• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

ocing doesn't boost speed at all!

B

binkgle

Limp Gawd
Joined
Oct 16, 2004
Messages
161
just got my new hd last night, and windows is up and running and i'm ready to oc again. the settings i've tested are cpu and ram @ stock, cpu @ 225 htt with 1:1 ram @ 2.5-3-3-10-2T, and cpu @ 235 with ram set to run at ddr333 (so it ends up something like 190 mhz) at 2-3-2-10-1T. all of these seem stable.

to test these all, i've tried running the sisoft sandra 2005 memory bandwidth test, which showed little noticeable difference between the three settings.
i've run 3dmark2001se at defaults settings and there really hasn't been a drastic change between any of these settings, only a couple hundred points. i've run 3dmark05's cpu test, and scores have all been within 100 points of each other.

therefore, is it even worth ocing with such close results? ocing my vid card helps a lot, but ocing the cpu seems to do nothing. i used to be able to set cpu @ 240, with ram at 200 2-3-3-10-1T, but no more, and from looking at what i've just found, i bet it wouldn't have changed much. am i missing something? you'd think that even if the ram were a bit slower than spec, if the cpu was oced by 235 mhz that there would be some difference!

i can't run any in-game benchmarks yet as i now only have half-life and cs, which aren't exactly taxing on my rig. i will have hl2 soon, but probably only during the upcoming winter vacation.

should i be seeing improvments? is it even worth wearing out the cpu if it doesn't help at all?

also, i now run a sata hd, and i've heard that the via sata controller doesn't do well past 220 mhz. is this true? could this be affecting my oc? this is a nioce hd (a WD Raptor), and there's no way i'm changing it after having just spent $170 on it.
 
I agree with rogue4mula. just use a different benchmark. and hey, why do benchmarks matter so much? do you see better performance in games or applications than you did before? If so then why bother with benchmark programs?
 
only game i have right now is hl and cs, so i don't yet have any games which stress cpu enough to test with them. i'll be getting hl2 soon.
 
here's my last post in the version of this thread i started on xtremesystems (http://xtremesystems.org/forums/showthread.php?t=47070&page=1&pp=25):

"270x9 1.62v seemed ok. mem @ 221 3-3-3-10-2T or 2.5-3-3-10-2T, 2.79v (though i might pump in an extra .1v). sandra cpu arithmetic was 11062, 3969/5157, mem bandwidth benchmark was 3008, 3013 when ram was @ 3-3-3-10. when i lowered ram to 2.5-3-3-10, mem benchie went up to 3025, 3026, but cpu went down to 10401, 3970/5160.

i'll test further for stability (so far i've only done sandra and a 1M calculation with super pi), but which of these ram timing setings means better perf?

EDIT: as a reference, with cpu @ 204x10 1.52v, mem @ 204 2-3-3-10-1T 2.74v, sandra cu arithmetic was 8779, 3332/4329, mem bandwidth was 3137, 3137"

an answer to this question would be great, seeing as i'm getting hl2 tomorrow or the day after and don't want to be worrying about ocing while playing, but instead getting best perf possible for my sys.

thanks to all!
 
You must log in or register to reply here.
Back
Top