I've started getting some artifacts (Colored geometric shapes, like Pink triangles and green or blue rectangles, etc. flashing all over the screen) in BFBC2 (Battlefield Bad Company 2) all of a sudden and not sure what the culprit is. I'm hoping it's not my GPU. 🙁 Never got artifacts in any other games, but i've mostly just been playing BFBC2 for the past 2 months and it's the most demanding, it seems.
I built the comp. about 6 months ago.
Intel Core i5-3570K Ivy Bridge 3.4GHz
ZALMAN CNPS9500A-LED 92mm Cooler
GELID Solutions GC-Extreme Thermal Compound
Kingston HyperX Black 8GB (2 x 4GB) DDR3 1600
SAMSUNG 840 Pro Series 128GB SATA III MLC Internal Solid State Drive
ASRock Z77 Extreme4
Microsoft Windows 7 Home Premium 64-bit
MSI HAWK Radeon HD 7870 GHz Edition 2GB 256-bit GDDR5 PCI Express 3.0
ASUS VW224U 22 inch 2ms(GTG) Widescreen LCD 1680x1050 16:10
Creative Soundblaster X-Fi Go! Pro USB Audio System
Antec Nine Hundred w/ Antec EarthWatts Green 750W PSU
I originally had the CPU overclocked to 4.5 GHz and was stable with Intel Thermal Analysis Tool (TAT) (though I think it hit 80c once.), except BFBC2 would crash intermittently until I lowered it to 4.2. It ran at 4.2 all this time with no problems. The cooler and thermal paste don't seem adequate. @ 4.2 GHz the CPU would hit about 67c max with BFBC2. CPU voltages were basically stock, not much overvoltage.
The GPU hits about 60c I think, with CCC's worthless auto fan control. (Has latest driver - 13.4.) GPU is not overclocked, except factory. 1100/1200 Tried to overclock the GPU when I first got it and it wouldn't. It didn't post or black screen or something, can't remember really, so I left it alone, since i've never had such a powerful/expensive card before.
I had a Raidmax Smilodon case at first, then, after many months I started to get the artifacts and thought something was overheating. I eventually got the Antec Nine Hundred case with much better fans.
While switching cases I noticed that the plastic piece on my Kingston SSD's power connector had snapped off. I RMA'd it and got the Samsung. The Samsung was cloned from the Kingston and I have since tried to roll back my GPU drivers and got errors, so had to keep the latest driver installed. (I also try to keep the Intel onboard Graphics driver uninstalled but Windows update sometimes sneaks it in.)
I read that my RAM is actually rated @ 1333, even though Kingston sold it as 1600, so not sure if that is the problem. I've just now dropped the RAM speed back to 1333 and have not tested it like that yet. Been running @ 1600 since I built it.
Before that i've verified my BFBC2 Steam files, reapplied thermal paste using the pea method instead of the spread method, increased fan speeds in my case. Bought another fan for the side window. Started using manual GPU fan control to keep GPU temps under 60c. Reduced my CPU to stock speeds.
The artifacts didn't happen back in the winter. And they seem to happen after playing for a couple of hours, never right off the bat, which seems to indicate a heat issue, but i've read that the CPU and GPU can hit like 80 or 90c and they are both under 70. (GPU is usually under 60).
Other possibilities I thought of were: corrupt data on SSD from broken connector/RAM overheating/chipset overheating.
Well, that should be most of the necessary info, if anyone has any more ideas.. :/
I built the comp. about 6 months ago.
Intel Core i5-3570K Ivy Bridge 3.4GHz
ZALMAN CNPS9500A-LED 92mm Cooler
GELID Solutions GC-Extreme Thermal Compound
Kingston HyperX Black 8GB (2 x 4GB) DDR3 1600
SAMSUNG 840 Pro Series 128GB SATA III MLC Internal Solid State Drive
ASRock Z77 Extreme4
Microsoft Windows 7 Home Premium 64-bit
MSI HAWK Radeon HD 7870 GHz Edition 2GB 256-bit GDDR5 PCI Express 3.0
ASUS VW224U 22 inch 2ms(GTG) Widescreen LCD 1680x1050 16:10
Creative Soundblaster X-Fi Go! Pro USB Audio System
Antec Nine Hundred w/ Antec EarthWatts Green 750W PSU
I originally had the CPU overclocked to 4.5 GHz and was stable with Intel Thermal Analysis Tool (TAT) (though I think it hit 80c once.), except BFBC2 would crash intermittently until I lowered it to 4.2. It ran at 4.2 all this time with no problems. The cooler and thermal paste don't seem adequate. @ 4.2 GHz the CPU would hit about 67c max with BFBC2. CPU voltages were basically stock, not much overvoltage.
The GPU hits about 60c I think, with CCC's worthless auto fan control. (Has latest driver - 13.4.) GPU is not overclocked, except factory. 1100/1200 Tried to overclock the GPU when I first got it and it wouldn't. It didn't post or black screen or something, can't remember really, so I left it alone, since i've never had such a powerful/expensive card before.
I had a Raidmax Smilodon case at first, then, after many months I started to get the artifacts and thought something was overheating. I eventually got the Antec Nine Hundred case with much better fans.
While switching cases I noticed that the plastic piece on my Kingston SSD's power connector had snapped off. I RMA'd it and got the Samsung. The Samsung was cloned from the Kingston and I have since tried to roll back my GPU drivers and got errors, so had to keep the latest driver installed. (I also try to keep the Intel onboard Graphics driver uninstalled but Windows update sometimes sneaks it in.)
I read that my RAM is actually rated @ 1333, even though Kingston sold it as 1600, so not sure if that is the problem. I've just now dropped the RAM speed back to 1333 and have not tested it like that yet. Been running @ 1600 since I built it.
Before that i've verified my BFBC2 Steam files, reapplied thermal paste using the pea method instead of the spread method, increased fan speeds in my case. Bought another fan for the side window. Started using manual GPU fan control to keep GPU temps under 60c. Reduced my CPU to stock speeds.
The artifacts didn't happen back in the winter. And they seem to happen after playing for a couple of hours, never right off the bat, which seems to indicate a heat issue, but i've read that the CPU and GPU can hit like 80 or 90c and they are both under 70. (GPU is usually under 60).
Other possibilities I thought of were: corrupt data on SSD from broken connector/RAM overheating/chipset overheating.
Well, that should be most of the necessary info, if anyone has any more ideas.. :/
)
Facebook
Twitter
Instagram