- May 19, 2011
- 21,340
- 16,553
- 136
Intermittent issue, but I'm typically getting this BSOD when playing XCOM Enemy Within. Sometimes the game will show graphics corruptions then recover, sometimes it won't recover, sometimes I get this BSOD instead. I've had plenty of hours playing time without any problems. The issue doesn't appear to be linked to running the game for a specific length of time.
PC spec:
AMD Phenom II X4 960T (3.0GHz, 3.4GHz turbo mode)
with two unlocked cores making it an X6
Arctic Cooling Freezer 13 HSF
ASUS M4A89GTD PRO/USB3
Kingston 4GB (2x2GB) HyperX DDR3-1600
ASUS GeForce GTX 750 Ti 2GB
Samsung 840 PRO 256GB (boot)
Western Digital Black 1TB
Samsung DVDRW SATA
LG DVDRW SATA
Corsair VX450W
Coolermaster Elite 330
Windows 7 Home Premium 64-bit
Iiyama E2208HDS 22" 1080p LCD
I think it's unlikely to be the graphics card because I think I had similar problems pre-dating it, just not anywhere near as often. It would happen from time to time in Skyrim. I've never had it in StarCraft 2 though.
On a previous occasion of playing XCOM and ending up with a dodgy scenario (probably the BSOD), I started troubleshooting. I undervolt the CPU normally, so I got rid of that setting (ie. set the mobo back to default CPU voltage settings). I also run the RAM at 1.5v despite the board always saying that I ought to let it increase the voltage to the modules if it needs to. I let it do this at the same time as losing the undervolt but I had another crash within a few days. I also noticed that the mobo was running the RAM at 1.8V at that point. After the crash I reinstated the undervolt and RAM voltage.
Fast forward to today, another BSOD during XCOM. I thought about the possibility of the extra cores being unlocked somehow being a problem despite the fact that I unlocked them something like a couple of years ago and I very likely prime95'd the machine for a good long while afterwards. So today I prime95'd it while browsing and downloading a file (an nvidia demo to stress the graphics card). Within probably ten minutes the machine BSOD'd again (same code), making me think I was on to something, however I've now been running another prime95 run (both times on 'maximum heat') for just over half an hour.
One really odd thing I've noticed is that because I spent a lot of time examining the machine's behaviour (stress testing etc) after undervolting it a few years ago, I know for a fact that its usual idle CPU voltage is 1.224v and its load voltage is usually (IIRC) 1.38v. However, today while under load the CPU voltage is a pretty consistent 1.176v (sometimes going up to 1.188v). Both HWMonitor and CPU-Z are reporting the same voltage. This board only allows me to reduce the voltage by approx -0.06v so this one puzzles me. When I reboot I'll see what the BIOS thinks the setting is, even though when I reinstated the undervolt before I know the BIOS said 1.224v in its monitoring info.
A side possibility I'm wondering about is perhaps the PSU isn't up to scratch any more, though it would surprise me if that was the case as I'd normally expect a power down, freeze or spontaneous reboot if the PSU was the problem.
Temps are all OK.
As a precaution I'll probably set it back to quad-core and see how it goes.
- edit - Going back to quad-core hasn't helped (as well as stock cpu voltage). I even had two recovered driver crashes during XCOM in the same session today. The event log has this to say about them:
The recoveries happened within 14 minutes of each other. I waited long enough after pretty patterns appeared on the screen then it allowed me to continue playing.
PC spec:
AMD Phenom II X4 960T (3.0GHz, 3.4GHz turbo mode)
with two unlocked cores making it an X6
Arctic Cooling Freezer 13 HSF
ASUS M4A89GTD PRO/USB3
Kingston 4GB (2x2GB) HyperX DDR3-1600
ASUS GeForce GTX 750 Ti 2GB
Samsung 840 PRO 256GB (boot)
Western Digital Black 1TB
Samsung DVDRW SATA
LG DVDRW SATA
Corsair VX450W
Coolermaster Elite 330
Windows 7 Home Premium 64-bit
Iiyama E2208HDS 22" 1080p LCD
I think it's unlikely to be the graphics card because I think I had similar problems pre-dating it, just not anywhere near as often. It would happen from time to time in Skyrim. I've never had it in StarCraft 2 though.
On a previous occasion of playing XCOM and ending up with a dodgy scenario (probably the BSOD), I started troubleshooting. I undervolt the CPU normally, so I got rid of that setting (ie. set the mobo back to default CPU voltage settings). I also run the RAM at 1.5v despite the board always saying that I ought to let it increase the voltage to the modules if it needs to. I let it do this at the same time as losing the undervolt but I had another crash within a few days. I also noticed that the mobo was running the RAM at 1.8V at that point. After the crash I reinstated the undervolt and RAM voltage.
Fast forward to today, another BSOD during XCOM. I thought about the possibility of the extra cores being unlocked somehow being a problem despite the fact that I unlocked them something like a couple of years ago and I very likely prime95'd the machine for a good long while afterwards. So today I prime95'd it while browsing and downloading a file (an nvidia demo to stress the graphics card). Within probably ten minutes the machine BSOD'd again (same code), making me think I was on to something, however I've now been running another prime95 run (both times on 'maximum heat') for just over half an hour.
One really odd thing I've noticed is that because I spent a lot of time examining the machine's behaviour (stress testing etc) after undervolting it a few years ago, I know for a fact that its usual idle CPU voltage is 1.224v and its load voltage is usually (IIRC) 1.38v. However, today while under load the CPU voltage is a pretty consistent 1.176v (sometimes going up to 1.188v). Both HWMonitor and CPU-Z are reporting the same voltage. This board only allows me to reduce the voltage by approx -0.06v so this one puzzles me. When I reboot I'll see what the BIOS thinks the setting is, even though when I reinstated the undervolt before I know the BIOS said 1.224v in its monitoring info.
A side possibility I'm wondering about is perhaps the PSU isn't up to scratch any more, though it would surprise me if that was the case as I'd normally expect a power down, freeze or spontaneous reboot if the PSU was the problem.
Temps are all OK.
As a precaution I'll probably set it back to quad-core and see how it goes.
- edit - Going back to quad-core hasn't helped (as well as stock cpu voltage). I even had two recovered driver crashes during XCOM in the same session today. The event log has this to say about them:
event log said:
event log said:
The recoveries happened within 14 minutes of each other. I waited long enough after pretty patterns appeared on the screen then it allowed me to continue playing.
Last edited:
Facebook
Twitter