for those with x6's

[SKYMTL]

Heat is the BYPRODUCT of increased power consumption. Basically, as efficiency goes down, power consumption AND heat production increase.

[CMetaphor]

Even so, I don't see how its possible for overall consumption to go up that much when you're only adding half a volt over stock... Must test.

Sky, whatever happened to the X6 OCing article you promised us?

[bojangles]

EDIT: Oops, I was thinking wind power lol. Power increases by the cubic of wind speed.

Power IS related by the square of voltage.

[sswilson]

I wish I could test this, but I don't have any form of power meter that will do in-line testing.

[SKYMTL]

Originally Posted by CMetaphor

Sky, whatever happened to the X6 OCing article you promised us?

I didn't promise it. That was all MAC.

However, prep for Computex + some other rush articles got in the way. ETA is unknown.

[CMetaphor]

Originally Posted by sswilson

I wish I could test this, but I don't have any form of power meter that will do in-line testing.

Testing now. Although I'm "only" at 3.6Ghz because i've been to busy to push beyond that thus far (plus it's been super hot recently)

Current OC:
3.6Ghz, Turbo core off, CnQ on - 1.38v
Rest of system:
Crosshair IV, 2x 5770s in Xfire, Ante TP-new 750w, (not too far from what TPUs setup)

Idle power (Read from my APC 1500VA load meter) = 160w
Full load on CPU = 270w

I suppose the only way to know for sure will be when I push to 4.0Ghz, but for now I don't see where TPU are getting their numbers from unless they were stressing Both the CPU and GPU at the same time by accident during a CPU review.

[Arinoth]

Originally Posted by Zero82z

It absolutely can be double.

Indeed it can. Here is the basic power formula
where V is voltage and I is current
Seeing as we're mainly playing with voltage and resistance (it never changes so its a known) we can move into the other power formulas

This shows that with a fixed resistance the power will actually increase quadratically making it possible for the power to double.

Stock 1.284V and the wattage is 144W which gives us an resistance of 0.008917 Ohms
Overclocking it to 1.5V and assuming the resistance stays the same (as it typically should though i dont know the actual properties) yields 252W.

This is an actual double in the power consumption, not near that of TPU's 440W, so either i'm off doing some basic calculations (again assuming resistance stays the same as the frequency increases) or someone at TPU dropped the ball on the actual wattage of this new cpu.

[sswilson]

We'd need some stock results to be able to get an idea of the delta. Keep in mind that increases in power usage are exponential so anything is possible.

What proc are you using?

[CMetaphor]

Originally Posted by sswilson

We'd need some stock results to be able to get an idea of the delta. Keep in mind that increases in power usage are exponential so anything is possible.

What proc are you using?

Ill throw it back to stock a bit later and do the same thing again. I'm running 1090T.

[CMetaphor]

Stock @ 3.2Ghz (Stock Volts)

Idle = 140w
CPU Load = 215w