Intel vs. AMD Performance-Per-Watt On Ubuntu 14.04 Linux

[phoronix]

Phoronix: Intel vs. AMD Performance-Per-Watt On Ubuntu 14.04 Linux

To complement the many Intel vs. AMD CPU/APU Linux benchmarks published earlier this week as part of our AMD A10-7850K "Kaveri" APU coverage, here's some results mostly examining the performance-per-Watt and overall system power consumption of the many different Intel and AMD processors running Ubuntu Linux.

http://www.phoronix.com/vr.php?view=19685

[nightmarex]

Yikes! I know there is a 65W version of the chip (7800k) that's 100mhz less. I would like to see if there's any difference there.

As it stands that's really surprising....

[devius]

These results are all wrong again.

Let's take the case of the Timed Kernel Compilation with the i3-4130 and the A10-7850. They both take about the same time to finish the test, but the i3 uses 40W less power on average to do so! That's about 34% less used by the i3 to do the same job in the same amount of time, which means the i3 is more power efficient than the A10. However on the bogus "performance-per-watt" chart the A10 appears with a much better score.

The C-Ray test is also clearly wrong. The i5-3470 and the i7-3770K have a very small difference in performance, but it's also accompanied by a similar difference in power usage, where the i7 uses a bit more power, so their performance-per-watt is very similar (as would be expected from two very similar CPUs using the same architecture), however on the obviously wrong "performance-per-watt" chart the i7 appears to be a lot more power efficient than the i5...

TL;DR

Everybody ignore this article because all the performance-per-watt numbers are wrong.

[UraniumDeer]

It can't be that much of a surprise that intel is more efficient than AMD.
Wasn't it roughly a 30% drop in wattage, or a 30% increase in performance at the same wattage, that intel reached, when they switched to the tri-gate design?

[zanny]

It can't be that much of a surprise that intel is more efficient than AMD.
Wasn't it roughly a 30% drop in wattage, or a 30% increase in performance at the same wattage, that intel reached, when they switched to the tri-gate design?

They went from 95w -> 77w on the top end, so it was a 19% drop, but the performance went up a bit too, so that sounds like an ok ballpark.

But that was more node shrink from 32 -> 22 than using trigates. Look at how Intel shrunk 45nm -> 32, going from 120w to 95. Same effect.

[curaga]

Michael, you put the same graph twice on the last page.

[GT220]

http://www.anandtech.com/show/7318/intel-demos-14nm-broadwell-up-to-30-lower-power-than-haswell

Performance per watt gap is gonna widen even more once Broadwell hits retail.

[brosis]

http://www.anandtech.com/show/7318/i...r-than-haswell

Performance per watt gap is gonna widen even more once Broadwell hits retail.

Native compiled applications please, not some generalized BS.

[verde]

It seems that AMD tried to pass Intel on Performance per Watt to convince OEMs to choose their products. Clever move..

[Daktyl198]

As it stands that's really surprising....

Is it really? We all know AMD is a power hog, it's always been that way and it will be that way for a while.

Does that stop me from buying AMD? No. I don't game much or do many other power-intensive tasks, so it stays relatively low for me for the cost.