Overclock.net › Forums › AMD › AMD - General › Horrible stuttering and weird problems (borderlands specifically)
New Posts  All Forums:Forum Nav:

Horrible stuttering and weird problems (borderlands specifically)

post #1 of 11
Thread Starter 
I am running borderlands at 3240x1980 with max settings... I play with 30+ frames for about 5 - 10 seconds and the came freezes, after 20 seconds it minimizes to the desktop, and then after another 20 seconds it brings my game back up. I get to play for 5-10 more seconds and same issue.

when it minimizes I have CCC open so I can see my GPU temps, and I have task manager open so I can see my cpu usage. my GPU is at 75c and my processor is under about 60% load, if I was really stressing my computer I could understand these kind of errors, if I lower the settings I can play with no issue, but my computer should be dominating this game or am I wrong.

I understand my CPU is bottlenecking my GPU, I will be upgrading when bulldozer comes out.


This same issue happens in battlefield bad company 2 as well, but not as often its every 10-20 minutes....

I haven't been able to play games I was told it was my powersupply and I have replaced that now and still can't play any games...
Edited by atticus589 - 5/31/11 at 12:02pm
post #2 of 11
Might be obvious but have you checked all drivers?
Small but Mighty
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom X4 @ 2.2ghz ASUS NARRA3 M2N68-LA 9800GT XFX 512MB OC 3gb DDR3 
Hard DriveOSMonitorPower
500gb & 1.5tb Samsung Vista x32 22" Samsung 3000:1 Jeantech Storm 700w 
  hide details  
Reply
Small but Mighty
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom X4 @ 2.2ghz ASUS NARRA3 M2N68-LA 9800GT XFX 512MB OC 3gb DDR3 
Hard DriveOSMonitorPower
500gb & 1.5tb Samsung Vista x32 22" Samsung 3000:1 Jeantech Storm 700w 
  hide details  
Reply
post #3 of 11
have you checked your memory? maybe 4 gigs not enough?
post #4 of 11
Can you tell with MSI Afterburner if your card is going back to 2D clocks?.
Good Enough Rig
(14 items)
 
  
CPUMotherboardGraphicsRAM
FX 8350 GA-990FXA-UD3 RX 480 Nitro+ (Passthrough) + XFX 5770 (Secondary) Kingston HyperX (4x4GB) 
Hard DriveCoolingOSMonitor
3x Samsung Spinpoint F3 Hyper 212 EVO Fedora 25, W8.1 virtualized with KVM + GPU Pass... ASUS VS228H-P 
PowerCaseAudio
Corsair TX650 CM Elite 330 Onboard ALC889 + HDMI Passthrough 
  hide details  
Reply
Good Enough Rig
(14 items)
 
  
CPUMotherboardGraphicsRAM
FX 8350 GA-990FXA-UD3 RX 480 Nitro+ (Passthrough) + XFX 5770 (Secondary) Kingston HyperX (4x4GB) 
Hard DriveCoolingOSMonitor
3x Samsung Spinpoint F3 Hyper 212 EVO Fedora 25, W8.1 virtualized with KVM + GPU Pass... ASUS VS228H-P 
PowerCaseAudio
Corsair TX650 CM Elite 330 Onboard ALC889 + HDMI Passthrough 
  hide details  
Reply
post #5 of 11
You must have something dodgy running in the background. Either that or it's your hard drive. Is that the 2TB 5400rpm Samsung F3/F4 you're running? If so, you made a big mistake by installing your OS on a low-power drive. Your Phenom shouldn't be "bottlenecking" any of those games, and I strongly recommend you run your OS and games off a 7200rpm performance drive at least. Can't believe no-one else noticed this. I have no idea why you wasted money on a PSU that's overkill by THAT much (a decent 500W would run that without breaking a sweat).

For your information, low power drives stop spinning/turn themselves off when not used for a few minutes. What you described sounds like the consequence of that feature kicking in, as most games only read data from the hard drive in bursts.

EDIT: Considering the massive resolution you're running, your CPU will never bottleneck a game as low-demanding as Borderlands. Running at high resolutions like that and getting a good framerate is entirely down to the GPU, going to a Bulldozer or even Ivy Bridge CPU with that GPU won't offer the noticeable increase in framerates that you're expecting.
Edited by Am* - 5/31/11 at 1:14pm
    
CPUMotherboardGraphicsRAM
Intel Celeron G530 Haven't decided yet. GTX 460 768MB 4GB Ripjaws 
Hard DriveOSMonitorPower
Samsung F3 1TB (SSD coming later) XP/Windows 7 Same old... Yet to decide. 
Case
Fractal Design Define R3 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel Celeron G530 Haven't decided yet. GTX 460 768MB 4GB Ripjaws 
Hard DriveOSMonitorPower
Samsung F3 1TB (SSD coming later) XP/Windows 7 Same old... Yet to decide. 
Case
Fractal Design Define R3 
  hide details  
Reply
post #6 of 11
Thread Starter 
How do I tell if its going back to 2d clocks,
it shows my gpu under 100% load and then 0% when it freezes.

