14 Jan
2002
14 Jan
'02
5:03 p.m.
It was thus said that the Great Richard Erlacher once stated:
I do believe the main reason for the 68010's appearance in what was previously
a number of 68000 applications was that it could support virtual memory, while
that was awkward on a 68K.
That is correct. The 68000 could not restart a faulted instruction since
not enough information was saved. The 68010 saved more of the internal
state of the CPU to restart the faulting instruction (and I think the saved
program counter changed too---mostly from past the faulting instruction to
the actual faulting instruction).
-spc (And they did speed up certain operations on the 68010)