Google Pixel 6 Pro-slow?

I was reading an article in the current c’t magazine. They did a head-to-head between the top Android handsets and the iPhone 13 Pro Max in the latest issue.

(Available in German, behind the heise+ paywall: https://www.heise.de/tests/Acht-High-End-Smartphones-im-Test-6307517.html )

They compared the “highend” smartphones – i.e. high end processors, not just high end prices, the cheapest was the 400€ Realme GT 5G, which still has the high end Qualcomm chip. They also included 3 phones which had their own custom designed CPUs, the iPhone, with its Apple A15 Bionic, the Samsung Galaxy S21 Ultra, with its Samsung Exynos 2100 and the Pixel 6 Pro, with the Google Tensor SoC.

There was no real surprise that the A15 ran rings around the Qualcomm and Samsung chips. (Geekbench 5, it was over 60% faster than the other phones). The only real weakness was in 3D gaming, where most of the competition beat it for playing time (only 11.8 hours, compared to 12-15.8 hours for the others, with the Galaxy winning that round.

What was a shock was just how badly the Pixel did. It came bottom in every performance test, by a big margin, and it only beat the Sony Xperia Pro-I on YouTube playback time (13.9 hours as opposed to 9.3, but the rest were all over 15 hours, with the Apple and Samsung pushing the 20 hour mark). It also reflected Paul’s experience, with charging, coming bottom of the pack by a margin of over 30 minutes!

On Geekbench 5 single core, the Pixel managed 928 points, with all the other Androids around the 1100 point mark (the iPhone was the leader at 4717).

The multicore test, the Pixel managed 2505, the other Androids all between 3400 and 3700 points (the iPhone walked away, again, with 4717).

3DMark Wild Life, the Pixel waddled across the line with 4311 points, the Android rivals with 5700-5800 points (the iPhone, again, with 9737).

If we ignore the Apple results, it is a totally different processor running a different OS, so there are too many factors to make a 100% accurate comparison and is not why I was surprised by the results, the real surprise is just how poorly the Google flagship performs, compared to its rivals using the same platform.

It came bottom in every test, usually by a very wide margin (all the other Android phones were within rounding errors of each other on performance). This seems to indicate that the Google Tensor SoC really isn’t ready for primetime at the moment. It is a first generation product from Google, but even so, if it compares so poorly to the SoCs it is trying to replace (Qualcomm Snapdragon and Samsung Exynos), should Google have held it back another generation, or do they really need the real-world feedback on its daily performance?

Would it have made more sense to test it internally, with employees, until they had a competitive product? Will it hurt the Pixel’s reputation? (Does it still have much of a reputation outside of die-hard Google fans any more, anyway?)

The phones in the test were: Apple iPhone 13 Pro max, Asus Zenfone 8, Google Pixel 6 Pro, OnePlus 9 Pro, Realme GT 5G, Samsung Galaxy S21 Ultra, Sony Xperia Pro-I and Xiaomi 11T Pro.

Conversation 14 comments

  • peterc

    Premium Member
    07 January, 2022 - 3:56 pm

    <p>To be honest i’m not at all surprised , some friends with the pixel 6 are reporting some shocking problems/performance/usage issues. Quite what google has been doing in testing is just bizarre. Saying that anyone whose updated their Oneplus 9 pro to Oxygen OS 12 is experiencing worse!! </p><p><br></p><p>How did the Xiaomi 11T pro and Asus Zenfone 8 turn out? Im likely to ditch my OP 9pro and pick up another android device to go with my iphone 12 and I was looking at those as well as the Motorola Edge 20 or maybe the pro version.</p>

    • wright_is

      Premium Member
      08 January, 2022 - 5:45 am

      <p>Due to its size, the Asus runs very hot, but they didn’t note any throttling. The Xiaomi was the fastest charger, with a 120W power supply, 50% in around 8 minutes, fully charged in under 20. It has 2 2500mAh batteries, which get charged in parallel. </p><p><br></p><p>The camera is the weak point with both of them. Asus is fine I’m daylight and well lit rooms, but poor in the dark. The Xiaomi is a little worse in daylight, but very bad in the dark. </p><p><br></p><p>the Xiaomi also only has IP52 rating, which isn’t full dust proof and only water splashes. </p><p><br></p><p>If photography is important, then it needs to be the Samsung, Sony, Google or Apple devices. Otherwise they are all pretty close with everything else. </p>

