On Mon, 2004-09-27 at 13:11, Cini, Richard wrote:
> IIRC from Bob's speech at VCFe2.0, he indicated that simulation speed is
> effectively unregulated, because why would you want to waste the benefits of
> a speedy host machine.
Wow, I'm shocked! Speed is a front-and-center major feature of a
machine, both subjectively and objectively. I might as well ask,
...because why would I want to run an old [slow] machine when I have
such a speedy host machine?
I have a kludgey simulator of my own, for a bit-serial machine, it
calculates memory reference accesses properly (fetch, execute, search,
etc) in any organization, so I guess I'll clean that up and make my own
LGP-21.
(It's a test kludge, the instruction decoder is a crock, amongst other
things.)
http://wps.com/projects/UM/code/sim
(If someone actually wants a new-design vacuum tube serial drum (or
switched-capacitor) computer built, and is willing to write checks, let
me know :-)
> On my Altair32 emulator, I have it set to run at the prototypical 2MHz speed
> (8080A).
Is that your code/patch in simh? I see a usleep() in the altairz80_cpu
code.
(It is certainly useful though to "unthrottle" for debugging and
testing! An 80MHz Z80 would be umm nice, huh :-)
Received on Mon Sep 27 2004 - 16:47:53 BST