Computer reboots during gaming

[prodystopian]

When I was playing Crysis, my computer rebooted a few times during gameplay. I brushed it off because I was near the end of the game. Recently, I downloaded nvidia's performance monitor and tonight while playing HL2 my cop rebooted again. Upon rebooting, I immediately started the monitor and my GPU temp was 70ºC. Is this high enough that it might be causing the reboot?

Specs: EVGA 8800gts 512 running at 670MHz core and 972MHz memory (I didn't OC at all). nforce fan is listed at 675 RPM.
e2160 @3GHZ OC (I tested this for 24 hours on prime95 so I'm fairly sure it isn't the cpu causing the reboot).

The rest is in my profile.

Edit: As I typed this message, the GPU temp went form 70º to 65º.
Edit 2: GPU is at 75º when I try playing again.

What can I do to fix this?

 

[Loser777]

Add a couple of fans in the case, maybe buying an aftermarket cooler for the GPU. Try seeing under your Nvidia Control panel what is the "warning" temperature. Usually it's something 90-115C.

 

[Xazax310]

Also download Rivatuner and see if you can UP the fan speed to exhaust more heat.

 

[prodystopian]

Any chance it is my PSU causing this? I have Corsair 450vx.

Loser777, I can't find the warning temperature. Do you mean in nTune?

I have rivatuner, but is there a way to increase fan speed only during gaming (under load). I'd prefer not to have it high all the time.

Tried changing fan speed via rivatuner. Went into power user > fan > current device and turned on applyatstartup and changed fanspeed to 725. When I restarted, nvidia monitoring saw the speed as 675 still.

 

[dragonsprayer]

turn off all back ground programs

open task manager and spead it too see ram and cpu usage

crysis uses almost all 2 gb of ram with vista

does crach after 20 mins each time or same period of time? remove the door - does it crash less?

its the psu

if thats not it:

does it crash randomly? its software - check settings on cpu voltage and other things

you nvidia ntune to set the fan to 70% it drop the temps from 75c t55-60c in game

i was running 85c with my 730mhz msi oc 8800GTS 512mb - changing the fan improve game play alot i was running 85c all the time during gme play now its 65c

close your browsers and increase the fan to 70%

 

[dragonsprayer]

take your door off and see if it last longer

450w is too small how my 12v do you have? but it still maybe what i said above

you kill that psu eventually its really too small

 

[prodystopian]

turn off all back ground programs

open task manager and spead it too see ram and cpu usage

crysis uses almost all 2 gb of ram with vista

does crach after 20 mins each time or same period of time? remove the door - does it crash less?

its the psu

if thats not it:

does it crash randomly? its software - check settings on cpu voltage and other things

you nvidia ntune to set the fan to 70% it drop the temps from 75c t55-60c in game

i was running 85c with my 730mhz msi oc 8800GTS 512mb - changing the fan improve game play alot i was running 85c all the time during gme play now its 65c

close your browsers and increase the fan to 70%

It's definitely random. I played for quite a few hours the other day with no problem and then all of a sudden last night it happened after a few minutes.

I finally found direct fan control in the ntune software so I will be using that to try to cool the card tonight.

 

[prodystopian]

Update.

I've played for a while tonight with no problems with the GPU at 76º. Looking around, this doesn't seem to be cause for the GPU to shut off (even 80º shouldn't be too high).

I did up the fan on my card later, which is really keeping my temps down.

Does anyone have any ideas on how to troubleshoot this? Could it be a software problem? It happened with both Crysis and HL2, and is very random.

 

[eck]

Update.

I've played for a while tonight with no problems with the GPU at 76º. Looking around, this doesn't seem to be cause for the GPU to shut off (even 80º shouldn't be too high).

I did up the fan on my card later, which is really keeping my temps down.

Does anyone have any ideas on how to troubleshoot this? Could it be a software problem? It happened with both Crysis and HL2, and is very random.

Start by checking event viewer [Control Panel -> Administrative Tools -> Event Viewer] and look in the System section for any errors. It can point you in the right direction if the crash is logged.

 

[partialsum]

Sounds like a virus. Does it only happen during gaming?

What is the recommended wattage for the 512? I thought it was 400. If it's 400 you ought to be okay. From what I understand the graphics card manufacturers set those required watts above the norm so people are not riding on the limit. So if it says 400, and you have 450, then you should be fine. Personally I think 500 would be better, but corsair is a good brand. Make sure the driver is correct and from nVidia. Again check into the virus thingy.

 

[nvalhalla]

I had the same problem with my 3850. turned out to be heat. had been fiddling with the overclocking settings when I first installed it and locked the fan at 50%. took me a lot of troubleshooting before I looked in there. thought it might be the drivers, the game, the modded vbios, the overclock. nope, just dumb ol me locking the fan at 50%. back to auto and so far so good.

