Overclock.net › Forums › Graphics Cards › NVIDIA › Stability testing GTX570
New Posts  All Forums:Forum Nav:

Stability testing GTX570

post #1 of 12
Thread Starter 
I have had my MSI GTX570 for a few months and decided to give a shot at OC'ing it. Settled on 850 core 2000 mem at 1.038v (temp mid to high 70s).

However what i noticed was that it was not Kombuster or my host of benchmark programs that really stressed this as after a few runs of 3Dmark11 and Kombuster that passed fine, it crashed in Cities in Motion after a few mins. Now that is not a demanding game and my GPU usage never went over 60%! And with Just Cause 2 the same thing, stable in benchmarks but not games.

So is there a benchmark as good as playing the games themselves?
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
6700k GA-Z170-Gaming K3 MSI GTX 1080 Ti Corsair Vengeance LPX DDR4 3200mhz 
Hard DriveOSMonitorKeyboard
Vertex 4 512gb , Samsung Pro 850 1TB Window 10 64 Bit Dell S2716DG Microsoft Sidewinder 
PowerCaseMouseMouse Pad
EVGA 1000W Fractal Design Define C Razer Death Adder Razer Manits Control 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
6700k GA-Z170-Gaming K3 MSI GTX 1080 Ti Corsair Vengeance LPX DDR4 3200mhz 
Hard DriveOSMonitorKeyboard
Vertex 4 512gb , Samsung Pro 850 1TB Window 10 64 Bit Dell S2716DG Microsoft Sidewinder 
PowerCaseMouseMouse Pad
EVGA 1000W Fractal Design Define C Razer Death Adder Razer Manits Control 
  hide details  
Reply
post #2 of 12
try furmark
Don't Panic
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i5 450M @ 2.4GHz Samsung Q430/Q530 Nvidia GeForce 310M Elpida 4GB DDR3 @ 533MHz 
Hard DriveOptical DriveOSMonitor
WD 500GB DVD Burner Windows 7 x64 Integrated 1336x786 LCD 
KeyboardPowerCaseMouse
Saitek Cyborg v.5 Battery Laptop Logitech G500 
  hide details  
Reply
Don't Panic
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i5 450M @ 2.4GHz Samsung Q430/Q530 Nvidia GeForce 310M Elpida 4GB DDR3 @ 533MHz 
Hard DriveOptical DriveOSMonitor
WD 500GB DVD Burner Windows 7 x64 Integrated 1336x786 LCD 
KeyboardPowerCaseMouse
Saitek Cyborg v.5 Battery Laptop Logitech G500 
  hide details  
Reply
post #3 of 12
Quote:
Originally Posted by ghost55 View Post
try furmark
NO DON'T RUN FURMARK UNLESS YOU WANT TO KILL YOUR GTX570..

There have been a great number of dead GTX570 while running Furmark and OCing..

Run 3DMark 11 loop, Heaven benchmark,Crysis benchmark loop for the best stress testing..

CHEERS..
     
CPUMotherboardGraphicsRAM
E6750 G0 @ 3.3GHz (1.25V LOAD) Asus P5KPL-VM G31 GeForce 8500GT @ 700/1600/445 Apacer 2x1GB 800MHz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar 160GB (7200,8MB) LG DVD Win 7 Ultimate 64bit build 7601 LG 19" 
KeyboardPowerCaseMouse
5$ keyboard Cheap stuff Linkworld crappy/modded SmartBox cheap 
Mouse Pad
N/A 
  hide details  
Reply
     
CPUMotherboardGraphicsRAM
E6750 G0 @ 3.3GHz (1.25V LOAD) Asus P5KPL-VM G31 GeForce 8500GT @ 700/1600/445 Apacer 2x1GB 800MHz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar 160GB (7200,8MB) LG DVD Win 7 Ultimate 64bit build 7601 LG 19" 
KeyboardPowerCaseMouse
5$ keyboard Cheap stuff Linkworld crappy/modded SmartBox cheap 
Mouse Pad
N/A 
  hide details  
Reply
post #4 of 12
Quote:
Originally Posted by KingT View Post
NO DON'T RUN FURMARK UNLESS YOU WANT TO KILL YOUR GTX570..

There have been a great number of dead GTX570 while running Furmark and OCing..

Run 3DMark 11 loop, Heaven benchmark,Crysis benchmark loop for the best stress testing..

CHEERS..

"Great Deal" is a bit of an exaggeration..
I would say about one in ten is a conservative number of GTX 570s will die when overvolted and stressed in such a way as furmark..


I do however agree with this comment for one reason. Furmark is stupid. It just pushes your card really hard, Pulls more power than gaming does. And yet does not let you know if its unstable. Ive seen 1 hr + stable furmark overclocks crash within minutes in game... My point is GAMING is the only true stress test IMO.
post #5 of 12
^This.

I bench for 99 loops in Crysis with everything to the max to basically stress the card and system out. Then I play the game since a benchmark is scripted and will never be comparable to actually gaming. Then I will either be happy or raise my OC up to compare frame rates.
post #6 of 12
Quote:
Originally Posted by KingT View Post
NO DON'T RUN FURMARK UNLESS YOU WANT TO KILL YOUR GTX570..

There have been a great number of dead GTX570 while running Furmark and OCing..

Run 3DMark 11 loop, Heaven benchmark,Crysis benchmark loop for the best stress testing..

