Hello everyone. I'm having some problems with my computer shutting down randomly. No BSOD. I just reset the BIOS and removed a single GPU, so I am no longer in SLI mode.
Thanks for any help you can give!
Specs below
Attachment 201223
Mind if I ask what steps you've taken so far? Given that you are one of the more advanced users, I do not want to post steps that have already been taken, i.e. temperature monitoring, checking for dust, etc. Especially since you put me through those steps when I first started on here, so I know that you know what to look for there.
No no, thats a good idea. Unfortunately I am not too adept at debugging crashes.
I have checked for dust, but not so much temperature. In fact, it could be temps causing hardware components to wear down. The 2 Nvidia 580 GPUs get very hot, and they crash on their own at times in Graphic intesive applications. They have a built in safety that keeps them from getting too hot, just as the CPU does. I have CPU water cooling and pretty good airflow, however the GPUs from gigabyte blow the hot air back into the system, and overall are not good at expelling heat. I was planning on running memtest tonight.
Forgot to mention that while heat would crash my GPUs at times, and there was a rare BSOD every now and again, I never had my computer just turn off in the middle of something, like just browsing the internet. I reset BIOS just in case I messed with it.
I have re-set hardware components, minus a single GPU which is out of my system. I'm going to play some games and see how they go.
Alright. One thing that is recommended with possible GPU related crashes is since they are not as good at expelling heat as you pointed out, sometimes opening the case up and running a house fan pointed directly at the system can help. If that does help, it probably is a heating issue. You can also monitor the GPU temperatures with various temperature monitoring software:
- Real Temp is a good CPU temperature monitor.
- Speccy - System Information - Free Download will monitor all hardware temperatures.
- HWiNFO, HWiNFO32 & HWiNFO64 - Hardware Information and Analysis Tools can be inaccurate for CPU temperatures, but is a good program for GPU temperature monitoring.
I also see a driver that is out of date that is known to cause crashes:Update the driver by:Code:ASACPI fffff880`0386a000 fffff880`03872000 Sun Mar 27 20:30:36 2005 (42476c4c) 00003c77 ASACPI.sys
- Going to the Asus motherboard support site
When you reach the website:- Scroll down the page and click Utilities
- Hold Ctrl and press f (ctrl+f) to enter the browser's find feature
- Search for "ATK0110 driver for WindowsXP/Vista/Win7 32&64-bit" (without quotes)
- Download and install the driver.
- After installation is complete, verify that it installed correctly.
- Click Start Menu
- Click My Computer
- Go to C:\WIndows\System32\drivers\
- Verify that the ASACPI.SYS file is dated 2009 or newer (2010,etc.)
Thanks to JMH and zigzag3143 for the above information.
Your crash is a MEMORY_MANAGEMENT crash, and it points to either memory problems or driver conflicts. The memory problems could include graphics card memory, as well. Use the |MG| Video Memory Stress Test 1.7.116 Download to test your graphics card memory.
You may also want to test your RAM itself. Run the boot version of Memtest86+ paying close attention to Parts 2 and 3 of the tutorial. Also, in case Memtest86+ misses anything and comes up with no errors, run the extended version of the Windows Memory Diagnostics Tool for at least five passes. These you may want to run overnight since they take a long time to complete (run them an hour before bed each of the next two nights and check before going to sleep that they are still running).
You're definately on to something here. Given me some peace of mind too, to have the issue narrowed down. Had artifacts with both SLI GPUs playing Civ V, but with a single GPU I can play Civ V no problem on highest settings. I wonder if the problem lies with the GPU I no longer have in my system. I've installed the driver correctly, and am working on all the other things.
CPU has never really been hot. With Speccy -
CPU 33C
RAM-unknown (Has heat sink and fan, though)
motherboard 27C
Graphics 31C
SSDs/HDDs 27-31C
But all of this will change if I start playing A Graphicly intense game. Allow me to do this in a bit.
With the video mem stress test, am i testing anything specific? I've tested the first GPU and here is the score:
Attachment 201236
Allow me some time to test the other, and do I test it according to how I tested the first (see picture)?
58C after playing Dragon age:Origins for 15mins or so, not bad.
Highest settings, no crashes. I couldn't do that with 2 GPU cards. I wondered why graphics had to decrease so much, but I couldn't pinpoint the problem and never took the time to research. These problems have been happening from the getgo, should have RMA'd the card.
The newer card (and the one I've set aside for now) is the one with problems, though I have yet to hook it up. Have to see if I still have all the paperwork. will test some more when I have the time. A big thanks for revealing to me what negligence costs on my end o.0
talk to you again as soon as I can
Looks like your graphics memory came up clean, and yes, that is the correct method for testing, but you may want to run more passes to be more thorough; we recommend seven passes with memtest86+, so that may be a good number to start with. Your temperatures are excellent, so that is not likely the problem unless they drastically change with the second card in.
We've seen a lot of problems with the SLI bridges on this forum; sometimes it is due to the bridge itself or the motherboard not working with the bridge, and sometimes it is due to one of the cards being bad. There is always more of a risk of things going wrong with additional components added to the mix, but I would imagine a very high percentage of the time, those four parts (2x cards, bridge, and motherboard) work well together or they wouldn't be produced that way.
In your case, I suspect the problem is with the card. Might be worth checking both slots to make sure they are both good, though. Let us know what happens when the newer card runs by itself. Also, what was your setup before the new card?
So far I've done a single pass with both cards (one at a time), and they come out clean. I played Newer games on the highest graphics settings for each card, and they both perform well, without any crashes. I suspect the SLI is what is causing the problems, but that would (*insert choice word*) because now I have 2 cards and SLI doesn't work.
Was having trouble loading the Memtest because I need to set my USB (keyboard) to legacy so I can use it to select boot device. Gave up and went to bed. I'll change the BIOS setting and post the results perhaps tomorrow (running it tonight), but I am 80% sure that the problem lies either with the SLI bridge with the fault being the motherboard (tried 2 different SLI connectors), or the 2nd PCIe slot is broke. I worked all day today so I'll be testing some more tomorrow. Thanks for leading me in the right direction, Mike. I'll let you know what I find.
No problem Dusty. I am inclined to agree that it is probably SLI related or a bad PCIe slot based on your steps and description. Have you had a chance to test both PCIe slots?
A single crash after 2 hours gameplay on PCIe slot #2, but a single pass of vmt had no errors for the same slot. I still have graphics at near highest settings. I could run it for more passes, but right now the computer isn't acting so poorly, except for the single crash. With SLI, it was frequent, especially with high graphic settings.
I've done 2 and a half passes of Memtest (took 5 hours, perhaps because I have a lot of RAM) without errors, and I do not suspect my RAM. It may be my computer somehow and for some reason isn't doing well in the SLI configuration.
Think I should run more passes on the slots/cards? takes about 15 mins each pass
Perhaps I should upgrade to X79 type motherboard and CPU
Quote