Facebook Twitter Flickr

Low light shots from the Nokia 808 PureView, Lumia 1020 and Samsung Galaxy S5

You asked for it – and I’m always glad to serve you here at the PureViewClub: low light shots captured with the Samsung Galaxy S5, compared with the Nokia Lumia 1020 and – by popular demand – the Nokia 808 PureView.

In previous posts – in much better light conditions – we’ve seen the Galaxy S5 performing very well. Most remarkable was the Galaxy S5 shots turned out a bit brighter, losing some contrast along the way.

One of the interesting “discoveries” was that when you use the “selective focus” option the Galaxy S5 will give you a >20MB file (read about it in this earlier post). Also, I’ve been impressed with the way it shows you what HDR does to your shot before capturing it.

I think on almost every occasion readers have been asking for a low light comparison – including the Nokia 808 PureView if possible. During a family visit this weekend I managed to capture this simple, but effective scene outside. I hope it will offer what you were looking for.

As usual I’ll share these shots with a few disclaimers: these were captured in between “family affairs”, so I guess that’s why I wasn’t concentrated enough to put the Lumia 1020 in 16:9 aspect ratio. It’s not something you notice fast when you’re shooting very dark scenes either, so I discovered I didn’t only a day later.

There is one advantage though: in 4:3 you’ll get most of the Lumia sensor, whereas the Galaxy S5 gives you the full 16MP in 16:9. One more thing: I only used the Nokia 808 PureView in 8MP PureView mode. I will bring the Nokia 808 PureView on a next short holiday soon, I promise you I’ll make some low light shots in full resolution with it as well.

Let me start with the Nokia 808 PureView since you appear to have missed it so much (I must admit it was fun to work with it again :-) You’ll see two results (both resized from the 8MP originals) – first in ISO 400 (1/2 econd)

2 Nokia 808 PureView - Balcony ISO 400Next: Nokia 808 PureView – ISO 800 (1/8 second) 140320_Nokia_BT_728x90_MB

2 Nokia 808 PureView - Balcony ISO 800

The atmosphere is very accurate, as expected from the (former) King of Camphones. The light in the corridor is white, the one next to the balcony yellowish, and you can even see the last light of the day leaving.

Next the Nokia Lumia 1020 – as I’ve written before, I’m sorry to say in the dark I didn’t actually notice it was on 4:3 – but it will give you the best of the sensor anyway.

Here we go – again with two different ISO settings: ISO 400 first (1/2 second), next ISO 800 (1/5 second).

2 Nokia Lumia 1020 - Balcony ISO 400 2 Nokia Lumia 1020 - Balcony ISO 800

As with the Nokia 808 PureView, I like both results. How does the Galaxy S5 capture the same scene?

I’ll show you two shots once more – first manually put to ISO 400, next the one in which the S5 chose ISO 2000 by itself. Both shots are captured in 1/17 second, so no wonder the ISO 400 shot gives the darkest result:

2 Samsung Galaxy S5 - Balcony ISO 400

And this is what the Samsung Galaxy S5 captures with ISO 2000.

2 Samsung Galaxy S5 - Balcony ISO 2000

To compare just a few crops, I chose the shots with the highest ISO of all three cameras. Please keep in mind the Nokia 808 PureView is on 8MP PureView, the Lumia 1020 on 5MP PureView and the Samsug Galaxy S5 on 16MP. It’s all very different but it will give you an impression.

2 Nokia 808 PureView - Balcony ISO 800 2 Nokia Lumia 1020 - Balcony ISO 800 2 Samsung Galaxy S5 - Balcony ISO 2000

Although you can still make out enough of the details, I think it’s pretty obvious the Samsung loses in this case.

Now of course you’re wondering if the high-res (38MP) result of the Lumia 1020 is any better than what the Galaxy S5 captures in 16MP. Well, here is another crop – without any oversampling, the result actually appears to be quite a bit better indeed.

2 Nokia Lumia 1020 - Balcony ISO 800 Pro__highres (2)

I must admit I like the results coming from the Nokia Lumia 1020 most. OIS probably helped me out here once again. I will share these shots on Flickr as you are used to. In a next comparsion, to satisfy both your and my own curiosity, I’ll make some full resolution shots with the Nokia 808 PureView as well.

It will take some more time, since I’ll be without my laptop for a while, traveling with my family on a short trip this Easter holiday. Hope to update you from there not just here at the Club, but also, and even more so, on TwitterFacebook and Instagram: please do follow me there as well!

Please join PureViewClub on TwitterFacebookFlickrGoogle+ and/or Instagram.

Lightning Ad 3


 Share on Facebook Share on Twitter Share on Reddit Share on LinkedIn

47 Responses



  2. mirco

    Just a short comment on this one:

    >> “There is one advantage though: in 4:3 you’ll get most of the Lumia sensor”

    One reads simillar statements very often on several places but it is simply wrong. The image quality, especially the level of resolved detail, doesn’t change at all. How could it alter the “resolution” if you just crop at the sides of an image? Because this is simply what’s happening here: You have the full sensor and crop different parts to get the desired aspect ratio (and in this case it doesn’t even matter if it is a multi format sensor like in the 1020 or a 4:3 sensor). The total number of pixels used for the image changes but not the resolving power of the camera, i.e. the ability to distinguish between two points.

