Resurrecting my dead GTX 460 after constant BSOD's

Peter Palvolgyi

Honorable
Jul 19, 2013
13
0
10,510
(I know my story is long, but please bear with me.)

Hello, I've had my computer for over 8 years now, I replaced nearly everything in my computer (except for the mobo, case, cooling etc...) and after all that pain I went through with my PC for over 8 years, this one took the cake.

2 days ago, after letting my PC on for 3 days straight (yes I do this alot), I started playing a match of League of Legends (video game). On a sudden moment after the match ended, a static type of screen full of fuzzy blocks and lines with all kinds of colors appeared and the whole screen froze up just as it was and I had to manually restart my computer. I thought my PC was overheating cause of being on for 3 days but unfortunately, after another match of league of legends the day after, the same thing happened, not a single sign of "too much heat" was going on, hell I even touched the card while the computer was on and it felt below body-temperature tbh.

I restarted my computer, logged in, and downloaded a few programs for monitoring and testing. I used a "memtest.exe" to stress out my ram to see if ram caused my GPU crashing. It wasn't the case. The weird thing is, I downloaded FURMARK, which is a GPU benchmarking/stressing utility that comes with live temperature & fanspeed monitoring and I started that aswell, and no problems occured even after 80°C which is weird, because when I play games like Call of Duty or League of Legends (way less graphic intensive then furmark) My temperature doesn't even go higher then 65°C-70°C and yet, my GPU freezes up and goes full static (red,green lines & blocks). And at Furmark, I can do the whole test without problems.

At that moment, I got mad. I searched up what it could be, I found out that removing my drivers and reinstalling it would help. So I went and checked if there were any updates and Yes, there were driver updates to be done. I did the driver updates. After the driver updates + additional "Nvidia Experience software", my PC didn't ask for a restart, so I didn't restart my pc but went and started Call of Duty to see if the problem still presists. And yes it did, but this time, the screen froze up and then it went black and came back normal with this error in the right corner of my screen "Display driver stopped responding and has recovered". After a few more times doing this, my pc finally got a BSOD and restarted. From that moment, I couldn't even get past the "starting windows logo" without having a moment of no signal followed by a BSOD having to do with "nvlddmkm.sys". So I went on and searched for more options. A guy said that it might have been a malware called "svchost.exe - bitcoin miner" so I went safe-mode (it works since it uses integrated VGA instead of my PCI-E), and I scanned my pc using Malwarebytes. And guess what, it did found that same exact malware I was looking for, I went on deleted it, Yes, I made sure that it was gone COMPLETELY (made several scans).

After that I went and reinstalled my drivers again, same problem, didn't fix it. But I had a strong feeling that this malware that ate my GPU had alot to do with my BSOD's. Anyways, since there was no more malware's on my pc, it could only be 2 things: Not a malware, or a malware that left a trace/scar.

After searching a lot, I somehow managed to start my computer in normal mode by using F8 and clicking on "Use Last known good configuration (advanced)". This way, I'm guessing the last known good configuration was the one without a driver. And yes, the computer started normally, but without any VGA drivers. My monitor was plugged in the GTX 460 and it was up and running, the fan was running, there was some heat on my card, but there were no drivers so the card wasn't recognized either, and as soon as I put drivers on my PC, I get a BSOD after "windows is starting" logo.

After another chunk of time searching, I managed to find a guy on a forum saying that he fixed his problem by resetting the CMOS and removing the battery & video card, leaving it there for a min or two, this way, the voltage going through the GPU would reset too. I tried this as well, no luck.

Now I'm all out of ideas, the last thing i read on forums, is a guy that literally baked his GPU for the soldering to melt again. The problem with this is, I have a strong feeling that my GPU isn't dead yet. There's just a malfunction that I didn't notice or my card is left with a scar from this ***** ***** **** **** * ******** Malware.

At this moment, I can only use my Integrated VGA in order to use my computer at all. If I try to install drivers, I get BSOD's at the windows logo.

If there's a very slight hope in me, then it's that someone else out there in the dark empty world with the same dangerous die-hard adventure lives, (like me) who could help me.

Thanks for any help in advance, Peter.

