I dug out some info somewhere on the net, where in DOS Games, the Cyrix
P133+ performed slower than the Intel 133. Here's what I found from an
Altavista search:
Identical configurations (all I do is swap the CPU out and change the
jumper for clock rate). TS54P AIO motherboard (with 6x86 bios support,
advanced features o* #9 motion771 video (S3-968 chipset) 2MB VRAM, 32
MB 60ns RAM, GUS soundcard. 25* PB cache on motherboard.
Doom v1.9 (ultimate doom) -timedemo demo3:
Screen at maximum size, no status bars shown. Hi res, kbd control
only.
Processor Gameticks RealTicks FPS (35*gt/rt)
6x86-100 3863 2939 46.00
6x86-120 3863 2453 55.12
Pentium-133 3863 2354 57.43
Quaketest 1, DOS version, UNIVBE loaded (no univbe above, my eval
period ran out when I ran the test), same machine configuration:
Processor TIMEREFRESH reported FPS
6x86-100 21
6x86-120 26
Pentium-133 46
(the 6x86-120, in both cases, is actually a -100 overclocked to 120,
my
6x86 wo* even boot at 133, otherwise I'd report numbers for that too).
> I'm contemplating an upgrade for my lowly 486dx2/66 system to Pentium
> level, and was wondering if anyone knew if the Cyrix 6x86s were close
> to the genuine Intel Pentium processors in their ability to run IndyCar
> Racing II. There's apparently parity on most benchmarks, say between
> the 133Mhz Cyrix chip (labeled 150+) and a Pentium 150 MHz, but there's
> soem talk that floating point is not as comparable. Does that affect
> the ability to render graphics at reasonable frame rates? Is anyone
> running such combinations that might have some observations?
> Thanks,