Skip to main content

Google Pixel 6 battery life tested — we have bad news

Google Pixel 6
(Image credit: Tom's Guide)

If you’re wondering if the Pixel 6 and Pixel 6 Pro are worth buying, the short answer is “yes.” As you’ll see in our Pixel 6 review and Pixel 6 Pro review, both phones deliver great cameras, responsive displays and plenty of power through the new Tensor chip.

The only major caveat for the Pixel 6 series is battery life. Both phones turned in below-average endurance on the Tom’s Guide battery test, which involves continuous web surfing over 5G at 150 nits of screen brightness.

For context, the phones on our best phone battery life list last all last 11 hours or more, and we consider 10 hours or longer above average. 

The Pixel 6’s 4,614 mAh battery lasted only 8 hours and 1 minute on our test with the screen set to static or 60Hz mode. On adaptive mode, which sales the refresh rate from 60Hz up to 90Hz, we saw as high as 8:13. Not good. 

ModelBattery life (hours:mins)
Google Pixel 68:13
Google Pixel 6 Pro7:53
iPhone 13 Pro Max12:16
iPhone 13 Pro11:42
iPhone 13 10:33
iPhone 13 mini8:41
Samsung Galaxy S219:53
Samsung Galaxy S21 Plus9:50
Samsung Galaxy S21 Ultra11:25
OnePlus 9 Pro12:48

Surprisingly, the Pixel 6 Pro turned in an even worse 7 hours and 53 minutes on our battery test, despite having a larger 5,000 mAh battery. In adaptive mode the Pixel 6 Pro lasted 7:46 or about the same time.

By comparison, the iPhone 13 lasted 10 hours and 33 minutes on our web surfing test and the iPhone 13 Pro Max hit 12:16. The Samsung Galaxy S21 endured for 9:53 and the Galaxy S21 Ultra endured for 11:25. Last but not least, the OnePlus 9 Pro lasted an excellent 12:48.

Interestingly, we tested the regular Pixel 6 a couple of times over 4G as well and it averaged 10 hours and 52 minutes. That's great, but we would like to see better results over 5G.

So what’s going on here? We have a couple of theories. For one, we tested the Google Pixel 6 and Pixel 6 Pro at 77% brightness for our battery test, as that’s what it took to get to our required 150 nits. Other phones tend to reach that nit level at around 50% brightness, but we wanted to keep the comparisons fair.

Second, the Google Pixel 6 and Pixel 6 Pro reportedly use an older Samsung modem found in the Galaxy S20 series. It's over a year old at this point. We suspect that this modem is not nearly as efficient as new Qualcomm modems in today’s flagships.

To be fair, no one is going to surf the web continuously until the battery dies, and our test is not designed to mirror real-world use. In fact, thanks to Google’s Tensor chip and tools like Extreme Energy Saver mode, you should be able to get through most of the day (if not more) on a charge.

But when it comes to tasks that utilize 5G on the go, the Pixel 6 and Pixel 6 Pro are definitely a a step behind the competition when it comes to battery life. 

Mark Spoonauer

Mark Spoonauer is the global editor in chief of Tom's Guide and has covered technology for nearly 20 years. In addition to overseeing the direction of Tom's Guide, Mark specializes in covering all things mobile, having reviewed dozens of smartphones and other gadgets. He has spoken at key industry events and appears regularly on TV to discuss the latest trends. Mark was previously editor in chief of Laptop Mag, and his work has appeared in Wired, Popular Science and Inc. Follow him on Twitter at @mspoonauer.

  • georGGGG
    There are plenty of reasons the 5G radio might run for long periods, like listening to music while on the go. Seems like this is a big problem, but many people will just turn off 5G as there just doesn't seem much point right now.
    Reply
  • Istolla
    "Other phones tend to reach that nit level at around 50% brightness, but we wanted to keep the comparisons fair."

