How to check
I tested the processors in three modes - one is a mode I call "stock mode". In this mode the processors operate by default setting Intel
And the motherboard runs in a default configuration. That is, original frequencies of action, original tensions of action.
The second situation is the state of net tension. As with the first test mode, the default processor frequency and the default settings of the motherboard are also used here. The only change made is a change in the processor's operating voltage. Here I decided to check the lowest power consumption on which the processor can operate in the default settings. All I did was go into Bussus and begin to gradually lower the operating voltage while checking the stability of the effort program. I do not want to display the operating voltages which I used simply because of the reason it varies between Processors
Different, even from the same model.
The third situation is the situation The fast
. In this situation I did The fast
Traditional processors K
. This means that I have not touched on power saving technologies but I have added base frequency and voltage multipliers as required to achieve stability in sane operating temperatures. The processors i7 2600K
Operate in this 4.5 GHz mode and the processors i5 3570K
Operates in 4.7 GHz.
For the test I used an AC power meter connected directly to the power supply so that the power consumption displayed is directly from the wall outlet.
Here is the bar that shows the power consumption of the processors at rest. The power consumption displayed is of the entire system for testing the use of the built-in graphics core Intel
In a processor that is at rest.
As you can see, the original rest state of processors Intel
It is the most economical power, even if a change in voltages is made downward, this mode will simply override the econometric default settings and therefore in the mode of cleaning the tension to stability in effort the processors consume a little more at rest. The power consumption at the rest of the processors is in a state of shock, even though they operate at a minimum frequency, simply high, actually double.
The following graph shows you the power consumption of the test system when you test a CPU-type effort Prime95 Small FFT.
In the test of processor effort, the differences in power consumption can be seen in all three situations. Note the significant power differences after cleaning the voltages at the default frequency. Even those who are not interested in processor acceleration and want low operating temperatures and reduced power consumption will expect up to 20 watts (!) Of power in processors after cleaning the processor voltage up to the lowest voltage at which the processor can operate at its original speed (including turbo).
The situation under Overclock
Not so encouraging for the honest truth. For thei7
A maximum power consumption of 50 and more watts was observed. I expected that at a similar acceleration rate and with the same voltage as the CPUi7 3770K
Will be more frugal, but depending on the improvement as well as extra power consumption has increased. Power consumption in processors i5
It also increased significantly but the rate of precipitation is higher as a result of lower operating temperature.
Now it's the turn of the video cards. I tested the video cards in four modes - while at rest, while straining in the default settings, while The fast
To the maximum operating frequency without changing the graphics core voltage and time The fast
Maximum. The state of effort of the graphics cards is during game play Crysis 2
Under Maximum Settings. To-HD7750
You can not change the core voltage as an example of video cards, so in the fourth mode I added the extraPowerTune
Which enables higher power consumption than the maximum. The speed values can be found in reviews of each of these video cards. I again mention that the results are of the entire test system. The results are as follows:
Represents all his brothers, since he carries the same architecture and form of action under Overclock
. In his case the assumption that electricity consumption did not increase significantly without the addition of voltage is absolutely correct. In tension Core
Of the Volt 1.3 and higher frequency of power consumption simply leapt up. This, even without the tuning of PowerTune
Switch to the maximum power settings.
The performance generation of the NVIDIA
Represents Geforce GTX670
. Although there is less space to perform overclocking and performance improvement is not greater than that of the presenter HD7970
But electricity consumption is also not significantly higher, if hardly at all. Additional video cards GTX600
Show similar numbers under The fast
Even with the addition of a graphic core voltage, the power consumption differences are simply not large and stand on a few watts up to about 20 watts in extreme cases. For comparison, the video card GTX570
The previous generation also introduced an extra significant consumption under The fast
But when I added the 0.1 volts that are enabled to program in order to reach a higher frequency of operation, the power consumption increased by about 50 watts and more.
HD7750? Simply refuses to consume more electricity. A complete test system is established Core i7 2600K It has a clean power and a video card HD7750 Gorgeous shows while playing Crysis 2 Power consumption of 102 watts. Alessandro Volta turns over in his grave.