old guys did this with tubes! It's quite terrifying.

tomj



On Tue, 2004-01-20 at 01:22, Brent Hilpert wrote:
> Somewhere in my list of (too many) projects to work on some day is to implement one of the first (tube) computers in SSI. One I had done some design on was the SSEM/Manchester "Baby" ("the first machine to run a stored program"/1948). I did enough of the logic design to implement it in a home-brew logic simulator (executable in browser via shockwave at http://www.cs.ubc.ca/~hilpert/tih/SSEM), but haven't got around to doing a physical implementation. Figured CMOS SSI would be the way to go. For the "Baby", a scope/CRT with Z-axis control could be used to emulate the original 32-bit-by-32-word storage-tube-memory monitor. I have an HP digital signal analyser from 1969 (basically an early digital oscilloscope) which has all the CRT support and power supplies and figured the new/old processor could be built into the HP DSA with a new front panel, making an interesting amalgam of hardware and architecture spanning several decades of computing. The same idea could apply to some
 o!
> ther 'scope equipment of course.
>
> Another interesting machine to implement in SSI might be the IAS machine, or perhaps something like the IBM 709.
> One could get carried away making scale-model front-panels with lots of LEDs and miniature toggle switches.
Received on Tue Jan 20 2004 - 14:36:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:36:47 BST