iPhone 5 vs Galaxy S3 benchmarking – which one is better?

Update

But what about benchmarking? Pre-release iPhone 5 benchmark tests revealed the iPhone 5 is by far one of the best performing iPhones out there, and that it can match and even surpass some of the hottest Android handsets in town, Galaxy S3 included.

Once the iPhone 5 was actually launched, several websites put the iPhone 5 through various tests to see what it’s capable of. Naturally, one of the Android devices that was immediately compared with the iPhone 5, when it comes to performance, was the Galaxy S3, the best-selling Android handset of the year, although other devices such as the HTC One X or the newly launched LG Optimus G popped up for comparison in some of the tests.

The results? Like it or not, the iPhone 5 is clearly one of the best phones of the year, even if it packs only a dual-core processor. But is it better than the Galaxy S3? As you expect, there’s plenty of disagreement on the matter. In what follows, we’ll take a look at various reports from AnandTech and PCMag, two publications that have independently tested the device. We’re also going to look at a CNET report mentioning those tests, although it’s worth saying that the latter has not actually performed the tests.

AnandTech tests

AnandTech put the iPhone 5 through a variety of tests including Geekbench, BrowserMark, SunSpider, Google’s V8 and Octane benchmarks and GLBenchmark 2.5. In every test (see screenshots attached in the gallery below) the iPhone 5 did a lot better than any of its predecessors.

Before showing the results, several explanations are offered on the A6 processor and its performance in tests:

As always, our performance analysis starts out on the CPU. Although we originally thought the A6 ran its two CPU cores at 1GHz, it looks like max clocks range between 800MHz and 1.2GHz depending on load. Geekbench reports clock speed at launch, which varied depending on CPU load. With an app download process in the background I got Geekbench to report a 1.2GHz clock speed, and with everything quiet in the background the A6 reported 800MHz after being queried. This isn’t anything new as dynamic voltage/frequency adjustment is in all smartphones, but we do now have a better idea of the range.

The other thing I noticed is that without a network active I’m able to get another ~10% performance boost over the standard results while on a network. Take the BrowserMark results below for example, the first two runs are without the iPhone 5 being active on AT&T’s network while the latter two are after I’d migrated my account over. The same was true for SunSpider performance, I saw numbers in the low 810ms range before I registered the device with AT&T.

The publication notes that the A6 chip inside the new iPhone is able to deliver twice the performance of the A5 found in the iPhone 4S. As for GPU performance, the iPhone 5 almost matches the iPad 3.

Unfortunately for Galaxy S3 fans that would like to hear the Galaxy S3 does better than the iPhone 5, the tests performed showed that in all cases the Galaxy S3 is mentioned, the iPhone 5 does better than the Android handset. It’s worth pointing out that the article was not a comparison with Android devices but rather testing against the previous-generation iPhone, which was also updated to iOS 6.

However, an other Android handsets surprised in tests, coming out ahead of the iPhone 5 in a few GLBenchmark 2.5 tests, but also ahead of the Galaxy S3.

[nggallery id=23]

PCMag tests

Unlike AnandTech, which compared the iPhone 5 mostly to the iPhone 4S throughout its article, PCMag went ahead and declared the smartphone “the fastest smartphone in the land:”

We crown the iPhone 5 speed king after pitting it against every other iPhone we’ve tested, along with today’s top Android phones.

The iPhone 5 looks to be the fastest smartphone we’ve ever tested at PCMag.com.

The publication used five tests to measure the performance of the new iPhone 5: SunSpider, BrowserMark, GUIMark 3 Bitmap, Geekbench and GLBencmark 2.5 see the gallery below for actual results.

PCMag concludes:

A phone’s hardware performance can’t be taken in isolation, but it’s definitely a piece of the puzzle. Based on these benchmarks, the iPhone 5 lives up to the promise of being twice as fast as the iPhone 4S. It’s also, for now, the fastest handheld computer sold in the US.

[nggallery id=24]

Controversy

A CNET iPhone 5 benchmark report stirred some controversy on Google Plus where xda-member Adam Outler called out the publication on Google+ for misrepresenting iPhone 5 tests:

CNET posted a bogus report on benchmarks. I had problems believing that a dual core from apple would be faster than every Android device so I decided to go to the source. Turns out the Galaxy S3 and the Nexus7 both score higher benchmarks than the iPhone 5.

