New GPU driver = BSOD?

3rd October 2006, 03:11

I recently updated my Forceware driver for my 7800GT. I wish I would have written down which dirver I was using before the update, but I updated to the 92.91 driver set. I think I saw a slight performance upgrade, but during a LAN (about 4 days after updating) my computer froze during some games. Then it started to BSOD during normal operations. I first thought this might be the nForce 4 Chipset drivers which I updated. That didn't help. So I downgradded my graphics card driver to the 91.52 set. That didn't work. So I reset my OCed settings in the BIOS back to default, and now everything runs smooth.

The problem is that I don't know what the problem is! I just don't understand how a graphics card driver would send my entire oced configuration out the window. I've tried minor voltage increases here and there to see if maybe it cause some unstability, but the outcomes where even worst. I just don't know what to do now. I even upgraded the BIOS which was almost a year old and it still didn't help. The system runs smooth with sock settings, but why would it now, after running stable for like 4 months since overclocking crash on me?

I run an A64 X2 3800+ with the Abit Fatal1ty AN8 SLI, 2gb Crosair DDR, and the 7800GT.

Anyone have any ideas what the problem might be?

3rd October 2006, 07:18

I doubt it has something to do with the drivers since they've worked before the LAN.
My guess is that something went wrong with your cooling when you moved your case.

3rd October 2006, 13:16

Well the LAN was at my house actually so the only movement the case recieved was about a 2inch slid over towards the edge of my desk. As for the cooling, minotrs showed the same report as always and temps where at a safe level.

jmke · 3rd October 2006, 13:45

Quote:

So I reset my OCed settings in the BIOS back to default, and now everything runs smooth.

The problem is that I don't know what the problem is!

the OC is the problem, of course

restart from scratch for finding the maximum stable overclock, not every OS, driver, application is as sensitive to OC as the other, might well be that the new driver and bios and chipset driver are more prone to crash when errors are dedected through OC. It's not because it worked well for 4 months that the system was 100% stable, just maybe no errors popping up "worse" enough to crash the system.

Prime95+3DMark2001SE loops, Folding@Home and MEmtest86+ should help you to maximise your system OC without sacrificing stability

6th October 2006, 03:37

Back at

2.6Ghz @ 1.575v
HT 4x

but now something else seems to be bothering me. I'm trying to run my ram at DDR-333 CAS: 3-3-3-8 @ 2.7v. It seems to be working until windows GUI and other things start to mess-up. So increase the voltage right? After all stock runs at 2.6v so we have some nice overhead room. So I bump the sucker up to 2.75v and the system shuts-down in mid air. And it does this for every voltage level (in .05v increments) up to 2.95v (I can go to 3.0v, but MOBO starts to set off alarms after that) and the same thing happens. The system just shuts down mid-air right when it's loading Windows.

So, 2.7v barley works keeping it stable enough to boot and run a few apps maybe. But anything below that voltage level and system won't boot, anything above and the system dies while loading Windows.

Is it me, or does that sound a bit strange? Is there some factor that changes (that I don't know off) when I increase the ram voltage levels that's causing this to happen?

I'm a bit lost here, cause I even got my CPU running 100Mhz+ from my last OC, and the ram, which at DDR-333 runs at 217Mhz with timming of 3-3-3-8 should be able to handle thoes settings with a bit of extra voltage.

11th October 2006, 03:42

I think I solved the problem. I had the nf4 oced from 1.5v to 1.6v, and when I installed the new nf4 drivers it must have had a nervous breakdown. Anyways, back at 2.5Ghz @ 1.575v and ram running in DDR333 CAS: 2.5-3-3-8. Stable by memtest, OCCT, and RMA Stability analizer.

I was wondering. I think I could run the ram at 3-3-3-6 or what I have it at now 2.5-3-3-8. Which one would be more effective in theory?

jmke · 11th October 2006, 06:18

on AMD lower timings do impact performance, but less than you be able to detect; I'd say, run your favourite game + FRAPS and check the difference

