PC crashes when graphics card is under stress help pls.

[Renokami]

My PC keeps crashing whenever my graphics card is under stress. It was fully working for past 8 months but recently (2 days) it kept crashing.. resolving in grey screen with stripes or black screen, no DVI- singal screen and all sorts of crazy screen coloros and shapes. After which I have to do hard reset. I reinstalled driver but it still keeps crashing. Help pls

My GPU is MSI R6870 Hawk.
Microsoft Windows [Verze 6.1.7601]
Copyright (c) 2009 Microsoft Corporation. Všechna práva vyhrazena.

C:\Users\Ren>systeminfo

Název hostitele: REN-PC
Název operačního systému: Microsoft Windows 7 Ultimate
Verze operačního systému: 6.1.7601 Service Pack 1 Sestavení 7601
Výrobce operačního systému: Microsoft Corporation
Konfigurace operačního systému: Samostatná pracovní stanice
Typ sestavení operačního systému: Multiprocessor Free
Registrovaný vlastník: Ren
Registrovaná společnost:
ID produktu: ---
Datum původní instalace: 18.9.2012, 2:55:54
Čas spuštění systému: 23.6.2013, 14:52:00
Výrobce systému: Gigabyte Technology Co., Ltd.
Model systému: GA-990XA-UD3
Typ systému: x64-based PC
Procesory: Počet nainstalovaných procesorů: 1
[01]: AMD64 Family 16 Model 4 Stepping 3 Au
thenticAMD ~2176 MHz
Verze systému BIOS: Award Software International, Inc. F9, 13.1
0.2011
Adresář systému Windows: C:\Windows
Systémový adresář: C:\Windows\system32
Spouštěcí zařízení: \Device\HarddiskVolume1
Místní nastavení systému: cs;Čeština
Národní prostředí: cs;Čeština
Časové pásmo: (UTC+01:00) Praha, Bratislava, Budapešť, Bě
lehrad, Lublaň
Celková kapacita fyzické paměti: 8 189 MB
Volná fyzická paměť: 5 694 MB
Virtuální paměť: maximální velikost: 16 377 MB
Virtuální paměť: k dispozici: 13 401 MB
Virtuální paměť: používána: 2 976 MB
Umístění souborů virtuální paměti: C:\pagefile.sys
Doména: WORKGROUP
Přihlašovací server: \\REN-PC
Opravy Hotfix: Počet nainstalovaných oprav Hotfix: 112
[01]: KB2764913
[02]: KB2764913
[03]: KB2764916
[04]: KB2764916
[05]: KB2718695
[06]: KB2718695
[07]: KB2670838
[08]: KB971033
[09]: KB2305420
[10]: KB2393802
[11]: KB2425227
[12]: KB2479943
[13]: KB2491683
[14]: KB2506014
[15]: KB2506212
[16]: KB2507618
[17]: KB2509553
[18]: KB2511455
[19]: KB2532531
[20]: KB2533552
[21]: KB2536275
[22]: KB2536276
[23]: KB2544893
[24]: KB2552343
[25]: KB2560656
[26]: KB2564958
[27]: KB2567680
[28]: KB2570947
[29]: KB2579686
[30]: KB2584146
[31]: KB2585542
[32]: KB2604115
[33]: KB2618451
[34]: KB2619339
[35]: KB2620704
[36]: KB2620712
[37]: KB2621440
[38]: KB2631813
[39]: KB2633952
[40]: KB2644615
[41]: KB2645640
[42]: KB2653956
[43]: KB2654428
[44]: KB2655992
[45]: KB2656356
[46]: KB2656411
[47]: KB2658846
[48]: KB2659262
[49]: KB2660649
[50]: KB2661254
[51]: KB2667402
[52]: KB2676562
[53]: KB2677070
[54]: KB2685939
[55]: KB2688338
[56]: KB2690533
[57]: KB2691442
[58]: KB2698365
[59]: KB2705219
[60]: KB2706045
[61]: KB2709715
[62]: KB2712808
[63]: KB2718704
[64]: KB2719985
[65]: KB2722913
[66]: KB2724197
[67]: KB2727528
[68]: KB2729094
[69]: KB2729452
[70]: KB2731847
[71]: KB2736233
[72]: KB2736422
[73]: KB2742599
[74]: KB2743555
[75]: KB2749655
[76]: KB2753842
[77]: KB2756822
[78]: KB2756921
[79]: KB2757638
[80]: KB2758857
[81]: KB2761226
[82]: KB2769369
[83]: KB2770660
[84]: KB2778344
[85]: KB2778930
[86]: KB2779030
[87]: KB2779562
[88]: KB2785220
[89]: KB2786081
[90]: KB2789645
[91]: KB2790113
[92]: KB2790655
[93]: KB2798162
[94]: KB2799494
[95]: KB2804579
[96]: KB2807986
[97]: KB2808735
[98]: KB2813170
[99]: KB2813347
[100]: KB2813430
[101]: KB2820197
[102]: KB2829361
[103]: KB2830290
[104]: KB2838727
[105]: KB2839894
[106]: KB2840149
[107]: KB2845690
[108]: KB2859903
[109]: KB958488
[110]: KB976002
[111]: KB976902
[112]: KB976932
Síťové karty: Počet nainstalovaných adaptérů NIC: 2
[01]: Realtek PCIe GBE Family Controller
Název připojení: Připojení k místní s
íti
DHCP povoleno: Ano
Server DHCP: 10.0.0.138
Adresy IP
[01]: 10.0.0.3
[02]: fe80::1c31:f541:e230:cc6d
[02]: TAP-Win32 Adapter V9 (Tunngle)
Název připojení: Tunngle
Stav: Médium je odpojené.