 

[spoonboy]

When I was playing Crysis, my computer rebooted a few times during gameplay. I brushed it off because I was near the end of the game. Recently, I downloaded nvidia's performance monitor and tonight while playing HL2 my cop rebooted again. Upon rebooting, I immediately started the monitor and my GPU temp was 70ºC. Is this high enough that it might be causing the reboot?

Specs: EVGA 8800gts 512 running at 670MHz core and 972MHz memory (I didn't OC at all). nforce fan is listed at 675 RPM.
e2160 @3GHZ OC (I tested this for 24 hours on prime95 so I'm fairly sure it isn't the cpu causing the reboot).

The rest is in my profile.

Edit: As I typed this message, the GPU temp went form 70º to 65º.
Edit 2: GPU is at 75º when I try playing again.

What can I do to fix this?

I had a kinda similar problem, STALKER started crashing one day, then quickly deterioated until it crashed after 30 seconds or so of a level being loaded, all of this in an afternoon, tried other games and they crashed to. Long story short I flashed to a new bios on my motherboard. Hey presto everything worked again. It was a motherboard/ram compatibility issue - googled asus p5b and corsair memory, not the only one who has had this problem - when a recent overclock on my vid card seemed the likely culprit (it was not, clocked the vid card higher since all fixed and it worked totally fine). Point is, if you find yourself stuck after all else fails, try bringing your cpu and fsb speeds down a bit. Say to 2.8ghz or so. See what that does. If your sure its not the psu, you could as a LAST RESORT try flashing your motherboard bios to the latest version, but be warned if it goes wrong for whatever reason your up a certain creek without a certain paddle. I only did it cos I was close to buying a new motherboard. Bios flashing takes 5 minutes but it is risky.

 

[prodystopian]

I think memory or my OC might be the issue so I tried the burn in tests on SiSoft Sandra Lite. Everything passed but one thing that seems weird from the report:

Board Temperature: 39.0°C (Min 39.0°C; Avg 39.0°C; Max 39.0°C)
CPU Temperature: 27.0°C (Min 27.0°C; Avg 27.0°C; Max 27.0°C)
System Fan: 978rpm (Min 978rpm; Avg 978rpm; Max 978rpm)
CPU DC Line: 1.28V (Min 1.28V; Avg 1.28V; Max 1.28V)
Aux DC Line: 1.97V (Min 1.97V; Avg 1.97V; Max 1.97V)
+3.3V DC Line: 3.25V (Min 3.25V; Avg 3.25V; Max 3.25V)
+5V DC Line: 4.87V (Min 4.87V; Avg 4.87V; Max 4.87V)
+12V DC Line: 1.62V (Min 1.62V; Avg 1.62V; Max 1.62V)
Standby DC Line: 5.08V (Min 5.08V; Avg 5.08V; Max 5.08V)
Battery DC Line: 3.17V (Min 3.17V; Avg 3.17V; Max 3.17V)
CPU Core Power: 69W (Min 69W; Avg 69W; Max 69W)

WHy is my 12v line only taking 1.62 volts? Is that normal?

 

[prodystopian]

Bump. Can anyone tell me if this is at all something I should worry about?

 

[jjblanche]

The 12V line should be taking 12V (or thereabouts), by my understanding. Get rid of that PSU. PC Power and Cooling.

 

[prodystopian]

The 12V line should be taking 12V (or thereabouts), by my understanding. Get rid of that PSU. PC Power and Cooling.

What does that mean?

Also, any chance the software is wrong? I think I may have fixed my problem (I think it was the memory timings).

 

[Conumdrum]

Your pushing your hardware. It's more likely a heat issue and your case air flow. Turn the GPU up to 100% fan and leave the door off and get a house fan to blow air in the case. Bet it stops. The inside of your case after 30+ minutes of playing is probably warm, thus the fans have only warm air to blow over the components AND thats the air being used by your PS to keep it cool.

Many peeps get a 8800GT and their case air flow can't keep up with it. You have 2-3 120mm fans on your case?=good Or 2 80mm fans?=no way. Next is the PS causing problems due to working too hard and not able to stay cool.

 

[jjblanche]

PC Power and Cooling make the best power supplies. Get one.

 

[jaguarskx]

The Corsair VX 450 offers 33 amps on the 12v rail so you have more than enough to power your system.

However, 1.62v on the 12v rail is unusually low. What does your BIOS tell you?

 

[jjblanche]

I would say critically low...

 

[prodystopian]

The Corsair VX 450 offers 33 amps on the 12v rail so you have more than enough to power your system.

However, 1.62v on the 12v rail is unusually low. What does your BIOS tell you?

Thanks, jaguar. I'm fairly sure this PSU is just fine. In the BIOS under PC Health Status, it says 12v...OK. SO I assume this software test was wrong. Is there any other way to check (somewhere else in the BIOS or some other software)?

As for my problem, I have relaxed my memory timings from 444 12 to 555 15 and haven't had a problem since, so I think that was the problem. Before that I tried upping the GPU fan and even at 60% (the GPU was at about 55º under load) I was still having problems.