Disappointed by the Z5 camera pictures quality: a critical comparison!

Search This thread

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
Dear all,

Like they say "the devil's in the details" and "the beauty is in the eye of the beholder"... Nonetheless this is not meant to inflame anyone but rather to open your eyes for what is to be seen.
It's not about fanboys wars but about waking up Sony until it's not too late. The idea for the pictures below is simple: they represent 100% zooms in comparisons of some of the currently available "heads on" camera tests between Sony Z5 and other models.
So far I will save my comments for later on. I should add I am a pro photographer for more than 15 year and - aside from the artistic misconceptions I might have - at least I properly understand what digital image quality about.

Source of images used in this comparison:
http://www.eprice.com.tw/mobile/talk...ompact-review/
AND
http://www.phonearena.com/news/First...arison_id73408

Provided other comparisons will show-up on the web I will take my time to add more of such 100% match-up. Enjoy and... spread the love! :highfive:

>>>>
proof_1.jpg

>>>>
proof_2.jpg

>>>>
proof_3.jpg

>>>>
proof_4.jpg

>>>> Correction: the caption on the right should read Galaxy S6 not Note5. My bad! :(
proof_5.jpg

>>>>
proof_6.jpg

>>>>
proof_7.jpg

>>>>
proof_8.jpg

>>>>
proof_9.jpg

>>>>
proof_10.jpg

>>>>
proof_11.jpg

>>>>
proof_12.jpg

>>>>
proof_13.jpg
 
Last edited:

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
I know dxomarks rushed review has given enormously high hopes to everyone looking at Sony but let's take that review with a pinch of salt. I highly dislike big praising reviews where no "pixel peeping" proofs are made available so that upon having finished reading the report everyone can get a good grasp at the proofs in the file and go home thoroughly accepting the results.
Not the case unfortunately with dxomarks and their last two "shouted" reviews: Sony's Z5 and Huawei's 6P... :(
 
  • Like
Reactions: AlexBurnout77

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
Lol. No EXIF data. Nothing except Adobe Photoshop data in EXIF. http://regex.info/exif.cgi?imgurl=http://www.zoomzoom.ro/z5/proof_1.jpg
Why do we must ever speak with you?

Let's not bring this to a kindergarten quarrel leveled, ok?
It is not about how we feel about brands and personal affinities. I have done a 100% fare comparison by opening the available pictures at the already discussed threads on xda Sony Z5 General sections. So, help yourself! ;)

http://www.eprice.com.tw/mobile/talk/4523/4953454/1/rv/sony-xperia-z5-compact-review/
AND

http://www.phonearena.com/news/Firs...vs-Z3-vs-G4-camera-samples-comparison_id73408
 
Last edited:

joe_dude

Senior Member
Oct 30, 2006
298
93
Ah guys, it's a 23mm wide-angle lens

The Xperia Z5 uses a 23mm wide-angle lens vs. the Note 5's 28mm lens. So, the Z5 pics are probably like 100% zoomed in further. I'm no photography expert (I'm a tech guy), but I think that speaks for itself. Those pics are not a fair comparison.
:eek:

Edit: Correction, 24mm lens on the Z5.
 
Last edited:

jetbruceli

Senior Member
May 8, 2011
2,916
608
Bamberg, Germany
edit:Nevermind these arent your own photos. Cant take this an end all situation

Where did you get the photos? I mean they seem to be taken from many places and are you comparing a 16mp vs 8? or 23mp actual to 16mp
 

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
It's not about focal lenghts

The Xperia Z5 uses a 23mm wide-angle lens vs. the Note 5's 28mm lens. So, the Z5 pics are probably like 100% zoomed in further. I'm no photography expert (I'm a tech guy), but I think that speaks for itself. Those pics are not a fair comparison.
:eek:

Hi @joe_dude,

It's not about focal lens here, in this comparison. Z5's 23mm vs a 28mm (Note5 or other camera) would result more in distortions and image degradation towards the image left/right sides, where Z5 seems to exhibit strong errors (as already pointed out in other threads). The 100% zoomed in crops I used in this comparison where mainly taken in the center area of the original picture so that it doesn't affect the overall balance of evaluation. So, NO, this is not where I tried to evaluate focal lengths and how their induced distortion was corrected by Sony's Z5. :cool:
 

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
edit:Nevermind these arent your own photos. Cant take this an end all situation

Where did you get the photos? I mean they seem to be taken from many places and are you comparing a 16mp vs 8? or 23mp actual to 16mp

Hi @jetbruceli,
I have updated the initial post with the links to the original pictures. All the pictures are 23mpx to the actual size of the compared cameras (16mpx for Note5 / S6 / G4).
Try to focus on the actual pictures quality and leave aside and supposed wrongdoing tentative on my behalf while proposing this comparison. :)
 
Last edited:

IT.Adviser

Member
Feb 19, 2008
22
7
I have done a 100% fare comparison by opening the available pictures at the already discussed threads on xda Sony Z5 General sections.

