You really mean 10-20% only between the two compiler options -m68020-60
and -m68060? That's very optimistic I think (I speak about the compiler,
not hand optimized assembler).
I'm afraid you're right.
And, please don't get me wrong. I'm not against special optimized
software. But it must have a noticeable effect. Otherwise you spent lot of
time for nothing. So optimizing the handful of tools there it make sense
is enough.
I agree with you on this after all, maybe there's really m68020-60 vs m68060 difference out of question. However, I have one experience where at least 68000 vs m680[20-]60 had not some but DRASTIC effect -- and that was gcc. When I ported gcc 3.3.6 using Patrice's tutorial, I've made brutal 060 optimization (in terms of raping makefiles ;-), I didn't care about multilib and stuff, I just added 060 everywhere. When I've tried this 060 gcc, that speed increase was about 30-40%! I'm not kidding, I couldn't believe my eyes. It was newer (=slower) gcc than original 2.95 and still with 060 it was this much, much faster. So I think when we want to be as universal as possible, let's make at least two "branches" -- 68000 and 68020-60... or let's leave m68000 stuff forever.. so when you want to reduce number of mintlib/kernel targets, this is my proposal -- nobody is using mint on atari st anymore .