New iPad’s A5X Beats The Pants Off Tegra 3 Graphics [Video]

Apple’s not exactly the kind of company that boasts lightly. That’s not to say they don’t boast a lot — they’re probably the most bragging of all the companies in tech, and for damn good reason — but every boast is weighted against genuine success, not numbers fudging.

So when Apple debuted the new iPad a couple weeks ago and claimed that their tablet — powered by a dual-core CPU and quad-core graphics — outperformed the quad-core CPU and 12-core graphics of the NVIDIA Tegra 3 SoC, a lot of people arched their eyebrows. NVIDIA raised a stink, saying it couldn’t possibly be true. But we quietly suspected that Apple would be proven right.

So guess which is faster in independent benchmarks?

The boys over at Laptop Mag put the new iPad through its paces against the Tegra 3 powered ASUS Eee Pad Transformer Pride.

The results? Apple exaggerated a little that the graphics of the A5X were four-times better than the Tegra 3, but not by much.

When GLBenchmark ran the Geometric test (vertex weighted), which measures low-level shader performance, the new iPad processed 7,530,524 frames at a rate of 57 fps while the Tegra 3-based Transformer Prime processed just 3,523,926 at a rate of 27 fps. We saw the same trend on the program’s Fill test, which measures texturing speed, as the iPad processed 1.98 billion textels per second to the Prime’s 404.61 million.

So about 60% better on low level shader performance, but almost five times better in texturing speed. Considering the raw power and multiple core advantage of the Tegra 3, that’s quite the feat, and can likely all be chalked up to Apple’s incredible efforts towards optimization. By controlling both the hardware and the software running on it, they’re outperforming a far more powerful chiclet of silicon.

Of course, with raw CPU performance, Laptop Mag found that the Tegra 3 had the advantage over the A5X. That’s interesting, but graphics are the bottleneck these days — especially as we move towards Retina Displays. And NVIDIA, a graphics company, is hardly likely to brag that their Tegra 3 CPU can crunch more numbers than Apple’s.

Related
  • nukemhill

    So about 60% better on low level shader performance, but almost five times better in texturing speed.

    No.  To quote:

    the new iPad processed 7,530,524 frames at a rate of 57 fps while the Tegra 3-based Transformer Prime processed just 3,523,926 at a rate of 27 fps.

    That would roughly translate to the Tegra processing 1,750,000 frames at 54 fps.  Which means the A5X is over 4x faster.

    Simple arithmetic, gang.  Try it.  You’ll like it.

  • steffenjobbs

    I think that’s enough to prove that Apple didn’t lie about the benchmarks.  Not that it really should make any difference.  I don’t think Apple should have mentioned the Tegra 3 in their marketing.  Doing such a thing will only cause animosity and prove very little.  As long as users are satisfied with everyday performance, that should be enough.

  • warrengonline

    Did I miss the point of this video?  WHO CARES!?!?  Now my thing is can I play those games for 5+ hours on both devices on a full charge?  Shadowgun has been crashing on my iPad2…  Looking forward to the arrival of my iPad 3rd Gen 64Gb to play it FINALLY as my iPod touch 4th Gen it would not even load completely.  World of Warcraft Lite for iPad2 anyone?  $4 a month?  I’m in.

  • CaveMan5464

    Muhahahahha! U rock man. :)

  • alextheukrainian

    While interesting video to compare the actual games, it’s simply inaccurate when comparing slates. For example, Infinity Blade 2 has no issues rendering flags flapping in the wind so the fact that Shadowgun doesn’t have them in iPad version is a game limitation, not iPad’s. Similarly, Sky Gamblers: Air Superiority does a great job of rendering rain drops splashing onto camera, so again that’s game issue, not A5X issue.

  • nukemhill

    Jeebus.  And they haven’t corrected it yet.

    Way to live up to journalistic standards, boys.

    Oh, wait–

  • Anthony Graham

    The reason they haven’t corrected it is because your maths does not translate to this situation. The benchmark is ran over a period of time. At the end of that period of time the results of average fps and total frames is given. For your statement to be true it would require the Tegra to A) be able to run at 54 fps and B) to be given approximately one quarter of the time of the A5X. (Edited time derp up)

About the author

John BrownleeJohn Brownlee is a Contributing Editor. He has also written for Wired, Playboy, Boing Boing, Popular Mechanics, VentureBeat, and Gizmodo. He lives in Boston with his girlfriend and two parakeets. You can follow him here on Twitter.

(sorry, you need Javascript to see this e-mail address)| Read more posts by .

Posted in News | Tagged: , , , |