Maximum PC

I have the common P7801-u video card crashing (to BSOD) problem. The cause of the crash, according the BSOD, is nvlddmkm.s When playing a game or video for any length of time, the video card crashes. I am trying to increase my video card's stability so I can get in a couple of hours of gaming before it crashes.

So far, I've underclocked it by 20% (now at a clock of 480 instead of 600 for the GPU core), undervolted it to .97V (using the bios described here: http://forum.notebookreview.com/gateway ... 7805u.html)) and turned off PowerMizer. I keep the laptop on top of a fan. It idles at 45-50 C and never reaches temps above 80 C.

I'm using the driver supplied by the laptop manufacturer, which is 186.32. Perhaps there is a driver that might be more stable I could try?

I'm running Windows 7 x64. I haven't altered the hardware this model comes with.
Intel Core 2 Duo CPU P8400 2.26GHz. 4 GB RAM.

Please supply whatever possibly useful ideas you can come up with! I've been working on this off and on by myself for about a year and am quite frustrated. Thanks in advance!

Definitely try updating the driver. That might help... you really should have RMAd the laptop when you found you couldn't fix the problem easily.

Others have attempted to do that with no success - also, it was already out of warranty by then.

I'm going to try the latest nvidia driver and the latest laptopvideo2go inf-modified driver. I've done that in the past to no avail, but there are newer ones now...

Tried both. Both still crash.

I would like to see what happens when its stock+updated. Go back to the orginal BIOS and stock clocks, clean out the fans (if your confident add new thermal paste) and update to the lastest nVidia drivers here. You also might want to check the temps of your CPU too.

Same problem with original BIOS and clocks, even at low temperatures. Temperature appears to be unrelated.

My CPU temps are 20-25 C for either of the three BIOSes I have tried (original, updated, customized.)

If there was a way to get a new video card, I'd probably do that. I've tried everything I can think of and am sooooo frustrated.

Maybe a clean Windows install? If all else fails you can probably RMA it, or get a replacement 9800m gts.

It's out of warranty. Do you think an RMA would still be possible at all?

The 9800M GTS is "integrated," meaning they'd have to replace the whole motherboard - I think.

I highly doubt that its integrated. Even on the newegg page it say dedicated.

You can either see if the manufacture can sell you a replacement or get one off of eBay.

Hmm! It appears I misread something at some point. I can't seem to find any information about how to replace the video card in this laptop though. Any ideas?

Also, I've done a clean windows install several times with no luck. It simply lengthens the amount of time before a crash from about 20-60 seconds of usage to 2 hours of usage. Within about a week, I'm always back down to about 20-60 seconds again.

Do you folks think he may have a thermal sensor problem. I note that he reports that his CPU temp is 20-25 degC. You would just about have to be sitting in a refrigerator to get that temp even at idle. Especially with a laptop.

I don't know if the sensors concerned are on the mobo or built into the CPU. If the latter he may be able to just get away with changing out the CPU. Assuming he wants to spend the money trying to repair a laptop.

You may also want to get a copy of HWMon and run it to see what your real time temps are doing.

I'm not willing to spend a LOT on fixing my laptop. As I've said before, this is a common problem with this model - maybe they distributed a lot of faulty vid cards or there's something incompatible with the card in the motherboard? I'd spend some, maybe enough for a new vid card if I could determine that was the problem and knew how to safely replace the card.

New information: I'm able to run modern 3d games in WINE (something like windows emulation in Linux) without any problems. Problems only occur in Windows.

My laptop rests on a Rosewill notebook cooler, which lowers the temps a bit. (See it here) I use HWmonitor and GPU-Z already. Here are my current idle(+Firefox) HWmon temps and some other info:

Hardware Monitors
-------------------------------------------------------------------------

Hardware monitor ACPI
Temperature 0 50°C (122°F) [0xCA0] (TZS0)
Temperature 1 46°C (114°F) [0xC78] (TZS1)

Hardware monitor Intel Mobile Core 2 Duo P8400
Temperature 0 26°C (78°F) [0x4A] (Core #0)
Temperature 1 26°C (78°F) [0x4A] (Core #1)

Hardware monitor NVIDIA GeForce 9800M GTS
Temperature 0 54°C (129°F) (GPU Core)

Hardware monitor Hitachi HTS723232L9A360
Temperature 0 36°C (96°F) [0x24] (Assembly)

Hardware monitor Battery 1
Voltage 0 12.45 Volts [0x30A1] (Current Voltage)
Capacity 0 86580 mWh [0x15234] (Designed Capacity)
Capacity 1 75602 mWh [0x12752] (Full Charge Capacity)
Capacity 2 75602 mWh [0x12752] (Current Capacity)
Level 0 13 pc [0x57] (Wear Level)
Level 1 100 pc [0x64] (Charge Level)

With everything at stock and w/o the cooler what are the temps? Even with undervolting and underclocking my P8600 hits 37C idle. They do seem oddly low. (That with HWMonitor also, try RealTemp for readings too)

Edit: Actually, I got 22C from a cold boot. Where do your temps fall to after being warmed up?

Still right about there. Same temps.

Did you run Driver Cleaner after you uninstalled each driver? If not, try it. Seems like a driver related problem - a lot of the pages I read about the solutions to the nvlddmkm.sys BSOD are because of old driver problems.

My Google search results