From: owner-hecnet at Update.UU.SE [mailto:owner-
hecnet at Update.UU.SE] On Behalf Of Dave McGuire
Far be it for me to disparage Nanny Beeb, but this
type of dependency
is indicative of poor design.
Oh yes. Poor design is certainly the case here. In order to keep everything simplistic
they wanted to avoid BUS contention while running the CPU at 2MHz. They had trouble
finding RAM which would match this design and Hitachi was the only manufacturer with a
suitable component. The NS 81LS95 multiplexer was the only component that was stable in
this configuration.
To make matters worse, they were exceeding the CPU's specifications which cause the
machine to crash. The designers worked out if they put their finger on a certain spot on
the board the machine became stable. They updated the design to include a resistor pack
across the BUS which kept the machine stable. This kludge was nick named "the
engineers fingers".
A dependency on a particular manufacturer's
implementation of a
standard and common design pattern like the 81LS95 is a bug. I sure
hope they fixed it in a later revision.
Absolutely not. They produced around 1.5 million machines with both of these kludge fixes
in place and the machine made it to end of life with no revision changes.
Cheers, Wiz!!