games freeze do to overclocking?

Â

∞Velocitymaster∞

Gawd
Joined
Dec 12, 2004
Messages
584
setup as follows...
QX6700 OC 3.2ghz FSB 320X10 Zalman9700 Volts 1.55 set in BIOS
Mobo Evga 680i SLI A1 with new BIOS P30
Ram 2 gigs Corsair Dominator 1066 OC to 1120mhz 2.2volts "manufacture voltage" with stock timing set in the BIOS.
2 Evga 8800GTX SLI drivers 158.22
Creative X-fi gamer
Enermax 850W PSU
Win XP 32bit


other config
FSB 1.4v
HT(SPP-MCP) 1.2v "auto"
nfoce SPP 1.5
Nfoce at 1.5v auto
all this ran stable for 2 months.
my OC is done only in the BIOS

I disable all what needs to be disable in the BIOS before overclocking on the 680i mobo oc guide and same applies to the C2d OC guide.

The computer was built 2 half months ago and run very stable on all games. I ran Orthos for about 9 hours in this OC settings with no lockups very stable, screen capture here http://img382.imageshack.us/my.php?image=stresstestcc9.jpg I did also ran memtest for 10 hours with the OC setting and got no errors. But that was before this situation.

Since last week I was playing Call of Duty 2 find then it lockup, only way was to hit reset to reboot, I try all other games like far cry, serious sam II, Stalker and BF2142 they will freeze, after I reboot and reboot, I try any of this games they will hang or freeze. It can be any where from few secs in the game or 4 min even 10 min in.... I turn my computer off for 10 min or so "unplug it" then it will run fine in a game for 2-3 hours fine, then the next day I came back from work ran a game boom there it goes again.

Before this all happen since I had my machine up and running, games would ran for hours with any problems.
First I thought might be my software, reinstall the video drivers current and the new beta one, reinstall windows, try 1 video card then the other but still the same problem. I figure it out that was not any of my video cards, I also try my onboard sound for a while then this problem happen again. The computer runs fine when encoding videos and nothing happen when not gaming.

What I did was set the nfoce spp to 1.5v and move up the CPU volt to 1.6v ran for 2 days smooth and this it lockup again. I notice after changing my OC values then it runs fine for a day or 2. Now im very worry. At this moment I back out of overclocking first time and running my CPU ram and voltage at default speed. For sure this is not a software or driver issue. Hopefully this problem wont come back or I will have to RMA my mobo :( I have to run this default hardware setting for 2-3 days just to be sure is not overclocking related
Anyone heard of this problem?
 
Finding the cause of such things can be very difficult I have no answer but will comment on a couple of things.

Your CPU voltage is very high, at or above the Intel absolute maximum Vcore values. If you have tested and such high Vcore are required to make your 3.2GHz OC I would consider reducing your OC and being happy with whatever you can get stable at 1.45V. If you have not tested I would work on doing stability testing while adjusting Vcore to the lowest possible value to obtain your desired OC.

I would overnight memtest the machine again.

I suggest re-establishing your OC settings using a rational testing methodology to either confirm your present settings or establish new ones.

Sending the board back will probally not help you.

Run speedfan in temp logging mode during a gameing session and see what your coretemp readings are. Be aware that while the Digital Temp Sensor embedded in the cores is very accurate and good, there can be hot spots, if you are near the edge on temps as reported by the DTS, there may be a hot spot that is causing the failure.

Check your Windows System Event Viewer for any failure messages.

It is my personal opinion that the high CPU voltage is probally the issue. At the least you are compromising the life of the processor. Even if it does not resolve the issue, reducing the cpu voltage to something a little more reasonable will be better long term even if you have to reduce your OC some.

Sorry I cannot point out one thing and say "that is it", its a complex system and you have it running at the edge, its just not possible to point to one thing. Methodical testing is the only answer. Testing the CPU, memory, and video as standalone components in OC situations, then retesting as a system. It will require a lot of time and note taking and having a plan is critical to success. Or the next poster will have your answer and you can ignore all this. :D
 
Games are way more complex than Orthos.
What are your temperatures doing while gaming????
I have to agree, at that Vcore you are going to kill the CPU soon.

You need to start over and OC with less of a Vcore.......:eek:

It aint the MB.
 
