Cycle accuracy question

[Anarchist420]

Is it true that a 486 can't be emulated with cycle accuracy?
If so, then why can't it be?

Is there any way that a 486 DX2 66 and VGA graphics could be replicated (i.e., 100% emulation accuracy) on today's CPUs?

I personally think that DOSBOX doesn't do a very good job emulating the DOS version of MK3 because of the checker pattern in the health and run bars and because the speed is totally inconsistent.

 

[ChronoReverse]

If you can't adjust DOSbox's cycle settings to your liking (http://www.dosbox.com/wiki/4.77_MHz) you could try using Bochs instead (which is much slower).

 

[Exophase]

Of course it can be emulated with cycle accuracy. Is your question whether or not anything out there currently does emulate it with cycle answer? If that's the case then I doubt it, or at least definitely not any PC emulator I've ever heard of. Intel may internally have software emulators capable of this, not counting lower level design simulation tools.

But emulating a 486 with cycle accuracy doesn't mean that much when software of the time was already designed to run on a variety of computers of different speeds, including x86 CPUs not made by Intel. It'd be very odd for PC software to be so sensitive to the timing of a particular 486 model that it needs to be emulated with cycle accuracy. I doubt the problems you mention has much to do with that.