discussion Overclocking the ASUS GTX 580 Voltage Tweak Edition

[SKYMTL]

Unigine: Heaven v2.0 (DX11)

Unigine: Heaven v2.0 (DX11)

Unigine’s Heaven benchmark is currently the de-facto standard when it comes to simple, straightforward DX11 performance estimates. While it is considered a synthetic benchmark by many, it is important to remember that no less than four games based on this engine will be released within the next year or so. In this test we will be using a standard benchmark run with and without tessellation enabled at three resolutions,


1680 x 1050

1920 x 1200

2560 x 1600

 

[SKYMTL]

8x MSAA Testing

8x MSAA Testing

In this section we take a number of games we have tested previously in this review and bring things to the next level by pushing the in-game MSAA up to 8x. All other methodologies remain the same.

BattleField: Bad Company 2 (DX11)

Note that 8x MSAA is enabled via the game’s config file for the NVIDIA cards since it is not a selectable option within the game menu

Dirt 2 (DX11)

Just Cause 2 (DX10)

 

[SKYMTL]

Folding @ Home Performance & Power Consumption

Folding @ Home Performance & Power Consumption

One of the cornerstones of consumer GPGPU performance has always been a card’s ability to crunch numbers for distributed computing projects like Stanford’s Folding @ Home program.

In this test, we use the latest GPU3 client on a number of cards and let it run for 24 hours in order to get an approximate PPD average. Since this is the beta of GPU3, the quantity of projects is limited which helps give a slightly more accurate comparison between the GPUs. In order to monitor PPD, we used the HFM.net program. Note that results are based off of three Work Units with 611, 912 and 925 points.

Folding @ Home performance is quite frankly incredible once the core is overclocked. However, we have to caution you about GDDR5 memory overclocking when Folding since the returns are next to nil when you compare it to the increase in overall power consumption.

Naturally, with a massive increase in Folding power comes an associated boost in power consumption. Even with the excellent cooling solution installed onto the GTX 580, this highly overclocked card still consumes slightly more than a stock GTX 480. This may be considered “high” by some but in our opinion, the performance increase in this GPGPU application is more than worth it.

 

[SKYMTL]

Core Temperature & Acoustics / Power Consumption

Core Temperature & Acoustics

For all temperature testing, the cards were placed on an open test bench with a single 120mm 1200RPM fan placed ~8” away from the heatsink. The ambient temperature was kept at a constant 22°C (+/- 0.5°C). If the ambient temperatures rose above 23°C at any time throughout the test, all benchmarking was stopped. For this test we use the 3DMark Batch Size test at it highest triangle count with 4xAA and 16xAF enabled and looped it for one hour to determine the peak load temperature as measured by GPU-Z.

For Idle tests, we let the system idle at the Windows 7 desktop for 15 minutes and recorded the peak temperature.

Note that for this test, the fan on the overclocked GTX 580 was set to 67%.

When overclocked, the GTX 580 does tend to push the limits of the vapor chamber heatsink; even with the fan running at slightly over 65%. However, due to the modifications NVIDIA did to the fan on this card, any fan speed below 70% can be considered relatively quiet when compared to the racket the GTX 480 put out under certain conditions.

System Power Consumption

For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we once again use the Batch Render test in 3DMark06 and let it run for 30 minutes to determine the peak power consumption while letting the card sit at a stable Windows desktop for 30 minutes to determine the peak idle power consumption. We have also included several other tests as well.

Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the power meter.

It should go without saying that increasing clock speeds AND boosting voltage will have a negative impact upon power consumption. Naturally, that’s exactly what happened here but the actual increase isn’t all that bad when you consider the overclocked performance against the framerates displayed by a reference GTX 480.

 

[SKYMTL]

Conclusion: Maniacal Performance for the People

Conclusion: Maniacal Performance for the People

In its stock form the GTX 580 is already a hugely powerful video card but when you add overclocking to the equation, the results are simply mind blowing. The performance increases we achieved led to significant framerate gains across every single game and in some situations pushed playability to new heights.

We’re reasonably sure that with some time, effort, proper fan speeds and a bucket full of patience, most buyers of GTX 580 cards will be able to reach the levels we did in this article. These new NVIDIA cards are overclocking monsters and some phenomenal numbers can be squeezed out of them without resorting to over the top cooling methods. High clock speeds do lead to some over-the-top power consumption numbers but when you are pushing the limits, who the heck cares about something as trivial as a power bill?

The ASUS GTX 580 we used for this test deserves special mention since it was a more than willing participant in our little overclocking experiment. It wasn’t unique either since all of our GTX 580 cards hit approximately the same speeds when a bit of extra voltage was applied.

Believe it or not we’re actually starting to feel at home with the Smart Doctor software, no matter how much it has been bad-mouthed around here in the past. Sure its sliders are minuscule and the OSD is dominated by a mostly pointless readout but ASUS consistently has one of the only pieces of software that supports overclocking AND voltage tweaking on new architectures. To us, that counts for a lot even though it is locked to ASUS products.

There has been quite a bit of fuss made about NVIDIA’s implementation of current draw protection on their GTX 580 cards. It is supposed to limit the amount of power the GF110 core can consume but it really does seem to be a non-issue here - even when pushing the card above and beyond its reference clocks. From our experience, the casual overclockers among us who stick to air cooling will likely run into a heat wall long before NVIDIA’s supposed limiter kicks in. Once pushed to extremes, there may very well be a time when the protection routines scream “enough is enough” but even at 951Mhz, we didn’t see any type of limitation whatsoever. For the time being though, we have yet to come up with any evidence to suggest throttling will kick in when using any application outside of FurMark or OCCT’s GPU stress tool.

In all too many situations, trying to push the clock frequencies of many current graphics cards can often lead to a lesson in frustration. This isn’t the case with the GTX 580. Considering the easily attainable and massive performance boost an overclocked GTX 580 can put forward, we’re sure this will draw quite a few more people into waiting embrace of overclocking. If anything, the abilities of this new card will allow it to act like a shot of adrenalin for many enthusiast circles and for that alone it deserves our praise.

http://www.hardwarecanucks.com/foru...580-voltage-tweak-edition-comment-thread.html​