I have got this problem over a year now and I tried everything I could think of.
I got two Sapphire Tri-X R9 290X OC in AMD Crossfire mode.
Fixes I tried:
- New powersupply (from 700W to 1000W)
- Updated BIOS
- Updated all drivers
- EventVwr (nothing weird, only critical error as soon as I switch of the power when I get a blackscreen)
- Use stabile voltage for VRAM
- Use stabile frequency for GPU
- installed A LOT of new fans to the PC (overheating won't be the problem)
- and more...
The PC runs perfectly when not playing games or running GPU intense operations (it crashes in Blender 2.8 too for example- only when I am in realtime rendering mode!). There is also never a bluescreen. Most of the times I can even play the games for a few minutes until the problems occur. I don't think that overheating is a problem tho, since my GPU's maximum temperature before crashing is about 70°C.
Current GPU driver: 18.104.22.168-200421a-354309E-RadeonSoftwareAdrenalin2020
(downloaded and updated with Radeon Software)
Motherboard: MSI Z97 Gaming 5 (MSI MS-7917)
CPU: Intel Core i7-4790K @ 4GHz
GPU: 2x AMD Radeon R9 200 Series (Sapphire R9 290X Tri-X OC)
RAM: 16GB DDR3 (idk the frequency)
Power Supply: be quiet! Straight Power 11 1000W (brand new)
OS: Win7 Home Premium 64-bit
System runs on a 2TB SSD
I would really appreciate help since I have this problem for a very long time now. If there is any information missing you need to know, just let me know!
Thank you very much in advance.
Take one card out, test and switch cards and test again. See if it is the XFire configuration and/or one card causing the issue.
Since CrossFire, if a tower case the upper card will run hotter due to hot air rising from the bottom one when both cards are working. Anyways those cards are up there in age and the thermal compound is probably hard/degraded. Recommend you remove both cards, take them apart, clean and repaste. It may not be the GPU that is over heating it could be the ram as well.