Random Restarts -- I'm desperate

Joined
May 12, 2007
Messages
60
I just built a computer about a week ago; it's working great except it restarts "randomly".

Firstly, here are my system specs:

- Western Digital Caviar SE WD2000JS 200GB 7200 RPM SATA 3.0Gb/s Hard Drive
- Intel Core 2 Duo E4300 Allendale 1.8GHz LGA 775 Processor Model BX80557E4300
- G.SKILL 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400) Dual Channel Kit Desktop Memory Model F2-6400CL5D-2GBNQ
- LITE-ON Combo Drive Black SATA Model SHC-52S7K-05
- GIGABYTE GA-965P-DS3 LGA 775 Intel P965 Express ATX Intel Motherboard
- EVGA 256-P2-N624-AR GeForce 7900GS 256MB GDDR3 PCI Express x16 KO Video Card
- OCZ GameXStream OCZ600GXSSLI ATX12V 600W Power Supply

So far, I've noticed two types of restarts: The first case is when I'm playing Oblivion. Within minutes of playing the game, it reboots. This has been consistent and has happened every time I attempt to play the game. The second type of reboot occurs randomly; whether I'm playing a game or browsing the internet. It usually occurs within a few hours of being on the computer.

The restarts occur without the BSOD or any other error message. I disabled auto-restart under the system failure setting. I also can't find anything relating to the reboots in the event log.

So far I've done the following in an attempt to pinpoint the problem:

- Took off side panel and put large room fan to cool the system. It still rebooted.
- Checked temperatures using Everest; everything is cool (usually 30-40 C).
- Bought a new PSU and graphics card. Problem still persisted.
- Checked mobo for swollen or leaking capacitors; couldn't find any.
- Tested RAM with MemTest for 2 hours (2.5 passes) with no errors.
- Took out one stick of ram; problem still occurred. Repeated with the other stick.
- Placed RAM in different slots.
- CHKDSK to see if there were any hard drive problems; none showed up.
- Torture test with Orthos for an hour. Temperatures remained reasonable; CPU at about 50 C

However, after doing all that, the problem still remains. :(

After all those tests, I don't know what to think. Could it still be the RAM? Could the board be short circuiting on the risers? Is the chipset overheating (the heatsink feels quite hot)? Is there anything I can do in the BIOS to help solve the problem (e.g. raise voltages)?

Anyway, I'm exhausted from this. If anyone here has any ideas, please post; it's greatly appreciated. Thanks.

edit: Oh yeah, I haven't overclocked anything.
 
I had this problem a few years back, I did the same I replaced just about every part in that machine I finally narrowed the problem to a bad video card which was also an EVGA :mad: (6600 gt) try another video card, if you can get a friend to lend you a card to test with MADE by a different company that'd be great.

this was the most frustrating problem I ever face cuz it would do it even outside of a game so i thought it was bad ram, then with new ram in I switch psu 2x! and even swapped cpus and mother boards as well.

Good luck.
 
I had this problem a few years back, I did the same I replaced just about every part in that machine I finally narrowed the problem to a bad video card which was also an EVGA :mad: (6600 gt) try another video card, if you can get a friend to lend you a card to test with MADE by a different company that'd be great.

this was the most frustrating problem I ever face cuz it would do it even outside of a game so i thought it was bad ram, then with new ram in I switch psu 2x! and even swapped cpus and mother boards as well.

Good luck.

I had a Gigabyte 8600 GT before this one and the computer was still restarting. I got rid of it because I thought it could be overheating (I didn't trust that silent piping). But even before I bought the EVGA, I tried my brother's card and it still restarted. Could it be a bad PCI slot? That could explain why the problem didn't go away when I got a new graphics card.
 
install/replace the chipset fan with a good ball bearing fan, update chipset drivers and flash BIOS to latest revision.
 
install/replace the chipset fan with a good ball bearing fan, update chipset drivers and flash BIOS to latest revision.

I updated the chipset drivers and I'll order a chipset fan (probably a new heatsink as well). As for the BIOS, I'm not sure how I'm going to flash to the latest version as I don't have a floppy drive. I was reading the manual and they listed some other methods (USB stick / HDD, but it requires a FAT system) and a way to update it through windows. However, I'm not sure how well that will work.
 
Aye, sounds like a driver or chipset issue. Do you know how to change the Windows setting so it does not automatically reboot at BSOD? If so, do that and let us know what message appears on the BSOD, specifically the text in bold white coloring.
 
Aye, sounds like a driver or chipset issue. Do you know how to change the Windows setting so it does not automatically reboot at BSOD? If so, do that and let us know what message appears on the BSOD, specifically the text in bold white coloring.

Already did that. I don't get a BSOD or any other error messages. It just restarts; there's no lag, it doesn't freeze up, it just reboots.
 
Already did that. I don't get a BSOD or any other error messages. It just restarts; there's no lag, it doesn't freeze up, it just reboots.

In that case I'm going to vote hardware issue. If the OS doesn't get the chance to toss an error, it's likely the northbridge overheating or a faulty motherboard.
 
In that case I'm going to vote hardware issue. If the OS doesn't get the chance to toss an error, it's likely the northbridge overheating or a faulty motherboard.

I just ordered a $37.00 copper heatsink with 40mm fan (as opposed to the aluminum stock w/out fan). That should hopefully solve the problem (unless it's a faulty mobo, in which case, I'll cry because I don't want to have to rebuild everything).
 
