2 Comments

Summary:

The iPad 2 is a beast under the hood, thanks to the new A5 system-on-a-chip designed in-house by Apple. But what does that mean in terms of the hard numbers of actual performance measurement, and how does it stack up to other iOS devices?

a5-ipad-feature

Looking at the iOS 4.3 update for existing iOS devices is a useful comparison to determine if performance trade-offs are worth the added features that come with a software update for existing hardware.  Now that the iPad 2 has been released, comparing iOS 4.3 running on each of the iOS devices shows what a difference refreshed hardware can make.  It wasn’t surprising to see the new iPad 2 outperform the original iPad.  What was surprising (at least to me) was that the iPhone 4 ended up ranking last in every category.

iOS Hardware Tested on 4.3

In every test, the original iPad outperformed the iPhone 4 on iOS 4.3. That was not the result I was expecting, especially given how just much the iPad outperformed the iPhone 4 in iOS 4.3.  When I originally got my iPad, I hardly touched my iPhone 3G. And when I got my iPhone 4, I started using my iPad less and less. I was convinced my iPhone 4 outperformed my iPad in nearly every regard, but it looks like my perception was incorrect.

Don’t Take Just My Word For It

I was curious if the results I was seeing with my very small sample size of one iPad 2, one original iPad and two iPhone 4s was representative of the greater population of iOS devices.  So I turned my attention to Primate Labs Geekbench and started checking against their online benchmark results.  I expanded my quest to include all iOS devices currently supported by iOS version 4.3 including the iPhone 3GS, iPhone 4, iPod touch 3rd Gen., iPod touch 4th Gen., iPad, and iPad 2.  I took the average of the benchmark results for the latest twenty-five tests submitted for each device, and here’s what I found:

The results were consistent with the results from my sample size of personally owned iOS devices.  And again, the performance gains on the iPad 2 are stunning.  They are in fact on par (in terms of percentage improvement over previous models) with the performance gains of the recently released MacBook Pro updates.

By comparison though, as impressive as the iPad 2 results are, according to the Mac Geekbench Results, with a score of 740, the iPad 2 is still only as powerful as a 15-inch PowerBook G4 1.33 GHz and just a little more powerful than the original Mac Mini G4 released in 2005.  But considering how much I can get done on each of these iOS devices, including what I can still do with my old iPhone 3G, I really don’t think raw performance alone is where the focus needs to be anymore.

You’re subscribed! If you like, you can update your settings

  1. I had a similar impression with the original iPad and iPhone 4. I agree, benchmarks do not always tell the whole story.

    Benchmarks and user experience are not always synonymous. Benchmarks are designed to test the raw processing power from one or more perspectives.

    As you conveyed the original iPad outperforms the iPhone in benchmark tests. From a user perspective, the iPad usually provides a much richer user experience. Depending on the application, the additional overhead of the richer user experience may be perceived as a reduction in performance.

    Take the Mail app as an example, in the list view the iPad displays twice as many messages in the pop over list view as the iphone displays. It may take longer to load those that are being initially displayed, but the iPad may be doing twice the work to support the additional content.

    Just a thought …

  2. iPhone 4 is equipped with 512MB RAM, vs original iPad with 256MB. That’s probably enough to make the iPhone 4 noticeably faster in practical use, even if the iPad has a faster CPU/GPU. If the benchmarks don’t take advantage of the iPhone 4’s extra RAM (and I doubt they would), they will not reflect that advantage in the scores.

Comments have been disabled for this post