Device Conclusions & Remarks

Finally wrapping up, hopefully we learned some interesting results today. These camera articles take a very large amount of time to do, and although we covered quite a few things, there’s still tons of other aspects of the cameras we haven’t really even touched upon. At the end of the day, we don’t have infinite time, and I prioritised on just covering some of the technical aspects of today’s varied smartphone camera solutions.

To wrap things up, I think what’s best to do is go over the various different conclusions of the various devices tested today, making note of their strengths and weaknesses.

Apple iPhone 12 Pro

Amongst the most traditional hardware setups of this review, the iPhone 12 Pro is still an outstanding performer in most circumstances. Apple’s biggest strength lies in daylight HDR processing that more often than not is able to achieve results that are far more accurate and pleasing than what other phones are able to achieve with technically superior hardware. Apple is also still the only vendor right now that can capture images in wide colour gamut, which results in much more representative and popping colours in well-lit scenes.

In low-light, the hardware disadvantages catch up with the phone, as while its processing is adequate, it notably falls behind the competition which has just better sensors.

Google Pixel 5

Google’s strength also remains largely on the software side of things. It’s actually quite close to the iPhone in terms of results many times, sometimes actually being more accurate with handling of HDR scenarios – there’s a definitive upgrade compared to past Pixel phones. The Pixel 5’s very simplistic dual-camera solution doesn’t help it in terms of versatility of the camera experience: it doesn’t go very wide, and for longer focal ranges it also simply has to rely on digital cropping and super-resolution upscaling, which isn’t a replacement for dedicated camera modules.

OnePlus 9 Pro

The newest flagship from OnePlus was announced with a lot of fanfare in regards to its new camera system, along with its collaboration with Hasselblad. This latter I only mention now because it’s mostly completely irrelevant to the phone’s capture experience and image results. In daylight shots, the one defining factor for the OP9Pro for me was its typical artificially darkened shadows, and a severe case of over-sharpening and overprocessing of details that make the results look extremely digital and artificial. It’s still an overall good camera experience, but it’s hardly a camera I would call accurate. In low-light scenarios, it actually competes surprisingly well in details, and does have the sensor capabilities to compete, which is definitely a bonus for the phone.

Xiaomi Mi 11

The Mi 11 was released earlier this year, and in terms of camera setup wasn’t a particularly exciting device as it simply carried forward the camera modules from the Mi 10 last year. Generally, we had reviewed this system quite a bit, and it’s mostly characterised by the 108MP main camera sensor which captures auto-mode images in 27MP. The results here are adequate, but generally fall behind in dynamic range or proper light gathering capabilities compared to the competition. The good points about the Mi 11 is a very natural image processing, but also has some downsides, as I’ll mention about the Mi 11 Ultra as well just below.

Xiaomi Mi 11 Ultra

We hadn’t reviewed the Mi 11 Ultra camera until now, so this was definitely something I was looking forward to for this review. Generally, the Mi 11 Ultra is definitely one of the strongest camera solutions out there today. The hardware setup, even though it’s “only” a 3-camera solution, smartly takes advantage of the quad-Bayer sensors, effectively turning the system into a “virtual” 5-camera setup, bar a quality gap from around 70-120mm.

The huge new 1/1.12” main sensor allows for extremely good retention of natural details – well actually, all the sensors have this common characteristic as Xiaomi seems to employ an extremely light touch (or even none) when it comes to postprocessing here, which is great.

The only one negative I would say is that its HDR processing isn’t quite as capable in recovering information as some of the competition’s implementations, so I found the phone is maybe too conservative there and doesn’t pull out the full capabilities that the hardware is capable of. This is especially valid for the periscope telephoto – while technically excellent, the image processing here in terms of exposure metering and HDR leaves things to be desired.

In low-light, the phone is a top-performer, but isn’t totally consistent in its ultra-wide. As I mention the UWA – the optics quality on this module isn’t quite satisfactory on the outer frame, and performs notably below that of other devices.

Samsung Galaxy S21

The smaller Galaxy S21 has the same camera setup as that of the regular S20 series devices, which actually isn’t necessarily a bad thing. I’ve noted quite a few times that I think what Samsung has concocted here in terms of hardware solutions is extremely smart, and amongst the most versatile smartphone cameras out there, even if it doesn’t have the super fancy telephoto capabilities of other higher-end phones.

