Have a pair of BFG GTX260 MaxCore OCX edition GPU's, CARD A is the 65nm version (Secondary Card) and CARD B is the 55nm refresh (Primary Card). CARD A is the one that seems faulty and was purchased November 2008, fortunatley I registerd this card with BFG so hopefully the 10 year warranty still applies even though they don't sell GPUs in Europe anymore!! Problems I get are a brief checkerboard pattern flashing up in the bottom half of some games, some games minimising to desktop while playing and when playing Metro 2033 a while back after 30 to 60mins a hard lock would occur!
Sytem Spec:
HAF932 Case with side panel off while messing about!
Asus P6T Deluxe V2 with BIOS 1003
Core i7 920 D0 SLBEJ @ 3.6GHz + Thermalright Ultra 120 eXtreme 1366RT with Push/Pull fans
2 x BFG GTX260 MaxCore OCX 216 Cores
1TB WD1001FALS HDD
6GB Kit Corsair Dominator RAM TR3X6G1600C8D
3 x Optical Drives
Creative X-Fi XtremeGamer Soundcard
PC Power & Cooling 750w PSU (60A on a single rail)
24" Samsung SyncMaster 2493HM
Mouse + Keyboard (or else how would I type and post this!!)
If I run FurMark 1.8.2 on CARD B for just 15 minutes all is fine (I use Stability test, PostFX on, 1920 x 1200 res for all runs), I can also run a full 1hr of OCCT 3.1.0 on this card in the GPU test @ constant 85°C. However, if I run the FurMark Multi GPU test the system locks with a blank screen when CARD A hits around 74°C? Pulled out both cards and just put CARD A in the first PCI-E Slot (Blue slot) on the board and ran FurMark again and I get the same lockup when the temp is in the mid to high 70's!! The puzzling thing is I can run OCCT 3.1.0 on this card for a full 1hr though as per CARD B straight after the lockup and a reboot with FurMark?
Ran EVGA Precision and manually put fan on full 100% until temp went down to a level 40°C, set EVGA Precision fan control back to Auto and exited then re-ran FurMark and the lockup occured at around 1min 40sec @ 78°C, repeated the above of EVGA etc and re-running FurMark and the lockup occured around 1min 26sec @ 73°C!!
So, does this mean I have a faulty GTX260? Is there any other test to be 100% sure this card really is faulty?
