Sign in with
Sign up | Sign in
Your question
Closed

Help! Facing system freeze during games

Last response: in Graphics & Displays
Share
July 21, 2011 12:26:45 PM

Hi,

This is my new build (some months old) and during that time I changed my multiple components due to BSOD. Recently I changed my motherboard which had issues and now system is stable from usage point-of-view as its not giving any BSOD.

Now since my bad motherboard (which I replaced with new one) I am facing system freeze issue during game-play after sometime. Previously I thought its due to my crappy motherboard but now yesterday after using my new board for some days when first time I started playing Crysis with everything maximum at 1080p around after 40 mins same freeze happened which I was facing earlier.

I can hear game sound but the screen-stuck and there is no artifact on screen (image is proper). If I press ctrl+alt+del and system became responsive although game was killed/crashed.

Yesterday I ran Furmark's Burn-In test for more than 1.5 hours and manually increased the FAN speed to 50% as without it the CPU temperature was touching 100C+ and Furmark was hanging in 7 minutes. The maximum temperature it was touching with 50% FAN speed was 71C.

Then again I ran Crysis after this Furmark and FAN speed adjustment and with-in an hour Crysis hanged and pressing ctrl+alt+del recovered my system and game crashed.

Can someone help me to identify issue or explaining the reason for this ? I have latest ATI driver and no GPU overclocking.

Thanks in advance
July 21, 2011 12:27:57 PM

Forgot to give details about my system. Here it is:

Specs:

-- Motherboard - GA-990FXA-UD3

CPU - Phenom II X6 1090T CPU 3.7GHz, CPU-NB 2.6GHz, HT 2GHz and CPU VCORE 1.38v; Cooler is: Coolermaster V6GT CPU cooler
RAM - 4 x 2GB G.Skill 1600MHz DDR3 7-8-7-24 @1.6v

-- GPU - Asus Radeon HD 6970 2GB GDDR5

Storage - 500GB WD 7200rpm (16MB cache) and 1TB WD 5400rpm (64MB cache)
Case - Coolermaster CM-690 II Advanced
PSU - Thermaltake XT-775W
Monitor - AOC 2436Vwa (1080p, 60Hz)
Score
0
July 22, 2011 9:13:20 AM

Updated my first post with some more details. Can anyone help me for this ?
Score
0
Related resources
a b V Motherboard
July 22, 2011 12:41:46 PM

overheat can also lead to freezing. make sure your case have good air flow because without it the heat will be circulate inside your case and the cpu/gpu cooling mechanism will fail to properly cool down your cpu /gpu. report back after checking your air flow
Score
0
July 22, 2011 12:56:40 PM

My air flow is good (I think); as the CPU temperature on Prime95 was touching 53C maximum after 4-hours ......... and GPU temperature after manual FAN settings to 50% is maximum 71C in Furmark Burn-In test.

If I don't set FAN speed manually then GPU temperature touches 105C because FAN is running at 30% _then_ changing it to 50% reduced the GPU temp to 71C.

Is these temperatures are fine ?

For air-flow you can look at this picture of my system which I posted some days ago in Gallery. I planned to manage cables this weekend but still I think the flow is fine; what is your suggestion ?




Score
0
a b V Motherboard
July 22, 2011 1:07:26 PM

fslateef said:
My air flow is good (I think); as the CPU temperature on Prime95 was touching 53C maximum after 4-hours ......... and GPU temperature after manual FAN settings to 50% is maximum 71C in Furmark Burn-In test.

If I don't set FAN speed manually then GPU temperature touches 105C because FAN is running at 30% _then_ changing it to 50% reduced the GPU temp to 71C.

Is these temperatures are fine ?

For air-flow you can look at this picture of my system which I posted some days ago in Gallery. I planned to manage cables this weekend but still I think the flow is fine; what is your suggestion ?



http://i1215.photobucket.com/albums/cc507/fslateef/SAM_0851.jpg


You've got do something with your cables... the video card breathes through the rear and after looking at the pic it looks as though it's not getting any air. Even if you turned your fan fan speed up to 100% it's prob. only getting 50% if you're lucky. It is imperative you fix the cables. 2nd, move the hdd's to the bottom of the cage and take out brackets you're not using and get get yourself another fan that'll fit in the hdd cage where it's blowing on the card(s).

Unplug everything and route the cables to the back, hook everything back up one by one, starting with the headers for the front of the case, all the boards power cables etc and so on...
Score
0
July 22, 2011 1:40:38 PM

Thanks for nice suggestions, I will follow them asap.

