Is my motherboard dying on me?

K

Komskies

Weaksauce
Joined
Nov 1, 2011
Messages
97
Greetings Hardforum. This is my first post here, and I am coming here seeking advice via my brother's recommendation. I will preface this by saying that while I use my computer a TON, I am not a wiz by any means, and sometimes I have trouble properly diagnosing or describing issues. Please bear with me here and I will do the best I can to provide all of the necessary information to provide a proper diagnosis (and hopefully not an autopsy report!)

To begin, my computer is around 2.5 years old, and built with the following components:
  • ASUS P6T mobo
  • Intel I7 2.66GHZ processor
  • ATI Radeon 4870X2 HD video card
  • 3x2GB of DDR3 ram
  • 2x120GB Intel X-25M SSDs, RAID 0 array (these are also the boot drive)
  • Windows 7 home edition
  • PC Power and Cooling 750W quad black PSU

Over the last several weeks I have been experiencing issues with my computer. These issues include but are not limited to:
  • computer randomly restarts while using Chrome or playing WoW
  • frequent BSODs (atikmpag.sys)
  • heavy graphical artifacting, which I will post images of below
  • on startup certain SATA ports are not recognized

These first three images show what happens when I start windows normally, ie not in safe mode:
http://i42.tinypic.com/2yuc9d5.jpg
http://i42.tinypic.com/iylt1i.jpg
http://i39.tinypic.com/2vt6pvo.jpg

This next image shows the Safe Mode loading screen:
http://i44.tinypic.com/6z7mfs.jpg

Windows error recovery screen: (scometimes the bars are white like the image, and sometimes they are red and fragmented looking)
http://i39.tinypic.com/jp9abq.jpg

BIOS with artifacts:
http://i40.tinypic.com/250858x.jpg


Thus far I have tried a wide range of troubleshooting methods, with varying degrees of success, and always temporary. These methods include: fully removing and reinstalling video card drivers, reformatting my hard drive and reinstalling Windows, reinstalling BIOS firmware. So far nothing has been able to provide a permanent fix to my issues. As of now I can only get the computer to a usable state in Safe Mode and Safe Mode + Networking, and even then it is only sometimes.

Last night the issues seemed to intensify, and I was having trouble even getting the computer to post. After turning it on it would sit at a black screen, while the fans were spinning. At one point this happened for around 20 restarts in a row before it finally posted.

My brother is fairly confident this these issues are motherboard related, and I am inclined to agree with him.

At that, I wish to reach out to the Hardforum community for help and advice on how to proceed. If any additional information is needed, I will be more than happy to provide whatever I can.

As an endnote, I sincerely apologize if I posted this in the wrong sub-forum, and kindly ask a mod to relocate this thread if need be.

Thank you very much in advance and I anxiously await any advice or recommendations!
 
Last edited:
Well, what power supply do you have?
I would think the highest possibilities could be: Very faulty power or MB going out on you. (At least from my general issues in the past anyways).
 
Sherkhan said:
Well, what power supply do you have?
I would think the highest possibilities could be: Very faulty power or MB going out on you. (At least from my general issues in the past anyways).
Click to expand...

Thanks for the quick reply Sherkhan. I will call home to find out my PSU information, as I am at work right now and do not know what it is off the top of my head. I believe it is PC Power and Cooling, but I do not remember the wattage. I will post when I find out.
 
Another vote for "what PSU do you have".

Odd, because the SATA port thing sounds most like motherboard issues, the graphic corruption sounds most like GPU (but could be power or motherboard issues as well)...

The combination of instability and graphical artifacts leads me more toward power / GPU than motherboard, but the SATA thing throws a wrench in it. I don't know.
 
quality power supply is key....especially since its an x2 card so even though its just 1 video card its crossfire'd......

anytime i have ever had that type of artifacting on an sli or xfire is when the power supply isnt quite up to it or the cables or power suppply had inadequate shielding or no filters @ end of cable.

BSOD's can easily happen if power is not quite up to it.....that would be 1st thing most of us would check in your case.

Also considering the age.......have you properly cleaned it out? Just a while ago there was a post with a nice rig with huge amounts of dust. Please check this.

Also do not discount that the board may be going bad.

I have a p6t7 asus board and have had 2 previous boards that have had sata issues as well and got properly rma'ed.....i believe asus now tests your board when you send it in and if its ok then it will come back to you.
 
Those things are rock solid and should handle that machine without any issues, though it's possible it may be failing.

I'm more likely to blame the board now.
 
Could be the power supply as 750 is around what I would use, the gpu alone could chew through around 280 of that, depending on the clock speeds of course, I hazard a guess that it is motherboard or driver conflicts, also just make sure that the settings are correct, such as clock speeds for cpu, for memory and the like.

Funny thing with Intel, less so with AMD based chips/motherboards, if they are over/undervolted for the clock speeds you want/need them to hit, they can sometimes flake out over time, so maybe it is not the motherboard being "faulty" but possibly the voltages that are out of whack, and the board can no longer make up for the shortfall??

