Windows 7 Having trouble playing games, constant BSOD, random crashes at times

Bortelli

New Member
Hello everyone. Before I' get into more detail I'll just put the .dmp files first, I know that's what your probably interested in.

Anyways, I've been having a lot of trouble keeping my computer stable. Not matter what driver I am on, stock, new, or old I will always receive a BSOD. It's almost instantaneous when I play Dead Space 3, when I disabled v -sync for more frames, and other games like Skyrim, but this isn't always the case, yet Devil May Cry works flawlessly.

Sometimes When I am using Firefox I will crash rarely. I'm not sure what the problem may be.

Sometimes when I receive an error report the codes given will vary. But usually complains about my drivers. This is extremely annoying.

Right now I'm using 13.2 beta drivers. And I am on Service Pack 1.

Here are Specs :
Operating System
Microsoft Windows 7 Ultimate 64-bit SP1
CPU
AMD Phenom II X4 965 47 °C
Deneb 45nm Technology
RAM
8.00 GB Dual-Channel DDR3 @ 666MHz (9-9-9-24)
Motherboard
MICRO-STAR INTERNATIONAL CO.,LTD 760GM -E51 (MS-7596) (CPU1) 36 °C
Graphics
TF-TV3207E* (1920x1080@30Hz)
AMD Radeon HD 6900 Series (Gigabyte)
Hard Drives
932GB Seagate ST31000524AS ATA Device (SATA) 27 °C
Optical Drives
ELBY CLONEDRIVE SCSI CdRom Device
Audio
AMD High Definition Audio Device
 

Attachments

  • Documents.rar
    129.3 KB · Views: 279
Majority of the crashdumps are generated straight from the AMD driver, in that the AMD driver is reporting an internal fault to Windows and is forcing a crash. This is very commonly caused by hardware failure or internal driver failure that it's reporting but for proprietary reasons it does not explain what it means (private documentation is required to decipher). By far the most common cause is hardware failure, often the GPU but sometimes the power supply can be involved.

The latest crashdump 3/02/13 shows your ATI driver being from Apr 19, 2011. In fact all the crashdumps in this report mention the same date. If you say that you have the latest version but this states otherwise, then there may be outdated fragments still lingering. Do a full uninstall (perhaps even with something like CCleaner or Driver Fusion) of the video drivers, then a complete install of the new ones. If the BSODs persist with the new driver, it's time to look for a new graphics card. Understand PSU can also be at fault so consider that as well.

Analysts:

AMD driver calling straight into KeBugCheckEx with its own proprietary bugcheck code. Without private documentation it will be very difficult to decipher what the bugcheck code means.

Code:
3: kd> !analyze -v
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Unknown bugcheck code (a0000001)
Unknown bugcheck description
Arguments:
Arg1: 0000000000000005
Arg2: 0000000000000000
Arg3: 0000000000000000
Arg4: 0000000000000000

Debugging Details:
------------------

TRIAGER: Could not open triage file : C:\Program Files (x86)\Windows Kits\8.0\Debuggers\x64\triage\modclass.ini, error 2

CUSTOMER_CRASH_COUNT:  1

DEFAULT_BUCKET_ID:  WIN7_DRIVER_FAULT

BUGCHECK_STR:  0xA0000001

PROCESS_NAME:  System

CURRENT_IRQL:  b

LAST_CONTROL_TRANSFER:  from fffff88007259db1 to fffff80001a8ac40

STACK_TEXT:  
fffff880`01ef6668 fffff880`07259db1 : 00000000`a0000001 00000000`00000005 00000000`00000000 00000000`00000000 : [COLOR=#ff0000]nt!KeBugCheckEx[/COLOR]
fffff880`01ef6670 00000000`a0000001 : 00000000`00000005 00000000`00000000 00000000`00000000 00000000`00000000 : [COLOR=#ff0000]atikmdag+0x4cdb1[/COLOR]
fffff880`01ef6678 00000000`00000005 : 00000000`00000000 00000000`00000000 00000000`00000000 fffffa80`00000000 : 0xa0000001
fffff880`01ef6680 00000000`00000000 : 00000000`00000000 00000000`00000000 fffffa80`00000000 00000000`00000000 : 0x5


STACK_COMMAND:  kb

FOLLOWUP_IP: 
atikmdag+4cdb1
fffff880`07259db1 ??              ???