(Additional info below if the story is too long ^_^)



Things I already tried:
- CMOS Reset
- Driver Update/reinstall (did it like 50 times)
- Malware Removal (suspect found, did not fix the problem though)
- Clear dust, basic stuff
- Drive sweeper (completely remove drivers)
- Memtest, Furmark - before bsod (Both went successful which is kinda weird)

Things I can't/won't try:
- Put my GPU in another computer (don't wanna risk other friends computers)
- Bake my GPU (I don't think it's time for that yet.)
- RMA (I can't, bought my GPU 2 years ago used on ebay.)
- Contact manufacturers (I'm sorry, but they are just plain retarded.)

PC Replacements:
* CPU - Replaced an Intel E6300 with an Intel Q6600 3-4 years ago
* GPU - Replaced a Geforce 9600GT with a Palit Geforce GTX 460 2GB 2 years ago
* RAM - Replaced 4x 512mb with 4x 1GB 2 years ago
* PSU - Replaced a stock dell 300 watt with a unknown cheap brand 750W PSU 2 years ago (together with my GPU)

Current Specs:
- Dell Dimension E520 case (original PC I bought 8 years ago)
- DELL WG864 Socket 775 Motherboard
- 4GB DDR2 Ram
- Palit Geforce GTX 460 2GB
- Intel Core 2 Quad Q6600 @ 2.4ghz
- EZCool PS-1000, 700Watt PSU
 
Solution
Malware will not damage your GPU, CPU, etc.

Can a friend loan you a spare GPU to try? Could also buy a cheap GPU locally to test with (and keep as a spare).

If your windows installation is really hosed (this may be the case as well), the best bet is to back up your data and reinstall the OS.
Malware will not damage your GPU, CPU, etc.

Can a friend loan you a spare GPU to try? Could also buy a cheap GPU locally to test with (and keep as a spare).

If your windows installation is really hosed (this may be the case as well), the best bet is to back up your data and reinstall the OS.
 
Solution

Peter Palvolgyi

Honorable
Jul 19, 2013
13
0
10,510
After my holidays, I will reinstall my OS, I have a strong feeling that it might solve it as well, and yes, I could buy a GPU to see if it works, but it is quite obvious that it has to do something with my Nvidia drivers, so as long as it's not my own card, I won't have any problems.

Thank you for your reply, didn't expect someone to actually read the whole thing, haha.
 

Peter Palvolgyi

Honorable
Jul 19, 2013
13
0
10,510


Thanks for the reply,

Yea, I had all of these options in mind, and i'm willing to try 2 of those things.
Thought about replacing my PSU, but then I always asked my self how did that thing work without any problems for more then 2 years, it never had any problems whatsoever, and currently, I'm just doing the same daily routines with my PC, playing the same games etc.. and yet my PSU died just now. Just seems a little weird.
 
You don't have to have a specific event or something traumatic for PSU or GPU to fail. They wear out over time and the cheaper ones with low quality parts can just do so faster.

Still, having problems seeing how the PSU could keep you from installing drivers. That's why I keep coming back to messed up OS or GPU. Oh, and baking the GPU can actually work. I limped my 800GTX along for nearly a year doing that (total of three bakings).
 

Peter Palvolgyi

Honorable
Jul 19, 2013
13
0
10,510


Well, I believe that baking my GPU should save it from death, but I just don't think there's anything wrong with my GPU at all. This is also why I don't really wan't to install a fresh OS because if it's fixed, then I'll never know what caused the problem.

I tested a year old driver (version 295.xx) from nvidia, and now, after the windows screen I had no signal, then for a gilmpse of a second I could see my mouse and then it immediately froze up and gave me the typical BSOD with the nvlddmkm.sys error 0x116 type. (as always)
 

Peter Palvolgyi

Honorable
Jul 19, 2013
13
0
10,510
Update:

I used TdrDelay and TdrDdiDelay registry keys to increase the time having to recover my drivers if it stops, and I managed to log in. I immediately downloaded MSI Afterburner to see my GPU Usage. And what I thought was true, every time I do something, wether it's opening google chrome, or just doing some simple stuff, the GPU Usage jumps to 98-99%. Anyone knows any cause of all this, and maybe any solutions?
 

radeonDEUS

Honorable
Jan 28, 2014
1
0
10,510
Hey Peter!

I have this same issue going on, using 2 EVGA GTX460's in SLI with a GT430 for physx. Able to boot using any single GPU in the 3rd PCIe slot, but as soon as i try to put ANY of the 3 cards in the first PCIe slot, BSOD after windows logo on boot. SUPER FRUSTRATING. I RMA'd my board (Gigabyte 990FX) and they returned it without issues. Weird thing is, all 3 cards work? WTH!!! Did you get your issue solved???