You can see here for yourself. IPhone scores 1590, Galaxy S3 scores 1766. These are the benchmarks mentioned in the article below.
Android benchmarks: http://browser.primatelabs.com/android-benchmarks
iPhone benchmarks: http://browser.primatelabs.com/ios-benchmarks

He is right to question the tests in the article, but only because that CNET story hit the print, so to speak, on September 16, six days ahead of the iPhone 5 launch. However, the article titled “iPhone 5 benchmarks surface: Performance doubles” does not say the iPhone 5 is better than the Galaxy S3, or any other Android device, in tests, but it looks only at a then-unconfirmed Geekbench result, and compares the iPhone 5 with its predecessor:

The iPhone 5’s A6 processor appears to be roughly twice as fast as any chip in an existing iOS product, if results posted by Geekbench prove to be accurate.

The results show a score of 1,601, beating the dual-core A5 and A5X processors in the iPhone 4s and third-generation iPad (Retina), respectively. […]

If these iPhone 5 benchmarks are legitimate, they would match Apple’s claims. “With the new A6 chip, just about everything you do on iPhone 5 is noticeably faster — up to twice as fast compared with the A5 chip,” Apple states on its iPhone 5 features page.

CNET’s story was (coincidentally or not) updated on the same day Outler posted his comments on Google Plus with this line:

For reference, the Samsung Galaxy S III with a Samsung Exynos 4412 quad-core chip rated at 1.4GHz landed at the top of Android benchmarks. [Links to http://browser.primatelabs.com/android-benchmarks]

But CNET did not say the Galaxy S3 did better than the iPhone 5, just as it didn’t say the iPhone 5 did better than the Galaxy S3 either before the update.

Also the Geekbench browser Outler mentions and CNET links to (see above), do not offer such a huge performance variation. The iPhone 5 is at 1588 right now in that test compared to 1590 (Outler’s Google+ post), 1601 (CNET’s iPhone 5 pre-release test) and 1640 (AnandTech’s test).

As for the Galaxy S3, here are the current scores for the different Galaxy S3 versions:

  • 1753 – Samsung Exynos 4412 1400 MHz (4 cores)
  • 1674 – Samsung Exynos 4412 1400 MHz (4 cores)
  • 1583 – Samsung Exynos 4412 1400 MHz (4 cores, LTE)
  • 1429 – Qualcomm Snapdragon S3 MSM8960 1500 MHz (2 cores) – North America
  • 1395 – Qualcomm Snapdragon S3 MSM8960 1500 MHz (2 cores) – T-Mobile
  • 1388 – Qualcomm Snapdragon S3 MSM8960 1500 MHz (2 cores) – Sprint
  • 1370 – Qualcomm Snapdragon S3 MSM8960 1500 MHz (2 cores) – Verizon

So the quad-core Galaxy S3 seems to be better than the iPhone 5, or on par with it in Geekbench Browser tests, depending on what version we’re talking about. As for the dual-core Galaxy S3, it’s still trailing behind the iPhone 5. We’ll certainly keep an eye on those tests and see how the Galaxy S3 performs once it receives the Jelly Bean update.

In the same test, the Asus Nexus 7 currently has a score of 1526.

Conclusion

After we’ve seen the first iPhone 5 benchmarks, we can conclude that the new iPhone is going to be a tough handset to beat.

Does this mean the Galaxy S3 is a worse choice than the iPhone 5 and therefore should not be purchased? Absolutely not.

Does this mean you have to look at benchmarks before buying any device? Not unless you’re specifically interested in such tests – that means “absolutely not” for average users who will most likely not observe any significant changes performance-wise during their day-to-day smartphone-related activities.

Will more iPhone 5-related benchmark comparisons be made in the near future? Absolutely yes, just wait for Jelly Bean to come out for the Galaxy S3, not to mention the launch of the Galaxy Note 2 and the LG Optimus G.

Finally, we’ll also tell you that we haven’t performed any of these tests ourselves, and that we’re looking at what other people are saying about these devices, and their performance as shown by their tests, or existing tests performed by others.

Comments

ncG1vNJzZmivp6x7orrDq6ainJGqwam70aKrsmaTpLpwtc%2BhpqedXWp6t7%2BMoJilmaiuerR%2FjJucp5uYoq6zt8innmavmJ6wqXnOp5xmoaNir6bA056pZmlhbH95gI4%3D