• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Random BSOD's

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
I just changed back to kernel memory dump at "%SystemRoot%\Minidump". I also changed the Total paging file size to 8194 Initial size (MB) and 8800 Maximum size (MB). I swapped it from system managed as it had 2 Mb under my RAM size.

EDIT: When I get the BSOD's, they tell me that a dump was created but when I search for it with WhoCrashed none is to be found.
 
Last edited:
????? do you analyse whit whocrached
also i dint see if you did a CHKDSK and a SFC/SCANNOW.open CMD and do one at the time
 
Last edited:
Qyratar said:
I just changed back to kernel memory dump at "%SystemRoot%\Minidump". I also changed the Total paging file size to 8194 Initial size (MB) and 8800 Maximum size (MB). I swapped it from system managed as it had 2 Mb under my RAM size.

EDIT: When I get the BSOD's, they tell me that a dump was created but when I search for it with WhoCrashed none is to be found.
Click to expand...

Now that you made those changes and reboot, does it create a dumpfile successfully?

System managed is generally safest when t-shooting.
 
When doing the scf /scannow (3 times) I got this message "Windows Resource Protection found corrupt files but was unable to fix some of them."
Details are included in the CBS.Log

Doing the chkdsk I got a message saying "Windows has checked the file system and found no problems" Should I try to repair anyway?

Have not crashed after getting the new settings yet. Will come back with results after the next crash.
 
Here is the file http://depositfiles.com/files/o01rhhuae

EDIT: Should I reinstall windows and see if that will fix the problems? it will at least fix the minidump problem.

EDIT#2:I found some files which it could not repair (as following): windeploy.exe.mui WinLGDep.dll.mui audit.exe.mui setup.exe.mui W32UIRes.dll.mui oobeldr.exe.mui (Was repeated twice) spwizimg.dll (Repeated twice) W32UIRes.dll.mui (Twice) iassdo.dll.mui (Twice)
Also said something about that it could not "reproject corrupted file to all of the files above"
 
Last edited:
Ok, I reinstalled Windows as I got real annoyed by the fact that I couldn't get any kind dumps no matter what I tried. I later installed as few drivers as I could in order to eliminate any kind of malfunctioning drivers. I reinstalled a game which frequently gave me BSOD's. I started playing and after 30 min I got a BSOD, this time my computer made a dump. I did not check what kind of dump it would make as I wanted to mess as little as possible with the dump system. It made a kernel dump which I later uploaded on depositfiles. (http://depositfiles.com/files/7bexz4u4v)

I have changed it to start dumping minidumps and will from now on post future minidumps here.
 
Since you've reinstalled Windows and are still getting crashes, then it's definitely a hardware issue. The two most likely components are the PSU (despite previous tests) and the GPU, in that order. You're gonna have to do some part swapping to figure out which.
 
Qyratar said:
I still have my old graphics card, could swap out my current one for my old one and see if the problem still occurs.

EDIT: also new minidump http://depositfiles.com/files/9jwe1122u
Click to expand...

That won't necessarily be conclusive: if the power draw of the old card is less than the new one, it may not have the same effect on the PSU. So it would be best if you could swap the GPU for the same model, or swap the new GPU into a known stable system, or swap the PSU.
 
My previous GPU was the SAPPHIRE VAPOR-X HD 4870 2GB GDDR5 PCI-E (http://www.sapphiretech.com/presentation/product/?psn=000101&pid=217&lid=1#) which recommended Wattage is the same (actually the GTX 660 SC's recommended minimum wattage is 450W) so I presume the power requirements would be the same?

I tried with only running one stick at a time and the problem still persisted so I think the memory is not the problem.

EDIT: I do still have the warranty for the new GPU so I guess that I could send it in and say that I have had a problem with it having BSOD's in some games.
 
Last edited:
Qyratar said:
My previous GPU was the SAPPHIRE VAPOR-X HD 4870 2GB GDDR5 PCI-E (http://www.sapphiretech.com/presenta...pid=217&lid=1#) which recommended Wattage is the same (actually the GTX 660 SC's recommended minimum wattage is 450W) so I presume the power requirements would be the same?
Click to expand...

According to this 4870 review, the 4870 used ~92W more under load than at idle. According to this 660 Ti roundup, the 660 Ti uses ~120W more under load than at idle. You are putting ~30W more stress on the PSU, and even though it should handle the load just fine, it may not be.

EDIT >> It may even be that the 660 Ti cannot handle the ripple from the PSU that the 4870 can.
 
Last edited:
So what are you recommending me to do? Test the PSU/Get a new PSU? Test my old graphics card?

Something which could be worth mentioning is that I ran AC 3 on maximized setting on this GPU without a single BSOD when I got it. Could reinstall it again and see if any BSOD's occurs.
 
The easiest/cheapest test would be to test the GPU in another capable and stable PC. If the crashes follow the card then it's the card for sure. If not, then it's most likely the PSU and you should just replace it unless you have access to another quality ~500W PSU that you can test with.