Having an extremely solid ultra-wide and main camera module without any gimmicks provide the S21 with great capabilities, along with image processing, that while sometimes can be a bit off compared to Apple or Google, still is amongst the best out there on the market.

The phone’s versatility comes through the secondary 64MP wide angle module on the phone, which is still to this day unique in the industry in terms of its implementation, and although the device doesn’t have a telephoto module per se, it allows for seamless full quality magnification up to around 3x, along with also leading quality images at 2x/50mm.

At low-light, the phone’s main camera is only outperformed by its Ultra siblings and other devices which have far larger sensors, but overall, still manages to get amongst the best results across a wide variety of scenarios. It’s generally been my favourite camera solution to date.

Samsung Galaxy S21 Ultra

While the S21 has a well-thought-out camera system, the S21 Ultra feels like it’s been designed by the company’s marketing department. Already last year we had identified that the company’s 108MP sensor that bins down to 12MP doesn’t actually represent an advantage to the photo experience, but actually is a handicap, being less versatile than other 48/50MP class implementations. Weaknesses here are for example no sensor or optical solution to achieve good 2x/50mm results, leaving a notable gap in the focal range of the phone. To make things worse, the phone does this same mistake in the periscope telephoto module which this year went from a native 4x/103mm optics to 10x/238mm. This results in a quality improvement at the far end, but comes at great compromises at the middle focal range. There’s a new 3x telephoto module added in into the mix, which gets good results, but can only help so much. Seeing today how the S21 Ultra got spanked by the Mi 11 Ultra’s telephoto solution, as well as the main sensor’s 2x native pixel mode, just further enforces my views that Samsung’s module combination choice was a mistake. The company needs to focus on results rather than dumb marketing metrics such as 100x space zoom – it’s hard to comprehend how the S21 and S21 Ultra can be so different here in terms of their approaches.

Nevertheless, as harsh as my critique is (just due to frustration of wasted potential), the S21 Ultra is still a good phone. The image processing advantages, while not perfect, are still in favour of the phone, and it does generally perform well – as long as you’re on a native focal range.

It’s also been several months now since the devices’ release, and software and firmware wise we’ve now seen low-light modes perform much better and more consistently, but there’s still a detail processing difference in favour of the Snapdragon variant in daylight shots.

ZTE Axon 30 Ultra

The Axon 30 Ultra was a wildcard device in our comparisons here. I was actually surprised by the phone’s daylight performance, sometimes capable to compete with the best devices out there. It’s certainly also a quite unusual hardware camera solution which I thought was worth covering, and indeed it can achieve some good results. ZTE’s image processing and HDR was quite good in many scenarios, and while I wouldn’t say it’s top-scoring, does outperform some other third-party vendor’s implementations.

The secondary wide-angle module at 31.45mm focal length was a bit perplexing, as I’m not sure it warrants its existence compared to the superior hardware of the main sensor. The small periscope telephoto is also functional, but doesn’t quite compete with the big boys such as the S21 Ultra or Mi 11 Ultra. Finally, low-light photography was a bit disappointing for the phone, mostly due to flat night mode results.

A year of telephotos and megapixels

So far in 2021, it seems that the big focus points for most devices have been around marketing extremely zoom or very high megapixel camera solutions. There are actually good implementations of this approach, such as the Mi 11 Ultra, but also bad ones, such as the S21 Ultra. What I hope to see more is the industry being able to actually rationalise their camera choices with actual benefits for the capture experience, rather than choices to enable marketing strategies. Some companies still struggle in balancing this equation, and that’s always been an age-old problem.

While hardware choices can’t be altered anymore, software and firmware are always moving targets, and we’ve seen that this is actually the norm rather than the exception for many vendors. Today’s review in this regard was simply a snapshot of the current situation (as tested end of May), and many of the aspects of today’s tested phones can be improved upon throughout the lifetime of the devices. Let’s hope that actually happens.

