Weird Pixels that lead to BSoD

Hello, I have had this issue for a couple months now and I still cannot figure out the cause. I believe it to be hardware related but cannot get error returns from any program I've tested with. Below I will post screen shots of what the issue appears as, links to BSoD dumps and more details.

I can play games that run direct x in full screen and never see an issue, never encounter a crash or BSoD, never see artifacts.

If I browse the net I can usually create the issue over some time, using a program such as premiere pro usually causes the issue and BSoD within 20-30 mins especially when rendering. If using vmware machines I can make the issue happen after some time as well. If I continue browsing the net when the issues begin to happen my browser will start to crash for no apparent reason as well. Windows Explorer will also crash randomly once it gets bad enough.

After I began to encounter issues I took the laptop apart and cleaned it out very well, how ever there was very little dust inside it and no vents were even partially clogged.

Here is system specs:
ASUS G73JH-X2
CPU i7-720QM Clarksfield
8GB 2GB*4 - PC3-10600 DDR3 SDRAM - ASU1333D3S9DR8/2G
ATI Mobility Radeon HD 5870 [ASUS] - 1GB DDR5
Hard Drive - Seagate ST9500420AS - 7200 RPM

I have Installed the Latest Bios on the Motherboard, Video Card, Hard drive

I have tried doing a system restore to factory settings, a full format and clean install using a windows 7 disk and all ends up with same issues, which is why I believe it to be a hardware issue.

I have ran Memtest86 for over 3 hours, 3 full cycles. Prime95 for 2+ hours. Many hours of other various test programs such as BurnInTest. None have ever failed or returned an error message.

Here are a few screen shots of what I see, Also one thing when I screen shot the issue sometimes its not exactly as I seen it when I took the screenshot.





Here are links to 2 of the dump files from BSoD's the uploaded said their an invalid file to be uploaded so I will just give links.
https://dl.dropbox.com/u/7677274/g73jh/112512-38766-01.dmp
https://dl.dropbox.com/u/7677274/g73jh/112512-38828-01.dmp

If there is anymore information I can provide to get this issue resolved I would be more than glad to provide it.
Thanks a lot for any help.

 

The graphics card is overheating?

I thought it was relatively easy to get to the insides of the g73. Maybe it's possible to salvage it with a repaste. The graphics card, at least. ..And ..I don't know.. 2 years old, and those warm components -- it could be about time for a repaste soon anyway.

 

Sorry I forgot to mention that I did do a re paste about 3 weeks ago both cpu and gpu. The temps stayed pretty much the same, idle temps dropped a bit, load temps stayed about the same (within 2-3C but that could even be due to ambient change). I found some Corsair ram on sale locally and went and picked that up today to give a shot. Its the only thing I can come up with honestly. The more I start to think of what causes it is having very memory intensive apps open or have a lot of windows open and such that would be loading into the ram. Games are not as ram intensive as a lot of other programs and I would think games clear ram constantly and don't really store it for extended periods of time.

I was thinking it to be video at first but if I can game 6-12 hrs a day and never see an artifact or crash during game play, never have extreme temperatures. The chance of it really being video card I think is slim. I will keep this post updated if the problem continues.

Thanks.

 

Then it's probably not that, no. The only other time I've ever seen something that sounds like what you're describing (the attachments don't show up) is when I overclocked the bus on an old system, and the southbridge got too warm. What made that so difficult to figure out then was that I got interrupt conflicts or random software hangs because of corrupted data. But when trying to reproduce it after a reboot, it didn't happen at exactly the same situation as before..

And gaming works, memtests work, etc. - the only times you will see corruption then (if it is that) is when you are running heavy IO, or the kind of tasks that shift data on the bus over a long period of time.

But I don't know if the southbridge is integrated on the die now..? Might be something to go on, though. Maybe it's the sata controller, or the wifi card, for example, that keep causing errors when direct IO to some buffer happens.

Doesn't necessarily have to be caused by overheating either - a new device-driver, a sound driver that doesn't use the wdm interfaces the right way. Wifi with too long input buffers.

..remember to make a post if you figure it out, though. Is interesting.

 

Thanks for the info, and I have fixed the screenshots.

Now your latest post has opened my eyes onto new areas and I will do more investigating. After Installing the new ram I have noticed a very big difference but did end up with one BSoD last night. I will attach the dump to this post of that BSoD but it was caused after about 30mins of some really hard use (starting up 6 vmwares and doing some updates and such maxing out the hard drive IO the whole time but was able to boot back up and never had a hiccup for the next hour of doing same task). Being that it feel very stable now compared to before I changed out the ram, I am also thinking I need to do a 100% full wipe of the hard drive and do clean install, maybe even write 0's to the whole thing. I have another hard drive from a laptop that isn't being used currently was thinking of pulling that out and slapping it in this laptop and giving it a go for a week or so.

https://dl.dropbox.com/u/7677274/g73jh/112712-25240-01.dmp


Thanks again for the help and I will be sure to keep posting my findings.

 

well found out one set of the ram I just purchased was bad so i returned that and got a replacement ran memtest86 all was good (I believe my old ram is still good for after a lot of memtest86 it never showed defective so it will be nice to have as spare, i like the 16gb I have now makes huge difference in vmware) then after beating my head off the wall for basically 2 days straight, I downgraded vbios, motherboard bios to stock, used a 2nd hard drive installed fresh copy of windows, installed only sp1, went and installed video drivers, opened browser and started to notice something that I could finally make happen every time. It can be seen here in this video Weird Blocks on mouse over links and text - YouTube. I did a search for archived video drivers and came to notice someone with a different model 5xxx series ati card said they had issues (never stated specifics) on a 12.6 driver and reverted back to 12.4 well I had nothing to loose so went and uninstalled my current driver which was the latest driver 12.10, ran ati crap cleaner, and installed 12.4 series driver and bam no more black block(could this be the resolution to my weird pixels as well? lets hope so), so now I have re updated the vbios and motherboard bios to current versions and in almost 24 hrs no bsod, no weird pixels, seams all is good now lets just hope it stays this way. Thanks for all the help and being inspirational for me to finally dig though this and resolve the issue.

The site I found the drivers on was An archive of AMD Mobility Catalyst Windows 7 x64 drivers – PCR's notepad

Thanks again.

 

Going to mark this issue as solved. 0 Issues since my last post.

 

Just curious - when you tested your RAM, did you test one module at a time, as well as together? A lot of video issues seem to arise from either:

A. Bad driver
B. Overclocking
C. Overheating

Glad to know you were able to resolve it though.