    Raising the brightness level to 77% send innocent enough, but the iPhones have lower resolution screens. You ended up not making it fair in that the phones with higher resolutions are doing more work at a brightness level that is not their respective 50%.
    Reply
  • iamstuff
    Take this with a grain of salt. Considering phonearena.com lists completely different results for what appears to be a similar test.

    They say:
    iPhone 13 Pro max gets 18 hour 52 mins at 120hz web browsing.
    Samsung Galaxy S21 Ultra gets 14h 43 mins at 120hz web browsing.

    Tomsguide says:

    iPhone 13 Pro max gets 12 hour 16 mins at UNKNOWN hz web browsing.
    Samsung Galaxy S21 Ultra gets 11h 25 mins at UNKNOWN hz web browsing.

    This is a massive discrepancy across the board. So /shrug.
    Reply
  • varase
    The Google Tensor SoC (built by Samsung) contains two Cortex-X1 cores - and I believe the general consensus is to only include one of these cores in a SoC due to their power requirements.

    As a consequence, the Snapdragon 888 only contains a single Cortex-X1 in a three level pyramid of cores.

    As I recall, the Cortex-X1 at 5nm ran pretty much neck-and-neck with Apple's A13 high performance Lightning cores at 7nm (used in the iPhone 11), with Lightning being overall a bit faster and Cortex-X1's floating point taking the lead.

    https://www.anandtech.com/show/15875/apple-lays-out-plans-to-transition-macs-from-x86-to-apple-socs
    I believe when Google announced they were "building their own chips" that what really meant is Google designed an NPU and seconded Samsung as their silicon proxy and told Samsung to put two Cortex-X1s in their SoC - despite any warnings Samsung may have given - in an attempt to beat Apple at their own game.

    All these companies saying they're "building their own chips" despite having no expertise in silicon design are just cousin' for a bruisin' - like Microsoft using their silicon proxy Qualcomm, and ironically Qualcomm attempting the same through their silicon proxy the recently acquired Nuvia (composed of ex-Apple silicon engineers).

    Qualcomm's expertise is picking standard ARM cores and putting them in a SoC with a few cache line changes - I don't recall if they've ever designed their own cores. That's why they acquired Nuvia, hoping to drain some of Apple's brain.

    And so ... the Pixel 6 may simply be suffering from Google's hubris in "making their own" Tensor SoC.
    Reply
  • iamstuff
    varase said:
    The Google Tensor SoC (built by Samsung) contains two Cortex-X1 cores - and I believe the general consensus is to only include one of these cores in a SoC due to their power requirements.

    As a consequence, the Snapdragon 888 only contains a single Cortex-X1 in a three level pyramid of cores.

    As I recall, the Cortex-X1 at 5nm ran pretty much neck-and-neck with Apple's A13 high performance Lightning cores at 7nm (used in the iPhone 11), with Lightning being overall a bit faster and Cortex-X1's floating point taking the lead.

    https://www.anandtech.com/show/15875/apple-lays-out-plans-to-transition-macs-from-x86-to-apple-socs
    I believe when Google announced they were "building their own chips" that what really meant is Google designed an NPU and seconded Samsung as their silicon proxy and told Samsung to put two Cortex-X1s in their SoC - despite any warnings Samsung may have given - in an attempt to beat Apple at their own game.

    All these companies saying they're "building their own chips" despite having no expertise in silicon design are just cousin' for a bruisin' - like Microsoft using their silicon proxy Qualcomm, and ironically Qualcomm attempting the same through their silicon proxy the recently acquired Nuvia (composed of ex-Apple silicon engineers).

    Qualcomm's expertise is picking standard ARM cores and putting them in a SoC with a few cache line changes - I don't recall if they've ever designed their own cores. That's why they acquired Nuvia, hoping to drain some of Apple's brain.

    And so ... the Pixel 6 may simply be suffering from Google's hubris in "making their own" Tensor SoC.
    Interesting points. I'm not sure if Google said they "made" this, or if they "designed" it and are claiming it as their own (the latter is fair to say in my opinion).
    Reply