Low-light: Large Sensors & Computational Night Modes
Comments Locked

60 Comments

View All Comments

  • s.yu - Friday, June 25, 2021 - link

    Bottom line, the "people" factor in photos actually translates to skin tones, and in that respect all phones should be fine since it's just a matter of color and the most easily repairable even from JPG output, so it's unlikely, with the exception of 1+'s artificial black crush, that skin tones would be mapped beyond the visible DR with all this automatic HDR.
    There's also the matter of accurately portraying skin tones, or automatically "enhancing" them making them subjectively more flattering, which is kind of beside the point of Anandtech reviews, which largely focuses on objective measurements and realism of the processing.
  • ottonis - Wednesday, June 23, 2021 - link

    For a state of the art review of phonecameras, this article is the reference in 2021.
    Incredible amount of work has been put into this meticulous comparison, and it shows.
    Hats off to Andrei and Anandtech and a big thank you for the great and truly useful work.
  • s.yu - Wednesday, June 23, 2021 - link

    Wow you've outdone yourself (and of course everybody else) Andrei! Next time you could add diffraction to the equation :)

    What I learned from this article is that Xiaomi now has a surprisingly robust night mode, and that Samsung finally gets the hang of properly utilizing that 108MP main, such that it generally matches the Mi11U.

    The darkest scene shows blotchy(in squares, or checkered) noise from the Fuji, which is new to me. It seems as though not even Fuji's own JPG algorithm could perfectly handle X-trans. You could try Fuji's own conversion software though, people who are serious about using Fuji can't go without it.

    I sometimes get the Exynos blurriness from my Gcam too, and some of the Pixel samples also seem to show that behavior(like the whole thing's out of focus, or the lens is decentered). I initially thought it was bad compatibility with my hardware but it looks like an inherent software issue now. It's not consistent, at least on Gcam, I could take multiple shots when the issue arises and would generally get a keeper.

    Minor addition though Mi11U's night mode switch seems entirely useless, not only does it automatically do night mode in auto, it automatically reverts to auto in night mode, the first night sample shows two nearly identical 1x shots from Mi11U, and examining the texture I believe they're from auto.
  • Andrei Frumusanu - Thursday, June 24, 2021 - link

    > The darkest scene shows blotchy(in squares, or checkered) noise from the Fuji, which is new to me. It seems as though not even Fuji's own JPG algorithm

    These are RAWs, not SOOC JPEGs, so the blotchiness would some of the CO noise reduction.
  • s.yu - Thursday, June 24, 2021 - link

    Oh! I thought you used OOCJPG for reference too. That's really weird because avoiding artifacts that exhibit a regular pattern (especially one that hints at x-trans specificity) should've been Fuji's top priority. It's a glaring issue. I thought the reference samples looked a bit too...refined, to be OOCJPG though.
  • Lilja - Wednesday, June 23, 2021 - link

    I'm again a bit disappointed that there's no comparisons of the RAW photos that these cameras produce :/
  • ottonis - Thursday, June 24, 2021 - link

    I think you would most probably be disappointed because I don't think that all the HDR(+), night-mode and other multi-exposure tricks are actually baked into a single raw file. Instead, the software algorithms merge several RAW files into one final piece of output, and this generally a JPEG file.
    So, testing RAW files would certainly be a great test of the hardware -/sensor-performance but would barely do justice to the computational side of things. And we all know that the success of phonecameras actually relies more on their software algorithms than on sensor size.
  • s.yu - Thursday, June 24, 2021 - link

    Testing RAW the Pixel should win hands down, though it introduces another level of variation because Gcam results greatly vary from the metering, while the metering is not subject to manual control, and is not entirely predictable, and there's absolutely no telling what comes out until the file is saved and both ends of the DR examined with sliders in LR.
  • s.yu - Thursday, June 24, 2021 - link

    I want to clarify that metering for a lower luminance level, and metering higher but manually pulling down the slider on the right till the screen apparently shows the same exposure, do not yield the same results, the former will increase the likelihood of blown out highlights but provide cleaner shadows, while the latter will retain more highlights but sacrifice shadows. One can only attempt to strike a balance somewhere in the middle, but not have both by, for example, enlarging the shutter speed gap between long and short exposures in the stacking sequence.
  • Rοb - Thursday, June 24, 2021 - link

    It great to see a thorough comparison that has some disagreement with DXOMark: https://www.dxomark.com/rankings/#smartphones-mobi...

    Where the dust usually settles is after a few updates of the Firmware squeeze the last bit of performance out of the hardware - on that front a couple of manufacturers do better than some others.

    I'm happy that my _old_ phone from a few years ago works well, and that in a couple of more years (hopefully my phone lives so long) we'll have new phones that significantly improve on these.

    Nothing like the convenience of a great camera in the compact size of a phone, now only if telephoto, HDR, and stabilization could be improved with computational photography to rival DSLRs of yesterday.

Log in

Don't have an account? Sign up now