FurMark and Overclocked GTX 480: 108 Amperes is not a Problem…



FurMark and an overclocked GTX 480: 108 amperes
FurMark and an overclocked GTX 480: the intensity of the current reaches 108 amperes!

GTX 480 cooled by Thermalright's SpitFire VGA cooler
GTX 480 cooled by Thermalright’s SpitFire VGA cooler



Yes 108 amperes! A current of 108 amperes goes through the GPU. But watch out my friends, the GPU we are talking about, is the one behind the GeForce GTX 480. To reach 108A, the GPU is overclocked to 797MHz (the reference clock is 700MHz) with a VDDC (GPU voltage) of around 1.04V (the non-oc VDDC is 1V).

The guys at pcbrain.it have FurMark-ed a GTX 480 with standard clocks and VDDC and with overclocked settings. They have done the test with the stock VGA cooler and with Thermalright’s SpitFire (official page HERE).



FurMark and a GTX 480 with default settings (clocks and VDDC) and stock VGA cooler

FurMark and GTX 480 with default clocks
FurMark and a GTX 480 with default clocks and VDDC

FurMark and GTX 480 with default clocks

FurMark and GTX 480 with default clocks



FurMark and a GTX 480 with overclocked settings (clocks and VDDC) and stock VGA cooler

FurMark and overclocked GTX 480
FurMark and an overclocked GTX 480

FurMark and overclocked GTX 480

FurMark and overclocked GTX 480

FurMark and overclocked GTX 480


These images clearly show the GPU that is throttling back when temperature is too high: the temperature curve does not follow a normal heat progression. This is the overheating protection of the GTX 480.



FurMark and a GTX 480 with overclocked settings (clocks and VDDC) and Sptifire VGA cooler

FurMark and overclocked GTX 480 with a SpitFire VGA cooler
FurMark and overclocked GTX 480 with a SpitFire VGA cooler (source)

FurMark and overclocked GTX 480 with a SpitFire VGA cooler

FurMark and overclocked GTX 480 with a SpitFire VGA cooler



I did a quick test with my EVGA GTX 480 (standard settings + stock VGA cooler) and I got similar results:

FurMark and an EVGA GTX 480 with default clocks
FurMark and a EVGA GTX 480 with default clocks and VDDC

As you can see, with a good VGA cooler, an overclocked GTX 480 can resist without problem to FurMark torture stress test.

More than 100 amps for a GPU… These numbers are just impressive. I wonder if the GTX 480 can support such a current during a long period… I hope so!



9 thoughts on “FurMark and Overclocked GTX 480: 108 Amperes is not a Problem…”

  1. Yahogie

    Get a clue and take high school physics. Why is this a shocking number at 1V? If it were 108 Amps at at 120V you would be pulling 12+ kilowatts.

  2. WacKEDmaN

    somethin up with the numbers there…

    Power (w) = Voltage(v) * Current(i)

    1.04v * 108.5a = 112.84watts

    or if we go the other way…

    Current(i) = Power(w)/Voltage(v)…

    105w / 1.04v = 109.2

    something definatly up with then numbers

  3. JeGX Post Author

    Actually the power consumption is calculated from GPU-Z sensors: VDDC and VDDC Current. And the VDDC value read by GPU-Z is a bit lower (less than 1V) than the value set by the tester (1.04). That explains the numbers.

  4. DrFPS

    Did you guys go to school at all? More than 100 amps. If that be the case you have to have a nuclear reactor for every home.
    Suggest you google OHMs law before you tattoo failed across your forehead.
    You might want to look at the output of the PSU.
    It’s 12 volts@105 watts=12.6 12v amps.
    Can’t wait to see how many amps it is overclocked
    Mo,Curly,Larry

  5. JeGX Post Author

    The 100 amps are under a voltage of around 1V, so no need a nuclar reactor to feed the GTX 480! Keep in mind that GPU voltage is not 12V or 5V but 1V.

    The important point is the energy and especially the transfer of energy. The VRM allows this transfer of energy from the PSU to the GPU. The VRM acts like a voltage transformer…

  6. WacKEDmaN

    @JeGX… would it be possable to show the actual value of VDDC + current up near the power consumption in next version please

  7. nevermind

    108 Amps is foolish! It would require a huge conductor, nevermind if it is just 1V.

Comments are closed.