- Mar 7, 2017
- 1
- 0
- 510
Hello all. I'm not sure if I'm supposed to add my asking-post to existing topic, or create a new one - and I'm unable to find forum rules, and I didn't see it in general Terms of Use. So, I decided to write new topic, I wouldn't mess topics of other users.
edit: Just found it, don't know how the hell I missed it! Silly me.
I read lots posts about similar issues, but I'm still unable to find solution, or source of problems at least.
The base of issue is as stated in title. When working on high load, the monitor goes black, probably to sleep mode. Usually I need to restart my computer manually, but sometimes display backs on - in that case, the source of high load (usually game) crashes. Also, usually I can see "Display driver stopped responding and has recovered" but I can't be sure if it's every time when game crashes. Sometimes another message shows: "application xxx.exe has denied access to graphics hardware". In AMD Radeon Settings I'm getting notification that WattMan settings return to default due to system instability - but nothing was changed in WattMan and I always use clean install for updates. I was tinkering with TDR, adding QWORD, but it doesn't work for me. Temperatures are OK (idle: CPU <= 30*C, GPU <= 45*C; on high load: CPU <= 55*C, GPU <= 75*C). Issue is - most likely - temperature-indenpendent (sometimes it can resist long-time >70*C, sometimes crashes on 60*C). Wattage seems good enough - just switched from Silentium Vero 500W to XFX XTR 650W and frequency of crashes didn't change. I was able to successfully perform stress-tests using Unigine Heaven and FurMark, it's strange for me. Altought, HWiNFO logs suggest that the problem is PSU and / or GPU - in the moment of crash, GPU wattage falls to 0. Also, other spikes are noticeable. I can attach a logfile if it could help. Unfortunately I'm unable to test other's GPU or PCI slots - just it, R9 380X is big, and mobo is... rather, say, small... Don't know if testing on integrated card and / or with single RAM have any sense if I'd be unable to perform high load tests.
I had similar problems year ago, but in my previous place of residence I had problems with eletricity in general. I'm using decent surge protector.
In general, I'd give PC to service for tests, but services in my country works like "we'll turn computer on, leave it for night".
So, what could be source of problems? Faulty GPU? Damaged motherboard? Or maybe do I need greater PSU?
Motherboard: ASRock B150M Pro4S
CPU: Intel Core i5-6500 Skylake 3.2 ghz
RAM: 2x4 GB DDR4
GPU: Gigabyte Radeon R9 380X 4gb
GPU Driver: Crimson 16.12.2
OS: up-to-date Windows 10
PSU: XFX XTR 650W
Best regards
Ved
edit: OK, just finished checking RAM. Three iterations of memtest, everything seems fine, no errors.
edit2: Intel Burn Test 2 also stable
edit: Just found it, don't know how the hell I missed it! Silly me.
I read lots posts about similar issues, but I'm still unable to find solution, or source of problems at least.
The base of issue is as stated in title. When working on high load, the monitor goes black, probably to sleep mode. Usually I need to restart my computer manually, but sometimes display backs on - in that case, the source of high load (usually game) crashes. Also, usually I can see "Display driver stopped responding and has recovered" but I can't be sure if it's every time when game crashes. Sometimes another message shows: "application xxx.exe has denied access to graphics hardware". In AMD Radeon Settings I'm getting notification that WattMan settings return to default due to system instability - but nothing was changed in WattMan and I always use clean install for updates. I was tinkering with TDR, adding QWORD, but it doesn't work for me. Temperatures are OK (idle: CPU <= 30*C, GPU <= 45*C; on high load: CPU <= 55*C, GPU <= 75*C). Issue is - most likely - temperature-indenpendent (sometimes it can resist long-time >70*C, sometimes crashes on 60*C). Wattage seems good enough - just switched from Silentium Vero 500W to XFX XTR 650W and frequency of crashes didn't change. I was able to successfully perform stress-tests using Unigine Heaven and FurMark, it's strange for me. Altought, HWiNFO logs suggest that the problem is PSU and / or GPU - in the moment of crash, GPU wattage falls to 0. Also, other spikes are noticeable. I can attach a logfile if it could help. Unfortunately I'm unable to test other's GPU or PCI slots - just it, R9 380X is big, and mobo is... rather, say, small... Don't know if testing on integrated card and / or with single RAM have any sense if I'd be unable to perform high load tests.
I had similar problems year ago, but in my previous place of residence I had problems with eletricity in general. I'm using decent surge protector.
In general, I'd give PC to service for tests, but services in my country works like "we'll turn computer on, leave it for night".
So, what could be source of problems? Faulty GPU? Damaged motherboard? Or maybe do I need greater PSU?
Motherboard: ASRock B150M Pro4S
CPU: Intel Core i5-6500 Skylake 3.2 ghz
RAM: 2x4 GB DDR4
GPU: Gigabyte Radeon R9 380X 4gb
GPU Driver: Crimson 16.12.2
OS: up-to-date Windows 10
PSU: XFX XTR 650W
Best regards
Ved
edit: OK, just finished checking RAM. Three iterations of memtest, everything seems fine, no errors.
edit2: Intel Burn Test 2 also stable
Facebook
Twitter
Instagram