I still have some confusions:

1) Moved HDDs to bottom of cage or up the cage ? I think I can take-out bottom-four HDD holder including the cage for those 4 HDDs holders (as the cage is divided into two already). Then isn't that front FAN which is taking AIR-INSIDE enough for GPU ? Or should I add another FAN on the bottom of casing (at the place of HDD cage after removal); if yes then the FAN should blow AIR INSIDE the case or outside ?

2) What is the normal full load temperature of high-end GPU like mine (6970) ? Is ~70C is too much ?

3) If Furmark is fine then can I assume that my GPU is ok from Hardware/Driver point-of-view ?

Waiting for response.

Thanks
Score
0
a b V Motherboard
July 22, 2011 2:44:55 PM

1) for me i used the front intake to keep my HDD cool. if i were you i will move the all the HDD up and remove the lower cage. it is also a good idea to add another fan at the bottom of your case. normally people will use the bottom fan as intake fan. i believe most if not all case use fan that located at the bottom/lower section of the case as an intake fan. hot air tends to go up that's why outtake fans usually located at the top rear case.

2) for high end gpu 80c-90c is normal when under full load. so 70c still considered as cool in my book. but when the temps starts to reach 90c+ you should start worried about it

3) if you can run the program without crashing or without producing any artifact that's mean your gpu should be stable with the clock being used

btw i think your problem might cause by automatic fan control which only raise the fan speed when the temperature already too high. try using MSI Afterburner and create fan profile. other than that do some cable management for your case.
Score
0
July 22, 2011 2:55:19 PM

Thanks for detailed reply. Previously I was thinking that 110C is normal for high-end GPU which was wrong :( 

Now I will add another fan, remove bottom cage and add some application to auto control fan properly, do cable management _and_ will post back my results.

thanks
Score
0
a b V Motherboard
July 22, 2011 3:03:52 PM

AFAIK most high end gpu able to reach 105c max before the system start crashing/shutdown or in worst case the gpu totally broken.
Score
0
July 23, 2011 1:15:07 PM

I haven't tested my system hang/crash issue but I managed to spend some time on fixing air-flow in my case.

Here are the pictures for feedback. I will post back about my issue as soon as I got some time to test it.


Score
0
a b V Motherboard
July 23, 2011 4:51:41 PM

looks much better than before :) 
Score
0
July 23, 2011 8:42:41 PM

I ran Furmark now and system crashed/BSOD with SERVICE_EXCEPTION stop code 3B :( 

Just before crash temperature was 89C, Fan was at 45% speed. I will try to install some proper FAN management application; but seems like some driver/hardware issue :( 

Score
0
a b V Motherboard
July 24, 2011 2:38:28 AM

have you tried using MSI afterburner? at such temperature some people will let their fan running at 100%. MSI Afterburner have fan profile and you can set your fan speed depending on your temperature reading
Score
0
July 24, 2011 8:02:36 AM

I installed Asus Smart Doctor which also has profile support, and after doing settings in it for fan speed my temperature in furmark was in 60s. After this I started facing weird problem: in Crysis I am getting only 5 FPS but earlier I was getting 35+. and in Crysis 2 just at the end of game loading application is crashing :(  even restart and exit from smart doctor is not fixing it. Although in furmark I was getting usual 20+ FPS.

I reinstalled Windows 7 64bit pro as after changing my motherboard I was using old installation, and at the moment running memtest86 to verify my RAM again.

Will update you soon. Thank you very much for bearing with me till now.
Score
0

Best solution

a b V Motherboard
July 24, 2011 10:08:35 AM

seems the temperature are fine right now but another problem emerge. i'm not sure if your performance decrease caused by Asus Smart Doctor but you can try MSI Afterburner. AFAIK MSI Afterburner is one of the commonly used 3rd party tool for gpu. there are other which also great such as EVGA Precision but i believe the software only work with nvidia cards unlike MSI Afterburner. one thing did i like with MSI AB is you can set your fan curve so the fan will adjust it's speed accordingly to the gpu core temperature.
Share
July 24, 2011 2:36:20 PM

FAN curve in MSI AB application will be really nice; thanks for letting me know. I will install it.

At the moment I am running memtest86 and during previous run of memtest I got one memory error after 13 hours in pass 6 .... then I stopped the test, increased RAM voltage to 1.61V from 1.6V which is default according to spec of RAM.

I have 4 RAM modules of 2GB each and was facing issues with my system even with old motherboard which I already returned. So seems like my RAM is behaving wrongly thus crashing application or system :( 

I am going to create another thread for getting suggestions for my RAM issue. As its seems like my RAM not graphics card.

Really thankful for your help.
Score
0
July 24, 2011 2:36:47 PM

Best answer selected by fslateef.
Score
0
a b V Motherboard
July 24, 2011 2:47:54 PM

This topic has been closed by Mousemonkey
Score
0
!