New Gaming PC crashes when I load games

[DeegansTroop]

Hey guys thanks for taking the time to check out my post. I recently built my first computer and everything seemed to work fine. I got it running with Windows 7 RC downloaded all the drivers and installed a few games. The computer works fine when I start it up and I can run programs like CPU-Z and Internet Explorer without a problem. I can leave it running for several hours and all is well. The problem is that the computer will completly shut down and reboot when I run certain programs. The first is Windows Meida Center, when I start it I see the loading screen for a few seconds then bam it shuts off. It also occurs when I play video games. I have a few Steam games loaded like CSS, TF2, DODS and I can start all these games and even get into a server but as soon as I finishing picking my team and class, BAM it crashes. I am not sure what is causing the problem. I've reinstalled Windows, removed my memory and tried 1, 2, 3, and 6 stick configurations. I've swaped the video cards and tried running one at a time. I've run MemTest86 and Windows memory diagnostics. I've checked the Event Viewer to the best of my ability but it doesn't look like anything is showing up thats causing the crash. I've played with different refresh rates on the monitor from 65 to 120 hz. I've check all the temps and they seem to be fine and nothing is getting super hot. I feel like I've tried everything but nothing has seemed to help. Please let me know if you guys have any suggestions.

System:
Windows 7 64bit RC
Intel Core I-7 920 (not overclocked)
Asus P6T Deluxe V2 Motherboard (supports 12g 1600memory)
Corsair I-7 1333 Dominator Ram 6 x 2048 1600mhz 8-8-8-20
Ultra Chill-Tec thermal electric cooler
Ultra X3 1600w PSU (running at 1200w with standard plug)
Two EVGA GTX285 FTW 1024mbs in SLI configuration
Seagate Barracuda 1.5TB HDD 7200rpm
LG DVD-RW Disk Drive
CoolerMaster HAF932 case with standard fans
Viewsonic 22" 3D Monitor
Nvidia Steroscopic 3D Vision Software and Hardware
NZXT Fan Controller
Logitec Ergonomic Keyboard
Razor Diamondback Mouse

 

[Outlander_04]

check the heat sinks

try with single gfx cards , but try both

 

[sirkillalot]

did you try playing under windows xp? let me know
also try a older graphic driver 185. or something like that
also is your bios up to date? it seems more of a software problem than a hardware 1

 

[selea]

First try a CPU/RAM stressing program as Prime95 or Intel Burn test. If the PC pass that use a benchmark also for the GPU, as PCMARK Vantage or ATi Tool Overclocking tester.

If the PC halts when using these last benchmarks it's a problem with a GPU and/or drivers or also of the PSU.

You have to go with steps in resolving these types of problems. You have to check components one by one with different tests to acknowledge what's wrong.

 

[Twoboxer]

Sounds like you know what you are doing, but please forgive me backing up a bit to make sure we don't skip past something:

1) You say nothing gets super hot. Did you use CoreTemp, RealTemp, or CPUID Hardware Monitor? Either way, can you give some numbers for temps as you near the time you crash?

2) If you are running through a UPS, please take it out of the power stream for a bit. You seem to be able to get it to crash almost at will. Does the behaviour change when the UPS is not there?

3) The only review I can find on the 1600W psu is that +12V was low (11.97V) but rock stable. That's certainly within spec. But I can't help suspecting the psu. We'll come back to that.

4) I saw memtest results were OK. No mention of Prime95. Have you stressed with Prime95 for an at least an hour with "Detect rounding errors" checked? My experiences here and at home says Prime95 will catch memory problems missed by memtest.

5) As Outlandr said, have you tried with only one vid card installed? This would also reduce power requirements. Then try the other.

6) Do you have a second psu capable of driving this system with one vid card? Don't know the cooler requirements, but 500W oughta do it.

7) Speaking of cooling . . . if you have a second psu, is it possible to power that chiller from a second psu (with the second psu touching this PCs case in some way for common ground)? Seems like you're crashing just about when it might be calling for cooling.

 

[DeegansTroop]

Thanks for all the feedback guys. I am using CPUID HW Monitor to check temps CPU stays in the 30-40 range idle up in the 50s under stress. The GPU's are around 50's idle and up to the 70's under stress. I dont' have the comp hooked up to a UPS its going straight into the wall outlet right now. Im not sure what you are trying to get to with the number three suggestion. Memtest results were okay and I also ran Prime95 with the "detect rounding errors" checked. I did the torture test and others but all tests came out sucessful. I have already tried running with just one video card and just one stick of ram. Same problems. I do have an Ultra 500w PSU in another computer but im not in the mood to take that apart at the moment. I did however get some interesting results when I ran PCmark05. It ran all tests but it seems to fail at the Video Card Memory Test somthing about 64 lines not being read. I tried going online to check results but I can't seem to find anthing that tells me what happened. I also forgot I bought PCMark06 a long time ago so Im dling that and Im going to see what happens.