BTW, did you ever try underclocking the GPU? You could use MSI Afterburner for most Nvidia cards.
 
I have unfortunately no access to another pc which can run this GPU as most of my friends prefer laptops.

I have not tried to underclock the card and I have never tried to underclock/overcklock anything in my rigs so far. I can try this but it would be very much appreciated and if it isnt too much to ask for if you could post some guide on how to do it.

EDIT: I'm starting to believe that the BSOD's are actually caused by the game (Sleeping Dogs) as I only started getting them after installing Sleeping Dogs. Before that I only got a couple of BSOD's which were caused by the fact that i had the options set incorrectly.
Also, after googling Sleeping Dogs BSOD I came to realise that there are a lot of people which get BSOD's in this game.
Maybe i'm just going for the easiest way out...
 
Last edited:
Just a suggestion - if you haven't done it yet....

Get memtest and load in on a bootable thumb drive. Boot into memtest with all of your memory sticks in place and let it run overnight.

If you don't get errors, scratch that off the list.

Next I would suggest putting the old card back in and running a really intense benchmark on it to see if you can get it to BSOD with the old card. That would indicate that it's probably not the new vid card and point to power on the board or PSU.

My first gut reaction is that it's a power issue since you moved from one card to another and it started, but you can't know that for sure.

good luck.
 
Qyratar said:
I have not tried to underclock the card and I have never tried to underclock/overcklock anything in my rigs so far. I can try this but it would be very much appreciated and if it isnt too much to ask for if you could post some guide on how to do it.
Click to expand...

It's simple, so don't sweat it. Install MSI Afterburner and launch it (you'll need admin privileges). You'll see several sliders, all set at default settings. Just slide the "Core Clock" to the left, lowering the MHz by ~20%. Then do the same for "Memory Clock". This should lower the power consumption close to that of the 4870, if not a little less. Leave everything else alone. Run the troublesome game again and see if it crashes.

Cr0nJ0b said:
Next I would suggest putting the old card back in and running a really intense benchmark on it to see if you can get it to BSOD with the old card. That would indicate that it's probably not the new vid card and point to power on the board or PSU.
Click to expand...

This is actually a good idea, especially if you overclock it (you can overclock with AMD's Engine Control Center). Getting it to similar power consumption of the 660 Ti should stress the PSU similarly.
 
Changing the Core clock and the Memory clock by 20% allowed me to play for 2h and 30min. Could have just been a coincidence but I dont think so.

Have yet to try to play with the old GPU, will change and see what happens.
 
Qyratar said:
Ok, I reinstalled Windows as I got real annoyed by the fact that I couldn't get any kind dumps no matter what I tried. I later installed as few drivers as I could in order to eliminate any kind of malfunctioning drivers. I reinstalled a game which frequently gave me BSOD's. I started playing and after 30 min I got a BSOD, this time my computer made a dump. I did not check what kind of dump it would make as I wanted to mess as little as possible with the dump system. It made a kernel dump which I later uploaded on depositfiles. (http://depositfiles.com/files/7bexz4u4v)

I have changed it to start dumping minidumps and will from now on post future minidumps here.
Click to expand...

OK, so what did WhoCrashed say about the dump? I'm not on a Windows box so can't look at the dump, but that's certainly a logical action to do that immediately....
 
@dclive: it said that hal.dll is to blame "WHEA_UNCORRECTABLE_ERROR" and the code 0x124. Not sure what you are after here. If this is what you are after, good enough. If not, tell me what you are looking for and I will search for it.

@Cr0nJ0b & Anonymouseuser: after running the old GPU at the highest overclock (790 core & 1100 mem) for an hour with furmark running, a movie in the background and the troublesome game as well I got a BSOD. It states that ntoskrnl.exe+80640 (the same error as the new GPU) is to blame, followed by 0x124 and WHEA_UNCORRECTABLE_ERROR. I will run the new GPU at those exact clocks and see if it crashes as well, clocks were prety much at the same frequency but results might possibly change.
 
Qyratar said:
@Cr0nJ0b & Anonymouseuser: after running the old GPU at the highest overclock (790 core & 1100 mem) for an hour with furmark running, a movie in the background and the troublesome game as well I got a BSOD. It states that ntoskrnl.exe+80640 (the same error as the new GPU) is to blame, followed by 0x124 and WHEA_UNCORRECTABLE_ERROR. I will run the new GPU at those exact clocks and see if it crashes as well, clocks were prety much at the same frequency but results might possibly change.
Click to expand...

It's leaning more and more towards a PSU issue. Thankfully, they aren't anywhere near as expensive as a 660 Ti. If the XFX PSU isn't still under warranty, a 500W Corsair builder series should be more than enough for what you're running.
 
Im planning to upgrade my system soon with some fans and additional RAM so I will spend 11 more gbp on the 600 CXM in order to not worry about the PSU's capabilities.

Also memtest cane up with a few errors while testing. at different tests 6: 730 7: 58 8: 65535+ 9: 48. (At pass 10)
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top