Geekzone: technology news, blogs, forums
Guest
Welcome Guest.
You haven't logged in yet. If you don't have an account you can register now.




279 posts

Ultimate Geek
+1 received by user: 39


Topic # 162090 28-Jan-2015 21:22
Send private message

Hey,

Anyone have any thoughts I seem to be getting crashing under load and I'm pretty sure it's not from overheating.

System:
Windows 10 Technical Preview
Intel i5-3570K
Corsair Vengeance 16GB DD.R3 (8GBx2)
Seagate Barracuda 2TB
Some ASUS DVD Drive
Asus Strix GTX 970
Corsair GS700
Some brand of 140mm fan, Corsair SP fan, Intel water cooling kit

Firstly I don't think its related to Windows 10 or the new build of 10, I've been using 10 as it's main OS for the past two months including for long gaming sessions and have never had an issue until today. I've been playing Far Cry 4 for the past week, today I attempted to play it and after around 20-30 mins the screens just turned off followed shortly by the computer rebooting. Nothing shown in event log other then unexpected reboot. Since then I've had around 8 more crashes. I tried to play Far Cry 4 twice more and ended up with a reboot both times.

Next I thought I would test something else ran EVGA OC Scanner X and Prime95 to stress test, everything ran fine. CPU at 100% load and GPU at 100% load I had graphics card at 70 degrees Celsius at 45% fan speed, motherboard at around 37 degrees Celsius and CPU at 56 degrees Celsius. After 15 minutes the screen turned off (idle screen off thing) and would not turn back on.

I also had it crash after a reboot not opening any apps except the ones which auto start, ASRock eXtreme Tuner, Asus GPU Tweak and Spotify.

Tried to play Far Cry 4 just now and was met with a lot of screen flashing (as in such as when you change resolution) and then a reboot. Once it rebooted ran torture test again and it was running fine until about 16 minutes at which point my left screen turned completely pink and the right screen turned yellow shortly followed by a reboot once again.

The only significant thing I can think of is I've had 3 losses of power during gaming in the past few days. However I played FarCry for ~2 hours yesterday and it was fine.

Unfortunately I can't go back to Windows 8/7 to test until the semester starts (can't access DreamSpark Premium until then).

View this topic in a long page with up to 500 replies per page Create new topic
 1 | 2
2028 posts

Uber Geek
+1 received by user: 794

Trusted

  Reply # 1224342 28-Jan-2015 21:24
Send private message

Is it possible your power supply is failing, or isn't able to provide enough power under load?

That can cause these sorts of issues.



279 posts

Ultimate Geek
+1 received by user: 39


  Reply # 1224362 28-Jan-2015 21:41
Send private message

muppet: Is it possible your power supply is failing, or isn't able to provide enough power under load?

That can cause these sorts of issues.


I'm thinking it's the PSU failing or my nvidia driver really has issues. It should definitely be able to provide enough power under max load as its a decent brand 700w Bronze and nothing has overclock on either. I'd rather not have to buy a new PSU unless I can confirm it though first, all the other PSUs around the house aren't really powerful enough to power a 970 though.

 
 
 
 


3503 posts

Uber Geek
+1 received by user: 1972

Trusted
Lifetime subscriber

  Reply # 1224390 28-Jan-2015 22:12
Send private message

How old is it thou? PSU's ability to handle load often deteriorates with age.




Information wants to be free. The Net interprets censorship as damage and routes around it.




279 posts

Ultimate Geek
+1 received by user: 39


  Reply # 1224404 28-Jan-2015 22:38
Send private message

Lias: How old is it thou? PSU's ability to handle load often deteriorates with age.


22 Months, wouldn't expect it to deteriorate quite that fast personally. The PSU has a 3 year warranty so if I can confirm it's that then I'll RMA it.

Mr Snotty
8089 posts

Uber Geek
+1 received by user: 4057

Moderator
Trusted
Lifetime subscriber

  Reply # 1224407 28-Jan-2015 22:44
Send private message

I would say your problem is the power supply, replace this with a better one as with this build you could be borderlining 700w under load with those specs.




782 posts

Ultimate Geek
+1 received by user: 178


  Reply # 1224409 28-Jan-2015 22:47
Send private message

I would blame the graphics card.



