Bugcheck 116 & B4, Sisoft Lockup

imrazor said:

I'm posting this from work, so I don't have the memory dumps from the BSODs but I will attempt to post them when I get home. I'm hoping someone will have seen a similar problem and now how to resolve it.

Up until 5 days ago I had XP64 installed on my primary system. It worked perfectly, except once every two or three months a JMRAID.SYS driver would blue screen the system. I then installed an MSDN copy of Win7 Home Premium in trial mode, but it immediately began blue screening. Here are the specs:

Gigabyte P35-DS3L
Intel Q6600 @ 2.4 GHz
3GB DDR2 667
XFX Radeon 5750 1GB
Western Digital 7200RPM 640GB SATA II
NEC 16X DVD-RW

The symptoms are that 1-2 hours after booting the system it will crash with a bugcheck B4 or 116 in drivers NTOSKRNL.SYS or DGXKRNL.SYS. The other symptom is that SiSoft Sandra will lock up in the GPGPU Encryption Test, but only when preceding it with other tests in burn-in mode. Running the test immediately after boot will work fine. "Lock up" means that the system freezes; even the mouse won't move. Upon reboot there are no events in event viewer that correspond with the time of lockup. Sisoft leaves a blank log, so there's no telling what part of the test is locking the system up.

What I've tried-

- Reinstalling Win7
- Swapping in an Nvidia 8800GT - this works fine, but is a very dated card.
- Driver verifier - detected bad behavior in AVG so installed Avast
- Running with only 2GB (2x1GB instead of 3x1GB)
- Trying the Radeon in a different system. Got a VPU recover error, but I was trying to run it on a system with 1.5GB of RAM.
- Monitoring temps. The CPU tops out at around 71C (but not during the GPU Encryption test) and the video card at 66C.
- Blowing the dust out of the video card (there was some but not a whole lot.)
- Installing an older version of the Catalyst drivers (11.3)
- Trying a different monitor

I wanted to try Win7 before buying it, and it looks like that was a good idea...

I'll try to force a crash by playing a video game and post the memory dump once I get a chance to get home. The old dumps have been wiped out by reinstalls of Win7.

Bugcheck 116 and 64 are TDR video timeouts.

"It's not a true crash, in the sense that the bluescreen was initiated only because the combination of video driver and video hardware was being unresponsive, and not because of any synchronous processing exception.

Since Vista, the "Timeout Detection and Recovery" (TDR) components of the OS video subsystem have been capable of doing some truly impressive things to try to recover from issues which would have caused earlier OSs like XP to crash.

As a last resort, the TDR subsystem sends the video driver a "please restart yourself now!" command and waits a few seconds.

If there's no response, the OS concludes that the video driver/hardware combo has truly collapsed in a heap, and it fires off that stop 0x116 BSOD.

If playing with video driver versions hasn't helped, make sure the box is not overheating.

Try removing a side panel and aiming a big mains fan straight at the motherboard and GPU.

Run it like that for a few hours or days - long enough to ascertain whether cooler temperatures make a difference.

If so, it might be as simple as dust buildup and subsequently inadequate cooling.

I would download cpu-z and gpu-z (both free) and keep an eye on the video temps Let us know if you need help STOP 0x116: VIDEO_TDR_ERROR troubleshooting - Windows 7 Forums STOP 0x116: VIDEO_TDR_ERROR troubleshooting

Try this and let us know if it doesn't help

I have been monitoring the temps with HWMonitor (from the CPU-Z/GPU-Z folks) and Speedfan. CPU temps are peaking at 71C and GPU at 66C. The GPU temp seems well within bounds, but the CPU seems just a tad high. Does that agree with your experience?

