CPU Performance

While there’s a great deal of ground to cover on the tablet as a whole, one of the most interesting aspects of the Nexus 9 is the SoC. While we’ve tested Tegra K1 before, we were looking at the more traditional Cortex A15 variant. The Denver variant (Tegra13x) is mostly similar to Tegra K1-32 (Tegra12x), but instead the CPU cores are a radically different design. In order to get an idea for how this translates into real world we can look at a few of our standard benchmarks in this area, although Google Octane couldn’t complete a full run. This build of Android clearly has AArch64 active, which means that we should be able to directly compare the Nexus 9 to the iPad Air 2 for performance.

SunSpider 1.0.2 Benchmark  (Chrome/Safari/IE)

Kraken 1.1 (Chrome/Safari/IE)

WebXPRT (Chrome/Safari/IE)

BaseMark OS II - Overall

BaseMark OS II - System

BaseMark OS II - Memory

BaseMark OS II - Graphics

BaseMark OS II - Web

As one can see, at least at this stage in development the Nexus 9 can show some level of promise at times, but can be a bit disappointing in others. In SunSpider, Denver is generally even slower than Krait. However, in a benchmark like Kraken the Nexus 9 easily pulls ahead to take the top spot. In Basemark OS II the Nexus 9 does well overall but this seems to be due to its graphics performance/GPU performance and storage performance rather than CPU-bound tests like the system and web tests. It seems that when the code morphing systems works as expected, Denver can deliver significant amounts of performance. However, when such code morphing falls flat its true performance with a dual core, 2.3 GHz configuration is around that of a four Krait core CPU system at similar clock speeds. Once again, it's important to emphasize that this build is far from complete so performance should improve across the board with launch software. The fact that Tegra13x can approach A8X in CPU performance in some tests is definitely interesting to see.

Battery Life

While Denver's performance is a bit mixed, it's worth taking a look at battery life to see how Denver performs in these areas. As always, our battery life tests are all run with the display calibrated to 200 nits.

Web Browsing Battery Life (WiFi)

While an early build, it seems that the Nexus 9 is reasonably competitive in battery life but I'm not sure that these results are perfectly accurate. At any rate, efficiency at this stage seems to be par for the course, which should bode well for shipping software. This is a mostly display-bound test though, so we'll look at Basemark OS II to get a better idea for compute-bound battery life.

BaseMark OS II Battery Life

BaseMark OS II Battery Score

As one can see, while the battery life of the Nexus 9 ends up on the bottom for phablets and tablets, the overall performance during the test is quite high. We're working on a better comparison for the final review, but this should give a good idea of what to expect in general.

 

Introduction GPU Performance and Initial Conclusions
Comments Locked

146 Comments

View All Comments

  • eiriklf - Monday, November 3, 2014 - link

    That is, the optimizations seem to be architecture specific.
  • Krysto - Monday, November 3, 2014 - link

    What I'd like to know - is the tablet actually taking advantage of the 64-bit/Aarch64 mode? Or is it running in 32-bit mode still? Because I think it was still running in 32-bit mode in early Geekbench tests. I don't know why Google would release it like that, but it would be disappointing if they did, since that could be cutting its CPU performance by 10-20 percent.
  • Aenean144 - Monday, November 3, 2014 - link

    Is it really a guarantee that 64-bit performance will be better than 32-bit performance with a VLIW architecture?

    The performance is going to be highly dependent on how well the code morpher can pack instructions into a VLIW instruction. So, if the VLIW instruction is 256 bits wide, does that mean it can only pack half the 64-bit instructions?
  • Andrei Frumusanu - Monday, November 3, 2014 - link

    It's running on a 64 bit kernel.
  • kron123456789 - Monday, November 3, 2014 - link

    But apps(benchmarks included) aren't 64 bit.
  • Marcos Stein - Monday, November 3, 2014 - link

    64bit optimization will not save it.
    http://arstechnica.com/gadgets/2014/11/nexus-9-rev...
  • kron123456789 - Monday, November 3, 2014 - link

    1) It's not final firmware on Nexus 9.
    2) It's pre-alpha (or something) build of 64-bit Geekbench.
    We'll see when the final version of 64-bit Geekbench will be released.
  • AnandTechUser99 - Monday, November 3, 2014 - link

    I would wait until these benchmarks can test AArch64 for Android before jumping to a conclusion on performance.

    Any chances of seeing performance in GFXBench 4.0? Or is that not ready yet?

    All of these GPU benchmarks test OpenGL ES 2.x or 3.x performance, which misses the point of Tegra K1.
  • Marcos Stein - Monday, November 3, 2014 - link

    Gfxbench's results posted here were very generous.
    Nexus 9's Gfxbench onscreen results has only 2048x1440. The results will be worse when Gfxbench update the software to process 2048x1536 (196608 more pixels).
    The same happened with Iphone 6
  • Fidelator - Monday, November 3, 2014 - link

    Seems pretty competitive with the Air 2 all around, what about some display calibration details? Those would be greatly apprecited

    Now, the real catch will be next year, when we get quad core Denver + a Maxwell GPU on a 20nm process, they were able to match the A8X in pretty much everything but some CPU bound tests with a bigger process

    Still, after seeing these results the display calibration and speaker sound quality will determine if I get it or not

Log in

Don't have an account? Sign up now