Q6600 overclock has affected game speed

nzpoopy

Distinguished
Mar 3, 2008
3
0
18,510
After I overclocked my Q6600 to 3.0Ghz the models in Counter-Strike: Source started to jitter as if I was playing with a high latency. When I went to spectate a player, the crosshair would jitter.

I then tried hosting a server and added bots. The game speed significantly increased but on console the gamespeed showed as 1.0.

I then tried WoW when i turned my character to the side it would skip into place instead of turning smoothly.

I followed CompuTronix's guide and temperatures are well below the safe zone. I'm guessing the increased frequency of the CPU has somehow affected the frequency of the games.

I used EasyTune5 to change the clock speed.
 

nzpoopy

Distinguished
Mar 3, 2008
3
0
18,510
Mobo: Gigabyte P35-DS3R
CPU: C2Q Q6600 G0 Stepping
CPU Fan: Thermaltake V1
GFX Card: XFX 8800 GTS 512
RAM: 2 x 1gb G.Skill DDR2-800 (4-4-4-12)
PSU: SilverStone Olympia 650W
 

Lupiron

Distinguished
Feb 9, 2008
1,711
0
19,780
Thats true, my Crysis needed an extra voltage notch beyond prime 95s small ftts. Would still BSOD if I didnt do it, though it was 8 hour stable on prime 95. Maybe it has to do with the game using a larger amount of over all power? It only brings core 4 to 100% But maybe the audio, video and all the other junk cranking away to play it is an over all power drain that impacts the VCore a lil? Got me! But one more notch solved it.

--Lupi
 

Thanatos421

Distinguished
Mar 26, 2007
549
0
18,990



This is why I always do a 24 hour stress test to ensure stability. I have had many OCs fail a core on Prime after 14 hours. A lot of people swear by 8 hours, and in some cases, it may be enough. However, 24 hours will run the entire gambit of the test. If it passes 24 hours, then it works, period :p
 

jay_l_a

Distinguished
Feb 15, 2008
576
0
18,990
Maybe that's what I should try also.
For some reason, red-lining my PC for 24 hours makes me uneasy.
Are there actually a finite number of tests in Prime95? I thought it was just doing the same calcs with different numbers to stress the system.

Is it asking the CPU progressively harder questions? ;)
 

Evilonigiri

Splendid
Jun 8, 2007
4,381
0
22,780

I usually recommend a bare minimum of 6hours for the cpu stress test. After all, most people don't use their pc running intense programs for 6hours straight.
 

blackpanther26

Distinguished
Nov 29, 2007
757
0
18,990
I've noticed something with the Phenom 9600 BE at 2.9GHz and Dawn of War Dark Crusade it would BSOD every time I go in battle but Prime95 would say its' stable. I'm going to see if I can get to 2.9GHz again. I really want to hit 3GHz.
 

blackpanther26

Distinguished
Nov 29, 2007
757
0
18,990
yeah got it stable at 1.4v with out crashing 61c load. But it goes to show you programs know if the system is stable or not. but did you fix your problem?
 

Lupiron

Distinguished
Feb 9, 2008
1,711
0
19,780
Hahaha, what one? I have been hard at work testing away! Looking for a minimum voltage for 3737 on my g0 right now. I know that some boards get less efficient as you make it put out more power. At 3.6 and a vcore of 4.500 its 1.40 in windows, 1.38 fully loaded. But when I get it up to 3825 it needs huge vcore, and the droop goes from .02 to .06. That makes it hard to stabilize, because thats a few extra notches to get its fully loaded dr0op where it wont bomb out. And this board is waaaaay worse than the p5k-e the g0 is on, its droop is .06 to start, and gets to .08 so far, and thats at 3.6 on this b3. Its way hard to get it stable with that droop, because its a high setting in the Bios, and I simply dont like doing it, though I will do it on occasion!

--Lupi!