The Tests, Test Setup, And A Side Note
The Tests
Many of today's tests are the same ones we used in our previous article, AMD's Radeon HD 5000 Series: Measuring Power Efficiency. As an example of a non-gaming workload, we are measuring power consumption during Cinebench's OpenGL viewport test. We’re using the results from Cinebench as a measure of desktop application performance.
We're also conducting tests with CyberLink's PowerDirector 8, making sure GPU acceleration is active for GPU-accelerated video filters. PowerDirector also supports video encoding with GPU acceleration.
Since PowerDirector 8 does not natively support hardware encoding on the GPU using AMD cards, we patched the application with the latest update (3022) and installed AMD’s Avivo transcoding tool.
The last test is H.264-accelerated video playback in CyberLink's PowerDVD 9.
Test Setup And A Side Note
We're using the same hardware setup from our previous article. This way, the results are directly comparable to those obtained previously.
We left the power-saving features of the Phenom II X4 955 BE enabled in the BIOS, and we set Windows’ power policy to Balanced. To lower the base system power consumption even more, we altered Cool'n'Quiet with K10Stat, allowing our processors to run at even lower voltages. Using these settings, our test platform’s base power consumption hovered around 55 watts at idle and 80 watts during H.264 video playback. This is right around the threshold where our platform's PSU efficiency drops off.
Our altered voltages mean this isn’t a stock setup, per se. So, these results are not directly comparable if you’re running default voltage settings. You can see the differences in system power consumption between running with default voltages and our undervolted settings below.
| Phenom II X4 955 BE & Integrated Radeon HD 3300 | ||||
|---|---|---|---|---|
| Idle | Crysis | CS 4 | Cinebench | |
| Undervolted | 55 W | 121 W | 120 W | 132 W |
| Default Voltage | 73 W | 160 W | 143 W | 195 W |
There is one minor consideration to bear in mind as you look at the power measurement results: variation between samples. Variations from one card to another exist. They may be due to the components used, board design, and even the graphics chip itself. Your own testing may consequently look a little different from what we have here.
All power consumption numbers are for a full system, sans monitor. They are recorded with a Watts Up! Pro power meter sampling at one-second intervals.
Related Stories :
- ARM: Intel Cannot Match Us in Power Consumption
- Power Profiles May Help Lower Power Consumption of Chips
- Power Supply Reference: Consumption, Savings, And More
- SanDisk Pushing For Low Power Consumption SSDs
- Power Consumption Solved: Magnetic Processors
- Display Power Consumption: CRTs Versus TFT-LCDs
- Solidata SSDs: Record Power Consumption
- The SSD Power Consumption Hoax
- Intel Cuts Power Consumption Rating Of "Skulltrail Xeon"
- Dell Aims To Cut Power Consumption Of PCs By 25%
Dude, I'm now feeling a sudden urge to watch The Matrix!
I think, considering those people using SLi and crossfire and higher end videocards, they don't really give a gat about how much elec. they are using. They can afford to buy two expensive PCBs, why would they care about extra 5~10 bucks per month? If poeple are focused on lower power consumption, they would go for lower performance components, arent they?
I guess the most important point of this review is that you can lower the cards voltage while running at stock speed. For example I'm running my GTX 460 (stock 675/1800@1.012V) at 777/2070@0.975V or if I wanted to use the stock speeds, I could lower the voltage to 0.875V. I've also lowered the fan speeds to allow the card to run almost silently even at full load.
@.@ there is no apple @.@
I wonder if this article might inspire someone to make an application. Come on open source dont fail me now >.
This is neat though
Could you do comparison of "the fastest VC" vs "entry level" and then show us how much money we might end up paying each month or day?
Manufacturers should find a way to implement this automatically, imagine the possibilities!
@OvaCer
I have 2 gfx cards pushing 3 displays, but I'm all for saving watts wherever I can. Our society has advanced to the point where sustainability is a very important buzzword that is widely ignored by mainstream media and many corporations, and this ignorance trickles down to the mainstream like Reaganomics. Minuscule reductions such as 30w savings across hundreds of thousands if not millions of users adds up to a significant reduction in carcinogenic emissions and saves valuable resources for future consumption.
Holy crap, I watched the Matrix again just before this topic was posted..
So when playing video, you risk your amd card going into uvd mode? What models does that apply to?
I want to know, cause for instance in a raid, I'd sometimes watch video content on another screen while waiting around for whatever there is to wait for. I already lose the crossfire performance because of window mode. I don't want to lose even more.
Does my ancient 4870x2 support uvd?
so... for the dual bios HD6900s, I can RBE one bios with my desired settings and just choose which bios to use before I power up my PC? hmmm... interesting.
Pretty pointless article which concludes the obvious.
This is like trying to save gas with a Ferrari.
This is like trying to save gas with a Ferrari.
With proper engine mapping, good throttle control, weight reduction from materials used, loosing carpets, spare tyre, radio etc. and the aerodynamic design of a Ferrari it isn't as stupid as you think.
Interesting. When I have time I will try lowering the voltage while maintaining the OC and see what happens.
I think, considering those people using SLi and crossfire and higher end videocards, they don't really give a gat about how much elec. they are using. They can afford to buy two expensive PCBs, why would they care about extra 5~10 bucks per month?
At a full 8 hours per day, every day per month, you're talking $2.40 or basically the cost of one 12 oz. latte at Starbucks.
This should all happen automatically with a number of parameters set by the user, like motherboards. To me, fan noise is the biggest enemy, and a small power savings takes back seat. One of the downsides I see is if some manufacturers used a new power saving architecture as an means to take shortcuts in cooling their cards properly at full throttle.
I undervolt most of my cpus anyway
Here I thought that my two 4000-series systems (4850 and 4890), which do not reduce memory frequency in either 2D or UVD mode, were truly behind the times. It appears even 6000-series cards don't do it while in UVD mode. That alone is something which could be enabled by AMD and it's board partners on all new cards by default. But, seeing as the default memory speed provides the best benchmark numbers, that's likely the reason why it's not done.
I typically use CCC's Overdrive to manually reduce memory speed while in 2D-mode, which works fine. Despite having been manually set, whenever it senses UVD playback (be it Flash, DivX, etc.) the memory frequency changes to it's default speed (1000MHz on the 4850, 975 on the 4890). Now I know why - once UVD mode is active, it's clocks reign supreme. It seems the only way to change that is to mod the BIOS.
When it comes to general users, the end result of all this tweaking seems to unfortunately only save rather minuscule amounts of power.
With proper engine mapping, good throttle control, weight reduction from materials used, loosing carpets, spare tyre, radio etc. and the aerodynamic design of a Ferrari it isn't as stupid as you think.
Those are performance modifications, less gas consumption might be a side effect but it is not the goal.
The 599 is stupid, the 430 is godly.