Sorry for the rather large bump, but I RMA'd the motherboard and got a replacement (same model). Anyway, after getting my PC back up and running again, I notice that my keyboard isn't working (the keyboard used the purple keyboard input; whatever it's called). So I grab my USB keyboard and that works.

Anyway, I ignore that issue and try Oblivion. Boom 1 minute into the game and it restarts. I also try using Photoshop (I've noticed it restarts within 5-10 minutes when using PS). Sure enough, it restarts. Still no BSOD, nothing.

So I put in Ultimate Boot disk and test RAM. Did 4 passes with no errors. Used one of the HDD diagnostic programs (forget which) and everything came up clean. Next, I try the Mersenne Prime Test.

Here's where things get interesting. After about an hour or so, I get this:

Code:
Test 5, 560000 Lucas-Lehmer iterations of M200705 using 10k FFT length.
CPU 0: Machine Check Exception: 0000000000000004
Bank 1: b200000000000175
Bank 2: b20000300c000e0f
Kernel Panic: CPU context corrupt

I do a search here on the forums (God, I love this place) and sure enough, I found a lonely thread pertaining to Core 2 duo's and my error:

http://hardforum.com/showthread.php?t=1185528&highlight=kernel+panic

So could this be the little bugger which has been driving me insane the past month? The issue is extremely rare, but I have replaced the mobo, gpu, psu, tried each stick of RAM individual, checked temps, done all sorts of tests, what else could it be? The symptoms are similar, especially the keyboard problem.

Anyway, I'm gonna go to Gigabyte's website and see if they have a new BIOS version which solves the problem. Hopefully this is it; my RMA period is up and I don't feel like spending more money on this. I'll post back here tomorrow or Friday to let everyone know what happens.
 
...the first thing, /first/ thing one should do with a new board is flash the latest BIOS. Always.
 
did you not try this before RMA???

I put a big 2 and a half foot room fan on high in front of the mobo, aiming it at the North Bridge. It still restarted. I updated the chipset drivers too. I didn't flash the bios because I was afraid it would restart during the bios update and then I would have been really screwed.

However, know that I know about that Intel error, I'm going to flash the bios and pray that it doesn't reboot during the process.
 
Flash from DOS via a CDR -- most motherboard co's have tiny ISOs to download for this purpose anymore, not sure about Gigabyte, or if you have a floppy somewhere, use that obviously. Less things running, less likelihood of hitting that bad register or whatever it is.
 
I put a big 2 and a half foot room fan on high in front of the mobo, aiming it at the North Bridge. It still restarted. I updated the chipset drivers too. I didn't flash the bios because I was afraid it would restart during the bios update and then I would have been really screwed.

However, know that I know about that Intel error, I'm going to flash the bios and pray that it doesn't reboot during the process.

Yuo definitely get an A+ for effort - you've tried just about everything to get your system up and running stable!

I vote that the BIOS update will help. BIOS updates are usually pretty safe - and a good thing to try with random reboots like you are encountering. A lot of motherboards have a failsafe in case of a bad BIOS update. You can also enter the BIOS and leave your system on overnight to see if it reboots before updating if you are concerned.
 
Thanks for the advice everyone. I just checked the Gigabyte BIOS page and there is a new version. However, the new version doesn't mention anything about a CPU microcode fix. Nonetheless, tomorrow I'll update it and see what happens.

Thanks again everyone, your help has been much appreciated. :D
 
For what its worth, I have had memory that will pass memtest for a few hours before it fails, If the BIOS flash doesn't fix your problem, I would recommend running memtest for a bit longer (~24hours or something). If you tried both sticks individually, I kinda doubt both of them are bad, but its worth a shot.

Also, could it be a problem with the power in your area?
 
What do you have your memory voltage set at? That memory is not "happy" at 1.8V
 
Currently my RAM voltage is set at the default (1.8v). On the last mobo I set it to 1.9v and the restarts still occurred.

I just flashed the BIOS and it installed fine. Now I'm just installing some updated realtek audio drivers. After that, I'm going to try Oblivion/Photoshop and see if the restarts still occur. If they do occur, I'll test the RAM longer like ilikecake said.

If all that fails, I'll probably call Intel and complain until I get a replacement CPU.

Ok... just was installing the realtek drivers and got a BSOD stop error. Here it is:

*** STOP: 0x0000007e (0xc0000005, 0xB660847B, 0xBACFB418, 0xBACFB114)
*** RtkHDAud.sys - Address B660847B base at B61EF000, Datestamp 44e16c86

This PC is messed up. I have a feeling I'm going to see a format in my future.

edit: Tried installing it again and it worked.

edit2: Just tried Oblivion and amazingly enough, It didn't restart. I only played for 5-10 minutes though.

edit3: Going on about 20 minutes now.

edit4: Stopped playing Oblivion and booted up PS. Created a 4000 x 4000 pixel document at 300 dpi and been messing around with the brushes. There is a lag (even at very small canvas sizes) but I heard that's due to a bug with PS and intel processors. Anyway, it looks like everything is working good. I'm just going to assume that the stop error was just a fluke and I'll continue to monitor and test my computer.

edit5: Keyboard now works. On an unrelated note, I also finally fixed a problem with BF2 and it works now. All I need now is to buy a ball-bearing fan and plop it on my north bridge.
 
Back
Top