I had the problem since I upgraded my graphic card from Msi Gtx 980 ti 6g to Msi gtx 1080 ti 11G. The problem is my system is completely shut off and then turned on again when entering graphic intensive area in games, I play on 4k resolution @60hz monitor with v-sync on, game settings are ranging from high to very high, it crashed certainly and specifically in: 1. Ghost Recon Wildlands, in the mission where you had to extract El Yayo from his house, and the crash is occurred when I was entering his coca field 2. In witcher 3, especially in the deep forest and I used piercing cold Aard, my system is likely to shut off. When I was using my 980 ti, It never shut off, it probably hung or locked up the pc so I had to hard reset the system. 3. In the Mass Effect Andromeda Title. Yes in the title, even I'm not able to start the game. Meanwhile I can started the game with my gtx 980 ti before even on 30-40fps The thing I had done to troubleshoot this issue were: 1. I already stressed the psu, gpu and cpu at the same time using OCCT Psu stress and the system was rocking fine for 3 hours without having any issues (graphic card using 117-120%TDP, and cpu dram etc using ~95watt TDP). 2. I already RMA my first msi gtx 1080 Ti with the new one, and the problem still persists. 3. Monitoring all the temps when the crash is likely to occur, my cpu and gpu never exceed 65 degrees Celcius 4. The crashed happened even I'm downclocking the gpu to its base clock, but take more time to crash and decrease the likelihood to crash. 5. The crashed happened on various gpu power draw, in gr wildlands it's never exceeded 105%, in mass effect andromeda it probably exceeded the power limit >117%, and the witcher 3 is uncertain, the power draw are ranging while using Aard. 6. Countless time for using DDU for clean install driver and no avail 7. Using different pcie power connector slot on psu. 8. Make sure pcie power connector is connected and secured. Tomorrow I gotta RMA this psu with the new one. If the problem still persist I don't know what to do anymore. Do you guys have any idea about what's wrong with my system? My system specs are: Gigabyte Z97X Gaming 5 Motherboard i7 4790k @4.4Ghz Corsair vengeance 1600mhz 2x8 GB RAM MSI Gtx 1080 Ti Gaming 11G Corsair H110i Corsair HX850i Platinum power supply Samsung 850 Evo SSD Seagate Barracuda 2Tb 7200rpm. Windows 10 Pro 64bit
I upgraded my motherboard bios from gigabyte to F5 since 2 years ago and never upgraded since then. Should I upgraded to F7? The description just said it is for Broadwell compatibility
It makes sense, I will RMA the psu tomorrow. And I will try to play Mass effect andromeda with my gtx 1080 Ti on my friend's rig and see what happens.
If it hard freezed by 980ti it could be some sort of voltage issue as well, my guess mostly on cpu power phase or vccsa - cpu system agent voltage too low.. Now with 1080ti it stresses that part further and just shuts down and or reboots. Yes update to latest bios and readjust those two values. Cpu power phase raise for +2steps (should be in mv, max is 4steps) Vccsa 0.050v offset +
Yes, but it very very rare occasion when I was using my 980 Ti even playing in 4k @40-55 fps resolution with 99% gpu usage and 110% TDP. Thanks anyway, I'll try to flash newer bios for my motherboard and see what will happens.
I have upgraded Bios to the latest version and RMA my psu with the new one with new cable too! Aaaaandd no luck, it still shut down and turned on again whenever I enter Mass Effect Andromeda Title I guess, my motherboard is aging and it's time to upgrade the whole system :bang::bang::bang:. Anyway, I noticed something weird, every time it shut down and turned back on, my profile in msi afterburner always reset to the default. Maybe this issue is driver related?
Hi, first post here. I used to assemble pc's and sometimes fix them, I haven't seen anyone mention the memory, you could take them out and pop them back in, or test them one at the time. Memory was almost always one of the first thing we would be looking at with black screen error and I have seen a lot of errors with vengeance memory. If it's possible you could also try slot 3 and 4 on the motherboard for your memory
Nah dude, thanks for reply anyway. My motherboard has some weird issue if I'm messing up with my ram, reseat it and Bam, the pc won't even post, so I must remove the processor as well and leave it for about 20 minutes to discharge all juice in the capacitors and put all back in to reboot. Yep, it is always like this but it works. Somehow, I just recently managed to fix my shut down issue by adding more 1 pcie cable connector to power other 8 pin on my MSI gtx 1080 ti. Before that, I used 1 Pcie cable to power 2x8 Pin on the gpu. So when the current spikes, the over current protection in psu kick in and my pc shut off. That is just my theory by the way. I thought all pcie connector on psu are powered by the same rail. I'am not sure. But it works! Thanks for all of your reply guys!
