System Performance

While subjective judgments of performance may be possible to make when the delta is significant, when the gap gets increasingly close within the range of perceivable performance differences it becomes important to rely on more precise and accurate methods of measuring the overall performance of the mobile device. For the most part, when we’re discussing system performance the single biggest factor is often the SoC, which makes sense given that an SoC contains the CPU, GPU, video encode and decode blocks, memory bus, and DSPs. There are other aspects of the device that determine the overall perception of performance and things that can have a meaningful effect on performance, but the SoC is often the gating factor.

In order to test this we run mobile devices through our standard suite of benchmarks. In the case of the Galaxy Note5 and Galaxy S6 edge+, there shouldn’t be too many surprises given the commonality in components with the Galaxy S6.

Kraken 1.1 (Chrome/Safari/IE)

Google Octane v2  (Chrome/Safari/IE)

WebXPRT 2013 (Chrome/Safari/IE)

WebXPRT 2015 (Chrome/Safari/IE)

Basemark OS II 2.0 - Overall

Basemark OS II 2.0 - System

Basemark OS II 2.0 - Memory

Basemark OS II 2.0 - Graphics

Basemark OS II 2.0 - Web

PCMark - Work Performance Overall

PCMark - Web Browsing

PCMark - Video Playback

PCMark - Writing

PCMark - Photo Editing

If you guessed that performance in these benchmarks would be similar to the Galaxy S6, you'd be right. Given the shared SoC and general commonality in components performance remains as high as it is with the Galaxy S6. In some cases we see improvements, likely a combination of changes to Chrome and changes to areas like the frequency governor to respond faster to changes in load. It's probably fair to say that the Exynos 7420 will continue to be the best SoC for Android mobile devices in 2015, although it's likely that we'll see significantly increased competition for 2016.

Display System Performance Cont'd and NAND Performance
Comments Locked


View All Comments

  • ciderrules - Saturday, October 3, 2015 - link

    Who said anything about overclocking?

    Geekbench aggregates thousands of scores to get their numbers. I think I'll accept their results over those posted online. Especially by someone like Kuzi, who seems to hate Apple so much that he feels the need to lie online to make a point.

    BTW, I also refer to Geekbench themselves for iPhone scores, before someone claims bias.
  • Kuzi - Monday, October 5, 2015 - link

    Lie about what? You can find Geekbench multicore numbers for Galaxy devices on many review sites, an example is below:
  • Kuzi - Monday, October 5, 2015 - link

    More Geekbench numbers here:
  • Kuzi - Saturday, October 3, 2015 - link

    I just tested Galaxy S6 and got 1505 single and 5264 multi on Geekbench.

    Running Android 5.1.1 here.
  • Kuzi - Saturday, October 3, 2015 - link

    GSM Arena also getting same numbers as me:
  • Tech_guy - Friday, October 2, 2015 - link

    GFXBench is a graphics test. Geekbench is a cpu test so that's why the scores are direct, they're feeding student things. Where did you get 5600 Geekbench multi-core score from? That's an inflated number vs Geekbench's own charts.
  • mercucu1111 - Friday, October 2, 2015 - link

    I mean CPU Perf of Exynos is much higher but GPU Perf of Apple Ax chip is much higher.
    So We can't say which is better
  • ciderrules - Saturday, October 3, 2015 - link

    Not according to Primate Labs. You know, the people who actually CREATED Geekbench.
  • flashbacck - Friday, October 2, 2015 - link

    What ever happened to the benchmark cheating that was happening with some android phones? Is that no longer happening?
  • KoolAidMan1 - Saturday, October 3, 2015 - link

    I assume its not happening, otherwise their results wouldn't be there in the first place.

    Maybe the slipped through but I doubt it. Anandtech is incredibly thorough.

Log in

Don't have an account? Sign up now