Information Technology Reference
In-Depth Information
order. That was a very time-consuming process. Furthermore,
the addition of new memory or interface boards was simply im-
possible without redesigning the entire thermal-protected pack
of
PCB
casings.
The
R
2
E
Micral, on the other hand, was designed from the
start with the objective of allowing flexible operation with a
variety of peripherals that might require their own individual
interface cards installed in the computer. What Gernelle, the
main hardware engineer at
R
2
E
, came up with was a mother-
board with a number of sockets mounted on it. One could plug
all sorts of boards into these sockets, from
CPU
and memory
boards to interface boards. (For those readers to whom com-
puter architecture is not a mystery, the Micral had a 60-bit
single data bus, called Pluribus, which could handle up to seven
channels for fast peripherals, each channel having its own buffer
memory.) To replace a Micral's faulty board or replace one with
an upgrade, one had only to pull out the board to be replaced
and push a new one into the desired socket. The operation took
seconds. And yes, the Micral was equipped with an internal fan.
Ramer's
Development Policy
document was, of course, just
a proposal, prepared under the rather heavy clouds that had
started to descend upon
MCM
. It would not be worth an ex-
tended discussion if, in the end,
MCM
had not taken the very
path he recommended - but with rather different results from
those forecast in his report. The
MCM
/77 concept was re-
leased in 1976 as the
MCM
/800. It was ten times faster than
the
MCM
/70, but its
CPU
was neither the 8080 nor any other
microprocessor. The
MCM
/800 had to be packaged into a new
case and it did have an internal fan. Because of its late release
and lack of outstanding features, the computer was mostly un-
noticed in the (by then) crowded market of small computers.
Microprocessors had not disappeared, but instead had become
even faster and even more popular. The Intel 8080
A
, Motorola
