Jump to content

System at stock not stable


Recommended Posts

  • Replies 86
  • Created
  • Last Reply

Top Posters In This Topic

Idle values

5V idle

5.02V

 

12V idle

12.24V

 

3.3V idle

3.31V

 

Load values:

First value is during test 1-2, 2nd value is during test 3, 3rd is during test 4 where worker #3 failed(round error) and 4th is test 5

 

12V molex yellow 

 

12.18V

12.27V/12.28V(alternating between two values)

12.26V

12.27V/12.28V

 

5V molex red

 

5V

5V

5V

4.99/5.00V

 

3.3V measured from 24 atx orange pins 

3.30V

3.30V

3.30V

3.30V

 

Here's the requested 1.54 CPUz pics, alongside 1.67 and in some HWMonitor at 1.25V in bios for CPU, 1.327V for chipset.

http://s604.photobucket.com/user/Dan_The_Gamer/embed/slideshow/cpuz154pics

Edited by DanTheGamer11

Share this post


Link to post
Share on other sites

Power supply tests look good.

 

What is your HT Link speed set to in BIOS?

 

Also, lets back off of Prime stress testing for just a bit and try OCCT large data test.  I'd like to see if failure is repeatable.  Let it run for 30 minutes if it doesn't fail, but watch CPU temps.

 

You can get it here if you don't already have it.

 

http://www.ocbase.com/

Share this post


Link to post
Share on other sites

I wouldn't worry about the 12V.  It's well within the +/- 5% ATX spec.

 

I was a little befuddled by the HT Link speed shown in CPUz @ 1000, because looking back at my CPUz shots mine read 2000 (it is probably just a difference in detection / reporting though).

 

Let me know what happens with OCCT test.  

 

Honestly your CPU shouldn't have any problems running stock frequency / w stock vcore, RAM settings look OK, and as you're not running high speed memory even the inability to bump NB and CPU-NB voltages shouldn't impact stability.

 

At this point I'm running out of ideas and might have to start considering that you've got a bum cpu, motherboard or possibly even memory.

 

Did you ever get around to running Memtest 86+ on each stick?

Share this post


Link to post
Share on other sites

OCCT?

 

Yeah, I ran Memtest86+ and Memtest at 800Mhz and 1066Mhz, no errors. One weird thing I noticed was that when I have either memory stick in socket two it shows up as DDR400 in Memtest86+... odd.

 

I've ordered a stick of that Crucial Smart Ballistix RAM, has a temp sensor so I can monitor that among other things. If it doesn't solve the issue I think I might return both the CPU and mobo and get a new FX 6300 and some AM3+ DDR3 only mobo, which would have the proper HT rate :/

 

I think PCs hate me :(

Share this post


Link to post
Share on other sites

OCCT?

 

Yeah, I ran Memtest86+ and Memtest at 800Mhz and 1066Mhz, no errors. One weird thing I noticed was that when I have either memory stick in socket two it shows up as DDR400 in Memtest86+... odd.

 

I've ordered a stick of that Crucial Smart Ballistix RAM, has a temp sensor so I can monitor that among other things. If it doesn't solve the issue I think I might return both the CPU and mobo and get a new FX 6300 and some AM3+ DDR3 only mobo, which would have the proper HT rate :/

 

I think PCs hate me :(

Memtest sometimes reads the wrong speeds, don't pay it any mind.

 

Does the computer crash in anything other than extreme torture tests like Prime95?? If not and the RAM doesn't fix it...I'd just live with it. Unless it's crashing on games or productive tasks then I see reason to worry

Share this post


Link to post
Share on other sites

It BSODd in the weekend when I wanted to restart the PC after gaming.

I used Prime in the first place to see if the PC was unstable as I was getting fps drop spikes in Crysis 3... Which in the end I found to be the C State setting by looking at CPUZ and seeing the clock.speeds drop to 800Mhz every so often under load but I also discovered that the PC wasn't too stable since I was getting BSODs every Prime run. Though now I only get failure workers, seems to have gotten better(latest Prime still BSODs every few tests tho). I also don't know why the PC freezes when I run prime with one stick of ram

 

When the ram arrives, if the problem is fixed do you think the problem is memory controller or mobo?

Edited by DanTheGamer11

Share this post


Link to post
Share on other sites

Sounds like the PSU to me. Not much else will cause a hard shutdown (versus a hang or reboot).

 

I wouldn't expect a multimeter to catch a sag under load. Do you know anyone with a PSU you could use for testing purposes?

Share this post


Link to post
Share on other sites

Sounds like the PSU to me. Not much else will cause a hard shutdown (versus a hang or reboot).

 

I wouldn't expect a multimeter to catch a sag under load. Do you know anyone with a PSU you could use for testing purposes?

It's certainly possible, I remember mine kicked it...I had the best luck in the world and was building up my cousin's machine and thought...."hey why don't I try her PSU in my machine" and of when I did, it ran fine.

 

I then tried my PSU in her computer and it ran but had random errors. I'm guessing since her machine (X4 620 and GTX460) versus mine (Q9450 and 4870x2+8800GT) was more efficient, my PSU simply couldn't run the kinda loads my system had.

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×
×
  • Create New...