OllieLaSouris.9521

I get no errors from Windows Event Log before the crash. I get errors after the crash stating that the last shut down was unexpected, with both OCCT and GW2.I've done more testing on my current hardware (will list specs) and GPU testing with OCCT also seems cause crashes, although it takes slightly more time . edit: the gpu test also needs to be set at the highest intensity, or else no crash. Current specs: i7 4790, with a zalman cnps8900 cooler4gb x 2 ddr3 1333 mushkin ramnvidia gtx 760, from evgaSeasonic G series 650 wattGigabyte GA-Z97N wifi mini-itx motherboardcorsair 250D case3 SSD's, 250gb Adata sx900, 500gb Intel, 120gb Samsung evo1 tb Western Digital Caviar Green HDDGW2 is installed on the samsung ssd (it's the only thing on it along with GW1). GW2 also used to be on the intel SSD when crashes first started. I moved it and the crashes still occur. Temps are fine and RAM was tested for about 10 hours with memtest86, no errors. I am in the process of trying out an older video card (gtx 570) and an older PSU (corsair tx650) that I already have, I will come back with results. update: I crash even faster with my old gtx 570! Much faster! 5 seconds into OCCT default GPU test! (still using the seasonic PSU) Never had a problem with this card before. 2nd update: No crashes yet using my gtx 570 and my old PSU. Seems my problem might really be the seasonic PSU. Will put my gtx 760 back in, with my corsair PSU. If the seasonic PSU truly is at fault here, I'm going to have to install a non modular PSU in my mini-ITX case...what a joy! 3rd update: No crashes in OCCT using my old PSU and gtx 760, maxing out cpu, gpu and ram simultaneously. I'll go try GW2 using this setup.

I've been having a similar issue, where my computer shuts down while playing gw2, for about one month. I did the usual (this is not the first time I've had to troubleshoot a pc) tempurature checks and everything is OK (nothing above 60 celsius). I even re-installed my CPU and GPU coolers to make sure. I then proceeded to do some benchmarks (prime95, memtest, furmark) but none have been showing any sign errors. Tonight I've managed to successfully recreate this crash (twice!) using OCCT on "power supply" mode and I suggest that you all give it a try. Something is certainly going wrong with our PSU's! Also if you want to monitor/log tempuratures and voltages, try out speedfan.