17 Dec
1998
17 Dec
'98
9:47 a.m.
A mechanical von Neumann machine? I'm not sure I
know of any machine, toy
or not, that can claim that (and that's mostly due to my ignorance of the
early relay-based machines, probably). But there are certainly electronic
von Neumann toys, such as the Science Fair Microprocessor Trainer I
mentioned earlier.
True. They're essentially small SBC's, right? (And an SBC is a small
computer which happens to be on a single board.) Electronics don't have the
twisted amusement of moving rods and rolling marbles, unfortunately.
OK, I'll spill the beans somewhat. Normally, the
only reason I don't
spill beans is because I hope to find an instance of the machine myself
before it becomes well-known and "valuable," but I have little hope of
finding this one.
You could build one! I don't think you will find one either.
The first kit was called SIMON (in honor of Simple
Simon). I don't have
all the details yet, but I believe it was relay based, and could be
programmed via paper tape! I have a picture of it which I'll scan and
publish someday soon.
It is a very nice machine, though I wouldn't call it a computer in the
strict sense. Berkeley wrote some books (and there may have been some
Scientific American articles) too.
Of course, I could be underestimating the power of Berkeley's machines.
Berkeley also wrote a nice book about LISP (which may be the only reason why
you can run PDP-1 LISP on your PDP-1 emulator -- the source is published in
the book and it would be VERY hard to find otherwise). But I digress.
Sutherland wrote a division routine for it on paper
tape, and also created
a storage unit for it called MAGDUM (for Magnetic Drum). I have a copy of
the MAGDUM plans which I'll also be web-publishing. I also have an early
Berkeley Enterprises catalog, including info about SIMON, MAGDUM, GENIAC,
and the early robots.
I heard about the division routine somewhere but not MAGDUM.
I dunno know about you, but when I learned of this, I
was dumbfounded.
A personal computer made in 1950 created by one of the early pioneers and
which heavily influenced one of the most influential computer scientists
of our time!
Actually, it is amazing, isn't it? I shouldn't be disappointed even if it
turns out to be a "not-quite-a-computer" class machine.
-- Derek