[Tested and Burned] ASUS GeForce GTX 570 DirectCU II Review

ASUS GeForce GTX 570 DirectCU II review index

6 – ASUS GTX 570 DirectCU II Power Consumption and Overclocking

ASUS GTX 570 DirectCU II stressed by FurMark 1.9.0
ASUS GTX 570 DirectCU II stressed by FurMark 1.9.0

For the power consumption and overclocking test, I used FurMark 1.9.0 (finished now, will be released in few days released!).

The total power consumption of the testbed at idle with the GTX 570 is 120W.

The GPU temperature at idle is 32°C.

With default settings (GPU @ 742MHz, Vcore @ 0.950V), the total power consumption of the testbed stressed by FurMark (FurMark settings: 1920×1080, fullscreen, Burn-in mode, dynamic background, no AA, no postfx) is 384W with a max GPU temperature of 75°C.
P = (384-120) * 0.9
P = 237 watts

where 0.9 is the PSU efficiency factor. For the Corsair AX1200 PSU, this factor is around 0.9 (see this article, there is a graph of the AX1200 efficiency).

To overclock the GTX 570, I used ASUS’s SmartDoctor for setting both GPU core clock and GPU voltage.

ASUS SmartDoctor

As you know, GTX 500 cards have a hardware power limiter (mixed with a black list of stress test applications…). But the good news, is that ASUS has disabled this hardware power limiter on its GTX 500 DirectCU II series. How can you quickly test the presence of the power limiter? Simple, just benchmark your card with FurMark 1.8.2 (1980×1080 fullscreen, 60 seconds, no AA, no PostFX). Why FurMark 1.8.2? Because this tool is black listed in NVIDIA drivers.

The GTX 570 has the similar performance than the GTX 480. On my test bench, a GTX 480 has 6478 points (or 108 FPS) in FurMark 1.8.2 test and the GTX 570 DC2 has 6499 points (108 FPS). Then the power limiter is disabled. What’s more another clue you can use is the power consumption. A GTX 570 (with default clocks) stressed by FurMark has a power consumption of around 230W.

With auto fan speed, the max stable OC settings I found for the GTX 570 DC2 under FurMark 1.9.0 are: {GPU: 810MHz ; VDDC: 0.963V}. With these settings, the power consumption of the test bench is 410W while the GPU temperature reached 78°C.

410W for the test bench means the power draw of the GTX 570 DC alone is: (410-120)*0.9 = 261W.


Can we reach higher GPU clock speed under FurMark 1.9.0? Yes, by setting the fan speed to 90%. In this situation, the DirectCU II cooler is noisy but the GPU is perfectly cooled.

Then with a fan speed of 90%, I managed to stress test the GTX 570 DC with FurMark 1.9.0 with this OC setting: {GPU: 843MHz ; VDDC: 1.013V}. With this high OC setting, the power consumption of the testbed was 450W but thanks to the high fan speed, the GPU temperature didn’t exceed 69°C. A high fan speed allows also a better VRMs cooling. That’s why the GTX 570 was stable.

450W for the test bench means the power draw of the GTX 570 DC alone is: (450-120)*0.9 = 297W

This OC setting deserves another badge:


Okay, let’s see now with normal applications. With the auto fan speed, the card was stable under 3DMark11, Unigine Heaven 2.1 and MAFIA 2 with the following OC setting:
{GPU: 921MHz ; VDDC: 1.088V}.

Here are the scores:

  • 3DMark11: P6033
  • Unigine Heaven (DX11, 1920×1080 fullscreen, tessellation normal, 4X MSAA, 16X aniso): FPS: 47.6, Scores: 1199
  • MAFIA 2 (benchmark settings: high): 106.1 FPS

And here are the sames test with default clock speeds:

  • 3DMark11: P5253
  • Unigine Heaven (DX11, 1920×1080 fullscreen, tessellation normal, 4X MSAA, 16X aniso): FPS: 41.6, Scores: 1048
  • MAFIA 2 (benchmark settings: high): 71.0 FPS

Nice overclocking!