C:\Users\Ren>

 

[DjDafiDak]

Check your GPU temps under stress with Hwmonitor.

 

[Renokami]

Sadly when I turn Hwmonitor on, my pc instantly crashes. Could it be due to overheating? (I've cleansed my grpahics card yesterday, right after first 5 cca crashes) ATM Catalyst shows me that when my GPU is kinda Idle 0-40% of usage its cca around 51°C

 

[DjDafiDak]

51 c is definitely pretty high when idle, and it might be overheating.

but opening hw monitor causing ur pc to crash is kinda strange, so first when we need to check if its ur gpu or something else.

what you need to do is, remove your gpu, connect your montior to onboard grapichs( directly to ur motherboard) and then launch hwmonitor and see if ur pc crashes, if it does, it has nothing to do with ur gpu.

but try again and with other programs to tell me ur temps under stress, even after opening a game for a minute, hopefully u can catch it before it crashes.

 

[Renokami]

Check your GPU temps under stress with Hwmonitor.

Sadly when I turn Hwmonitor on, my pc instantly crashes. Could it be due to overheating? (I've cleansed my grpahics card yesterday, right after first 5 cca crashes) ATM Catalyst shows me that when my GPU is kinda Idle 0-40% of usage its cca around 51°C

 

[slamer80]

Download MSI Afterburner, you can see your gpu temerature there too.

 

[Renokami]

Sadly (again) I dont have any VGA output on my motherboard so I am completly dependand on my GPU http://www.czc.cz/gigabyte-ga-990xa-ud3-amd-990x/92546/produkt as it seems my PC keeps crashing only because of my GPU, since its only logical ending I can come up with... I've left my PC turned off for 30-50 mins. and right after boot up Ive had to let my fan go on 100% (in Catalyst) so it doesnt crash right away. (and with 100% on my fan I am still around 50°C)

 

[Renokami]

https://fbcdn-sphotos-g-a.akamaihd.net/hphotos-ak-prn1/1015773_561374473900725_889271795_o.jpg

 

[Renokami]

https://fbcdn-sphotos-a-a.akamaihd.net/hphotos-ak-ash3/1052644_561375720567267_2101736600_o.jpg
another one after leting my fan do whatever it wants.. atm just waiting for an inevitable crash

 

[Renokami]

I cant test it in stress situation since even when I get slightly closely to one it just crashes.. my screen goes black (or grey with stripes about 4x times, once it has gone red, few times it even stopped detecting DVI, and so on, with no error msg on it) after that audio goes on for few seconds and then it freezes also.. ending in complete pc freeze, where only option is hard restart

 

[Renokami]

even a youtube video crashes it -.-

 

[slamer80]

try plug in another card, maybe borrow one from a friend. or try your graphic card in a friends pc

 

[DjDafiDak]

Wait is it overclocked? if so thats obviously the reason.

