Android and iphone performance differences

[Draconian Guppy]

So after official numbers on the new iphone performance numbers and how iphone is "OMGZ0rs" doubles and triples android scores for geekbench and antutu. Are these number comparable in the sense that one platform handles differently than the other, gpu drivers, kernels and all those OS differences? Or in other words, how significant are these numbers: http://bit.ly/2w5ERJC Like people always rave on how iOS is more optimized to specific hardware and all that jazz ?

 

Say randombenchmark software Android score:6 iOS score:10 How is the 6 comparable to the 10 given apples closed ecosystem, custom designed processor. I think a good analogy would be mazdas rotary 1.3 engine is not equivalent to a normal piston engine at 1.3?

[Draconian Guppy]

found this researching the subject, this explains a lot for me as an average joe I guess, die sieze!

 

[adrynalyne]

iOS tends to have a smoother performance, but if you get stock Android, it too is quite smooth these days. OEMs tend to muck it up with their own skins and frameworks though.

 

Both are very capable platforms overall and more often than not, it boils down to user preference over performance.

[Draconian Guppy]

7 minutes ago, adrynalyne said:

iOS tends to have a smoother performance, but if you get stock Android, it too is quite smooth these days. OEMs tend to muck it up with their own skins and frameworks though.

 

Both are very capable platforms overall and more often than not, it boils down to user preference over performance.

oh Agree on user preference, but just the numbers are almost double in some occasions it really boggles me:

 

 

 

 

 

 

[adrynalyne]

Yeah but synthetic tests do that often. Can you really tell is what matters the most. 

[Draconian Guppy]

16 minutes ago, adrynalyne said:

Yeah but synthetic tests do that often. Can you really tell is what matters the most. 

According to a user on reddit:

Quote

 

 

Geekbench uses a lot of crypto ###### built into the iPhone chipset so the results are heavily skewed in their favor.

Edit: Linus Torvalds on Geekbench:

Geekbench is ######.

It actually seems to have gotten worse with version 3, which you should be aware of. On ARM64, that SHA1 performance is hardware-assisted. I don't know if SHA2 is too, but Aarch64 does apparently do SHA256 in the crypto unit, so it might be fully or partially so.

And on both ARM and x86, the AES numbers are similarly just about the crypto unit.

So basically a quarter to a third of the "integer" workloads are just utter BS. They are not comparable across architectures due to the crypto units, and even within one architecture the numbers just don't mean much of anything.

And quite frankly, it's not even just the crypto ones. Looking at the other GB3 "benchmarks", they are mainly small kernels: not really much different from dhrystone. I suspect most of them have a code footprint that basically fits in a L1I cache.