Here is a comparative table of the power consumption of the card ALONE (not the total power consumption of the system):

Power: 378W – EVGA GTX 580 SC (core: 870MHz, Vcore:1.125V) – OC settings
Power: 330W – ASUS ENGTX580 (core: 871MHz, Vcore:1.088V) – OC settings
Power: 324W – EVGA GTX 580 SC (core: 797MHz, Vcore:1.082V – default settings)
Power: 298W – ASUS Radeon HD 6950 (core: 840MHz, Vcore: 1.300V) – OC settings
Power: 290W – ASUS ENGTX580 (default settings)
Power: 276W – SAPPHIRE Radeon HD 6970 (core: 940MHz, Vcore: 1.175V) – OC settings
Power: 272W – ASUS Radeon HD 6950 DirectCU II (core: 950MHz, Vcore: 1.150V)
Power: 272W – EVGA GTX 480
Power: 260W – SAPPHIRE Radeon HD 6970 (core: 880MHz, Vcore: 1.175V)
Power: 237W – ASUS ENGTX570 DirectCU II (core: 742MHz, Vcore:0.950V)
Power: 225W – ASUS ENGTX560 Ti DirectCU II TOP (core: 900MHz, Vcore:1.025V)
Power: 220W – SAPPHIRE Radeon HD 6870 (core: 1000MHz, Vcore: 1.300V) – OC settings
Power: 204W – ASUS Radeon HD 6950 DirectCU II (core: 810MHz, Vcore: 1.100V)
Power: 197W – ASUS Radeon HD 6950 (core: 810MHz, Vcore: 1.100V)
Power: 185W – ATI Radeon HD 5870
Power: 155W – SAPPHIRE Radeon HD 6870 (core: 900MHz, Vcore: 1.175V)
Power: 147W – ASUS EAH6870
Power: 135W – MSI N460GTX Cyclone 768D5 OC
Power: 81W – ASUS GeForce GT 440 @ 823MHz

ASUS GeForce GTX 570 DirectCU II review index

13 thoughts on “[Tested and Burned] ASUS GeForce GTX 570 DirectCU II Review”

  1. Zibri

    I think you should test this card with EVGA OC Scanner (the one I modified) just to check the real artifact-free OC.

  2. Pingback: GTX 570 overclocking thread - Page 47 - Overclock.net - Overclocking.net

  3. Mark Lev

    hi.. i wanna buy this card..i got a core 2 quad 3.0 ghz.. i have a 750 watt antec power supply.. will it be okay ?

  4. Wisico

    Is it normal that my card when it pass the 60% fan speed (with afterburner) it start to make a loud noise?? :O
    But there is Nothing In the fan…
    What should I do??

  5. JEswe

    Mark Lev yes youre setup will run this card.
    Wisico youre card is probebly geting hot and protecting the GPU or you got dust in there.

  6. Chris

    what are those slots on the front with the plastic covers?
    should I connect them to anything?
    I got a small flat cable with my card
    but I dont understand where to put it?
    plese help!

  7. Nick

    I’m not sure what I’m missing here. I recently got my hands on one of these cards and knew it was going to be a true force to be reckoned with. While playing Skyrim with the high res pac I get jumpy frames 60fps-0fps; Yet my hardware isn’t even trying! (the fan speed stays the same and the temp never rises) I’m confused why it’s holding back! I’ve played Crysis 2 DX11 and Arkham City on full specs without a problem – why little DX10 Skyrim has low frames I’m baffled. I was going to save the benchmarking until I was able to finish my build off with 8 more gigs of ram and an SSD but the recent speedbump in gaming brought me to it early. I was astounded how low I was ranking, to say the least. >=( Furmark was the first and I’m about to try unigine and 3dmark and probably be twice as disappointed.

    I was wondering how the numbers here and on every other benchmark were attained with less of an overclock than I even have, when I haven’t touched overclocking yet since I wasn’t worried until now! furmark put me at just over 2,000 (36FPS) >=(!!!
    Hardware: (everything @ stock specifications)
    i5 2500k @3.3GHz w/ a Zalman fan cooler
    Asus P8Z68-V Pro mobo
    2 4GB corsair vengeance @ 9-9-9-24
    1080W PS

  8. Nick

    >>>And yes, I forgot to mention, I have the updated drivers. Even the Nvidia one specifically for the Skyrim high res pack.

  9. Brad

    Hello nick, dont have any experiance with your card, but running skyrim on full specs with all the GFX and hi res mods, on a setup less powerfull than yours. The only thing I had to do is patch the main executable so it allows more then 2GB ram usage. Try doing that there are guides online, let us know how you get on.

  10. Jim

    Thank’s! please add Assassin’s Creed II Gaming benchmark!
    Thank’s Again!

Comments are closed.