Apple's A7 processor vs Bay Trail on Benchmarks

[snowComet]

Anandtech posted their review on the iPhone 5s. In this review, they bench-marked the A7 processor, run it against other phones/desktop cpus, and it showed some interesting results.

 

 

(*Update from Anandtech: Intel responded with a Bay Trail run under IE11, which comes in at 329.6 ms.)

 

 

You can read the full review here:

http://www.anandtech.com/show/7335/the-iphone-5s-review

[Stonep11]

Its impressive, but mobile power has been a huge focus lately so it really isn't all that surprising.

[beebskadoo]

it's all about software optimizations...

[Stonep11]

it's all about software optimizations...

Very true to a point, and I don't think we have reached that point yet, but its like consoles, they can get more out of those than I would have ever thought possible, but it does have a limit.

[beebskadoo]

Very true to a point, and I don't think we have reached that point yet, but its like consoles, they can get more out of those than I would have ever thought possible, but it does have a limit.

I was speaking of Apples optimizations. Wish it were true on the Android side too. It's just too fragmented...

[LAwLz]

Don't take browser based benchmarks too seriously. They are a horrible way of comparing CPU performance.

You could think of it this way, would you trust a comparison of the FX 8350 vs the i7-4770K if the benchmark was done with one CPU on an Ubuntu computer using Firefox, and the other CPU was running Windows Vista and IE8? That's basically what is happening here, and that's why I have always been against Brian's testing methods of mobile CPUs. He doesn't even like it himself, but think it is the best we got currently (which he is probably right about, but it's still crap).

[chriswills]

very impressive, but honestly i would rather see companies focusing on advancing battery technology rather than cpu performance

[swifty7]

according to those charts, the new 5s murders the s4....now!!! if only the 5s had a 5" 1080p screen

[SirPaul]

Don't take browser based benchmarks too seriously. They are a horrible way of comparing CPU performance.

You could think of it this way, would you trust a comparison of the FX 8350 vs the i7-4770K if the benchmark was done with one CPU on an Ubuntu computer using Firefox, and the other CPU was running Windows Vista and IE8? That's basically what is happening here, and that's why I have always been against Brian's testing methods of mobile CPUs. He doesn't even like it himself, but think it is the best we got currently (which he is probably right about, but it's still crap).

 

 

Cant they just code some program to do some math and based on :speed: they calculate the benchmark ?

[qwertywarrior]

ehh those are browser benchmark

[LAwLz]

Cant they just code some program to do some math and based on :speed: they calculate the benchmark ?

Well they could, but my guess is that it's really hard to do a fair cross platform benchmark which simulates real world performance.

[pit5000]

impressive, but it's not a big deal. the real question is how does it work in everyday operations?

[LAwLz]

impressive, but it's not a big deal. the real question is how does it work in everyday operations?

Well those are everyday operations.

 

By the way, the Intel Atom is not a desktop processor. It is a phone and tablet processor just like the Apple A7.