 

[DeegansTroop]

Ok interesting update. Im running ATI Tool and when I scan for artifacts it takes about 10 -15 seconds and then it does the same thing. The whole comp turns off all the leds in the comp turn off and then it reboots. The first time ran the test and it crashed I got an interesting view of windows in 4 quadrants on my monitor. I promply restarted and problem solved. I also hear a wierd sound coming from the case. My PSU is bottom mounted and right next to my SLI cards so Im not sure which is making the noise. Its hard to discribe but its not very loud and also Im not sure If Im hearing the GPU fans fully kick into gear I've watched videos and other ppls computers seem to be a lot noisier. Am I just going crazy? Im thinking about taking it into a computer shop or calling EVGA tech support. Lemme know what you guys think. Thanks

 

[Twoboxer]

Thanks for the complete reply. The only thing that may be missing is running with one video card - with each video card? It would be odd to have two bad video cards, eh?

If neither vid card runs successfully by itself . . . drivers, psu, and mobo are left.

I know its a PITA, but trying with one vid card with that other psu - and powering that cooler (supposedly 50W draw at max) from a second psu - would allow you to conclude "2 bad vid cards or mobo".

 

[DeegansTroop]

OK well this morning I removed the video card that was in the primary slot and moved the card from the secondary slot up to primary. I ran the ATI tool for about 7 mins and the card got to 81 degrees without crashing. So I turned of the comp and swaped cards putting the original back in all by itself. Again I ran ATI tool and same result. Got to 82 degrees after about 7 mins no crashing. Then I thought about my PSU, It is modular and has only two 6 pin PCI slots. Each of my GTX cards requires two 6 pin connectors and just having one pluged in means that my 6 pin slots are fully populated. The other card I had to improvise and I used the 8pin to 6 pin converters. I was worried that might be one of my problems but I got the same results with ATI tool no crashing after 10min. I jumped on to one of my Steam games Day of Defeat: Source and was acually able to play an entire map no problem. Then I started Crysis warhead with one video card and the 8 pin to 6 pin connectors and it crashed before I even got through the advertisements. So I restarted, powered down, then pluged in the straight 6 pin connectors. Loaded Crysis got all the way to the menu and pressed Single player and CRASH. I will try and take out that 500w PSU from my other comp but im not sure what to do. Any help?

 

[Twoboxer]

Well, that was a fair amount of testing - but unless I missed something not enough keeping the test approach consistant.

Can you run Crysis with Card "A" and 6- or 8-pin? No.
Can you run Crysis with Card "B" and 6- or 8-pin? Unknown.

If the answer for Card "B" is "Yes", the problem is most likely Card "A".
If the answer for Card "B" is also "No", the problem is most likely psu.

Keep going, you're on a good path to a solution now.

That 50W cooler still bother me though lol.

 

[DeegansTroop]

Twoboxer thank you for all the help I really appriciate everyones responses. Ok so I installed card b with 6 pin connector and it still crashed. Then I pluged in the 8 pin connector and I was acually able to play for a couple of minutes. It is the first time thats happened since I put this computer together. I was acually playing for about 10 mins and thought I was in the clear. Then it crashed again. So I dont think its the video cards is there anything else you might suggest

 

[Twoboxer]

OK, the idea you could have two bad video cards is unacceptable.

You had that funny noise from the bottom area . . . psu's can make noises (not good), video cards do not.

Its time to try that other psu with a single graphics card. I think you have a bad psu.

 

[DeegansTroop]

Well its been awhile since I've worked on the computer. I finally got around to working on it today and Im still having problems. This thread is pretty old so Im probably gonna start a new thread if nobody replies. I started by starting the computer and playing TF2 untill my comp crashed. Then I removed the PSU from my old computer its an Ultra 500w PSU. I also removed one GPU since I didn't want to overload it and I've already eliminated them as the cause. I then started playing games and again it still crashed Twoboxer was concerned about the Thermal Electric Cooler so I removed that and then installed the stock I7 cooler. Ran computer again and still its crashing. So far I've eliminated the memory, graphics cards, power supply, and CPU cooler. Unfortunaly I dont have a spare I7 or 1366 Mobo lying around so Im not sure where to go from here. Any help? Much appriciated everyone Thanks!

 

[Twoboxer]

One more quick test - if your case has a reset switch, remove the jumper for the reset switch from the motherboard. Assuming nothing changes . . .

Whether or not the motherboard is the issue, I think you've got enough test data to RMA the motherboard.

1. List Configuration.

2. You have crashes in Media Center and Many if not all games. In this case, "crash" is defined as the PC powering down and then rebooting.

3. You have eliminated cpu heat as the cause using CPUID Hardware Monitor to show cpu temps 30-40 at idle and 50s under stress (that sounds low btw, maybe not enough stress yet. Or maybe you have a great cooler). Also not Vidcard heat with temps 50s idle and 70s under stress (again pretty low stress temps, maybe not enough stress).

4. You experience the same results even when running with:

- 1, 2, 3, and 6 stick configurations,
- 2xVidcard as well as either Vidcard run alone,
- 2 different psus (1600W and 500W) running a single Vidcard,

This seems to leave only a very odd short (mobo to case somewhere), cpu, or motherboard. IOW, this leaves the mobo as the most likely issue.

 

[DeegansTroop]

Hey Twoboxer thats a clever idea with the reset jumper. I gave it a shot but it wasn't the problem. I've emailed ASUS a service request for the mobo to see what they say and I should be getting an email from them soon. I'll let you know how it turns out.

 

[Twoboxer]

Please do.