no, this is not the 1st of april ... x86 beats ARM at power efficency in Tablets

Windows, MS, you know what i meant.

MS has added support for ARM into the legacy windows. They made it work. That doesnt mean it is coded as a effeciently as a x86 version.
And I don't really know what rock you are living under, but we are at A15 at the moment.
If you meant they support ARM with their mobile OS? Yes they do. And it is terrible...

My point was, DrWho stated x86 was more power efficient then ARM, showing 1 screenshot with win8 while disregarding results from other OS. Which is total BS. Anyone with a brain knows this. If you want to state your hardware arch is better, then compare it with different software archs on top and similar specs of peripheric hardware.

Quote:

---

Originally Posted by BoredByLife

Windows, MS, you know what i meant.

MS has added support for ARM into the legacy windows. They made it work. That doesnt mean it is coded as a effeciently as a x86 version.
And I don't really know what rock you are living under, but we are at A15 at the moment.
If you meant they support ARM with their mobile OS? Yes they do. And it is terrible...

My point was, DrWho stated x86 was more power efficient then ARM, showing 1 screenshot with win8 while disregarding results from other OS. Which is total BS. Anyone with a brain knows this. If you want to state your hardware arch is better, then compare it with different software archs on top and similar specs of peripheric hardware.

---

Don't get defensive.

I just point out that the data set is lacking, and you you are making a conclusion where there is nothing to conclude upon. You claim clover trail "appears better" because the OS is inefficient on ARM, there is nothing you can provide in terms of data or facts to support that conclusion, you are projecting this statement to rationalize an observation that you either disagree with or do not like.

In short, data that is available is not supporting your thesis, Anand's single plot, as you say, simply shows that a clover trail tablet gets better battery life than an ARM tablet, on the surface (pardon the pun) I agree with you that that is not enough to proclaim 'more efficient' victory, however anyone with a brain can look at other data and see a trend or append the single plot observation. For example ...

-the ARM tablet has a battery of 31.5 WHr, the clover trail system has 30 WHr
-the ARM tablet has a 10.6" screen, the covert trail tablet has a 11.5" screen

Then there are a smattering of clover trail data from Razr i showing similar performance and much longer battery life. Other ARM optimized OSes are also showing clover trail meeting or exceeding ARM in efficiency.

It wasn't that long ago that many people said it wasn't possible for x86 to get down to sub 5 Watt profiles for tablets or phones yet here we are debating the fine points whether or not the OS is responsible for the observations. Dr Who's point was that the myth x86 could not be made to fit these power envelopes is obviously not true.

I am not totally convinced that this is the x86 ticket to mobile, but neither am I so closed minded that I think it is impossible. A15 will come and certainly shake it up, some devices are already showing that but we are not at A15 yet, only a few devices are showing up with that core, and again the data is lacking. Then again, 2013 will see the clover trail equivalent arrive on 22 nm with an OoO engine.

I find your assertion that MS simply added support to windows for ARM, could you explain that in more detail?

Edit:

Quote:

---

If you meant they support ARM with their mobile OS? Yes they do. And it is terrible...

---

Again, by what criteria....what is your measurable observable and what value do you affix to that observable to conclude it is 'terrible'? You have a strong affixation on asserting an opinion as a conclusion.

Do we know the stats of each screen in power draw? Bigger screen does not necessarily mean more power draw.

Quote:

---

Originally Posted by STEvil

Do we know the stats of each screen in power draw? Bigger screen does not necessarily mean more power draw.

---

Indeed, but I'd does not mean it consumes less either, and that is sorta the point.... People seem quick to dismiss the possibility that Intel has a more efficient solution for what ever reason, and the reasons do not have supporting facts or testable scenarios to say one way or the other. No matter, many conclude with conviction, the superiority of ARM. I would challenge that assertion as unsupportable at this stage.

Regardless, there is not enough data one way or the other to state a conclusion one way or the other; however the data that is available contradicts the asserted opinion that ARM is better (higher performing less power) than Clover Trail solutions and people seem to be offended by that ... A15 is still not prevalent to find much data at all, but I certainly expect it to out shine clover trail as I suspect valley view will out shine A15.... The battle is shaping up to be quite interesting.

Here are some desktop Exynos 5 benchmarks http://www.phoronix.com/scan.php?pag...rome_a15&num=1
Looks like it's about as fast as the D525 in most benchmarks. The battle between Clovertrail and 28nm QC Exynos 5(and other 28nm ARM SoCs) should be very interesting. Although i got to admit, ValleyView is probably looking quite scary for most ARM partners right now, due in part to the process advantage Intel has