Also based on artifacting, either the gpu is acting up cause of temp related, that it can no longer sustain required clocks "stabily" or the power supply in question can no longer handle the setup

To me based on reading such, and seeing picks, as the ATIk driver bsod, that leads me to believe the gpu is flaking out, maybe if you have the option, take the card apart very very carefully, clean it well, and maybe change the core/Vram heat pads/paste, from my own experience, if the gpu is running to warm I see all kinds of corruption in menus, in game. 99% of time through experience with my 4870 and now 6870 corruption or artifacts was usually related to the clocks being pushed to hard or the temps being to high on the Vram, whereas BSOD with various gpu related codes usally came back to the core on the gpu being pushed to hard/warm

I would clean it well, and up the fan speed some, also keep an eye on the temps for gpu, cpu, motherboard.

GPU-Z is perfect for GPU monitoring, hardware monitor or openhardware monitor is very good to keep eye on motherboard, voltages, and SMART for hard drives, hope this helps some.
 
Last edited:
Thank you for all of the replies so far. It's great to see some good dialog on this.

@dragonstongue, since I built the computer in July 2009 I have been running the default BIOS settings (most of the tuners are set to auto). The only component change since the computer was new was the switch to SSDs in January 2011, and the only BIOS changes I made at that time was changing the drive settings from IDE to RAID.
 
Last edited:
It looks like bad ram or bad video card.

did you run memtest?

Test each stick individually, and if each one passes, then test all 3 sticks.

Did you test another video card or does that board have onboard video?

I'm betting your video card is at fault....it's pretty obvious that if you are getting graphic artifacting in the BIOS, before Windows even has a chance to load drivers, it's the card or the board.
 
drdoug99 said:
It looks like bad ram or bad video card.

did you run memtest?

Test each stick individually, and if each one passes, then test all 3 sticks.

Did you test another video card or does that board have onboard video?

I'm betting your video card is at fault....it's pretty obvious that if you are getting graphic artifacting in the BIOS, before Windows even has a chance to load drivers, it's the card or the board.
Click to expand...

I actually started a memtest86 before I went to bed last night, and in my post-sleeping in haste I didn't have a chance to check it before work this morning.

I do not have a secondary video card to test with, but I believe my board does have integrated video. If I were to test this, do I simply disable my video card in the BIOS, or do I have to physically remove the card from the PCI-E slot?

Edit: Ignore that italicized part, this board requires a dedicated GPU.
 
If you have a different video card, put it in. Then you can check off yes or no that the artifacts are from your video card right away.
 
Computers are quirky, so even if you had everything set to auto, that does not 100% mean the voltages were +/- what they should have been, my P5Q-E had its quirky moments, I had to change the voltages at least 4 times with the same clocks/speeds over a 2 year period, it happens, things wear out, software goes crazy etc :p
 
Last edited:
I checked my voltages in the BIOS.

CPU Volt: 1.104V
3.3V: 3.312V
5V: 4.998V
12V: 12.193V

I'll install some temperature monitoring software and check that.

EDIT: GPU-Z is reporting that my GPU(s) are sitting at 69C idling in Windows.
 
Voltage monitoring in software doesn't really mean anything. You need a multimeter to check that out.

See if you can borrow a graphics card from another machine. That artifacting issue looks more like graphics issue rather than motherboard.
 
I agree with Tsumi, it sounds alot more Graphics related, although the voltages you put, the only one that might be below where it should be is CPU, 1.15-1.28v stock is where it should be, it might actually be a hair lower voltage then it needs, that could cause issues as well, but 100% doubtfull it would cause artifacting, usually it would just not boot up, or would just keep crashing.

I would clean the video card as best you can without taking it apart, use Catalyst Control Center and hardware monitor/gpu-z to see temps of video card, maybe you just need to up the vid cards fans speed, if its temp related, it may fix those issues, also if you can, point a case fan towards gpu so it gets a bit more air, cause I know the 4k series allows the GPU to heat up quite a bit before fan ramps up to cool it off, this is hard on the fans, and plays hell on the cards memory and Vregs, both of which can cause crashing/artifacting if they get to hot.
 
Tsumi said:
Voltage monitoring in software doesn't really mean anything. You need a multimeter to check that out.

See if you can borrow a graphics card from another machine. That artifacting issue looks more like graphics issue rather than motherboard.
Click to expand...

Hi Tsumi, sadly I do not have a secondary graphics card to test your theory. Is there any kind of test or diagnosis software out there that can test if a graphics card is faulty?
 
Run Furmark for a while; if the machine crashes, it could be PSU or GPU. Then take out one card (or are you running a 4870X2? It's not clear how you wrote it...) and try again.
 
silent-circuit said:
Run Furmark for a while; if the machine crashes, it could be PSU or GPU. Then take out one card (or are you running a 4870X2? It's not clear how you wrote it...) and try again.
Click to expand...