 

[Renokami]

It wasnt, its AMD driver overclockin thingy so it should be safe.. but I've left it on default 99,9% of time and last time I had it overclocked was like 4 months ago.. ATM I have plugged in friends graphics card and everything seems to be working np, so the problem probably is only in my GPU not in the whole PC or software.

 

[Fishwithadeagle]

It doesn't matter if it is overclocked or not. I had a stock 6970 and that made my comp crash and I have a 460 watt good quality psu. There might be an issue with your current psu if it is supposed to be able to handle the card. Otherwise, go into afterburner and set the power level to -20, that should usually help you and determine if it is the card or the psu that is the issue. Once again, check to make sure that there is not any background programs that are running and causing the gpu to be used. There are some programs out there that use your computer as a bitcoin computer, which causes it to heat up. Lastly, make sure you clean out your case. I usually do this once a week, but once a month is fine.

My final answer would be that your psu is going bad especially if it is an offbrand, or a store brand. This was the case when I had the 6970, so I upgraded to the 7870, and have no problems with the power limit.

JUST SAW THIS: Your gpu should have voltage control unlocked since it is a hawk card. I originally had this problem where I could not control the voltage. Go into your settings and check allow voltage control. If this doesn't work. Uninstall the drivers and any programs that could influence your graphics card, shut down, restart, shutdown, and then let windows do the basic install. Then reinstall the catalyst drivers and then msi afterburner. You will see that you card will perform much better when in afterburner it detects it as a hawk card.

Sorry for the long message.

 

[Renokami]

It doesn't matter if it is overclocked or not. I had a stock 6970 and that made my comp crash and I have a 460 watt good quality psu. There might be an issue with your current psu if it is supposed to be able to handle the card. Otherwise, go into afterburner and set the power level to -20, that should usually help you and determine if it is the card or the psu that is the issue. Once again, check to make sure that there is not any background programs that are running and causing the gpu to be used. There are some programs out there that use your computer as a bitcoin computer, which causes it to heat up. Lastly, make sure you clean out your case. I usually do this once a week, but once a month is fine.

My final answer would be that your psu is going bad especially if it is an offbrand, or a store brand. This was the case when I had the 6970, so I upgraded to the 7870, and have no problems with the power limit.

JUST SAW THIS: Your gpu should have voltage control unlocked since it is a hawk card. I originally had this problem where I could not control the voltage. Go into your settings and check allow voltage control. If this doesn't work. Uninstall the drivers and any programs that could influence your graphics card, shut down, restart, shutdown, and then let windows do the basic install. Then reinstall the catalyst drivers and then msi afterburner. You will see that you card will perform much better when in afterburner it detects it as a hawk card.

Sorry for the long message.

My psu is S12II-520Bronze Seasonic. Anyway it was working completly fine for almost 9 months and since yesterday I have lost the count of crashes that occured. Therefore I will try what you have written and lets hope for the best.

 

[Fishwithadeagle]

Sometimes fate plays a bad game and your psu had just a little too much dust in it and it will overheat and cut the max output. I really wish you the best of luck. Some of the new corsair psu are nice because they can run with passive cooling, allowing it to be quiter and have less wear and tear.

 

[Renokami]

Ive unlocked voltage control and decreased it -20 (from 1185 to 1165) and it crashed about 10 seconds after Ive done so. COuld cleaning the psu solve everything? (Friends low end GPU was running fine so even the voltage drop shouldnt affect it, so my GPU is running almost max voltage of the psu and when the psu decides to drop the voltage due to dust=overheating my GPU is basically out of power to go on and crashes the whole PC, thats how I understood what you are trying to tell me)

 

[Fishwithadeagle]

That is odd, cleaning it shouldn't do much at this point. You could use a voltmeter to measure the current coming from the 12v rails, but other than that, I am at a loss. But basically what i said was that psu's drop their output when they reach a certain thermal point, sometimes this can be caused by a faulty psu or by dust.

 

[Renokami]

Yeah I kinda got that, anyway If everything is gonna go as I expect it to be, I wont have any problem with buying anything once I come back from my workplace (faaaaaaaaaaar away from my PC, which will be standing here still for 2 months straight and since I am leaving in Monday I am not really in mood to "do" anyhting. Thanks for everything, I will ask for more guidance in future (September).... as in what to do and what would be best. Thanks for all your support. Cya in 2 months