Hi,
Not completely on topic, but my boss wishes to build a computer using
modern knowledge, but only materials, IO etc that existed as at 1st Jan
1900. The basic premise is that had a need existed, and they knew how
to make one, could have it been done.
That initially seemed hard, and certainly isn't easy, in that I'm the
one doing the "building" :-) however we have fairly quickly identified
that magnetic logic is the only likely candidate. Currently I'm playing
with saturable reactors to try and get enough gain to make more than
simple logic viable. We keep on hitting little curve balls ie. the
highest frequency sinusoidal AC we can find stuff on as at 1900 was
20kHz. Modern core materials and 20 kHz don't totally work together...
This is to be a real computer, stored program, RAM, ALU etc. Ian has
already built 2 of those by hand, one with TTL and one with 4000A CMOS,
so has some idea of how to do it (personally I'd just get a micro), but
he has the determination and the resources complete this. I'll keep
the list posted if there is any interest.
Regards,
Gavin Melville
gavin.melville at acclipse.co.nz