http://appleinsider.com/articles/13...ed-new-audio-video-features-in-apps-and-games
While the new ARMv8 definitely contributes, you can't simply write off the shift to 64 bit in the amount of data that the chip can now push through. These developers make it very clear that this chip is a huge advancement.
This article doesn't really support many of the claims that it makes. From what I can see, none of the examples it puts forward actually distinguishes performance gains from the new architecture and the "64-bit-ness" of it.
"Chandrasekher's opinion is particularly suspect because the A7 is already known to enable key features of iPhone 5s, including its advanced camera features (powered by the A7's Image Signal Processor, using an architecture similar to dedicated point-and-shoot cameras) and Touch ID (which relies on what Apple calls the A7's Secure Enclave Processor). Both are integrated into the A7."
Ok ... assuming that that is actually true, all this talks about is how all of this is relying on the enhanced capabilities of the A7, not its ability to perform 64-bit arithmetic operations natively...
On the iPhone 5s, the new 64-bit architecture of the A7 provides immediate benefits to developers thanks to its "modern instruction set," known as ARMv8...
Once again ... ARMv8 architecture.
Then here's the part I'm most intrigued with:
There is a marked increase in performance observed in moving from 32-bit to 64-bit benchmarks on the same hardware, in addition to the baseline improvement of the A7 over the A6 seen in 32-bit benchmarks.
Then I looked for these benchmarks in the article and I couldn't find anything!
The only "benchmarks" provided compare A6 (and other 32-bit ARM variants) to A7. Not exactly the comparison we were looking for.
Once again, they could be right, they just haven't actually demonstrated it!
Karim Morsy of Germany's Algoriddim noted that "optimizing djay 2 for the 64-bit A7 chip has allowed us to bring desktop-class power to our iPhone app.
Morsy added that "djay's audio processing and analysis is up to 2x faster, which not only makes the whole UI and animations run smoother but also allowed us to introduce new features and effects that weren't possible before.
Essentially: "We can do more stuff on A7 than we could on A6". Once again, what part about that is exclusive to 64 bit? What kind of performance gains are we talking about that can be attributed to its ability to perform 64-bit arithmetic operations natively?
"With the unmatched power of the iPhone 5s and its A7 chip, we can now combine fullscreen rendering effects, tons of polygons, and advanced gameplay processing in one smooth package.
I'm pretty sure this is more due to the beefed up GPU (based on IP from Imagination Technologies) on the A7, rather than the 64-bit CPU, no?
According to Anand, "There’s more graphics horsepower under the hood of the iPhone 5s than there is in the iPad 4. While I don’t doubt the iPad 5 will once again widen that gap, keep in mind that the iPhone 5s has less than 1/4 the number of pixels as the iPad 4."
Read more about it here:
http://www.anandtech.com/show/7335/the-iphone-5s-review/7
All in all, nothing really in the article points specifically to the move to 64-bit as a huge performance gain. Rather, it speaks of general performance gains moving from A6 to A7.
Anand:
"
In the ARM world, the move to 64-bit is motivated primarily by the same factor: a desire for more memory. Remember that ARM and its partners have high hopes of eating into Intel’s high margin server business, and you really can’t play there without 64-bit support. ARM has already announced its first two 64-bit architectures: the Cortex A57 and Cortex A53. The ISA itself is referred to as ARMv8, a logical successor to the present day 32-bit ARMv7.
The motivation for Apple to go 64-bit isn’t necessarily one of needing more address space immediately. A look at Apple’s historical scaling of memory capacity tells us everything we need to know:
The soonest Apple would need 64-bit from a memory addressability standpoint in an iOS device would be 2015, and the latest would be 2016. Moving to 64-bit now preempts Apple’s hardware needs by 2 full years.
The more I think about it, the more the timing actually makes a lot of sense. The latest Xcode beta and LLVM compiler are both ARMv8 aware. Presumably all apps built starting with the official iOS 7 release and going forward could be built 64-bit aware.
By the time 2015/2016 rolls around and Apple starts bumping into 32-bit addressability concerns, not only will it have navigated the OS transition but a huge number of apps will already be built for 64-bit. Apple tends to do well with these sorts of transitions, so starting early like this isn’t unusual. The rest of the ARM ecosystem is expected to begin moving to ARMv8 next year."
See:
http://www.anandtech.com/show/7335/the-iphone-5s-review/4
edit:
I stand corrected. I did some more digging here, and actually came up with some quantitative charts from Anand's review that detail performance gains due exclusively to the 64 bit transition. He compares 64 bit and 32 bit processes on the same hardware. I'd link the charts, but they're not in image format, so I'm not sure how to post them. But the conclusion is here:
The DGEMM operations aren't vectorized under ARMv7, but they are under ARMv8 thanks to DP SIMD support so you get huge speedups there from the recompile. The SFFT workload benefits handsomely from the increased register space, significantly reducing the number of loads and stores (there's something like a 30% reduction in instructions for the A64 codepath compared to the A32 codepath here). The conclusion? There are definitely reasons outside of needing more memory to go 64-bit.