Based on some of the advice I've read here and elsewhere I intend to try these-

- Get rid of Catalyst Control Center (uninstall Catalyst, install driver only pkg)
- Disable ATI Event/Hotkey Poller services
- Hardcode CPU cores and memory at boot with MSCONFIG/BOOT.INI/Advanced
- Install ATI Tray Tools and set fan at 100% and/or disable power mgmt and/or set 2d clock speed to the same as 3d speed
- Disable unnecessary hardware on mobo (parallel port fingered by bugcheck B4)
- Update non-video drivers
- Run SFC /SCANNOW
- Check BIOS for "PlugNPlay OS" setting enabled, perhaps ACPI as well
- And, of course, grab the minidump next time it crashes.

Do any of these seem likely to work? Do you recommend any steps over the others? Any I should avoid?

Thanks...

Blew out the CPU and dropped CPU temp by 12C. However, SiSoft was still locking up on the GPGPU encryption test. Ran Furmark and got a BSOD again - this time bugcheck 9f. Minidump attached. The resulting crash corrupted my hard drive, though chkdsk seems to have recovered the partition. For now I've put in a different video card to prevent further corruption. Let me know if you need the full diagnostic info from jgriff's utilities, though it may not be an accurate picture without the 5750 installed.

imrazor said:

Blew out the CPU and dropped CPU temp by 12C. However, SiSoft was still locking up on the GPGPU encryption test. Ran Furmark and got a BSOD again - this time bugcheck 9f. Minidump attached. The resulting crash corrupted my hard drive, though chkdsk seems to have recovered the partition. For now I've put in a different video card to prevent further corruption. Let me know if you need the full diagnostic info from jgriff's utilities, though it may not be an accurate picture without the 5750 installed.

This single crash still points to your video driver (or subsystem). I would re-install the newest driver using this moethod and if this doesnt help help you need to start looking at temps, power, video card, or vram.

When upgrading your graphic driver you MUST remove all traces of the current driver. In order to do that we recommend using Phyxion.net - Driver Sweeper

When it is removed then download and install the fresh copy.

This single crash still points to your video driver (or subsystem). I would re-install the newest driver using this moethod and if this doesnt help help you need to start looking at temps, power, video card, or vram.

When upgrading your graphic driver you MUST remove all traces of the current driver. In order to do that we recommend using Phyxion.net - Driver Sweeper

When it is removed then download and install the fresh copy.

This weekend I had the opportunity to do a more comprehensive test. I had an eval copy of Server 2003 handy, so I put that on another partition to see if would be more stable than Win7. Indeed, it didn't blue screen during testing. After some initial hiccups, I got Catalyst 11.6 and DirectX installed and began testing with Furmark. The test went well for 25 minutes, then I started seeing serious anomalies. Funny thing is the temperature peaked at 77C about 10 minutes in, but the artifacts didn't appear until 25 minutes into the test. At any rate XFX has issued an RMA and I'll be returning the card Thursday, when I'll have time to do the RMA rigamarole. Thanks to those who posted.

Sorry to resurrect a dead thread, but I wanted to post a followup to my last post. After RMA'ing the card, XFX received and tested the card and pronounced it "NO PROBLEM FOUND". However, they went ahead and sent me a different card anyway. To my surprise, it has passed many hours of gameplay without BSODing, as well as a few hours of Furmark. Looking closely at the details of the card I discovered that it was a slightly different model number. The new card turned out to be PCI Express 2.0 instead of the PCIe 2.1 old card, which may explain why it works better in my PCIe 1.1 motherboard. A happy ending to the tale, but also a warning to folks with older motherboards....

imrazor said:

Sorry to resurrect a dead thread, but I wanted to post a followup to my last post. After RMA'ing the card, XFX received and tested the card and pronounced it "NO PROBLEM FOUND". However, they went ahead and sent me a different card anyway. To my surprise, it has passed many hours of gameplay without BSODing, as well as a few hours of Furmark. Looking closely at the details of the card I discovered that it was a slightly different model number. The new card turned out to be PCI Express 2.0 instead of the PCIe 2.1 old card, which may explain why it works better in my PCIe 1.1 motherboard. A happy ending to the tale, but also a warning to folks with older motherboards....

Thank you for reporting back.
Safe surfing.