Well, I don't like to argue. 😅
What I wrote was my personal experience (Double Space on XT), no more no less.
You don't need to believe me if you don't want to, of course.
@4 Yes. And on my ancient 286 it was likely 12MHz, even, not 8MHz (not "ISA" based yet).
What's so special on the 286, is that the PC/AT-Bus bases on the 286 front side bus, which all (most) devices must share.
Depending on the chipset (if any), memory is also coupled to that bus. So lowering i/o overhead generally makes sense.
Edit: Small fixes.
Edit: @2 I'm sorry, but I can't help you on that. 🙁
If you have faith in other human kinds, you perhaps may wish to look at the next link(s).
https://en.wikipedia.org/wiki/Disk_compressio … formance_impact
https://en.wikipedia.org/wiki/Lempel%E2%80%93 … v%E2%80%93Welch
Edit: @1 I agree you may be right with the cycle timing. I'm just a lyman, after all.
I believe why I thought 60ns (or less) would be recommend, is, that I have rarely used any tech that old (80-120ns, 2 micron) before,
When I started with computing 70ns was the worst I saw and 60ns were all around.
I guess this posting at forum.beyond3d.com describes it rather well (link):
"believe this choice of frequency predates even the 386 (the 33MHz variant first appeared in 1988).
60ns DRAM started to appear about 1985-86 along with the 1um process but then, again, I'm not sure if a clock speed of
33MHz (60nS equal 2T) was chosen because of that or the other way around; in fact 60ns is just the guaranteed time needed
to fetch data after the RAS line goes down (60ns RAMs are, in fact, 105-110ns RAMs if we count the RAS precharge time as
part of a 'true' random access and not the page mode)."
"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel
//My video channel//