  • Daishi

    Premium Member
    07 January, 2022 - 4:20 pm

    <p>My god Google are an awful company.</p><p><br></p><p>It blows my mind that people still try to use how bad Apple/Microsoft are as a reason for using Android/ChromeOS.</p>

  • jlmerrill

    07 January, 2022 - 5:11 pm

    <p>I’m glad I cancelled my Pixel 6 order. My iPhone 13pro and Pixel 5A do Justine.</p>

    • pecosbob04

      10 January, 2022 - 10:37 am

      <p>How does Justine feel about this? ;)</p>

  • atimms

    08 January, 2022 - 5:04 am

    <p><br></p>

  • atimms

    08 January, 2022 - 5:09 am

    <p>Just an observation but there’s obviously something up with the Geekbench single/multi-core iPhone results – how can they be the same?</p><p><br></p><p>Suspect the single core score (since so high) is actually multi-core. Either a testing or Geekbench bug. Just going on your text, didn’t read the article.</p>

    • wright_is

      Premium Member
      08 January, 2022 - 5:31 am

      <p>Sorry, that was a mistake on my part. The single vote for the iPhone was 1744.</p>

      • wright_is

        Premium Member
        08 January, 2022 - 5:32 am

        <p>Damned autocorrect, the single core score was 1744</p>

  • ghostrider

    08 January, 2022 - 9:22 am

    <p>On paper, the Pixel 6 isn’t the fastest, but I don’t think Google set out to lead the benchmarks. It’s certainly fast enough (heck, my Pixel 4a is fast enough for everything I want). The differentiator is the Tensor chip, which offloads lots of AI work from the CPU. Other benchmarks show the Pixel 6 on par with the S21 Ultra in some tests, and falling behind a bit in others. Benchmarks are a PITA sometimes because they don’t tell the whole story, and people somehow always believe fastest is best. If your phone is fast enough for you, then that’s fine. Is an application opening 0.5s fast really going to make that much difference?</p>

  • paul_nelson

    09 January, 2022 - 1:14 am

    <p>I moved from the Pixel 5 to the Pixel 6 Pro, I get none of the stuttering I used to get on the Pixel 5 and Minecraft plays a lot smoother than it used.</p><p><br></p><p>But, the Tensor processor isn’t about being at the top of benchmarks (which have never been a good indication of real world performance), it is about Machine Learning and AI.</p><p><br></p><p>As for charging, I don’t understand why people complain so much about it. It charges quickly up to 80% and then slows right down to protect the lifespan of the battery.</p>

    • Paul Thurrott

      Premium Member
      09 January, 2022 - 8:36 am

      People complain about the charging because it’s too slow. 🙂

      The first 50 percent is fast-ish, but it takes two hours to fully charge this handset. There’s nothing this slow in the flagship category.

  • nappin

    Premium Member
    09 January, 2022 - 12:25 pm

    <p>I am more than happy with a 2 hour charging time and a battery that has a longer life – I intend to keep this mobile phone a long time and do not have the incredible luxury of getting the latest and greatest every year. It is fast, with a great screen, great camera, etc, and I, like many others, have had none of the issues raised on these pages. What puzzles me is why there is such a difference in experiences and such visceral reactions to something like the charging time and benchmarks.</p><p><br></p><p>On the subject of benchmarks – I gave up looking at benchmarks for PCs many years ago and never took any notice of them for mobile phones or tablets. The overall user experience is more important, and to me, that makes benchmarks pretty much irrelevant. Benchmarks have their uses, but too often the focus is on "<em>it has to be the fastest</em>" or "<em>I want it now!!!</em>", rather than taking a more holistic approach around the product lifecycle and user experience. </p>

  • sanskaram

    24 January, 2022 - 6:23 am

    <p>The camera is the weak point with both of them. Asus is fine I’m daylight and well lit rooms, but poor in the dark. The Xiaomi is a little worse in daylight, but very bad in the dark.</p><p><br></p>

Windows Intelligence In Your Inbox

Sign up for our new free newsletter to get three time-saving tips each Friday

"*" indicates required fields

This field is for validation purposes and should be left unchanged.

Thurrott © 2024 Thurrott LLC