15 Dec
2007
15 Dec
'07
11:03 a.m.
On 15 Dec 2007 at Robert Nansel <bnansel at bigpond.net.au>:
* Delay lines: I have a junk magnetostrictive DL on
the way. I
don't know if it works.
Golly, if you're talking about producing your own drum, a
magnetostrictive delay line would seem to be like a walk in the park.
Find some nickel wire; make a large-looped coil with the turns
supported by something like neoprene spacers so that they're
mechanically isolated; put a coil on each end. Shock-mount the whole
affair in a box.
The Packard-Bell 250 was a great little machine along the lines that
you're contemplating. Used delay-line memory and had a 22-bit word.
You could plug it into a 115-volt wall outlet. Few transistors and
lots of diodes, IIRC. Really one of the earliest minicomputers
around. There are several that are extant. I recall seeing the
memory in a 250 and being surprised at its simplicity.
I was thinking I would implement a data stack and a
return stack,
each being made of short recirculating buffers to hold the top two or
three stack entries, with the older entries swapped to longer buffers
with corresponding longer access times. Allowing for the overall
insanity level of this project, is this seem a sane strategy?
Sounds pretty complicated to me. Why not a simple one-plus-one
architecture? It may not be fast, but then, no drum- or delay-line-
based machine is going to be a speed demon...
Cheers,
Chuck