The ARM vs x86 Wars Have Begun: In-Depth Power Analysis of Atom, Krait & Cortex A15
by Anand Lal Shimpi on January 4, 2013 7:32 AM EST- Posted in
- Tablets
- Intel
- Samsung
- Arm
- Cortex A15
- Smartphones
- Mobile
- SoCs
Krait: WebXPRT 2013 Community Preview 1
I also included Principled Technologies' new HTML5/js web test suite WebXPRT in our power analysis. Intel and Qualcomm remain quite close in these tests. I didn't run the Qualcomm tests at the same time as the Intel tests the charts graphs aren't perfectly aligned, as a result it looks like Intel took longer to complete the test when in reality the opposite is true. Once again at the platform level, the W510 beats the XPS 10, but at the CPU level Krait manages to do better than Atom. Looking at GPU power consumption alone, Intel/Imagination are once again more power efficient. As 3D performance doesn't matter much here, the Qualcomm/Adreno 225 3D performance advantage does nothing - it just costs more power.
Once again there's no contest when we include Tegra 3 in the comparison. Atom/Krait are in a different league. It'll be interesting to see how Tegra 4 will do here...
TouchXPRT 2013
As our first native client test, we turned to PT's TouchXPRT 2013. As there is no "run-all" functionality in the TouchXPRT benchmark, we had to present individual power curves for each benchmark. Unlike the previous tests where Qualcomm was consistently slower than Intel, many of the TouchXPRT tests show the two competitors performing quite similarly. This gives us a better idea of how these two fare when performance is equal. For the most part, Acer/Intel seem to win at the platform and GPU levels, while Qualcomm takes the win at the CPU level. Once again, it's not abundantly clear to me how much of Qualcomm's CPU core power advantage is due to the fact that we're not taking into account power consumption of the L2 cache.
Facebook
Twitter
RSS
140 Comments
View All Comments
Kidster3001 - Friday, January 4, 2013 - link
Samsung uses everyone's chips in their phones. Samsung, Qualcomm, TI... everyone's. I would not be surprised to see a Samsung phone with Atom in it eventually.jeffkibuule - Friday, January 4, 2013 - link
They've never used other non-Samsung SoCs by choice, especially in their high end phones. They only used Qualcomm MSM8960 in the US GS III because Qualcomm's separate baseband MDM9615 wasn't ready. As soon as it was, we saw the Galaxy Note II use Exynos again. Nvidia and TI chips have been used in the low end from Samsung, but that's not profitable to anyone.Intel needs a major design win from a tier one OEM willing to put its chip inside their flagship phone, and with most phone OEMs actually choosing to start designing their own ARM SoCs (including even LG and Huawei), that task is getting a lot harder than you might think.
felixyang - Saturday, January 5, 2013 - link
some versions of Samsung's GS2 use TI's OMAP.iwod - Saturday, January 5, 2013 - link
Exactly like what is said above. If they have a choice they would rather use everything they produce themselves. Simply Because Wasted Fabs Space is expensive.Icehawk - Friday, January 4, 2013 - link
I find these articles very interesting - however I'd really like to see an aggregate score/total for power usage, IOW what is the area under the curve? As discussed being quicker to complete at higher power can be more efficient - however when looking at a graph it is very hard to see what the total area is. Giving a total wattage used during the test (ie, area under curve) would give a much easier metric to read and it is the important #, not what the voltage maxes or minimums at but the overall usage over time/process IMO.extide - Friday, January 4, 2013 - link
There are indeed several graphs that display total power used in joules, which is the area under the curve of the watts graphs. Maybe you missed them ?jwcalla - Friday, January 4, 2013 - link
That's what the bar charts are showing.GeorgeH - Friday, January 4, 2013 - link
It's already there. A Watt is a Joule/Second, so the area under the power/time graphs is measured in Watts * Seconds = Joules.Veteranv2 - Friday, January 4, 2013 - link
Another Intel PR Article, it is getting really sad on this website.Now since you are still using Win8 which is garbage for ARM. Please us the correct software platform for ARM chips. I'd love to see those power measurements then.
Anandtech did it again. Pick the most favorable software platform for Intel, give the least favorable to ARM.
Way to go! Again....
Intel PR at its best...
Veteranv2 - Friday, January 4, 2013 - link
Oh wait its even better!They used totally different screens with almost 4 times the pixels on the nexus 10 and then says it requires more power to do benchmarks. Hahaha, this review gave me a good laugh. Even worse then the previous ones.
This might explain the lack of product overviews at the start.