From: der Mouse:
Were they 68000s, though? All the ones
double-fault-fix machines I
know of were, I think, using 68010s.
The early Apollo systems were, I believe. I recall talking to one of
the "guys behind the table" at Wescon when the 68000 came out. He
said that the most commonly asked question by customers concerned
using the large addressing space and fault mechainsm on the 68K for
virtual memory--and that he was getting tired of saying that it
couldn't be done. It was certainly one of the first questions out of
my mouth.
Although it was kind of a kludge, he was proved wrong if one used two
of the beasts.
Does anyone remember the speculation that the feature size of the 68K
was getting close to the limit of IC fabrication technology? It was,
what, 3 micron?
Cheers,
Chuck