SYMBOL_STACK_INDEX:  1

SYMBOL_NAME:  atikmdag+4cdb1

FOLLOWUP_NAME:  MachineOwner

MODULE_NAME: atikmdag

IMAGE_NAME:  atikmdag.sys

DEBUG_FLR_IMAGE_TIMESTAMP:  4dae3c99

FAILURE_BUCKET_ID:  X64_0xA0000001_atikmdag+4cdb1

BUCKET_ID:  X64_0xA0000001_atikmdag+4cdb1

Followup: MachineOwner
---------

3: kd> lmvm atikmdag
start             end                 module name
fffff880`0720d000 fffff880`07b41000   atikmdag T (no symbols)           
    Loaded symbol image file: atikmdag.sys
    Image path: \SystemRoot\system32\DRIVERS\atikmdag.sys
    Image name: atikmdag.sys
    Timestamp:       [COLOR=#ff0000] Tue Apr 19 21:53:29 2011[/COLOR] (4DAE3C99)
    CheckSum:         008E91DD
    ImageSize:        00934000
    Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4
 
Thanks for the reply Gnarius, Unfortunatlely, I recieved another BSOD. I was playing Dead Space 3 again and about 30-40 minutes in I crashed

I recieved this error:

BCCode: 50
BCP1: FFFFBC80116ABD58
BCP2: 0000000000000000
BCP3: FFFFF80001B46222
BCP4: 0000000000000007
OS Version: 6_1_7601
Service Pack: 1_0
Product: 256_1
 

Attachments

  • 030513-18968-01.dmp
    268.7 KB · Views: 481
I see the ATI driver showing up here too, even though yes, it is the new one this time. Plus, the actual cause of the problem (bit flipping) is very iconic with hardware failure. While I do suspect the graphics card, there may be something else involved here.

What brand/model is your power supply, btw?

Just to make sure we aren't running off trail here, I recommend doing some hardware tests:

CPU: Prime95 - Torture Test; Blend; run overnight
RAM: Memtest86+ - run overnight

That should do for now. Before you run Prime95, be very conscious of your temps and make sure the CPU cooling is doing its job well, since Prime95 runs the CPU very hot and if the cooling isn't adequate it could damage your CPU. Use HWInfo on Sensors only option to check temps while Prime95 is running. In fact, logging while running Prime95 will help find issues quite a bit, especially in checking voltage readings as well as temp. If temp stabilizes at a safe level (60C or less) after 30 minutes then it should be safe to run overnight. As for Memtest, it is very safe on hardware so don't worry about anything when you run it.
 
Thanks for the replies, I will run these tests as soon as possible, I know I reply late, school and such.

Anyways, my PSU is A Thermal-Take Brand.

Model: TR2 RX-850AH3CH

Amazon.com: Thermaltake Power Supply 240-Pin 850 Power Supply W0319RU: Electronics

I recieved another BSOD right after exiting Black Light Retribution, I received this error.

BCCode: 1000007e
BCP1: FFFFFFFFC0000005
BCP2: FFFFF8800739BCE5
BCP3: FFFFF88008794378
BCP4: FFFFF88008793BD0
OS Version: 6_1_7601
Service Pack: 1_0
Product: 256_1
 

Attachments

  • crash030613-20750-01.dmp
    272.7 KB · Views: 474
I reviewed your power supply, and I found it's part of Thermaltake's budget line of PSUs (and discontinued no doubt). Budget PSUs for any brand are very risky purchases in that they skim a lot on internal components to increase profit margins. The results are unreliable power supplies that will cough and sputter early in their life unless you get a very lucky purchase, plus their lifespan on average is very poor. When you purchase a PSU, what goes into factoring the cost of one isn't so much the wattage nor any features like modularity, but the reliability of the internals inside. Commonly the more power efficient parts are also better quality and can take a lot more punishment than their budget counterparts. So what I'm trying to say is, I would not doubt the power supply being an element in causing your instability problems here. An article I recommend to people on looking for a good PSU is here, which also adds some recommended items that's kept fairly updated.

Now, the problem is, the PSU cannot be tested without the proper equipment (like a multimeter) or swapping in a trustworthy one and testing the system on that. So there isn't really effective means of testing a power supply outside of spending some bucks on either having someone else test it or purchasing the right equipment and testing it yourself. I would not recommend any "PSU testers" advertised out there outside of the high-dollar ones, as the cheaper ones do not test the PSU effectively.

As for the recent crash, the crashdump is very similar to your previous ones, with the ATI driver showing up and video memory management operations being performed on your GPU. It only lays more suspicion on hardware failure. Continue running those hardware tests and look about getting a reliable replacement on that PSU. If that PSU isn't even causing your problems right now, I am confident it will soon.
 
Great, thanks for the reply again. I am getting a new job soon so I might have enough money to purchase a new psu, but I dont want to be hasty and jump right into purchasing one. And My pc is pretty dusty, I have not cleaned it, about a 6 month time frame.

