NVIDIA Will Perform Its Own Benchmarks On A5X CPU After Apple Claims It Outperforms Tegra 3 Chip

NVIDIA Will Perform Its Own Benchmarks On A5X CPU After Apple Claims It Outperforms Tegra 3 Chip

Image courtesy of GDGT

During Apple’s iPad keynote yesterday, Phil Schiller, its senior vice president of worldwide marketing, claimed the tablet’s new A5X processor offers 4X the graphics performance of NVIDIA’s quad-core Tegra 3 chip.

NVIDIA says that while it was “certainly flattering” to be called out by the Cupertino company, it will be performing its own benchmarks on the new iPad to see if Apple’s claims are really accurate.

The Tegra 3 processor has been making its way into Android-powered tablets in recent months, providing devices like the Asus Transformer Prime will impressive processing power and stunning 3D graphics. But according to Apple, it can’t match the graphical performance of the new A5X chip.

NVIDIA (obviously) disagrees. Ken Brown, a spokesman for the company, told ZDNet:

It was “certainly flattering” to be called out by Apple, but the performance claims are sketchy without more data. “We don’t have the benchmark information,” said Brown. “We have to understand what the application was that was used. Was it one or a variety of applications? What drivers were used? There are so many issues to get into with benchmark.”

It’s unlikely Apple is going to provide its benchmarks. So NVIDIA says it will perform its own when Apple’s new tablet goes on sale on March 16.

“At some point it will become more clear what the performance really is,” said Brown. “For now, Apple has a really generic statement.”

You can be sure that NVIDIA’s benchmarks will provide different results, and that the company will fight for its own product. But do they really matter? I mean, will they really have any influence on your decision to purchase either the new iPad or a Tegra 3-powered tablet?

Related
  • sn0wball

    A5X all the way

  • xiaobaizz

    so are u trying to say it’s okay to cheat under this kind of circumstances? standard apple fanboy #whenwillapplefanboybelessbiased

  • Michael Von Verrenkamp

    I never was much of a fan of Apples method of simplifying the entire graphics performance down to a single number but to be fair that is just because I’m not the target audience for how they are marketing these devices. Truth be told what ever Nvidia says and benchmarks, the number that Apple has given out will be what most people hear.

  • techgeek01

    I don’t remember where I read this, but they were questioning this already.  Apparently when they  ran their tests and such, they got different results.  They tested the Transformer Prime and end up finding that the GPU was on Par with the A5, if not completely blew it out of the water.

    Meaning? That the Graphic performance of the X5 may not be better or be only 2 times better.

    And then there is the question of how are they comparing….?  Probably, the most fair and best comparison would be is to put iOS or Android on BOTH a Tegra 3 and A5X  CPU and then test the graphics.  That way  you would get the best results.

    And dealing with this, apparently the Tegra 3 CPU isn’t as good as S4 from Qualcomm or the TI cpu that is shortly coming out.

  • Honyant

    Probably at about the same time as Apple haters.

  • Takeshi Hartono

    im guessing you meant when tegra 3 is using 4 cores it will out run A5 easily.  Too bad not many apps utilise having 4 cores. Is the OS even threaded to run quad cores? in most cases anyway, the other two cores are just sitting there doing nothing, or the ninja core is being used and A5 will be running as fast as it needs to be.

    In terms of graphics though, here’s the GLBenchmark

    http://www.fudzilla.com/home/i… 

    Yes I am an Apple fanboy

  • Luis Faria

    I think this is actually true, in the glbenchmark 2.1 – pro – offscreen 720p, the ipad 2 does indeed have twice the number of frames per second, compared to the tegra 3 gpu, so extrapolating the results, the 5X quad-core gpu should do four times the frames.

  • jpv41193

    Benchmark results don’t always mean better “real-world” results. All it takes is for the A5X to be 4x better at one thing and Apple can make this claim. I’m interested to see nVidia’s benchmarks though. Better yet, let’s have a third party do a wide variety of tests and get some real data we can use. Until then, I’ll consider Tegra3 and A5X to be equal. They’re both very impressive, to say the least.

  • JAG_P3

    Tegra 3 powered devices will have the same issues that the tegra 2 ones, simply Android hasn’t optimized its software to take full advantages of the hardware in the market, including nvidia’s chips. So, how many tegra 2 powered android devices have full use of the harware? LG P990 Optimus are example of that, just to show up a name.

  • GregsTechBlog

    Seeing as it would destroy Nvidia’s mobile business, I’m going to say that Nvidia’s is going to skew those results heavily.
    I’m waiting for 3rd party benchmarks

  • vivekr127

    http://www.anandtech.com/show/

    Here are the benchmark comparisions between iPad 2 with an A5 chip and the tegra 3 on the transformer prime. It definitely seems to be better. Going by this benchmark i’d assume that the A5X definitely is gonna score much higher on the benchmarks, but it might not be 4X better than the tegra 3.

  • vivekr127

    http://www.anandtech.com/show/

    Here are the benchmark comparisions between iPad 2 with an A5 chip and the tegra 3 on the transformer prime. It definitely seems to be better. Going by this benchmark i’d assume that the A5X definitely is gonna score much higher on the benchmarks, but it might not be 4X better than the tegra 3.

  • bmdonovan

    If/when there comes a day where the gaming experience on a tablet becomes spectacular (Haptic interface or whatever), and high end games become exclusive to a SOC because of the processing requirements, benchmarking might begin to matter to me. Until then, as long as the SOC can snappily run the tablet, and do it without gorging on the battery, I don’t think benchmarks really matter.

  • Rob Stevens

    Uh, the slide also showed that the A5 has twice the performance. So why do we have to wait? Does the current iPad outperform the Tegra 3? I thought that was already determined, so I think nVidia is just being defensive.

  • ClickMe

    TI OMAP5 will run circles around both.

    Who cares?!

  • Mohy

    Rightly said, i was wondering about the same…

  • Henry Zou

    That’s very true, but only when they are done “offscreen”.

  • Mike Rathjen

    No doubt Apple found some benchmark that is really favorable to Apple hardware. Also no doubt Nvidia will counter by finding some benchmark that is really favorable to Nvidia hardware.

    And while the giants argue, none of us consumers will care because specs based on a secret benchmark method are completely worthless in real life use.

About the author

Killian BellKillian Bell is a staff writer based in the U.K. He has an interest in all things tech and also covers Android over at CultofAndroid.com. You can follow him on Twitter via @killianbell.

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

Posted in News, Top stories | Tagged: , , , , , , , , , , , |