31 Aug
2008
31 Aug
'08
2:58 p.m.
ENIAC was a decimal machine, not binary.
That's right. Eckert used the decade ring counters
that he had experience with from the radar world for
the digits in the ENIAC. He did, however, design
other binary machines later.
It was also not
a stored program machine; it was rewired for each problem.
Yes and no. It was certainly not designed as a stored
program computer initially, and was indeed rewired for
each new "program" at first. However, in 1948, Betty
Jennings, Dick Clippinger, and Adele Goldstine wired
the machine in a configuration that allowed it to be
controlled by a sequence of instructions dialed into
the switch panel. In effect, the switch panel became
an instruction ROM where it's original purpose was a
data ROM. That was the configuration used for most of
the rest of its life. New problems were dialed into
the switch panel instead of being wired into the machine.
Well before that, however, Eckert had written his paper
where he suggests the stored program, and, of course,
the EDVAC work was well underway.
They were well into the design and construction when
they realized that
there was a better way to do this so their next machine, the UNIVAC I,
Actually, there were at least two other computers between
ENIAC and UNIVAC that Eckert was the primary designer for.
The first was the EDVAC which was fully designed by the
time ENIAC was delivered for production use. However,
before EDVAC was constructed, Eckert and Mauchly left
U Penn over the university's retroactive rules on patents.
They formed the Eckert-Mauchly Computer Company and set
out to do a commercial machine. Before the UNIVAC ever
saw the light of day, however, they designed, built, and
delivered the BINAC. There is considerable debate regarding
whether BINAC ever really worked as it should have. But
what really makes it interesting is that it was the first
dual processor machine. It was a proof-of-concept for
a computer that could fly on board an airplane. The
redundant processors ran in parallel comparing their
results. If there was a mismatch, you know at least one
of them had a fault. The BINAC was delivered in 1949
and the money from it kept EMCC afloat long enough to
be acquired by Remington-Rand who then brought out the
UNIVAC I in 1951.
ENIAC, The Triumphs and Tragedies of the World's
First Computer by Scott
McCartney is my favorite on Eckert and Mauchly and their computers.
As it turns out, my copy of that book is on the shelf
just behind my head as I type this.
BLS