2600K unstable @ stock clock with Prime95????

R

rosh

Limp Gawd
Joined
Jul 9, 2009
Messages
232
Just asemmbled my rig

1. 2600k - stock cooler
2. Asus P8P67 deluxe
3. 8GB Gskil ripjaw
4. Sapphire 4870 X 2
4. Corsair 850AX (850w)
5. Windows XP/7 32bit (install 64bit later due to memory limitation)

Just under regular settings my cpu temp - 34C (according to AI 2). Ran few benchmark and had no any issue, (but temp went up to 85C some cases, but was stable)

Ran prime95 (26.5v) with the first two options. (Small FFTs maximum fpu stress..., In-place large FTs...) and the system was stable. But when i use the 3rd option (Blend - test some of everything) the system restarts within just a couple of seconds, sometimes 2sec, 5 sec, 20sec was maximum)

Checked with another AX850 PSU and have exactly the same issue. So its not the PSU

Any thoughts?
 
Check the c:\windows\minidump files to see if there is a crash dump.
If there is, use microsoft debugging tools for windows, and get the proper symbol package (you may be able to download on demand with the srv* path or is it *srv), and use !analyze -v in the commandline.

Then you should see the cause of the BSOD.

However if you're crashing at stock speeds, there are only two possible reasons for this to happen, and they are not temp related:

1) RAM timings or RAM speeds may be problematic for your board: try 1333 if you're at 16.66. Or RAM may be defective

2) Defective IMC on the CPU.

Usually one or the other.
The fact that small FFT passes, while blend crashes *instantly* shows that the cpu cores themselves are fine, but something else is amiss.
 
Blend tests memory so have you run MEMTEST? Also, double-check your VDIMM and timings if you find an error.
 
I would definitely guess a memory issue over a brand new processor without seeing the logs.
 
Guys, thanks so much for giving your ideas in this frustrated situation.

I ran prime95 with one 4GB stick and had no any issue. However if i remove that and out the other on the same slot, im getting that restart/stuck issue again. So im thinking that one of the stick is bad. Am i correct?

Is there any better memory test program working under windows?. Currently i dont have a floppy or Sata CD drive to test memtest

ps. Timing of memory looks fine i guess, its working @ 1600 and the time thats on the memory specs
 
rosh said:
Guys, thanks so much for giving your ideas in this frustrated situation.

I ran prime95 with one 4GB stick and had no any issue. However if i remove that and out the other on the same slot, im getting that restart/stuck issue again. So im thinking that one of the stick is bad. Am i correct?

Is there any better memory test program working under windows?. Currently i dont have a floppy or Sata CD drive to test memtest
Click to expand...

MEMTEST can also run off of a USB stick but I think you've narrowed down the problem to one stick already. Good work.
 
rosh said:
Guys, thanks so much for giving your ideas in this frustrated situation.

I ran prime95 with one 4GB stick and had no any issue. However if i remove that and out the other on the same slot, im getting that restart/stuck issue again. So im thinking that one of the stick is bad. Am i correct?

Is there any better memory test program working under windows?. Currently i dont have a floppy or Sata CD drive to test memtest

ps. Timing of memory looks fine i guess, its working @ 1600 and the time thats on the memory specs
Click to expand...

You can try lowering the speed to 1333 and see if that other stick works. 1600 on a 4GB stick can be tricky (although if it's rated for 1600, then send it back).
 
Just go ahead and RMA. RAM that doesn't work at advertised speeds is defective RAM. Glad you found out the cause.
 
Sorry, I've got nothing to add to the thread except for a bit of humor. I thought the title read: "2600K unstoppable @ stock clock with Prime95????" Now that I've had my morning chuckle, time to get to work...:rolleyes:
 
I had the same issue with my MSI P67A-GD55. Would crash on stock clocks and error in stability programs. I found out that it was an issue with the "auto" voltage settings in UEFI/BIOS, bumped up the CPU PLL voltage a notch or two from the "stock" voltage, and it ran stable. Same voltage is keeping me stable @ 4.4 GHz for the PLL. Your issue may be the same, or it may be another voltage, I don't trust auto voltages too much. Hope this helps.
 
`br4dz- said:
I had the same issue with my MSI P67A-GD55. Would crash on stock clocks and error in stability programs. I found out that it was an issue with the "auto" voltage settings in UEFI/BIOS, bumped up the CPU PLL voltage a notch or two from the "stock" voltage, and it ran stable. Same voltage is keeping me stable @ 4.4 GHz for the PLL. Your issue may be the same, or it may be another voltage, I don't trust auto voltages too much. Hope this helps.
Click to expand...

What does the CPU PLL voltage do anyways?
 
Got a new Ripjaw kit and works the system perfectly. No more restarts/crashes.

Thanks for eveybody for helping me. Greatly appriciated! :)
 
You must log in or register to reply here.
Back
Top