Would it be ok If i just bought a new PSU, and kept my 6950, considering initially the GPU is to blame? I don't want to spend any money right away, but I will If I have to. Also, what do you consider is a good PSU for my build? Do I really need 850 watts of power? The reason why I bought that PSU is becuase I might crossfire in the future, and yes.. It was budget..

I also took a look at my motherboard.. And it does not seem to support 6000x series cards.., but I am not entirely sure. I've wondered if compatibility with my mobo and gpu would have anything to do with the BSOD's I am receiving.

List of VGA cards that are compatible:

http://us.msi.com/file/test_report/TR1_1926.pdf
 
Last edited:
Use the article linked in my previous post. Yes, 850 is a good number, perhaps even for Crossfire (depending on the cards you install). Thermaltake has a PSU calculator webpage that you can use to get a general idea of the necessary wattage you need. I normally take what it says and add about 100~150 watts extra to account for degradation (wattage output degrades overtime).

Now, whether it's the PSU or the GPU, I am not entirely 100% sure on that. Your mention of the motherboard also has me curious. The problem is, all of them are suspect, and there's really no way to isolate or test either without swapping hardware or isolating in another reliable environment.

I think it's best to start first with the item that's actually crying foul: the GPU itself. Since commonly the crashing is because of the ATI driver alerting the system of a GPU problem, I reckon it's best to start there first, then approach the motherboard since it has not listed the GPU as supported (it looks like a very limited and old list though), then follow up with the PSU.

That, or you could skip over all this and approach a local PC shop to have their techs look at it. They very well may have the right environment, resources and tools to properly approach this. Mention to them the details provided on this thread so far (especially the BSODs being straight from the ATI driver).
 
The person that posted on FB about this thread (I hate doing that by accident!) may be on to something. While I did mention using HWInfo during the Prime95 run to see if the system is overheating any, you never did actually provide us the HWInfo log, nor did you mention that you ran Prime95. Can you do that (perhaps run Hwinfo sensor logging while gaming too) for us? Thanks.
 
I will run prime95 soon. Sometime early next week Sorry for the inconvienence. I reformatted my harddrive. I am on 13.1 Drivers. I recieved this error.

BCCode: d1
BCP1: 0000000000000028
BCP2: 0000000000000002
BCP3: 0000000000000000
BCP4: FFFFF880014D1780
OS Version: 6_1_7601
Service Pack: 1_0
Product: 256_1



 

Attachments

  • 030913-17468-01.dmp
    268.6 KB · Views: 471
I'm not sure why you reformatted. Nothing here has been hinted at a driver being the cause of the issue but rather hardware, and no amount of reformatting is going to fix that, especially if that bad hardware isn't the hard drive itself. I think you may have unnecessarily inconvenienced yourself with this action, as is evident by the continuing instability despite the fact.

I believe this particular latest crash may actually be caused by drivers, particularly your Realtek network driver which is currently dated Feb 26, 2009. Update them and see if stability improves, though that's doubtful given your previous symptoms. However there's good possibility this particular crash is caused some hardware failure itself such as mobo or PSU. Again, crashdumps only tell half the story, so we can't discern beyond what it says.
 
I ran memtest86+, and thankfully I did not receive any errors. My ram seems fine, also I decided to reformat my hard drive again, and I'm just using stock drivers from windows 7.

Unfortunately. I have received more errors. I have read your post today, And I realize that formatting probably wont help, I will upgrade my realtek drivers.

I'm convinced that this is driver related too, as many other moderators from different forums also suggested to upgrade my realtek drivers .
 

Attachments

  • 031213-50500-01.zip
    34.6 KB · Views: 271
As I mentioned before, the realtek driver "bugs" may actually be manifesting as improper hardware activity. The network controller itself may actually have internal failures in it, and it's causing the drivers for it to respond in a manner that isn't stable. I've seen this a number of times before. Again, we will see where this goes with the driver update, but I will say that even the Win7 stock drivers shouldn't cause that much of a hassle either unless some hardware is on the fritz.
 
I have not crashed at all, After upgraded my realtek drivers. Also, I realized that my pc isnt to fault for Dead Space 3. It seems the game was poorly optimized, pretty much a console port, which pretty much sucks, 30 fps. I found a work around for fixing v sync by using radeon pro, so far I have not crashed at all with dead space 3. Thanks a lot.
 
Last edited:
Keep us in the loop for a bit. I reckon it best to wait about a week or so before really closing this case. The new updated drivers may just be alleviating the symptoms, not getting rid of them completely. While it sounds great they're working, it's best to make sure.
 
Back
Top