VOGONS


First post, by InbetweenDays

User metadata
Rank Newbie
Rank
Newbie

I just got a different 486 motherboard and, in the process of deciding whether to use it to upgrade my regular rig, thought I'd run some CPU benchmarks to see how it compares.

I know that Symantec Sysinfo (SI8) isn't exactly a proper benchmarking tool, but interestingly enough it gives exactly the same result (132) with both DX2-66 and DX4-100 overdrive CPUs. Landmark and SpeedSys both show an appropriate speed increase between the two CPUs, but not SI8. (And by the way, it is slower than my other board unfortunately).
The jumper settings are correct (and don't need changing between these two CPUs anyway, but I did still experiment with a couple of different settings). I tried two DX4 CPUs (DX4ODPR100 and DX4ODP100) with no change. I've also tried different BIOS settings (most notably write-through vs write-back cache, though I don't think this DX4 is write-back), all with no effect.

Given that the SpeedSys rating goes from a score of ~25 for the DX2-66 to ~39.3 for the DX4-100, the lack of a difference in the SI8 score doesn't really matter... but I'm just wondering if anyone can help explain this? I was interested in the SI8 score originally, because I get a score of 182 on my other DX4-100 (DX4ODPR100), which is lower than the 205 (ish?) I got with an AmDX4-100. I was curious to see how much difference the motherboard makes, "all other things being equal".

It don't mean a thing if it ain't got 5-pin DIN.
Roland addict and founding member of the Association Of Molex Haters

Reply 1 of 2, by Jo22

User metadata
Rank l33t++
Rank
l33t++

Interesting. I never used Symantec Sysinfo, but Central Point Sysinfo (SI,"System Information").
I wonder how the Symantec Sysinfo works. Did this happen with other CPUs before ? Does the cache or turbo button make anything different ?

"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//

Reply 2 of 2, by InbetweenDays

User metadata
Rank Newbie
Rank
Newbie

No, I've not experienced this before. Changing to a different speed CPU has always resulted in a change in all benchmarks. If there was no change with all of them (ie Landmark, SpeedSys as well) then there'd obviously be a problem... but it's just SI8 which is weird, so I'm assuming it's something to do with that. Nothing I've tried has increased the SI8 result... but the turbo button or disabling the L1 cache slows it down. 😀

It don't mean a thing if it ain't got 5-pin DIN.
Roland addict and founding member of the Association Of Molex Haters