3rd October 2006, 03:11	#1
Nessmaster Posts: n/a	New GPU driver = BSOD? I recently updated my Forceware driver for my 7800GT. I wish I would have written down which dirver I was using before the update, but I updated to the 92.91 driver set. I think I saw a slight performance upgrade, but during a LAN (about 4 days after updating) my computer froze during some games. Then it started to BSOD during normal operations. I first thought this might be the nForce 4 Chipset drivers which I updated. That didn't help. So I downgradded my graphics card driver to the 91.52 set. That didn't work. So I reset my OCed settings in the BIOS back to default, and now everything runs smooth. The problem is that I don't know what the problem is! I just don't understand how a graphics card driver would send my entire oced configuration out the window. I've tried minor voltage increases here and there to see if maybe it cause some unstability, but the outcomes where even worst. I just don't know what to do now. I even upgraded the BIOS which was almost a year old and it still didn't help. The system runs smooth with sock settings, but why would it now, after running stable for like 4 months since overclocking crash on me? I run an A64 X2 3800+ with the Abit Fatal1ty AN8 SLI, 2gb Crosair DDR, and the 7800GT. Anyone have any ideas what the problem might be?

11th October 2006, 06:18	#7
jmke Madshrimp Join Date: May 2002 Location: 7090/Belgium Posts: 79,021	on AMD lower timings do impact performance, but less than you be able to detect; I'd say, run your favourite game + FRAPS and check the difference __________________

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
AMD Unlocks 3D Internet Potential with OpenGL ES 2.0 Driver	jmke	WebNews	0	26th July 2010 08:56
NVIDIA Geforce 196.34 Beta Driver Fixed Bug in GPU Overclocking	jmke	WebNews	0	27th January 2010 10:12
Optimal GPU Overclocking with GIGABYTE’s New Super Overclock Series	jmke	WebNews	1	2nd September 2009 19:31
ATI Catalyst 9.8 driver available for download	jmke	WebNews	0	15th August 2009 07:41
AMD Supports Microsoft Windows® 7 in ATI Catalyst™ 9.3 Unified Driver	jmke	WebNews	0	18th March 2009 11:29
NVIDIA Forceware 177.92 Beta Driver Promises Performance Boost	jmke	WebNews	0	26th August 2008 17:14
Corsair Launches Nautilus GPU Water Cooling Blocks	jmke	WebNews	0	11th October 2006 21:49
GPU broken or not?	Pirre	Hardware/Software Problems, Bugs	0	20th July 2006 20:25
NVIDIA Unleashes World's Fastest Mobile GPU	Sidney	WebNews	0	25th February 2005 00:45
PlayStation 3 GPU: NV40 and NV50 Hybrid with XDR DRAM, Says Report	jmke	WebNews	0	28th December 2004 22:12

3rd October 2006, 07:18	#2
wutske Posts: n/a	I doubt it has something to do with the drivers since they've worked before the LAN. My guess is that something went wrong with your cooling when you moved your case.

3rd October 2006, 13:16	#3
Nessmaster Posts: n/a	Well the LAN was at my house actually so the only movement the case recieved was about a 2inch slid over towards the edge of my desk. As for the cooling, minotrs showed the same report as always and temps where at a safe level.

6th October 2006, 03:37	#5
Nessmaster Posts: n/a	Back at 2.6Ghz @ 1.575v HT 4x but now something else seems to be bothering me. I'm trying to run my ram at DDR-333 CAS: 3-3-3-8 @ 2.7v. It seems to be working until windows GUI and other things start to mess-up. So increase the voltage right? After all stock runs at 2.6v so we have some nice overhead room. So I bump the sucker up to 2.75v and the system shuts-down in mid air. And it does this for every voltage level (in .05v increments) up to 2.95v (I can go to 3.0v, but MOBO starts to set off alarms after that) and the same thing happens. The system just shuts down mid-air right when it's loading Windows. So, 2.7v barley works keeping it stable enough to boot and run a few apps maybe. But anything below that voltage level and system won't boot, anything above and the system dies while loading Windows. Is it me, or does that sound a bit strange? Is there some factor that changes (that I don't know off) when I increase the ram voltage levels that's causing this to happen? I'm a bit lost here, cause I even got my CPU running 100Mhz+ from my last OC, and the ram, which at DDR-333 runs at 217Mhz with timming of 3-3-3-8 should be able to handle thoes settings with a bit of extra voltage.

11th October 2006, 03:42	#6
Nessmaster Posts: n/a	I think I solved the problem. I had the nf4 oced from 1.5v to 1.6v, and when I installed the new nf4 drivers it must have had a nervous breakdown. Anyways, back at 2.5Ghz @ 1.575v and ram running in DDR333 CAS: 2.5-3-3-8. Stable by memtest, OCCT, and RMA Stability analizer. I was wondering. I think I could run the ram at 3-3-3-6 or what I have it at now 2.5-3-3-8. Which one would be more effective in theory?