I can't believe that I typed 1.55 sorry I actually meant 1.35v not how stupid I was. Some how i mix up with the FSB I think. BTW I was stable at 1.35 and 1.36 in orthos at 3.2, now Im back to stock speeds and my volts are down to 1.35 stock volts are 1.33. So far I can ran my games fine, I but is not for sure till after tomorrow I will be sure.
Could 1.35 or 1.36 be to low I was thinking, I don't think so because I ran Orthos stable for over 9 hours even after this problem happen. Also how come all games were running steady for 2 months or so and now no if this is voltage related.
 
Wow LOL that does make a difference, big one, well the stuff about it being a complex system still applies but yes, at 1.35ish V at 3.2GHz I would give the CPU voltage a bump, try the next increment available to you and see if it helps. Try it a couple of days, keep an eye on temps, and if the problem is still there, try another increment. The upper limit is up to you, I personally stop around 1.4V but 1.45 is not outrageous. Anything over 1.45 concerns me.

Where did I get that number ? Well an old obsolete rule of thumb (although it is surprising how many engineers still use it), is to design at least a 10% "safety" margin in, and 1.325 x 110% = 1.4575 and then round down just to be conservative. This really means nothing as there is no hard facts to support it but I have seldom had any problem "pushing" most quality electronic components to 110% as long as you keep an eye on it (temps).
 
Thanks for the info, I will post updates later when I get some next results of this issue.
I will try 1.35 at stock speed which at 2.66ghz. If it crash it sometime later then thats a bigger problem, if no I will bring the clock speeds back to 3.2 but I will try 1.37 and up by 1 value till 1.4 or stable.

In event viewer I don't see any messages related to crash or freeze, sometimes when i ctrl alt del or try to click on the mouse the speaker in the mobo makes a beep. My CPU temp when gaming in speed fan reading gets around 47-50 depends on the games some are more CPU intense. What is the safest highest temp that I can go or what temp in speed fan that I should not reach? is it 60? I reach 60 when running both Orthos.
But I don't think is temp wise. The freeze happen couple times when the game just start like few sec in, one time Call of Duty 2 froze when the game preview starts even before the game menu. Before this happened I can run games for hours I don't think is temperature.
 
Dam is back again, looks like this is not related to overclocking. I run 3 games many throughout the day today nothing happend, the games ran smooth 1-2 hours on far cry and call duty2. I exit it then ran the games again, ran serious sam 2 twice. But at 11:30pm I was in 7-10 min in serious sam II the computer freeze again. The sound was still playing or going, when i press fire i can hear the gun spinning WTF is this about now. I hit 8 key to change weapon and I heard the cannon firing. Looks like the graphics had stop as the game continue.

It freeze where I still see the image of the game but I notice on top of the screen the graphics is broken tiny white artifacts. I try clicking randomly and hit ctrl alt del and i hear speaker in the motherboard beep once. My mobo show FF on the LED which is normal. Any suggestions. I wish I can repost this topic some where else.
Now im gona memtest it all night to night if no errors im gona try flip the battery coin. If it persist I will have to RMA the board.
 
ok looks like i found something in event viewer. "The driver nv4_disp for the display device \Device\Video0 got stuck in an infinite loop. This usually indicates a problem with the device itself or with the device driver programming the hardware incorrectly. Please check with your hardware device vendor for any driver updates."


Now is it my video card? I tested both video card one at the time and the game will hang or freeze. So it can't be both cards bad. I have the latest 162.18 forceware driver , it even happen with my other driers as well. I did clean install of windows also.
Could this be related to my memory? right now Im using memtest v1.70 and testing the ram, going to run at for 12 hours at least.

http://forums.nvidia.com/lofiversion/index.php?t29680.html someone had similar problem but I can't find any help on google now
 
I had exactly the same problems when clocking my new Q6600. I traced it down to core volts.

I can boot into windows and run most applications at 1.35v and 3.3GHz (on water here) but its not 100% stable and produces the odd solid freeze when gaming but the freeze was totally random. I had to bump the voltage all the way to 1.45v to get 100% stable.

One thing I did do was run Orthos on cores 1 & 3 on one program and 2 & 4 on the other. This seemed to provide more stress than using 1 & 2 and 3 & 4.
 
I'm having the same problem, except when I tested both of my cards seperately , they worked fine. I was using the 162.18 beta drivers, which actually works fine for me, as long as I disable AA in the control panel. The WHQL version as well as the 158.22s don't work and I get freezing, and sometimes BSODs. My system has been running fine for a while now, till I installed vista for a bit, and then went back to xp. I've reinstalled xp again just to make sure. Very frustrating that I can't figure out what's wrong.
 
You must log in or register to reply here.
Back
Top