Disabling both caches gives a result of 386sx-33 in topbench, but with either one enabled, makes little to no difference. Winding back the settings to 233mhz in the bios, and dip switches, both fail to actually get a result of anywhere near 233mhz, with or without caches disabled. Maybe a p3 can only go down to about 466?
regular p3 cpus have locked multiplier, for example your 700 is fixed at 7x, so the slowest clock for it is 66*7=466.
the lowest multiplier for p3 is 4x clock of the 533eb, so the slowest clock you can go is 66*4=266.
some(but not all) engineer samples have unlocked multiplier but i think the lowest they support is also 4x.
if you want something slower than that, give the viac3 a try, its performance is a quarter of the p3 at same clock.
wrote:regular p3 cpus have locked multiplier, for example your 700 is fixed at 7x, so the slowest clock for it is 66*7=466. the lowest […]
regular p3 cpus have locked multiplier, for example your 700 is fixed at 7x, so the slowest clock for it is 66*7=466.
the lowest multiplier for p3 is 4x clock of the 533eb, so the slowest clock you can go is 66*4=266.
some(but not all) engineer samples have unlocked multiplier but i think the lowest they support is also 4x.if you want something slower than that, give the viac3 a try, its performance is a quarter of the p3 at same clock.
I just threw a p2-350 in, for some reason it benchmarks as a 286 with one cache on and one off! 😲
wrote:wrote:regular p3 cpus have locked multiplier, for example your 700 is fixed at 7x, so the slowest clock for it is 66*7=466. the lowest […]
regular p3 cpus have locked multiplier, for example your 700 is fixed at 7x, so the slowest clock for it is 66*7=466.
the lowest multiplier for p3 is 4x clock of the 533eb, so the slowest clock you can go is 66*4=266.
some(but not all) engineer samples have unlocked multiplier but i think the lowest they support is also 4x.if you want something slower than that, give the viac3 a try, its performance is a quarter of the p3 at same clock.
I just threw a p2-350 in, for some reason it benchmarks as a 286 with one cache on and one off! 😲
That is normal for a P2. Different CPUs behave differently when you do this. This thread has lots of information and benchmarks on many systems with cache on and off.:
Let's benchmark our systems with caches disabled
The more I learn, the more I realize how much I don't know.
OPL3 FM vs. Roland MT-32 vs. General MIDI DOS Game Comparison
Let's benchmark our systems with cache disabled
DOS PCI Graphics Card Benchmarks
