>>>
On 9/4/10 10:59 PM, Andy Holt
wrote:
I'm astonished that there doesn't seem to be a
Multics emulator (OK, the
hardware is the epitome of CISC, but it is one of the true classics).
There isn't quite enough information on the I/O system. A few people have
gotten
as far as trying to get the main CPU to boot.
<<<<
Odd, I'd have thought that the IOM would have been the one part of a Multics
systems that was essentially unchanged from a G(e)COS one - the SCM(?)s (the
boxes that contain memory and connect to processors and IOMs) would
presumably need to know a bit about the memory management and the Datanet
probably had entirely different code)
>>>
> There are also emulators for
some of the first and second generation
> machines (Manchester "Baby" for example, but, again regrettably,
apparently
not the Ferranti Atlas)
Does any of the code survive for Atlas? I was in communication with the
Cambridge
folks, and they were saddened to discover their only known copies of the
software
for Titan were discarded.
<<<<
I really don't know - after all there were only 6 Atlases in total:
Manchester, London, Chiltern, Cambridge (x2), and Aldermaston unless GCHQ
also had one or more. But it would be fun if a CPL compiler still existed.
(CPL cut down led to BCPL led to B, extended led to C)
Andy