279 posts

Ultimate Geek
+1 received by user: 39


  Reply # 1225719 30-Jan-2015 23:15
Send private message

Well I installed Windows 8.1, then the Nvidia Driver and then Furmark and it crashed within 5 minutes. Screen turned grey and system was unresponsive. Put a voltmeter on the PSU and it was showing solid 12.X Volts during stress test, also tried running the graphics card powered by a second PSU. Which would mean only drawing 75w from the main PSU but it still crashed so I'm thinking it's not the PSU at fault. The other computer in the house I physically cannot fit my card in to test because this card is so massive lol. 700w if by far enough for a single card especially Maxwell with no overclocking, 500w would be enough realistically with overclocking.

As much as I don't want it to be an issue with the graphics card I guess it is possible. It could also be a motherboard issue I guess.

834 posts

Ultimate Geek
+1 received by user: 260

Trusted

  Reply # 1225720 30-Jan-2015 23:24
One person supports this post
Send private message

Before you buy or replace anything, strip it apart and vacuum it all out.  I've fixed several systems that had foreign material (tobacco) in the card slots or had enough carpet and other fibres in them to stop proper cooling.

14293 posts

Uber Geek
+1 received by user: 2590

Trusted
Subscriber

  Reply # 1225747 31-Jan-2015 08:05
Send private message

Try the built in graphics card. Watch the CPU temperature once you start a test, if it climbs it's a likely culprit - reseat the CPU onto the cooler. Unsure why people are all saying it's the power supply, worth considering but not at the top of my list. 




AWS Certified Solution Architect Professional, Sysop Administrator Associate, and Developer Associate
TOGAF certified enterprise architect
Professional photographer


339 posts

Ultimate Geek
+1 received by user: 25

Trusted

  Reply # 1225767 31-Jan-2015 09:13
Send private message

While people don't usually think of this, but a culprit could be PSU temperature. Modern quality power supplies have overheating protection where they will shut down automatically if they get too hot.

I had a problem with my current build where the PSU exhaust was pointing to the floor, and I had the computer sitting on carpet. Needless to say, the PC would shut down under gaming conditions. It is now sitting on extra feet on a desk, and I've had no problems since.


Also, 700W is plenty. I'm running a AMD 7970 and a Core i5 4690K on a 450W PSU (this was the highest available in SFX form factor when I built this system)...

61 posts

Master Geek
+1 received by user: 7


  Reply # 1225770 31-Jan-2015 09:26
Send private message

Take side of case off, point a fan in there, plug in electrical type I mean, blow air across the psu and gfx card. Take ram out and test 1 stick at a time. Certainly sounds like a thermal issue.

TLD

695 posts

Ultimate Geek
+1 received by user: 153


  Reply # 1225772 31-Jan-2015 09:51
Send private message

I can't see mention of what case is being used?  That's a nice spec, and most of the temperatures are pretty good, but I'd want to use a full tower with a system like that.  I am also wondering about the PSU being a bronze standard.  700W sounds more than enough (I use a 700W Gold with a slightly higher spec, but with every drive bay occupied in a full tower).

I was going to ask if you are running Speccy, but I see it is no longer free.   It sounds like you are controlling the fans.  Are you using MSI Afterburner  CPUZ is another good system monitor and is still free.  Unfortunately they will all need care if you install them, if you want avoid a lot of crap on your system.




Trevor Dennis
Rapaura (near Blenheim)



279 posts

Ultimate Geek
+1 received by user: 39


  Reply # 1225846 31-Jan-2015 13:01
Send private message

Glassboy: Before you buy or replace anything, strip it apart and vacuum it all out.  I've fixed several systems that had foreign material (tobacco) in the card slots or had enough carpet and other fibres in them to stop proper cooling.


I clean out my system every few months with compressed air so it shouldn't be too bad. I checked the PCI-E slot and PSU dust filter the other day and they were pretty clean, I'll have another look later.

timmmay: Try the built in graphics card. Watch the CPU temperature once you start a test, if it climbs it's a likely culprit - reseat the CPU onto the cooler. Unsure why people are all saying it's the power supply, worth considering but not at the top of my list. 


