System randomly crashes

[exp(x)]

Ever since I built my computer several months ago, I have been having random crashes. I used WinXP Pro for a month or two, and that would just suddenly reboot. Now I'm using Ubuntu linux (the 64-bit version), and that just locks up.

 

I've run memtest dozens of cycles and have always passed, so I don't think that's the problem even though I have gathered from these forums that DFI motherboards don't get along with Corsair RAM. I know my hard drives aren't the problem because I tried an old IDE disk I had lying around, and the computer still crashed. I don't think the video card is the problem because I had the first one replaced thinking it was the problem (WinXP used to give errors related to the video card after some of the random crashes). I'm not overheating, so that's not the problem, and I highly doubt my DVD or floppy drives are the problem.

 

That leaves the power supply; I was just reading about how the motherboard needs a native 24-pin PSU. Is this my problem? If it is, could there have been any permanent damage done to my mobo? I would hate to spend money on a new PSU just to have the crashes still. In fact, if that happened I would probably take a baseball bat to my system. This has really gotten on my nerves after all these months.

 

Also, I don't know if this is related, but I get a bios error message (ehci_hcd 0000:00:02.1: BIOS handoff failed (160, 1010001)) when linux is starting up that doesn't appear if I disable USB2.0 support in the bios. Would this indicate something is wrong with my board or is this a known issue?

[SaFrOuT]

i would try to update the bios to 704-2BTA

try another cpu since the Tt psu are known to have crappy rails under load

 

also i would try to run prime for more than 4 hours and then run a benchmakr prg like 3DMARk03 and leave it looping for more than a couple of hours

 

it maybe a lot of things that is why we need to test ur full system not only the ram using memtest

[TeddyDigBick]

hi....

every time i played my games it would do this same thing.crash after crash.i fixed it be turning down the settings in my game.BAM...it stoped.i thought it was my ram.my psu.heat.every thing but my settuings in my game..o..i for got it did it to me when i was surfing the internet to. later

[Critter]

exp(x),

 

My corsair TWINX1024-3200XLPT 3208 v. 1.2 would pass memtest till the cows came home. I still had freezes. Changing some timings fixed the problem. This was on a 20 pin PSU, which I have since replaced to run within spec.

 

My bios is below and the timings I used for the new bios are found in the "Stock Speed Database".

 

Good luck.

[exp(x)]

also i would try to run prime for more than 4 hours and then run a benchmakr prg like 3DMARk03 and leave it looping for more than a couple of hours

I ran prime for 24 hours without any failures or warnings. I can't run 3dmark03 because I don't use Windows anymore.

 

My bios is below and the timings I used for the new bios are found in the "Stock Speed Database".

Thanks, I'll try those settings and keep my fingers crossed.

[exp(x)]

Unfortunately, the system is still crashing despite changing the bios settings.

[Sharp]

Hello,

 

Is your memory here?

http://www.dfi-street.com/forum/showthread.php?t=25447

If so, use the settings that Angry suggests.

[exp(x)]

Yes, I did that.

[exp(x)]

Hi, sorry to bump this thread, but I am still having problems even after upgrading my PSU. Lately my crashes have been happening after only around 15 minutes to a half-hour. Just to summarize my situation, I'm going to list what I know is not the problem:

 

• Western Digital Raptor 36.7GB SATA (2x in soft-RAID)
  (these are not the problem because I still had crashes when I tested a single IDE drive instead)
  
• BenQ DVD Burner DW1620
  (crashes even when this isn't plugged in anymore)
  
• NEC Black 1.44MB 3.5" Internal Floppy Drive
  (crashes even when this isn't plugged in anymore)
  
• OCZ Powerstream 520W Power Supply
  (replacing my 480W PSU with this didn't help)
  
• Thermaltake VENUS 12 80mm Ball Cooling Fan/Heatsink
  (I don't OC, so this keeps the CPU cool enough)
  
• Lian Li PC-61 Case
  (crashes even when I unplug the front USB and switch/led connectors
  

 

Here's what could be the problem still:

 

• LANPARTY UT nF4 SLI-DR (704-2BTA)
  
• AMD Athlon 64 3500+ ClawHammer
  
• CORSAIR XMS 1GB (2 x 512MB) TWINX1024-3200XLPT
  
• ASUS Geforce 6600 256MB PCI-E
  

 

I was going to put the video card in the first list because my system still crashes after replacing the first one, but I suppose it could be a hardware incompatibility or overheating; I have no other card to test this theory with. I'm pretty sure the RAM and CPU are fine because I ran 48+ hours in memtest and mprime with no errors. So that pretty much leaves the motherboard as the main suspect. I don't want to RMA a good board like I did with my video card, though. What are your thoughts? Is there anything else I should try? Any help is appreciated.

[wevsspot]

Enter BIOS.

 

Change PCIe frequency from 101 (you might also try 102)

 

Up your chipset voltage to 1.6v

 

What driver set are you using on your v/c?

[L3thal Dream]

dude that linux os is garbage period. it eats ur computer alive! get rid of it as fast as possible. if u wanna use linux then buy suse

[exp(x)]

Enter BIOS.

 

Change PCIe frequency from 101 (you might also try 102)

 

Up your chipset voltage to 1.6v

 

What driver set are you using on your v/c?

Thanks for the reply; I'll try your suggestions. I am using the 1.0-7676 AMD64 driver from nvidia's website. I see that a new version (1.0-8174) just came out recently, so I'll switch to that now.

 

dude that linux os is garbage period. it eats ur computer alive! get rid of it as fast as possible. if u wanna use linux then buy suse

Your incredibly well-worded advise has left me completely speechless.