OnePlus Admits to Throttling Apps to Save Battery Life

Posted on July 8, 2021 by Paul Thurrott in Android with 9 Comments

OnePlus was found to be throttling apps to save battery life on its OnePlus 9 series handsets, triggering a new controversy for the firm.

“Our top priority is always delivering a great user experience with our products, based in part on acting quickly on important user feedback,” a OnePlus statement reads. “As a result of feedback, our R&D team has been working over the past few months to optimize the performance [of the OnePlus 9 and OnePlus 9 Pro] when using many of the most popular apps, including Chrome, by matching the app’s processor requirements with the most appropriate power. This has helped to provide a smooth experience while reducing power consumption. While this may impact the devices’ performance in some benchmarking apps, our focus as always is to do what we can to improve the performance of the device for our users.”

This makes sense to me: One of the primary selling points of OnePlus smartphones is their highly optimized Android derivative, which OnePlus calls OxygenOS. And this change did occur in the wake of feedback: OnePlus says it didn’t initially ship the OnePlus 9 series with this particular optimization, it instead made the change after customers asked it to “improve the devices’ battery life and heat management.”

The problem for OnePlus, of course, is that it’s found itself in trouble in the past for allegedly cooking benchmark tests. So when Anandtech reported on the new OnePlus optimizations, Geekbench promptly delisted the new OnePlus handsets from its Android benchmark chart. How dare OnePlus optimize its phones.

“It’s disappointing to see OnePlus handsets making performance decisions based on application identifiers rather than application behavior,” Geekbench tweeted. “We view this as a form of benchmark manipulation. We’ve delisted the OnePlus 9 and OnePlus 9 Pro from our Android Benchmark chart. We will also test the other OnePlus handsets in our performance lab to see if these handsets also manipulate performance in the same way.  If they do, we will delist them from the Android Benchmark chart [too].”

What this means is that OnePlus optimized the 9 series handsets by slowing performance when specific apps—among them Chrome, Twitter, Zoom, Facebook, Instagram, and YouTube—are used rather than when the phone is heating up or maxing out the processor. Arguably, the latter approach would be more sophisticated and perhaps more in line with what the company’s customers were asking for.

It’s also worth pointing out that the OnePlus 9 series contains the most powerful components available to smartphones today, and that even when throttled to protect battery life, these apps all run extremely well. It’s not like customers noticed performance degradation—as had happened before with iPhone when Apple was found throttling those devices to trigger more hardware upgrades—and that’s what led to the Anandtech tests.

In the end, this highlights why benchmarks are so pointless. No one ever complained that Chrome, for example, ran poorly on a OnePlus 9 series handset. But Chrome reported “abnormal” benchmark scores for a handset with a Snapdragon 888 processor. And I find it odd that Geekbench, a company that exists solely to provide artificial and meaningless benchmarks, would be so taken aback by scores that are artificially low; OnePlus wasn’t trying to game some benchmark here, it was trying to give its users a better overall experience. This is literally the reason why customers buy OnePlus handsets in the first place.

Tagged with ,

Join the discussion!


Don't have a login but want to join the conversation? Become a Thurrott Premium or Basic User to participate

Comments (9)

9 responses to “OnePlus Admits to Throttling Apps to Save Battery Life”

  1. mikegalos

    Arguably OnePlus was trying to game battery life benchmarks instead of the usual deceptive practice of gaming speed benchmarks.

  2. lezmaka

    If you read the linked AnandTech article, the "optimizations" were present when the phone was first released and was found in the launch-day reviews. The iPhone optimizations were also to give users a better overall experience. Having your phone randomly shut down because an old battery couldn't provide extra power quickly enough is a bad experience.

  3. peterc

    I'm a week into using my OP9 pro. I'm getting almost 2 day battery life, and no over heating issues at all. I'd say the recent system updates that have been issued fix the battery life and overheating issues. As for speed of use the device doesn't appear to be running slow at all, quite the opposite!

  4. mikeharris123

    I have had the OP 9 pro since release day and I am happy that they are optimising the phone for use. It is never slow and almost never gets hot unless it is in the windscreen of my car on a stand (yes not smart from me).

    I would expect all companies that care about the customer experience to do this, and yet when they do the tech press jumps on them. We live in a strange world

  5. wright_is

    Windows does this constantly, my Galaxy S21 is always suggesting apps to sleep or throttle.

    Something like Twitter or YouTube doesn’t generally need much processing power, more it will be idling much of the time, just the Video decoder will be working. You can easily throttle the processor to save battery life.

    • wright_is

      Ah, I just read The Register story, there is more detail.

      It does seem that they are being naughty. I'd misunderstood Paul's article, that benchmarks were excluded from the throttling techniques and normal software could be throttled as required - that is what the OnePlus statement implies. Reading deeper, they were profiling the apps and saying that Chrome never needs full power, so we'll always run it throttled, regardless of what it is doing.

      That obviously is wrong and I take back my previous statement.

      • peterc

        well everyone will have an opinion on this, but there is one point which is kinda front and centre in all this for me which is .... when users "interact" with their phone performance is boosted, when they dont its restricted. I dont really see how benchmarks can "emulate" this. Out of curiosity I ran the browser bench tests and Brave and Duck Duck go performed infinitely better than woeful Chrome and Edge. When I use all these browsers in actual day to to day handset usage there "appears" no discernible difference between them, none whatsoever.

        The cynic in me says something else is driving this story.

  6. sykeward

    I think you could make the case that OnePlus was directly trying to give its users a better experience if the OnePlus 9 Pro offered similar baseline performance to other SnapDragon 888-based devices. It doesn't. Anandtech found that the OnePlus "suffers from notably worse power characteristics and worse power efficiency [than the Galaxy S21 Ultra]...In our web-browsing battery life test, even with this performance crippling mechanism in place, with both devices at 120Hz under the same test conditions, the OnePlus 9 Pro achieves 11.75 hours or runtime, vs the S21 Ultra's 13.98 hours, the latter which runs at the SoC's full performance potential."

    About general performance: "I hadn’t immediately noticed it myself...However, having it side-by-side to a Samsung Galaxy S21 Ultra, or a Xiaomi Mi11 (Ultra) and paying attention, I do very much notice that the OnePlus 9 Pro is less responsive. The problem with claims such as 'less responsive' is that we cannot quantify it properly...Our pedigree here at AnandTech is all about being able to justify those claims with objective measurements."

    And the problem that GeekBench has is not that the scores are artificially low. It's that OnePlus took pains to ensure that benchmark apps were exempted from these "optimizations" and ran at the hardware's full capabilities, making the devices look better than they actually are in everyday use.

    You could compare this in some ways to what Volkswagen did with the whole dieselgate scandal; their cars were certainly capable of running very cleanly, just not without destroying vital engine components if they did so all the time. So VW decided to make its vehicles perform well for benchmarks and default to substandard performance during everyday use. Like benchmarks, emissions tests don't necessarily reflect real world performance, and like the OnePlus Pro 9, nobody complained about their diesel VWs and Audis either (including you and your wife) until they found out what was actually going on. Obviously the situation and real-world effects on people are different by orders of magnitude, but the similarities are pretty striking, IMO.

  7. markbyrn

    In short, OnePlus was throttling all the apps one was most likely to use except for benchmarks apps. Obviously, OnePlus doesn't think benchmark apps are pointless.

Leave a Reply