  3. Branko

    If you take a look at picture from Nokia 808 PureView – ISO 800 there is a purple tint at right side of frame. For me (my 808 does this also) this spoils most night photos taken at high ISO…

    • mirco

      This effect is called amp glow. It is basically heat introduced noise in the imaging sensor. It is quite hard to avoid it in such tiny devices with the battery and the cpu so close to the camera.

  4. Junnior Reis

    Nokia 808 PureView Wins… Simple As That !!!

  5. Dominick

    Great work Marc! Though I would say that I prefer the colors on the 808 (400iso) and the detail of the bricks behind the hanging plants. Both Pureview devices certainly did way better S5.

  6. AreOut

    the result varies greatly because not everyone has the same “hand stability”, and for the best quality it’s best to use tripod/improvise stabilisation for all devices especially at night, that’s where 808 shines and that’s how you take night pictures anyway if you want high quality, so why bother with handheld shots of static objects…

    • Marc @PureViewClub

      Because I would need many different tripods, which I don’t have.

      • AreOut

        actually one would suffice, there are quite cheap tripods which will host devices of variable size

        • Marc @PureViewClub

          Like the Lumia 1520? Anyway, I prefer to make my comparisons from the “consumer point of view”. That’s usually the person walking around with a mobile in his pocket – not a tripod.

          • MF

            Yup. Most people take pictures with their mobile phones handheld. A smaller minority would bring around their DSLR without tripod. An even smaller minority brings around their DSLR with tripod. And a much much much smaller minority brings around a tripod for their mobile phone. For all that trouble, why not just bring a DSLR or a more compact mirrorless system?

            • steve

              A dslr is not going to shoot photo’s any better in low light it also needs a tripod. In fact the 808 or 1020 or 1520 are easier to hold still then a heavy dslr. The quality from the 808 is the same as most dslr’s the exception being the full frame dslr’s. Camera shake is a problem when your shutter speed gets below 60 on a dslr and below 15 on a cellphone. If you want great low light photos employ a tripod the difference is huge. I think the point of using a tripod to do low light camera tests is it eliminates user error and is equally fair to each device.

            • MF

              LOL… you serious? The 808 is not even any better than the 1020 in lowlight, much less a DSLR?

              Btw, any idea how much bigger is a DSLR’s sensor compared to the 808? Physics win?

            • Physics and the PureView algorithm in play, the scrunched up 41MP sensor should have bigger superpixels size – in 8MP, 5MP, or 2MP – than an average DSLR.

            • MF

              Really? How about some numbers to back it up?

            • w w w . allaboutsymbian dot com/images/features/pixels808-v3.jpg

              w w w . allaboutsymbian dot com/features/item/14347_PureView_pixels_versus_the_com.php

            • mirco

              Yet, the 808 (and the 1020… just for the record) looses against DSLRs (and high end P&S cameras) in ALL fields except maybe resolution in perfect light conditions.
              For example, the pixel binning (PureView is only little more than this) doesn’t improve the dynamic range which is the weak spot of every smartphone camera. Depth of field (or the lack of it) is another example, where the 808/1020 doesn’t stand a chance. Shall we also start talking about flexibility and expandability of DSLRs?

              One should simply not put the 808/1020 into general competition with DSLR.

            • steve

              Ive done the tests against the gh2 and rx100 I’ll put my 808 up against any dslr in low light and it will beat most.https://www.flickr.com/photos/90056083@N07/12617980994/

            • steve

              I would love to see any evidence for your claim that the 808 is behind in dynamic range I’ve found the opposite to be true the 808 is equal to most dslrs if you compare shadow cut off to highlight clipping you will find the 808 dose extremely well. Maybe your confusing the 808 with an average cell phone like an Iphone if so I agree they have very poor dynamic range. Again I’ve done the tests with my gh2 micro 4/3 which is an average dslr.

          • AreOut

            not every consumer has the same “hand stability”, also including tripod photos doesn’t mean you have to exclude handheld ones ;)

            also there are many “natural” ways of stabilising camera without using tripod, you just have to lean the phone against something, but I agree you will not have that opportunity in every occasion

  7. mike

    I also like these comparisons more than in daylight! Samsung really dissapoints me, I thought it would give average photos, but it didn’t. Samsung’s camera didn’t have many real improvements at all… Only software.

    It would be interesting to see the z2, compared with Nokia pureview devices in low light photography.

    I like the results of the black update in lumia 1020, it isn’t all yellowish anymore! And it doesn’t have the “unfocused” results of the king 808, although it produces more noise. But a noisier picture is usually better than a blurred one, so at least in low light photography, it is obvious that the “prince” lumia 1020 is better and more flexible (due to OIS) than the “king” 808!

