The first iPhone 5 reviews have lifted, confirming the leaked Geekbench data we saw in our earlier post. Apple's A6 appears to feature two custom ARM cores running at up to 1GHz. A new datapoint comes courtesy of our own Brian Klug who's currently visiting LG in Seoul, South Korea. He ran into Vincent Nguyen of Slashgear fame, who kindly let him run SunSpider 0.9.1 on Vincent's iPhone 5 review sample. The score? 914.7ms.

SunSpider is quickly outlasting its welcome as a smartphone benchmark, but it does do a great job of highlighting issues with the Cortex A9's memory interface. Intel originally hinted at issues in the A9's memory interface as being why Atom was able to so easily outperform other ARM based SoCs in SunSpider. As we surmised in our A6 Geekbench post, it looks like Apple specifically targeted improvements in the memory subsystem when designing the A6's CPU cores. The result is the fastest SunSpider test we've ever recorded on a smartphone - faster even than Intel's Atom Z2460.

This doesn't tell us much about the A6's architecture other than it's likely got a better cache/memory interface than ARM's Cortex A9. What we really need is for someone to port SPECint to iOS...

Comments Locked


View All Comments

  • doobydoo - Thursday, September 20, 2012 - link

    Ah - this explains my above analysis - thanks.
  • keylimesoda - Wednesday, September 19, 2012 - link

    Any sense of how this will compete with IE10 on WP8 on SnapDragon?
  • Aenean144 - Wednesday, September 19, 2012 - link

    Depends on the JavaScript JIT compiler in WP8 IE. The CPU core is basically the same as the US versions of the SGSIII, HTC One X, HTC One S, Evo 4G LTE, and the Optimus G, which are all on the chart in the article.

    I don't think MS will do a much better job than Apple or Google with the JIT compiler, so it's going to be in the 1500 to 1800 ms range, if not worse.

    You have to compare one-to-one with Sunspider. Same OS, same browser. So, you should be comparing results to the Lumia 900 or Titan or whatever. Even there, it's a problem as WP8 will only run on new phones.
  • john1506 - Sunday, November 11, 2012 - link

    I completely disagree, the 900 was released early this year and mango last year. You obviously don't understand how wp works. Wp8 was released alongside ios6 and the lumia 920/htc 8x/ Samsung ativ s are all current gen hardware released with it only 2 weeks after the iphone 5. I got a score well below the iphone 5, 910ms so yeah your guess was twice as long highly wrong
  • MUTINOUS - Wednesday, September 19, 2012 - link

    I had almost a year with a phone that still rocks and it was never built to be a "top of the line phone". You might be able to beat the specs but you can't beat the the GNex. LOL spec whores.

    Samsung Galaxy Nexus
    android version : 4.1.1
    Base Version : FF02/FG02
    Kernel Version : 3.0.38-04149-gldc555-dirty Trinity Kernel
    AOKP version : AOKP_toro_jb-build-1
    build number : JRO03H eng.
    Carrier : USA, Verizon LTE
    Hardware updates: Extended Battery 2100mAh + Door
    Release Date : November 17, 2011
  • humancyborg - Wednesday, September 19, 2012 - link

    Sure you can. I sold mine and pre-ordered the i5 last week.
  • yahyoh - Wednesday, September 19, 2012 - link

    one years old s2 with Jelly bean get 1400 ms @ stock cpu speed

    and get 1080ms @ 1.6

    so who care about Icrap , just wait for s3 to get Jelly bean and u will see who will win :D
  • red_dog007 - Wednesday, September 19, 2012 - link

    This benchmark seems to be a load of junk.
    My HTC Rezound scored just 300points higher than my Samsung Galaxy S3.
    Both using Chrome, both on ICS.

    Is there some kind of performance setting on the S3? I turned power saving off.

    I like peacekeeper better for a browser/cpu benchmark.
  • tom.jones - Wednesday, September 19, 2012 - link

    of all the tech blogs, i expected a more nuanced view from you Anand.

    anyone who knows the first thing about JS performance knows how dependent it is on (software) optimizations. just in the last few years, we got up to 50x improvements on various benchmarks *on the same hardware*.

    and anyone who knows the first thing about JIT optimizations know that they must be carefuly written for each architecture, even specific versions of each (x86 vs x64 vs ARM6 vs ARM7 vs..)

    so, TL;DR version is: comparing phones by JS benchmarks, and especially _across architectures_, tells you nothing about CPU performance..
  • dysonlu - Wednesday, September 19, 2012 - link


    Where did this stupid idea of using JS benchmarks as comparative of CPU performance come from???

Log in

Don't have an account? Sign up now