Yes, sorry, my card is the 4870X2. A single unit.

When I get home from work today I will try running Furmark as you have suggested, to see if I can rule out PSU/GPU.
 
You might be able to voltage test it with a multimeter, as most cards now usually have check points for that, but unless you know what you are looking for, it would be hard to tell.

If you`re motherboard is a decent one, in the bios it should have a tab to see the voltages
3v, 5v, 12v, cpu, etc.

They should always be +/- a tad, thats normal. example [email protected] or a little more is ok, but if 12v@10v then thats not good, board has to work to hard to compensate.

For GPU, as I said above, unless you know what the voltages should be, it would be hard to see this way.

I would like said prior, give the gpu a good cleaning, dust/dirt can cause all kind of strange issues, and can even ground/short things out sometimes.

give it a good cleaning, go into windows and up the gpu fan speed a bit, keep an eye on temps for motherboard, cpu, gpu when you are running a benchmark or game you like to play, do not run furmark, it is unreasonable temp load on cards components, and can cause issues of card being injured especially on the Vregs if they are "just" able to work properly in the first place.

Also because it is a X2 style card, you should be able to disable crossfire to run as a single gpu instead of dual gpu card??

Simply, if power supply is not giving the board what it requires in terms of power it can wreck havok, as obviously it requires a certain amount of power to function the way it should.

If its temp related, causing the issues, cleaning, or upping the fanspeed, put at stock clocks may fix the issue when its related to the card not cooling properly, which from my experience, to high of clocks, or bad cooling(dirty, not tightened down enough) can cause all kinds of rebooting, error codes, and especially artifacting in this regard. 99% of time is related to to high off temps, becuase of the fans not spinning up properly at the right time.

A good cleaning usually helps a ton, while you are cleaning, changing the thermal paste on the GPU core can help reduce the temps there, changing the Vreg and memory thermal pads/paste can help keep temps down as well, manufacturers have bad habit of using a ton of thick pads/paste in regards to thermal transfer material, when they start wearing down, they simply no longer can transfer the heat very well, temps spike, bsod/artifacting happens.

Long post, sorry, but best of luck, hope you can get it sorted out.

If you have to get a new card, I recommend dual 6850/6870 or single 6950 6970 depending on $ situation, for Nvidia dual GT460/560Ti or single 570/580

Not sure what power supply can handle over all, but this can be figured out when you decide if you need to go this route.
 
dragonstongue said:
You might be able to voltage test it with a multimeter, as most cards now usually have check points for that, but unless you know what you are looking for, it would be hard to tell.

If you`re motherboard is a decent one, in the bios it should have a tab to see the voltages
3v, 5v, 12v, cpu, etc.

They should always be +/- a tad, thats normal. example [email protected] or a little more is ok, but if 12v@10v then thats not good, board has to work to hard to compensate.

For GPU, as I said above, unless you know what the voltages should be, it would be hard to see this way.

I would like said prior, give the gpu a good cleaning, dust/dirt can cause all kind of strange issues, and can even ground/short things out sometimes.

give it a good cleaning, go into windows and up the gpu fan speed a bit, keep an eye on temps for motherboard, cpu, gpu when you are running a benchmark or game you like to play, do not run furmark, it is unreasonable temp load on cards components, and can cause issues of card being injured especially on the Vregs if they are "just" able to work properly in the first place.

Also because it is a X2 style card, you should be able to disable crossfire to run as a single gpu instead of dual gpu card??

Simply, if power supply is not giving the board what it requires in terms of power it can wreck havok, as obviously it requires a certain amount of power to function the way it should.

If its temp related, causing the issues, cleaning, or upping the fanspeed, put at stock clocks may fix the issue when its related to the card not cooling properly, which from my experience, to high of clocks, or bad cooling(dirty, not tightened down enough) can cause all kinds of rebooting, error codes, and especially artifacting in this regard. 99% of time is related to to high off temps, becuase of the fans not spinning up properly at the right time.

A good cleaning usually helps a ton, while you are cleaning, changing the thermal paste on the GPU core can help reduce the temps there, changing the Vreg and memory thermal pads/paste can help keep temps down as well, manufacturers have bad habit of using a ton of thick pads/paste in regards to thermal transfer material, when they start wearing down, they simply no longer can transfer the heat very well, temps spike, bsod/artifacting happens.

Long post, sorry, but best of luck, hope you can get it sorted out.

If you have to get a new card, I recommend dual 6850/6870 or single 6950 6970 depending on $ situation, for Nvidia dual GT460/560Ti or single 570/580

Not sure what power supply can handle over all, but this can be figured out when you decide if you need to go this route.
Click to expand...

No need to apologize for a long post and providing lots of information! I really appreciate it.

When I get home armed with this new information I will begin going through some of the recommendations to see what I can rule out.
 
You must log in or register to reply here.
Back
Top