The Claimed Benefits of GPU Turbo: Huawei’s Figures

Some of the most popular and widely quoted figures from the marketing slides during the release of GPU Turbo were the claims that the new feature would somehow achieve both up to 60% higher performance while also being able to reduce SoC power by up to 30%. Huawei (and Honor) are very very keen on these numbers, and show the results in the best possible light.

These figures were so widespread that they were again reiterated in last Thursday’s IFA global launch of the Honor Play. Over the last few days, I’ve seen plenty publications reporting these figures and attributing them to the benefits of GPU Turbo. However as of yet, no analysis has taken place.

Most slides we have seen look something like this, making it very easy to just put numbers into an article:

As AnandTech is very much a technical-oriented publication, from the get-go these numbers looked suspicious to us. We take pride and value in our own independent analysis, making sure that the numbers promoted are remotely achievable. After all, numbers like this often go against common-sense engineering advancements. And this is exactly where the marketing numbers fall apart. More on that later.

So What is GPU Turbo? We Finally Have Some Clarity

When we initially went looking for a detailed explanation of GPU Turbo, I (Andrei) first attempted to figure out by myself what GPU Turbo does. The lack of details, aside from projected performance improvements, is a very weak place to start. It also ended up being quite a nightmare of a task for one specific reason: despite years of allowing devices to be rooted, at around the same time as GPU Turbo was launched, Huawei and Honor stopped allowing bootloader unlocking. This prevented users from modifying the system’s firmware, and us included as we were unable to root the devices to help with benchmark profiling. Furthermore, all new firmware that ships with GPU Turbo is said to be vastly more locked down. It has not been made clear if the two items are directly related, though anyone who likes a good conspiracy might be inclined to think so.

As a result, in the context of GPU Turbo, what I had in mind was to actually profile the GPU via Arm’s Streamline tool, as this would give us exact information on the workloads that the GPU is processing. With these tools, we can deeply analyze what is going on under the hood. Either intentionally or unintentionally, the lockdown of the bootloader prevents us from doing this. Unfortunately because of the limitations, this was a dead-end in my testing.

I had started writing this article with no further detailed explanation, however we met with the EMUI software team at IFA, and we were able to finally get a more comprehensive explanation as to what GPU Turbo is. We discussed this technology with both the hardware and software teams, and had very different discussions with both.

With the hardware team – specifically with HiSilicon – they made it clear that this is purely a software technology. The mechanisms in GPU Turbo are aided by the controls they put in place, but the actual way that GPU Turbo works is all down to software. This is good to know, and also explains why Huawei is able to roll it out across all of its smartphone range. It also is not tied to the NPU, although having an NPU in the mix does help, apparently.

However, the public first hint of what it actually does was included in Friday’s Kirin 980 keynote presentation, referred to as “AI Loading Prediction Technology”.

What the slide tries to convey is that GPU Turbo allows the SoC to better follow the compute workload, adjusting the hardware performance states in order to better adapt to the workload. For example, when the CPU needs more power than the GPU, the power ratio availability can be adjusted to match these new requirements, improving both performance and efficiency.

We go into a bit more depth on the next page, where we finally have a good explanation of the mechanism. We have to thank Dr. Wang Chenglu, Huawei’s President of CBG (Smartphone) Software, for this information.

Huawei’s GPU Turbo: A Timeline The Detailed Explanation of GPU Turbo
POST A COMMENT

64 Comments

View All Comments

  • LiverpoolFC5903 - Wednesday, September 5, 2018 - link

    Very interesting to say the least. The improvements from this, although not as much as promised, are still tangible and will make a difference in supported games.

    Also, alarming to see the quality difference between an Adreno unit and a Mali unit, especially considering they are supposed to be close competitors. I have an S9 with the Mali g72mp18 unit and going by the results on PUBG, it performs much worse than its Adreno counterpart, both in render quality and framerate.

    Hisilicon and Samsung should consider using Powervr gpus again, given the clear inability of the Mali to keep up. I have noticed this in the real world as well, with my LG V20 with a Snapdragon 820 lasting MUCH longer than my S9 while running emulators (PSP and Neo Geo), despite being years old.
    Reply
  • Manch - Wednesday, September 5, 2018 - link

    Maybe its my screen but the Honor Play and the S9 pics make it look like the dude got no undies. LOL Reply
  • umano - Wednesday, September 5, 2018 - link

    Amazing article, thank you. Having a P20 pro ( I don't play games on phone ) that was particularly interesting and I really liked the "ethic" behind words, supporting both customers and the company, asking the latter to do the right thing. I think this is the way professional journalism has to be done.
    Chapeau
    Reply
  • AshokGupta - Wednesday, September 5, 2018 - link

    Hundreds of Huawei's competaters have tried round and round to prove GPU Turbo is a fake junk, and all of them failed. Now you take over their job. Good Luck, Man! Reply
  • GreenReaper - Wednesday, September 5, 2018 - link

    Fake? No. But the reality doesn't exactly match up to the marketing. Reply
  • s.yu - Friday, September 7, 2018 - link

    The reality doesn't match up to the marketing, AT ALL. Good as fake.
    Huawei in all practicality was trying to sell this off as a *universal* performance and efficiency gain of 60%, 30% respectively while in fact it only works on *a handful* of games for about *10%* each. When you're exaggerating your claims by 3x, 6x, it's lying, it's fake.
    Reply
  • AshokGupta - Saturday, September 8, 2018 - link

    If you read the Chinese media, what happens here is just repetition of what happened exactly right after the technology was launched in CHINA. Including this stupid guess of saying it only covers few games. Then approved by many independent tech media it's applicable for all. Your name indicates you are most probably from China. I suppose you should know it. Don't understand why you come here again giving the approved fake comments. Reply
  • s.yu - Monday, September 10, 2018 - link

    Because of the opaque operation of Chinese media. Obviously you're also from China, don't tell me you don't know about the fuss Huawei created buying ads on international sites and then buying fake journalism back in China.
    http://tech.ifeng.com/a/20180710/45057623_0.shtml
    This article was widely spread as legit news but the international content cited was intentionally twisted, it's highly misleading.
    When Huawei buys western ads at least the hosts declare bought articles, in China there's no way of telling real journalism from Huawei's smokescreen, so I put off reaching a conclusion until global availability of the technology.
    Now from Anandtech's analysis and *interview* it's obviously certain that the tech only works on a handful of games, Huawei even admitted that each profile is trained separately then pushed to devices.
    I know a Huawei troll when I see one, I'll be keeping an eye on you in the future.
    Reply
  • ET - Wednesday, September 5, 2018 - link

    What's with the annoying 'Buy the Right CPU' autoplaying video? Reply
  • psychobriggsy - Wednesday, September 5, 2018 - link

    It's really annoying, and on every article, on every page, and it doesn't remember if you pause it on one page then go to the next. Reply

Log in

Don't have an account? Sign up now