Google's Tensor inside of Pixel 6, Pixel 6 Pro: A Look into Performance & Efficiency
by Andrei Frumusanu on November 2, 2021 8:00 AM EST- Posted in
- Mobile
- Smartphones
- SoCs
- Pixel 6
- Pixel 6 Pro
- Google Tensor
Phone Efficiency & Battery Life
While not directly released to the Google Tensor, I also finished running the various battery tests for the Pixel 6 and Pixel 6 Pro, and there are some remarks to be made in regards to the power efficiency of the devices, and how the new SoC ends up in relation to the competition.
As a reminder, the Pixel 6 comes with a 4614mAh battery and a 6.4” 1080p 90Hz OLED screen, while the Pixel 6 Pro features a 5003mAh battery and a 6.71” 1440p 120Hz OLED display, with variable refresh rate from 10-120Hz.
Starting off with the 60Hz web browsing results, both Pixel phones end up extremely similar in their longevity, at 14 hours runtime. The regular Pixel 6 is hard to compare things to as we don’t have too many recent phones with 90Hz displays in our results set, however the Pixel 6 Pro should be a direct comparison point to the S21 Ultras, as both feature 5000mAh batteries and similar display characteristics. The P6Pro here ends up slightly ahead of the Exynos 2100 S21 Ultra, which might not be too surprising given that the Tensor chip does end up at somewhat lower CPU power levels, even if performance is lower. It’s still quite behind the Snapdragon 888 variant of the S21 Ultra – which is again quite representative of the SoC efficiency differences.
Running the phones at their respective max refresh rates, both devices see larger drops, however the Pixel 6 Pro especially sees a more substantial hit. This time around, the 6 Pro ends up significantly behind the Exynos 2100 S21 Ultra, which had only a minor drop in the 60 -> 120Hz results.
Shifting over to PCMark at 60Hz, we see that there’s a larger difference in favour of the Pixel 6, as the Pixel 6 Pro ends up behind it in longevity by almost two hours. The 6 Pro still ends up in line with the E2100 S21U, however that device showcases significantly higher performance numbers in the test, which acts both as a performance metric for device responsivity as well as a battery life test.
At 120Hz, the 6 Pro ends up worse than the E2100 S21U, and quite worse than the S888 S21U.
When I was investigating the phones, the 6 Pro’s power behaviour was quite weird to me, as I saw best-case baseline power figures of around 640mW, and sometimes this inexplicably would also end up at 774mW or even higher. What this reminded me of, was the power behaviour of the OnePlus 9 Pro, which also suffered from extremely high baseline power figures. Both the 6 Pro and the 9 Pro advertise themselves as having LPTO OLED panels, but both of them very clearly do not behave the same as what we’ve seen on the Note20Ultra or the S21Ultra phones. The 6 Pro also only goes up to up to 750 nits 100% APL peak brightness in auto-brightness mode under bright ambient light, which is significantly lower than the S21U’s 942 nits. I think what’s happening here is that the Pixel 6 Pro simply doesn’t have the most state-of-the-art display, and thus is quite less efficient as what we find on the competition. It does kind of make sense for the price-point of the phone, but also explains some of the battery behaviour.
Naturally, the Tensor SoC also just doesn’t appear to be as efficient. Particularly many UI workloads would be run on the A76 cores of the chip, which just outright have a 30% perf/W disadvantage. The phone ends up OK in terms of absolute battery life, however performance metrics are lower than other devices.
I think the regular Pixel 6 here is just a much better device as it doesn’t seem to have any particular issues in display efficiency, even if it’s just a 1080 90Hz panel. There are naturally experience compromises, but it’s also a $599 phone, so the value here is very good.
US readers who are used to Qualcomm phones might also encounter efficiency regressions when under cellular data – we abandoned doing testing here many years ago due to the impossible task to get consistent test environments.
Facebook
Twitter
RSS
108 Comments
View All Comments
Alistair - Tuesday, November 2, 2021 - link
It's very irritating how slow Android SOCs are. I'll just keep on waiting. Won't give up my existing Android phone until actual performance improvements arrive. Hopefully Samsung x AMD will make a difference next year.Speedfriend - Thursday, November 4, 2021 - link
Looking at the excellent battery life of the iPhone 13 (which I am currently waiting for as my work phone) does iPhone till kill suspend background tasks. When I used to day trade, my iPhone would stop prices updating in the background, very annoying when I would flick to the app to check prices and unwittingly see prices hours old.ksec - Tuesday, November 2, 2021 - link
Av1 hardware decoder having design problem again?Where have I heard of this before?
Peskarik - Tuesday, November 2, 2021 - link
preplanned obsolescencetuxRoller - Tuesday, November 2, 2021 - link
I wonder if Google is using the panfrost open source driver for Mali? That might account for some of the performance issues.TheinsanegamerN - Tuesday, November 2, 2021 - link
Seems to me based on thermals that the pixel 6/pro suffer from thermal throttling, and thus have power power budgets, then they should have given the internal hardware, leading to poor results.Makes me wonder what one of these chips could do in a better designed chassis.
name99 - Tuesday, November 2, 2021 - link
I'd like to ask a question that's not rooted in any particular company, whether it's x86, Google, or Apple, namely: how different *really* are all these AI acceleration tools, and what sort of timelines can we expect for what?Here are the kinda use cases I'm aware of:
For vision we have
- various photo improvement stuff (deblur, bokeh, night vision etc). Works at a level people consider OK, getting better every year.
