cs4g6ag@maccs.dcss.mcmaster.ca (Stephen M. Dunn) (03/20/90)
In article <27842@cup.portal.com> Duel@cup.portal.com (Omid M Farr) writes:
$If a 386 runs software made for the 286 (assuming they are at the same clock
$rate) at similar speeds, then I take it the 386/486 would follow the same
$pattern.
No. Intel has heavily optimized the 486, especially in the floating
point processor, so that it easily beats a 386 or 386+387 of the same clock
speed.
$So this is what I think is going on, Please correct me if I am wrong...
$A 25 MHz 486 can run software written for the 386 specifically, but a
$25 MHz 386 would run the same program at about the same speed, meaning a 486
$really doesn't (today) have any advantages unless you intend on writing
$software for it.
Well-designed 25 MHz 486 machines have similar or slightly better
performance than 33 MHz 386 machines for non-floating point operations,
and they outperform a 33 MHz 386+387 combination by a factor of two or
better.
In the future, some software may be available that has been optimized
for the 486's pipelining, which can produce a 15-20% speedup over running
386 software on the 486. However, this is not a consideration now, and
personally, I'm not so sure that there will ever be that much software
that's aimed specifically at the 486 (but hey, I could be wrong).
--
Stephen M. Dunn cs4g6ag@maccs.dcss.mcmaster.ca
<std_disclaimer.h> = "\nI'm only an undergraduate!!!\n";
****************************************************************************
"So sorry, I never meant to break your heart ... but you broke mine."