CPU Performance

For our cross-platform CPU performance tests we turn to the usual collection of Javascript and HTML5 based browser tests. Most of our comparison targets here are smartphones with two exceptions: Intel's Bay Trail FFRD and Qualcomm's MSM8974 Snapdragon 800 MDP/T. Both of those platforms are test tablets, leveraging higher TDP silicon in a tablet form factor. The gap between the TDP of Apple's A7 and those two SoCs isn't huge, but there is a gap. I only include those platforms as a reference point. As you're about to see, the work that Apple has put into the A7 makes the iPhone 5s performance competitive with both. In many cases the A7 delivers better performance than one or both of them. A truly competitive A7 here also gives an early indication of the baseline to expect from the next-generation iPad.

We start with SunSpider's latest iteration, measuring the performance of the browser's js engine as well as the underlying hardware. It's possible to get good performance gains by exploiting advantages in both hardware and software here. As of late SunSpider has turned into a bit of a serious optimization target for all browser and hardware vendors, but it can be a good measure of an improving memory subsystem assuming the software doesn't get in the way of the hardware.

SunSpider Javascript Benchmark 1.0 - Stock Browser

Bay Trail's performance crown lasted all of a week, and even less than that if you count when we actually ran this benchmark.  The dual-core A7 is now the fastest SoC we've tested under SunSpider, even outpacing Qualcomm's Snapdragon 800 and ARM's Cortex A15. Apple doesn't quite hit the 2x increase in CPU performance here, but it's very close at a 75% perf increase compared to the iPhone 5. Update: Intel responded with a Bay Trail run under IE11, which comes in at 329.6 ms.

Next up is Kraken, a heavier js benchmark designed to stress more forward looking algorithms. Once again we run the risk of the benchmark becoming an optimization target, but in the case of Kraken I haven't seen too much attention paid to it. I hope it continues to fly under the radar as I've liked it as a benchmark thus far.

Mozilla Kraken Benchmark - 1.1

The A7 falls second only to Intel's Atom Z3770. Although I haven't yet published these results, the 5s performs very similarly to an Atom Z3740 - a more modestly clocked Bay Trail SKU from Intel. Given the relatively low CPU frequency I'm not at all surprised that the A7 can't compete with the fastest Bay Trail but instead is better matched for a middle of the road SKU. Either way, A7's performance here is downright amazing. Once again there's a performance advantage over Snapdragon 800 and Cortex A15, both running at much higher peak frequencies (and likely higher power levels too, although that's speculation until we can tear down an S800 platform and a 5s to compare).

Compared to the iPhone 5, the 5s shows up at over 2.3x the speed of last year's flagship.

Next up is Google's Octane benchmark, yet another js test but this time really used as a design target for Google's own V8 js engine. Devices that can run Chrome tend to do the best here, potentially putting the 5s at a disadvantage.

Google Octane Benchmark v1

Bay Trail takes the lead here once again, but again I expect the Z3740 to be a closer match for the A7 in the 5s at least (it remains to be seen how high the iPad 5 version of Cyclone will be clocked). The performance advantage over the iPhone 5 is a staggering 92%, and obviously there are big gains over all of the competing ARM based CPU architectures. Apple is benefitting slightly from Mobile Safari being a 64-bit binary, however I don't know if it's actually getting any benefit other than access to increased register space.

Our final browser test is arguably the most interesting. Rather than focusing on js code snippets, Browsermark 2.0 attempts to be a more holistic browser benchmark. The result is much less peaky performance and a better view at the sort of moderate gains you'd see in actual usage.

Browsermark 2.0

There's a fair amount of clustering around 2500 with very little differentiation between a lot of the devices. The unique standouts are the Snapdragon 800 based G2 from LG, and of course the iPhone 5s. Here we see the most modest example of the A7's performance superiority at roughly 25% better than the iPhone 5. Not to understate the performance of the iPhone 5s, but depending on workload you'll see a wide range of performance improvements.

The Move to 64-bit iPhone Generational Performance & iPhone 5s vs. Bay Trail


View All Comments

  • dylan522p - Tuesday, September 17, 2013 - link

    Adreno has a larger thermal headroom. If Apple moves to 5" they would be able to scale performance much much higher. Reply
  • blanarahul - Tuesday, September 17, 2013 - link

    The key word is "if". I would be really happy if they do move to 5 inch.

    BTW, these chips = Snapdragon 800.
  • tuxRoller - Wednesday, September 18, 2013 - link

    Yes, though the battery performance seems quite good for snapdragon 800. Reply
  • robbertbobbertson - Tuesday, September 17, 2013 - link

    So going off these theoretical numbers, the new iPhone 5S GPU is 4.36% as powerful as the one in the Playstation 4, and thats considered a weak GPU from the enthusiasts perspective. How is this is a "desktop" class chip. People thinking mobile will overtake everything are dreaming. Reply
  • dylan522p - Tuesday, September 17, 2013 - link

    You do realize that a Core Duo from years back is more than enough for most people on GPU and CPU, not the PS4 or some mid or high end chip now. Desktop class was simply marketing referring to 64-bit though. Reply
  • ScienceNOW - Tuesday, September 17, 2013 - link

    Are you kidding me? We are almost there. 4,36% is less than 5 doublings from 100% (PS4 GPU performance). In 5 years mobile GPU will be 40% MORE powerful than PS4. 8-9 years, and it'll equal todays' GTX 780 Reply
  • lowlymarine - Tuesday, September 17, 2013 - link

    Obligatory XKCD: http://xkcd.com/605/

    You're assuming that mobile GPU performance will continue to double each year indefinitely, which is patently absurd. There was a time each year's new desktop GPU doubled performance, too, but you reach a point where the laws of physics make that impractical.
  • dylan522p - Tuesday, September 17, 2013 - link

    True, but mobile GPU has been doubling-trippling every year for the past like 6 years. It shows no indication of slowing down either. or dropping below doubling. No doubt though, the mobile industry will be above the PS4 performance before it's update. Reply
  • A5 - Tuesday, September 17, 2013 - link

    A lot of that has been fueled by catching up on process tech. That party is almost over - 20/22nm class parts are still a year+ away, and ~14nm parts for anyone that isn't Intel are even farther out. Reply
  • melgross - Tuesday, September 17, 2013 - link

    Well, if Intel's mobile line is now at 22nm, and Apple's is at 28nm, that a problem for Intel. With the A7 proving to be about equal to what Intel is producing, going by Anand's tests here, then Intel's tech isn't all that great.

    Indeed, Intel has always depended upon its better process fabrication, and being a generation ahead for its superior performance. It's not just chip design. So if Apple can catch up in performance being a half generation behind in node, then Apple's designs are superior to Intel's. and then, Intel had better watch out.

Log in

Don't have an account? Sign up now