    Also I have to say that the building and the scenery looks like vacations in Greece! :)

    • Marc @PureViewClub

      I’m still waiting for a review sample of the Z2 – it’s not even released, so that might still take a while… And it’s the center of the Netherlands you’re looking at, not Greece (yet ;-)

  8. Richard

    Lumia 1020 really does a good job here~
    However, after reading so many awesome comparisons from pureview club, I just realize why some people keeps saying nokia 808 is better than lumia 1020 in some ways.
    I went back to check the daylight comparison of “Lumia 1020 vs Lumia 1020″(It also has nokia 808 in the comparison) I really can’t tell the big difference between nokia 808 and lumia 1020 from the full resolution pictures.

    I think the biggest difference is on taking pictures of people, especially something like portrait or anything has skin in the photo.
    No matter how close I shoot faces or a person, when you zoom in to see the details, especially on the eyes and skin, it has heavy noise even it’s on iso 100 with tripod. Whereas the nokia 808 could take almost noise free portrait.
    I guess it’s because the different algorithm. The nokia team probably test more on capturing objects like buildings and landscapes.

    OK…Sorry for writing too much, all I want to say is….


    I think it would be great to have a comparison like this since most comparisons in the past are focusing more on “landscape”.

    Again, thanks for the awesome comparison ~ :)

    • spaceyjase

      I do wonder if the same level of noise would be present if the 1020 had the option to disable OIS? I have a compact that has OIS and disabling the feature (for use on a tripod, for example) enables a far higher image quality at the same ISO setting (less noise!).

      Nokia could add an option to disable OIS.

      • Interesting idea!

      • Marc @PureViewClub

        Good idea, but I don’t think it will happen. I’ve read the next update will even take out the “double capture” possibility (so no more 5MP ánd high-res or .DNG combined). I really detest that, but hey: no-one’s asking me :-)

        • bigs

          really Marc, why would they take out the dual capture feature?? Is it possible that the latest update will allow for post zoom crop from a RAW DNG? Or maybe the ability to process DNG in phone to JPG?

          • I hope that they will disable dual capture since I shot only in RAW! But to be able to take only RAW you need support for viewing RAW in phone! :)

          • Marc @PureViewClub

            I understood they plan to take it out with the Cyan update to realize less “time between shots”. I’d prefer having a choice though… :-/

    • Marc @PureViewClub

      I’d have to find a model willing to pose for the PureViewClub. I didn’t manage to find one so far, maybe I’ll try to find someone through this site or on Twitter. There must be some Dutch models interested to be captured in PureView I’d say.

  9. Givi

    Wow I’m quite astonished by 1020 shots . I knew that S5 has no chanse with this amazing camera phones , but in my opinion and for my tasted 1020 did better job that 808 . Despite 808 is in 8MP, 1020 did better with 5MP shot . Details are better and color reproduction , I always thought 1020 had yellowish tint but, of 1020 is obviously better than 808 . Does it mean now 1020 has crown ?
    We need one comparison between 1020 vs 808 , daylight and night shots . and DNG on 1020 to show it’s full potential

    • Steve

      No The 1020 is not better then the 808. It only appears better because Mark chose equal settings for the 808 and 1020 ISO. The 808 200 ISO is equal to 400 on the 1020 and to take proper night shots a tripod should be used and the 808 should be set to 50 thru 200 ISO with 34mp on, not 8mp pureview. If your going to shoot hand held the 1020 ois will give better results at low shutter speeds. What is clear is that the s5 has no way of lowering the shutter speeds to match the 808 or 1020 so will never know how it could do in low light.

      • Dominick

        Yup, that’s 100% correct

      • Marc @PureViewClub

        That is partially correct I guess. Hope you’ll like what my next post brings – even without using a tripod (since I don’t own four different cameraholders)

      • bigs

        Agree Steve, not sure why the 1020′s ISO is about 1 stop difference compared to the 808. The only reason for this I would imagine would be for marketing purposes. The ability to advertise ISO 1600 capability gives the impression that the phone camera is better than the 808 and competing dedicated digicams.

      • Ron

        ” tripod should be used ”
        Why? Because the 808 do not have OIS? Maybe Sony QX100 should be used with the S5 because the S5 do not have big sensor, Eh? Ridiculous? Of course.

  10. walli

    It would be very interesting to put these 3 phones on an tripod and shot on lowest iso. I did my best 808 pics with long exposure.

  11. Praveen

    I don’t understand why people accuse you of being inclined towards Pureview devices. It is quite obvious that pureview quality is unmatched atleast till now.

    Thanks for the comparision Marc. Keep up the good work :)

  12. Nice! I like those comparisons much better than in daylight!

    I know you can’t take all phones with you, but I just wanted to see how 1520 compares with S5 since they are more similar than phones with huge sensors – I think they are in different category!
    In my tests 1520 has higher ISO with slower shutter speeds compared to 808, but here looks like 1020 has about the same ISO and shutter speed compared to 808! So I just am really curious! ;)