• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

random blue screens not under load

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

trents

Senior Member
Joined
Dec 27, 2008
A couple of times lately I've gotten blue screens. They come when the computer is idling or doing light tasks such as browsing the internet. The system is stable when running Prime95 blend, however, which I've checked lately.

Here is what the mini crash dump file turns up. I don't know anything about crash dump files and I'm not so sure they are always helpful but if this gives any insight, here it is. Seems to be indicating the cause is a faulty driver if I understand it correctly.
 

Attachments

  • debug.JPG
    debug.JPG
    89.2 KB · Views: 65
Well, I ran 5+ hours and 4+ passes of Memtest86+ on the ram and no errors. I hate this kind of problem, it can be so many things. I did put in a new PSU recently that is giving me lower 12v+ rail voltages than the previous one. The new one is giving me 11.55, still within acceptable range but about .2v less than the old one.
 
"Trents" zip the dump file and attach it to the forum and will take a look. That is a picture and cannot copy and paste any part of it.
 
Hope this works. I just changed the .dmp ext to .zip as I couldn't find a freeware zipper that would recognize the .dmp extension. I reckon you'll have to change it back to view it.
 

Attachments

  • 062912-16239-01.zip
    285.2 KB · Views: 14
RGone said he couldn't open the file after I changed the extension to .zip so I have actually compressed it with 7zip. Here it is again:
 

Attachments

  • 062912-16239-01.zip
    22.9 KB · Views: 14
Could it be the chip maybe undervolting a little too far while idling?

Yeah, I wondered that too. That's why I mentioned the PSU thing and it is true that I also have LLC enabled which can contribute to that. Unfortunately, there is not adjustment to the LLC in my bios. It's either enabled or disabled. I could increase the baseline voltage I guess or lower the frequency.
 
Well, I ran 5+ hours and 4+ passes of Memtest86+ on the ram and no errors. I hate this kind of problem, it can be so many things. I did put in a new PSU recently that is giving me lower 12v+ rail voltages than the previous one. The new one is giving me 11.55, still within acceptable range but about .2v less than the old one.
If it were giving 11.55, your machine would not be running.
 
11.55 is what HWMonitor reports and I do realize software voltage reporting is not especially accurate but HWMonitor was also used to report voltage when using the other PSU which should be at least comparing apples with apples. When using the other PSU HWMonitor was reporting 11.74v I believe for the 12v+ rail, so almost a .2v difference.

And I'm not sure I agree with you that if it were actually putting out 11.55 volts that the system would not be running. I have heard from knowledgeable people a number of times that + or - 5% is acceptable and 11.55 is certainly within that range. 12x95=11.4.
 
+- 5% is ATX Specification, not exactly acceptable.
If you had 11.55v idle, then on load it would most surely dip to 11.4v or lower. Tighter voltage tolerance to 12v like +-2% is acceptable for decent PSUs, most of the time manufacturers tune them to 12.0-12.2v so they never leave 11.8-12.2v when under load, even on cheaper group regulated units.

If I had a PSU giving 11.55v on the 12v rail it would be absolutely unacceptable, would be returned. Not all parts work that low, mostly GPUs and hard drives can be finicky with too low or too high voltages.

Regardless, HWMonitor is wrong if you have an ASUS motherboard. Check your bios ;)
 
What does AI Say? I think the Motherboard software is more accurate than HWMonitor.

Look at the difference with mine, i seriously doubt a 750w Sea Sonic made PSU has that much Vdroop on it.... The test i read on it before i bought it had it in a sweet box pulling 800w + and it refused to budge by more than 1%

I think HWMonitor is good but a little inaccurate for volts.
 

Attachments

  • ai.PNG
    ai.PNG
    44.1 KB · Views: 50
Last edited:
Before I begin using any Monitoring "software", I check with a Fluke DMM. Then I know for real what the power supply voltages are going to be.

But based on what "Frakk" shows, HWMonitor appears to read a little low. So add back the difference and not a pure power supply problem. Even though the newer P/S seems lower even than the one it replaced in +12V HWM reporting.

What does AI Say? I think the Motherboard software is more accurate than HWMonitor.

Look at the difference with mine, i seriously doubt a 750w Sea Sonic made PSU has that much Vdroop on it.... The test i read on it before i bought it had it in a sweet box pulling 800w + and it refused to budge by more than 1%

I think HWMonitor is good but a little inaccurate for volts.
 
Yeah, I wondered that too. That's why I mentioned the PSU thing and it is true that I also have LLC enabled which can contribute to that. Unfortunately, there is not adjustment to the LLC in my bios. It's either enabled or disabled. I could increase the baseline voltage I guess or lower the frequency.
Winner winner, chicken dinner. My guess is you tweaked it a bit too low. Raise your vcore up a notch and see if that goes away. :thup:

Vdroop though is usually a reference to vcore when loaded, not when talk about PSU voltages.. though I suppose they can be interchangeable. That threw me off and had to reread the thread. :p

Vdrop = Set voltage in the bios to windows idle
vdroop = idle voltage in windows to load.
 
"E_D" declares the chicken dinner winner. Let's all go eat. Hehehe. I want white gravy on my smashed taters.


White Gravy? is that chicken stock or something?

BTW... did i say sweet box? i meat sweat box... lol :rofl: :clap: hoi.......:screwy: To much Chicken roast on my mind.

But i will tell you what ternts, i think this Motherboard has its limitations, or perhaps the odd batch do, not only can i not get my CPU over 1.4v... i can not get my GPU over 1.1v without it crashing, there is absolutely no reason why it should not, Sapphire provide Overclocking software and a BIOS switch for 'enthusiast mode' the thing should run @ 1Ghz / 1500Mhz at high fan speeds and i can not get it (completely stable) over what you see in my sig, because as soon as i touch the volts its gone, exactly the same thing with the CPU.
 
Last edited:
BTW... did i say sweet box? i meat sweat box... lol hoi....... To much Chicken roast on my mind.

U shore as London fog haf sumtin onur mind.

Asus M5A97-PRO = at tang gat any slot powur sattings in duh beos? Hiffen hit duz whyunt you raise hit un duh slotz u oozing an I mat eveN disable duh unNot oozed slots. Hit wurth mite takenuh trie.

Chicken? Chicken? I hearem say chickin?
 
What did I just read? I am so confused.

Loosely translated (I think): "If the bios has controls to turn off power to the PCI slots, try disabling the unused ones and see if that puts less drain on the 12+ rail."

I think it was in Cajun. He's from Mississippi.
 
Back