I am using a 750gb 7200rpm older hdd right now I haven't actually installed the 2tb yet. sorry about that, but is there not a way to disable that?
post #7 of 11
Quote:
Originally Posted by atticus589 View Post
How do I tell if its going back to 2d clocks,
it shows my gpu under 100% load and then 0% when it freezes.

I am using a 750gb 7200rpm older hdd right now I haven't actually installed the 2tb yet. sorry about that, but is there not a way to disable that?
So is Windows 7 installed on the 7200rpm drive right now?

Also, see the attached image. I've underlined what you need to look at when you're running Afterburner in game. If you see these values drop, it's an underclocking problem. Post back here if you see these drop in-game or not.
    
CPUMotherboardGraphicsRAM
Intel Celeron G530 Haven't decided yet. GTX 460 768MB 4GB Ripjaws 
Hard DriveOSMonitorPower
Samsung F3 1TB (SSD coming later) XP/Windows 7 Same old... Yet to decide. 
Case
Fractal Design Define R3 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel Celeron G530 Haven't decided yet. GTX 460 768MB 4GB Ripjaws 
Hard DriveOSMonitorPower
Samsung F3 1TB (SSD coming later) XP/Windows 7 Same old... Yet to decide. 
Case
Fractal Design Define R3 
  hide details  
Reply
post #8 of 11
Thread Starter 
Yes, its windows 7 its this exact rig, I was going to install the 2tb when the psu got in, but I figured I'd make sure everything worked before I went and wiped everything, so I have halted on the 2tb.

The game screen froze so it didn't allow the MSI OSD to refresh, but looking at the logs, you can see that the GPU usage dropped to 0%, and for a second the core clock did as well, and then it stepped back up to 500, then went back to 890, then dropped back to 500 after I aft+f4'd the game.

What does it mean when the clocks drop to 0, it seems like that is where the initial freeze starts.

post #9 of 11
Thread Starter 
I ran prime95 for the hell of it, and it seems as though my second core has developed a problem.
Code:
[Wed Jun 01 09:19:17 2011]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
That's the error that it gave, the stress.txt file is nothing but a guideline doesn't give specifics as to what the fault is.

This actually makes sense, could this be causing these issues?

Is there a way to disable only 1 of the 4 cores?

I set the affinity to use all but the 2nd core, and I notice that when the lag spike hits the 2nd core is at 100% even when the game isn't using it, and on the crash the 2nd core also drops back to 0% where all the other cores are still at 60-80%

The first red line is the initial freeze, the second red line is where it minimizes the game... When it maximizes the game again I alt f4 it because the freeze will happen again seconds later. Setting the affinity to use not use the core doesn't seem to slow or speed up the crashes at all.


Is this turning out to be a more serious hardware error? or am I just being paranoid?






------------------------- EDIT -------------------------
I noticed via the OSD from afterburner that the crash happens AS SOON as my gpu hits 80c, and that the second crash happens as soon as the gpu hits 80c again. To confirm that I cranked the fan up to 70% didn't get over 65c and the game didn't crash or freeze or anything at all.


First red line is the first crash, you can see the temps drop back down to about 75, then as soon as they hit 80 crash at the second red line and the temps drop back off because I alt+f4'd.


Are both my GPU and proc faulty?
Edited by atticus589 - 6/1/11 at 7:57am
post #10 of 11
Thread Starter 
I did more testing on borderlands and furmark, I am not able to get the game to crash with the fan keeping the card cool, it seems as though as soon as my card reaches 80c, it becomes unstable. Furmark can run it up to 85c before the card does the same thing.

Furmarks crashes leave me with more information, furmark is running in windowed mode, the whole 3 monitor setup freezes not just furmark.

When borderlands normally minimizes furmark blacked out all 3 monitors. When the screens come back up, furmark is no longer responding.

I can audibly notice that the fan on the graphics card drops its rpms to either 0 or a range that is not detectable by the human ear, as if the graphics card was reset, so my card is seizing and recovering or the CCC is crashing and recovering either way this cannot be good.


Pictures of the temperature during furmark red line being the point when it crashed.


Picture of the clock speeds, it seems as though it returns to 2d speeds for the crash as well, but it doesn't ever hit 0 like it did with borderlands.

Any input anyone can offer would be of great help...



After monitoring the temperatures for the graphics card, I decided to check my temps for the processor, and it is running really hot, sitting at 48c idle, and 60c under load, isn't it only safe up to 55c?

Furmark had a process still runinng leaving one of my cores at 100%, I closed that, and I changed around 2 of my case fans from sucking air out to blowing air in, and my core dropped to 43c idle now, 57c under load, except now core 1 gets an error not core 2.
Edited by atticus589 - 6/1/11 at 9:49am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD - General
Overclock.net › Forums › AMD › AMD - General › Horrible stuttering and weird problems (borderlands specifically)