Overclock.net › Forums › Graphics Cards › NVIDIA › DirectX 10, PhysX enabled game crashing, despite burn in test from Furmark
New Posts  All Forums:Forum Nav:

DirectX 10, PhysX enabled game crashing, despite burn in test from Furmark

post #1 of 5
Thread Starter 
I've been overclocking my ECS 9600GT (1GMU-F) card for a while now, and from what I could tell with Furmark, I've arrived on what should be a stable clock speed of 714MHz Core ROP, 1800MHz shader, and 990MHz memory.

I've been running a burn in test for 27 minutes now in Furmark, and my card's temperature seems to rest at 70 degrees celsius. That seems to indicate a good amount of stability to me, especially when the driver crashes within 20 seconds if I chose a more unstable core clock speed.

Now, if I launch Metro 2033 under the current clock speeds and start playing a game, the driver will crash within 5 seconds. At stock speeds it will do fine, of course, but my question is: What is it that Metro 2033 is pushing, that Furmark isn't? Do I need to benchmark PhysX in specifc? Or is it something else?

Thanks in advance for any help on this issue.
post #2 of 5
Furmark doesn't really test stability, max stable in furmark is usually completely unstable in anything else.
Best is to play the games you play & overclock as you go. Or you can loop some kinda 3d benches for a starting point & go from there.
    
CPUMotherboardGraphicsRAM
i7 990x/3930k/3770k/4770k GB x58a-OC/Asus RIVE/x79 gd65/z77 mpower/z77 MV... gtx480s/gtx580s/7970s/680s/Titan Hyper/PSC/BBSE/Hynix/Samsung 
Hard DriveOptical DriveCoolingOS
HDDs/SSDs DVD Air/water/phase/copper pot Windowz 
MonitorKeyboardPowerCase
samsung 2320 Razer Lycosa ax1200 carboard box 
Mouse
Trackball 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
i7 990x/3930k/3770k/4770k GB x58a-OC/Asus RIVE/x79 gd65/z77 mpower/z77 MV... gtx480s/gtx580s/7970s/680s/Titan Hyper/PSC/BBSE/Hynix/Samsung 
Hard DriveOptical DriveCoolingOS
HDDs/SSDs DVD Air/water/phase/copper pot Windowz 
MonitorKeyboardPowerCase
samsung 2320 Razer Lycosa ax1200 carboard box 
Mouse
Trackball 
  hide details  
Reply
post #3 of 5
Furmark isn't really that good at weeding out unstable overclocks. It's pretty obvious that you aren't stable at this point. You can either test with OCCT error checking or just play your usual games and make sure you aren't artifacting. Temperature really isn't the main decidimg factor of stability like it tends to be with CPU overclocking.
Edited by Talon95 - 5/25/11 at 2:47pm
Bulldozing
(16 items)
 
  
CPUMotherboardGraphicsRAM
AMD Phenom II X6 1075T @ 3.8 GHz Biostar TA890FXE AMD Radeon HD 7970 @ 1125/1575 MHz Corsair TR3X6G1333C9 4GB 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 256GB SSD Western Digital Caviar Black 500GB Asus DVD Burner 24x Cooler Master Hyper 212 Evo 
OSMonitorKeyboardPower
Windows 7 Ultimate x64 Asus 1920x1200 25.5" 2ms Logitech G11 Thermaltake Toughpower XT 775w 
CaseMouseMouse PadAudio
Antec 1200 Logitech G500 Razer Goliathus Speed Corsair Vengeance 1500 
  hide details  
Reply
Bulldozing
(16 items)
 
  
CPUMotherboardGraphicsRAM
AMD Phenom II X6 1075T @ 3.8 GHz Biostar TA890FXE AMD Radeon HD 7970 @ 1125/1575 MHz Corsair TR3X6G1333C9 4GB 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 256GB SSD Western Digital Caviar Black 500GB Asus DVD Burner 24x Cooler Master Hyper 212 Evo 
OSMonitorKeyboardPower
Windows 7 Ultimate x64 Asus 1920x1200 25.5" 2ms Logitech G11 Thermaltake Toughpower XT 775w 
CaseMouseMouse PadAudio
Antec 1200 Logitech G500 Razer Goliathus Speed Corsair Vengeance 1500 
  hide details  
Reply
post #4 of 5
Thread Starter 
Thanks for your replies, guys.

I will give OCCT a try, thanks I would prefer to use a program that starts up fast, like Furmark, or perhaps OCCT once I try it, just to streamline the process of overclocking my graphics card. Artifacting isn't something that I've encountered much at all really, except when I push my memory clock a little high. Is that normal?

EDIT: OCCT really seems to be hitting the spot here, I never would have expected that from a program with the original intention of only checking CPU overclocks

I'll start adjusting my clocks based on OCCT then, thank you both.

As a side thought, I know that maximum clocks can differ within the same model of graphics cards, but this is just pitiful. With almost all owners of the 9600GT seemingly breaking 70+MHz from stock core ROP clock, it looks like I'll have to start looking at the 50MHz's and below. Also, my graphics card cooling fan, (stock, by the way) seems to be hardly breaking a sweat down there, despite the temperature rising, which concerns me. RivaTuner reports the "reference fan speed" as slowly increasing upwards to 60%, although I can't hear any difference.

It seems to be a common issue among 9600GT owners, where they can't control their fanspeed manually, although I think it would be a problem if the fanspeed isn't changing, period. Is that also a common problem?
Edited by tgp1994 - 5/25/11 at 3:30pm
post #5 of 5
Thread Starter 
Well, it almost seems now that I'm only experiencing PhysX-related crashes when overclocking. Putting my card back at a relatively faster 700MHz for the ROP (while keeping in mind that the memory clock is still locked to this overclock), I experience no issues playing Assasin's Creed 1, which seems to put a bit of a strain on my graphics card. And of course, any PhysX application (the Crazy Machines II tech demo, playing on a level involving water) or Metro 2033 (in general) crashes.

However, reverting back to stock speeds allows all games, PhysX or not, to play normally and without issue. Does anyone see any kind of reason to this?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › DirectX 10, PhysX enabled game crashing, despite burn in test from Furmark