Presumably the next step is similar improvement applied to video.
- recognition. Objects, OCR. I'd say the Apple stuff is "acceptable". The OCR is genuinely useful (eg search for "covid" will bring up a scan of my covid card without me ever having tagged it or whatever), and the object recognition gets better every year. Basics like "cat" or person recognition work well, the newest stuff (like recognizing plant species) seems to be accurate, but the current UI is idiotic and needs to be fixed (irrelevant for our purposes).
On the one hand, you can say Google has had this for years. On the other hand my practical experience with Google Lens and recognition is that the app has been through so many rounds of "it's on iOS, no it isn't; it's available in the browser, no it isn't" that I've lost all interest in trying to figure out where it now lives when I want that sort of functionality. So I've no idea whether it's better than Apple along any important dimensions.
For audio we have
- speech recognition, and speech synth. Both of these have been moved over the years from Apple servers to Apple HW, and honestly both are now remarkably good. The only time speech recognition serves me poorly is when there is a mic issue (like my watch is covered by something, or I'm using the mic associated with my car head unit, not the iPhone mic).
You only realize how impressive this is when you hear voice synth from older platforms, like the last time I used Tesla maybe 3 yrs ago the voice synth was noticeably more grating and "synthetic" than Apple. I assume Google is at essentially Apple level -- less HW and worse mics to throw at the problem, but probably better models.
- maybe there's some AI now powering Shazam? Regardless it always worked well, but gets better and faster every year.
For misc we have
- various pose/motion recognition stuff. Apple does this for recognizing types of exercises, or handwashing, and it works fine. I don't know if Google does anything similar. It does need a watch. Not clear how much further this can go. You can fantasize about weird gesture UIs, but I'm not sure the world cares.
- AI-powered keyboards. In the case of Apple this seems an utter disaster. They've been at it for years, it seems no better now with 100x the HW than it was five years ago, and I think everyone hates it. Not sure what's going on here.
Maybe it's just a bad UI for indicating that the "recognition" is tentative and may be revised as you go further?
Maybe the model is (not quite, but almost entirely) single-word based rather than grammar and semantic based?
Maybe the model simply does not learn, ever, from how I write?
Maybe the model is too much trained by the actual writing of cretins and illiterates, and tries to force my language down to that level?
Regardless, it's just terrible.
What's this like in Google world? no "AI"-powered keyboards?, or they exist and are hated? or they exist and work really well?
Finally we have language.
Translation seems to have crossed into "good enough" territory. I just compared Chinese->English for both Apple and Google and while both were good enough, neither was yet at fluent level. (Honestly I was impressed at the Apple quality which I rate as notably better than Google -- not what I expected!)
I've not yet had occasion to test Apple in translating images; when I tried this with Google, last time maybe 4 yrs ago, it worked but pretty terribly. The translation itself kept changing, like there was no intelligence being applied to use the "persistence" fact that the image was always of the same sign or item in a shop or whatever; and the presentation of the image, trying to overlay the original text and match font/size/style was so hit or miss as to be distracting.
Beyond translation we have semantic tasks (most obviously in the form of asking Siri/Google "knowledge" questions). I'm not interested in "which is a more useful assistant" type comparisons, rather which does a better job of faking semantic knowledge. Anecdotally Google is far ahead here, Alexa somewhat behind, and Apple even worse than Alexa; but I'm not sure those "rate the assistant" tests really get at what I am after. I'm more interested in the sorts of tests where you feed the AI a little story then ask it "common sense" questions, or related tasks like smart text summarization. At this level of language sophistication, everybody seems to be hopeless apart from huge experimental models.
So to recalibrate:
Google (and Apple, and QC) are putting lots of AI compute onto their SoCs. Where is it used, and how does it help?
Vision and video are, I think clear answers and we know what's happening there.
Audio (recognition and synth) are less clear because it's not as clear what's done locally and what's shipped off to a server. But quality has clearly become a lot better, and at least some of that I think happens locally.
Translation I'm extremely unclear how much happens locally vs remotely.
And semantics/content/language (even at just the basic smart secretary level) seems hopeless, nothing like intelligent summaries of piles of text, or actually useful understanding of my interests. Recommendation systems, for example, seem utterly hopeless, no matter the field or the company.
So, eg, we have Tensor with the ability to run a small BERT-style model at higher performance than anyone else. Do we have ways today in which that is used? Ways in which it will be used in future that aren't gimmicks? (For example there was supposed to be that thing with Google answering the phone and taking orders or whatever it was doing, but that seems to have vanished without a trace.)
As I said, none of this is supposed to be confrontational. I just want a feel for various aspects of the landscape today -- who's good at what? are certain skills limited by lack of inference or by model size? what are surprising successes and failures?
dotjaz - Tuesday, November 2, 2021 - link
" but I do think it’s likely that at the time of design of the chip, Samsung didn’t have newer IP ready for integration"Come on. Even A77 was ready wayyyy before G78 and X1, how is it even remotely possible to have A76 not by choice?
Andrei Frumusanu - Wednesday, November 3, 2021 - link
Samsung never used A77.anonym - Sunday, November 7, 2021 - link
Exynos 980 uses Cortex-A77