Stability issue

[Kewlx25]

I have a peculiar situation with my computer being unstable.

First off, here's my original stats: Dell i7-920 stock, 4GB DDR3, ATI4850, 320watt PSU, WD 640GB Blue (MSDN Win7 Ult)

I upgraded my PSU a year ago to an Antec NEW TP750.

Just this tax season I decided to get a new Case(Antec 900), Video Card(XFX 6950 2GB), and memory(Patriot Viper II Sector 7 1600mhz 9-9-9-24).

Quick note: Dell mobo has no OC features, so memory is running at 1066 1.5v

All seemed well, but then my computer BSOD'd with a DirectX driver. I shrugged it off as the unstable 11.1a drivers on a new card. Then later another BSOD, but this time in ntfs.sys. First thing that came to mind... ohh crap, memory is bad. Mem test. 48 hours later and 35 passes, no errors. Next thought, CPU issue. Prime95 on CPU burn in for 24 hours full bore on 8 threads. No errors. Corrupted System files ma'b?. Checkdsk full scan for bad sectors. Nothing.

No single test will give me errors. Not only that, the computer will run solid for a few days at a time. I have BOINC that I run at night for distributed, I run GPU data crunching clients, I play hours upon hours of Bad Company2 with graphics cranked. The computer will run fine and then all of a sudden, everything goes bad at once. Games will act funny or close out unexpectedly with no error, Chrome tabs will crash, and it will keep doing this until a BSOD finally takes it down or I shutdown. If I just restart the computer, the errors will continue, but once I do a full power down, everything goes back to working until the next time.

WTF? I checked temps and my CPU runs only 35c during games or 40c max with prime95, the GPU runs only 48c even during BC2 or 3DMark 2011.

I've been building custom computer for other and myself since I was 11, and now I'm almost 28. I've worked as IT for 2.5 year fixing an average of 7 computers per day for a University of 800 faculty computers(I didn't do lab computers).

For the life of me, I cannot figure this out. The only thing I can think of is this PoS Dell mobo can't handle the new memory/video card and it just gives up after a bit.

Thanks for any thoughts

 

[cpatel1987]

I have a peculiar situation with my computer being unstable.

First off, here's my original stats: Dell i7-920 stock, 4GB DDR3, ATI4850, 320watt PSU, WD 640GB Blue (MSDN Win7 Ult)

I upgraded my PSU a year ago to an Antec NEW TP750.

Just this tax season I decided to get a new Case(Antec 900), Video Card(XFX 6950 2GB), and memory(Patriot Viper II Sector 7 1600mhz 9-9-9-24).

Quick note: Dell mobo has no OC features, so memory is running at 1066 1.5v

All seemed well, but then my computer BSOD'd with a DirectX driver. I shrugged it off as the unstable 11.1a drivers on a new card. Then later another BSOD, but this time in ntfs.sys. First thing that came to mind... ohh crap, memory is bad. Mem test. 48 hours later and 35 passes, no errors. Next thought, CPU issue. Prime95 on CPU burn in for 24 hours full bore on 8 threads. No errors. Corrupted System files ma'b?. Checkdsk full scan for bad sectors. Nothing.

No single test will give me errors. Not only that, the computer will run solid for a few days at a time. I have BOINC that I run at night for distributed, I run GPU data crunching clients, I play hours upon hours of Bad Company2 with graphics cranked. The computer will run fine and then all of a sudden, everything goes bad at once. Games will act funny or close out unexpectedly with no error, Chrome tabs will crash, and it will keep doing this until a BSOD finally takes it down or I shutdown. If I just restart the computer, the errors will continue, but once I do a full power down, everything goes back to working until the next time.

WTF? I checked temps and my CPU runs only 35c during games or 40c max with prime95, the GPU runs only 48c even during BC2 or 3DMark 2011.

I've been building custom computer for other and myself since I was 11, and now I'm almost 28. I've worked as IT for 2.5 year fixing an average of 7 computers per day for a University of 800 faculty computers(I didn't do lab computers).

For the life of me, I cannot figure this out. The only thing I can think of is this PoS Dell mobo can't handle the new memory/video card and it just gives up after a bit.

Thanks for any thoughts

I know you'll probably rage for asking this, but are you certain the memory is running at that voltage? Because 1066 memory by default goes to 2.1-2.2v if not influenced by the BIOS in any way, not 1.5v, although I could be wrong about that too. I ran into a similar situation with Ballistix memory. It required 2.0 voltages at its highest clock speed, but was being under supplied by the motherboard BIOS since there was no way for me to adjust the voltage and it just made assumptions by the clock speed. I lowered the clock speed setting of the memory, which I could adjust on my BIOS, which in turn lowered the voltages, and haven't had problems since. I had random failures similar to yours.

 

[Kewlx25]

rawr... /rage or something



Actually, 1.5v is the DDR3 standard and 1.65v is considered the safe high voltage as 10% over is "safe" for nearly everything.

The memory is rated for 1.65v at 1600mhz, but it's only running 1066 and the actual SPD info in the chip itself lists 1066 @ 1.5v with 7-7-7-20-1 timings. Since I can't control any timings/speeds, the mobo just uses what Patriot put in the SPD data.