BSOD and reboots

[drmarvin]

I have been having many stability issues with my P4 2.8gig system. I previously thought it was in issue with the PSU, but buying a new Thermaltake 420W PSU didn't help at all.

Then, I thought it might be a problem with my Soyo SY-P4VGA mobo might be the reason, so I grabbed an ASUS P4S800 mobo, and there seems to be slightly different symptoms. The Soyo board always froze the system completely, and the ASUS board either reboots all the time (once Windows is installed) and during installation, many BSOD and Stop Errors.

I have tested the RAM, with no problems, I have no hardware in the system except my video card (AGP ATI Radeon 9000 Pro) and it seems to make it more unstable than and old PCI TNT card. My other AGP card seem to be older and don't fit in the AGP slot on the ASUS board so I can't test that. Does this look like a problem with the CPU? Should I get a new one?

I'm really getting upset and broke about this whole problem, since the video card works great in my Athlon 650 system. Today, I completely disconnected and remounted and reconnected the motherboard in the case, and it doesn't seem to help. I have remounted the CPU several times, making sure the thermal paste is set properly between the CPU and the Aero 4 HSF. When I can monitor the temperature, there seems to be no problem there. I have tried setting the VCore at several different settings with no luck. AGP Fast Writes on and off, AGP speed 4x,2x,1x. No changes.

Please let me know if you can think of any reason it is doing this. Thanks.

DrMarvin

 

[Black_Cat]

You might want to try cranking up the RAM voltage a bit. Try slowing the RAM timings a bit first, though.

You've tried and failed. The lesson here is, never try again. -- Homer Simpson.<P ID="edit"><FONT SIZE=-1><EM>Edited by Black_Cat on 01/21/05 05:49 PM.</EM></FONT></P>

 

[Crashman]

Multiple issues with a variety of hardware tend to point to RAM, which you tested, or the power supply.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[drmarvin]

So I have done a few more tests, and I seem to have ruled out the RAM. Every RAM test I do completes flawlessly. When I do a diagnostic startup with msconfig, it seems to be ok, but when I start services, things start to go a little crazy. I'm slowly adding services, hoping to find the culprit. If it's not the RAM, could it be the CPU?

The one test it does seem to freak out on is Prime95, and it has some sort of Fatal Rounding error. Does this mean anything? Also, everyone in a while when it's not crashing, an application will just close, for no reason. Thanks again.

DrMarvin

<P ID="edit"><FONT SIZE=-1><EM>Edited by drmarvin on 01/24/05 09:39 AM.</EM></FONT></P>

 

[tweebel]

Faulty CPUs are rare. Do a complete re-check of voltages, RAM-timings and temperatures and make sure you log them. After a crash you might be able to find something in the logs.
Applications closing for no reason at all might also be caused by a virus. If you can rule everything out it might be the CPU. You could inspect the cpu for discoloration, which means it might have been overheated.
Of course (don't know if it has been mentioned before) don't forget the BIOS flash, mobo and graphics card drivers.

 

[drmarvin]

Voltages are fine. Temps are fine. I don't really know how to adjust RAM timings, but i'm willing to try. I have the newest BIOS flash, and the latest ATI drivers and ASUS drivers.

There does seem to be some video issues sometimes though. I just had it flip off, and give the the "Sync out of range" message on the monitor, but the computer was still on (Num-lock active). I also just got several ATI VPU recoveries. Now I know that you might say that because of this, it's most likely a video card issue, and that's what I thought, but it's not any better with an old nVidia TNT card.

It's a fresh install of Windows XP SP2, on a freshly formatted drive so I don't figure it's a virus. CPU looks good.

The only thing that shows up in the event log is a bugcheck
0x1000000a or 0x10000050. Searching hasn't cleared up what these mean for me.

I'm really confused at this point. Any input is appreciated.

 

[tweebel]

RAM-timings can be set in the BIOS, it's usually quite easy. If you can't find it, look it up in the manual.

CPUZ might also tell you something about current timings, or try Sisoft Sandra.

 

[drmarvin]

I can find the RAM timing settings, but I'm not sure what I should set them to. Should I just slow all of them down?

Current settings
SDRAM CAS Latency 2.5T
SDRAM RAS to CAS Delay 3T
SDRAM RAS Precharge Time 3T
SDRAM RAS Active Time 7T
DDR Voltage is at 2.6V, but can also be 2.5 or 2.7V

I'm not sure why the above settings say SDRAM since I have DDR, but hey.

I'll try anything at this point, because I'd really like to avaoid taking it to a tech if I don't have to.

Thanks

 

[tweebel]

Try the CAS at 3.0.
If it doesn't work, try slowing down the RAM. What MHz is it running now?

Voltage is fine, don't mind it being named SDRAM.

 

[drmarvin]

So now the CAS is 3T, and the RAM is running at 266mhz, which is the slowest it can go. It was previously on AUTO so I'm not sure what it was before.

Now, so far, with the TNT card, this seems to have made a difference, so I'm going to try my Radeon. Hopefully it stays stable, and if so, thanks for your help. I'll let you know, or be back, asking for more help.

Thanks

DrMarvin

 

[drmarvin]

Update:

So far, seems to be dandy with the new RAM settings. I'm using the PC with the Radeon, with no reboots, and I'm running 3DMark 03 with beautiful results. I thank you for your help, and really appreciate your knowledge. Thanks

Further Update:

1.5 hours of running 3DMark 03 @ 1024x768 and still trucking!!!!!

DrMarvin