At 100% CPU Load with no graphics card system is stable, temperature is 56 degrees after an hour. I think the CPU fan speeds up from 30% to around 70-80% during the test. I haven't tried a Furmark test on the integrated if that was what you were meaning but I can't imagine it wouldn't get hotter then a 100% loaded CPU.

shermanp: While people don't usually think of this, but a culprit could be PSU temperature. Modern quality power supplies have overheating protection where they will shut down automatically if they get too hot.

I had a problem with my current build where the PSU exhaust was pointing to the floor, and I had the computer sitting on carpet. Needless to say, the PC would shut down under gaming conditions. It is now sitting on extra feet on a desk, and I've had no problems since.


Also, 700W is plenty. I'm running a AMD 7970 and a Core i5 4690K on a 450W PSU (this was the highest available in SFX form factor when I built this system)...


I was considering if the PSU is overheating. The GS700 has a Zero RPM fan mode, supposed to start up when it reaches high temperature/load. I've found it really hard to actually get it to spin up, the fan does it's test spin on startup so it obviously is still working. It could have a faulty temperature sensor I presume, the PC does get pretty warm but I would of thought the PSU fan would actually start spinning before overheating.

Kiwiuk: Take side of case off, point a fan in there, plug in electrical type I mean, blow air across the psu and gfx card. Take ram out and test 1 stick at a time. Certainly sounds like a thermal issue.


If I find a fan I will try

TLD: I can't see mention of what case is being used?  That's a nice spec, and most of the temperatures are pretty good, but I'd want to use a full tower with a system like that.  I am also wondering about the PSU being a bronze standard.  700W sounds more than enough (I use a 700W Gold with a slightly higher spec, but with every drive bay occupied in a full tower).

I was going to ask if you are running Speccy, but I see it is no longer free.   It sounds like you are controlling the fans.  Are you using MSI Afterburner  CPUZ is another good system monitor and is still free.  Unfortunately they will all need care if you install them, if you want avoid a lot of crap on your system.


It's a Raidmax Serian, not the best case ever but it's full ATX and has okay airflow. GS700 is Bronze 80+ but those standards are just about power consumption anyway (power consumption = heat to an extent though).

The CPU fans are controlled with ASRock eXtreme Tuner, by default they are on level 3 but will increase in speed to meet 55 degrees Celsius target temp. Usually when gaming they would very rarely go above level 3, they get noticeably loud very fast because there is a high RPM static pressure fan behind the radiator.
The GPU fan is controlled with Asus GPU Tweak, they are on full auto which seems to be below 60 degrees fan is off, above 70 degrees the fan increases in speed to keep it at 70. It get's to about 50% fan speed under full load to stay at 70, I've forced it to 70% fan speed and it sat at like 57 or so but still crashed.
The PSU fan just does it's own thing.

On the new Windows 8.1 install there is no controlling software but everything behaves the same. I had GPU-ID on the Windows 10 install, Furmark shows the GPU temperature also.



14293 posts

Uber Geek
+1 received by user: 2590

Trusted
Subscriber

  Reply # 1225853 31-Jan-2015 13:11
Send private message

If it works perfectly under load when the graphics card isn't in use but then is unstable when it's in use that's a clue.

You could try installing windows onto a different disk, see if it's the OS playing up.




AWS Certified Solution Architect Professional, Sysop Administrator Associate, and Developer Associate
TOGAF certified enterprise architect
Professional photographer


4373 posts

Uber Geek
+1 received by user: 2683

Lifetime subscriber

  Reply # 1225866 31-Jan-2015 13:25
Send private message

TLD: ... I was going to ask if you are running Speccy, but I see it is no longer free.  ...


Speccy is still free - I downloaded it today (v.1.28.709) - excellent diagnostic software.




Sideface


 1 | 2
View this topic in a long page with up to 500 replies per page Create new topic



Twitter »

Follow us to receive Twitter updates when new discussions are posted in our forums:



Follow us to receive Twitter updates when news items and blogs are posted in our frontpage:



Follow us to receive Twitter updates when tech item prices are listed in our price comparison site:



Geekzone Live »

Try automatic live updates from Geekzone directly in your browser, without refreshing the page, with Geekzone Live now.


Geekzone Live »

Our community of supporters help make Geekzone possible. Click the button below to join them.

Support Geezone on PressPatron



Are you subscribed to our RSS feed? You can download the latest headlines and summaries from our stories directly to your computer or smartphone by using a feed reader.

Alternatively, you can receive a daily email with Geekzone updates.