CHEERS..
I agree with this, and to the guy that thinks this is a small issue lets count how many 570s are dead on this forum. Thread of Dead 570s
Stealth Assassin
(16 items)
 
  
CPUMotherboardGraphicsRAM
i5 6600K Asus ROG Maximus VIII Hero Alpha EVGA 980ti Classified  Corsair LPX DDR4 3000 
Hard DriveCoolingCoolingCooling
Samsung 850 Pro Noctua D15s Cooler Master Silencio PWM 120MM 2400RPM Cooler Master Silencio PWM 120MM 1400RPM 
OSMonitorKeyboardPower
Windows 10 Samsung Curved S34E790C Corsair Strafe RGB EVGA SuperNOVA P2 650w 
CaseMouseAudioAudio
Master Case 5 Corsair Scimitar Sennheiser HD 558 Corsair Void USB 
  hide details  
Reply
Stealth Assassin
(16 items)
 
  
CPUMotherboardGraphicsRAM
i5 6600K Asus ROG Maximus VIII Hero Alpha EVGA 980ti Classified  Corsair LPX DDR4 3000 
Hard DriveCoolingCoolingCooling
Samsung 850 Pro Noctua D15s Cooler Master Silencio PWM 120MM 2400RPM Cooler Master Silencio PWM 120MM 1400RPM 
OSMonitorKeyboardPower
Windows 10 Samsung Curved S34E790C Corsair Strafe RGB EVGA SuperNOVA P2 650w 
CaseMouseAudioAudio
Master Case 5 Corsair Scimitar Sennheiser HD 558 Corsair Void USB 
  hide details  
Reply
post #7 of 12
Then account for how many were actually sold his logic is SOUND.
And when i think overvolting i think going over what afterburner sets as the threshold.
And some of them on that dead 570 forum weren't even using furmark just playing a game.
Its not furmark they dont like its the overvolting in general.
Edited by solar0987 - 3/5/11 at 1:39pm
post #8 of 12
furmark is just silly , unless you use ur gfx card for folding n such for long periods of time theres no point in risking that damage to ur gfx card .. furmark really burns your graphics card up and 570 wont like that , no wonder heaps of them are dead ..people push them beyond the limit and run them at 100% in furmark for hours .. that is just rediculous . For what reason ? to be cool .

I will never use furmark on my 6950 , while its overclocked on the 6970 bios (different ram timings etc) almost at their limit ... that suicide ..

I also use crysis to check my gfx stability , as all i do is play games .. so if its stable in game , its stable enough for me =) and i know my card wont die during the process .
Edited by Xristo - 3/5/11 at 1:43pm
    
CPUMotherboardGraphicsRAM
Intel i5 7600k @ 5Ghz, 1.312v Asus Prime Z270-AR msi GTX 1080 32GB Corsair Vengeance DDR4 3Ghz 
Hard DriveCoolingOSMonitor
Transcend SSD220S 120GB  Corsair H80i V2 Windows 10 Pro 64bit AOC AG322FCX 32" Curved LED 144hz 1920x1080 
KeyboardPowerCaseMouse
Razer Blackwidow Ultimate Antec Quattro 1000w Thermaltake Versa N27 msi Gaming series 
Mouse PadAudio
Razer Goliathus Steelseries Siberia V2  
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel i5 7600k @ 5Ghz, 1.312v Asus Prime Z270-AR msi GTX 1080 32GB Corsair Vengeance DDR4 3Ghz 
Hard DriveCoolingOSMonitor
Transcend SSD220S 120GB  Corsair H80i V2 Windows 10 Pro 64bit AOC AG322FCX 32" Curved LED 144hz 1920x1080 
KeyboardPowerCaseMouse
Razer Blackwidow Ultimate Antec Quattro 1000w Thermaltake Versa N27 msi Gaming series 
Mouse PadAudio
Razer Goliathus Steelseries Siberia V2  
  hide details  
Reply
post #9 of 12
Thread Starter 
Thanks guy for the input. I thought Furmark was the deal when it came to stability testing. But it just goes to show nothing beat a good session on a number of games to prove it stable. If Cities in motion can crash a Furmark stable 1hr card in 5 mins, something is wrong with Furmark!
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
6700k GA-Z170-Gaming K3 MSI GTX 1080 Ti Corsair Vengeance LPX DDR4 3200mhz 
Hard DriveOSMonitorKeyboard
Vertex 4 512gb , Samsung Pro 850 1TB Window 10 64 Bit Dell S2716DG Microsoft Sidewinder 
PowerCaseMouseMouse Pad
EVGA 1000W Fractal Design Define C Razer Death Adder Razer Manits Control 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
6700k GA-Z170-Gaming K3 MSI GTX 1080 Ti Corsair Vengeance LPX DDR4 3200mhz 
Hard DriveOSMonitorKeyboard
Vertex 4 512gb , Samsung Pro 850 1TB Window 10 64 Bit Dell S2716DG Microsoft Sidewinder 
PowerCaseMouseMouse Pad
EVGA 1000W Fractal Design Define C Razer Death Adder Razer Manits Control 
  hide details  
Reply
post #10 of 12
Quote:
Originally Posted by tgawn View Post
Thanks guy for the input. I thought Furmark was the deal when it came to stability testing. But it just goes to show nothing beat a good session on a number of games to prove it stable. If Cities in motion can crash a Furmark stable 1hr card in 5 mins, something is wrong with Furmark!
Furmark tends to just heat the card up imo, and fermi cards throttle down during furmark to prevent damage.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › Stability testing GTX570