Do you think, that we can compare totally different files with absolute different size?

http://regex.info/exif.cgi?dummy=on..._4523_d3ed90416a666e289d48f11080b9c155ori.JPG

5,520 × 4,140

http://regex.info/exif.cgi?dummy=on..._4523_f99d567409097f8c93322a5ed1978f9bori.jpg

5,312 × 2,988

I can take pictures with 100x100 size on 10 years old device and they will look superb comparing with 5000x5000 photos. If you want to compare set exactly same conditions and file setups.
 

joe_dude

Senior Member
Oct 30, 2006
298
93
Hi @joe_dude,

It's not about focal lens here, in this comparison. Z5's 23mm vs a 28mm (Note5 or other camera) would result more in distortions and image degradation towards the image left/right sides, where Z5 seems to exhibit strong errors (as already pointed out in other threads). The 100% zoomed in crops I used in this comparison where mainly taken in the center area of the original picture so that it doesn't affect the overall balance of evaluation. So, NO, this is not where I tried to evaluate focal lengths and how their induced distortion was corrected by Sony's Z5. :cool:

But you're comparing the cars, for example, at the same size. In the original pic, the car is smaller with the Z5. All the pics taken with the Z5 are zoomed out further. You will lose quality by zooming in closer. A fair test would require the shooter to walk slightly closer, so that the size of the objects with each camera matches up. That said, the iphone pics are zoomed in much farther. Now I understand why the iphone gets decent quality - by sacrificing FOV.
 
Last edited:

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
Do you think, that we can compare totally different files with absolute different size?

http://regex.info/exif.cgi?dummy=on..._4523_d3ed90416a666e289d48f11080b9c155ori.JPG

5,520 × 4,140

http://regex.info/exif.cgi?dummy=on..._4523_f99d567409097f8c93322a5ed1978f9bori.jpg

5,312 × 2,988

I can take pictures with 100x100 size on 10 years old device and they will look superb comparing with 5000x5000 photos. If you want to compare set exactly same conditions and file setups.

Hi @IT.Adviser,

Your reasoning is puzzling! :)) Sorry to be that blunt!-aspect
"Do you think, that we can compare totally different files with absolute different size?".... Man, this is what the camera sensors of two different smartphones are throwing out. One is a multi-aspect (Z5 is both 4:3 and 16:9) and one is a 16:9 (on Note5). What is the problem then? I opened each and every picture in Photoshop, zoomed to 100% and chose same size crop (1.000x1.000 pixels) in pretty much the same area, so the comparison is absolutely fare! Recheck your reasoning, pls! ;)


"According to some new evidence, it looks like the new sensor on the Sony Xperia Z5 series actually has a resolution of 25MP, although the photos you can extract from it are limited to 23MP due to the difference between the two standard aspect ratios. When shooting in portrait mode - a 4:3 aspect ratio - the sensor is capable of taking photos at 5,520 x 4,140 pixels (23MP). In landscape - 16:9 aspect ratio - mode, the camera takes photos at 5,984 x 3,366 pixels (20MP).
Read more at http://www.phonearena.com/news/The-Sony-Xperia-Z5-series-actually-has-a-25MP-multi-aspect-primary-camera-sensor_id73325#ofcKoe4y1dz93ck5.99"
 
  • Like
Reactions: lucker#1

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
But you're comparing the cars, for example, at the same size. In the original pic, the car is smaller with the Z5. All the pics taken with the Z5 are zoomed out further. You will lose quality by zooming in closer. A fair test would require the shooter to walk slightly closer, so that the size of the objects with each camera matches up. That said, the iphone pics are zoomed in much farther. Now I understand why the iphone gets decent quality - by sacrificing FOV.

Really? Really, really, really? I am not sure if you keep attacking the methodology used here or just asking for details in an offensive manner... Yet, the picture you're talking about has not been re-sized, fact you could easily observe from the attachment below. The EXIF bears the original details of the file that happen to match the Z5 capabilities (resolution, etc.)
really.jpg
 

dehnhaide

Senior Member
Mar 16, 2009
183
151
Bucharest
Not sure why...

I agree with you, that Sony photos are bad. But, I disagree about such comparison of this photos.

Why would you? I am open to describe the step-by-step workflow I followed when performing the assessment just to make sure that everybody is confident about the result. Unfortunately I am no at home, else I could even provide you with the .PSD file I have created with all the layers used in the comparison, so that you could let go the paranoia about the methodology.
I am not a fan of any brand (had owned quite too many of them actually to be a fan o any in particular) but i think the most important thing here is to let the community know (aside from the paid reviews by "precious" 3rd parties) about the current standing of image quality in Z5's camera. Agree?
 

a.mubher

New member
Oct 5, 2015
3
1
Depends on where you look, You can easily nitpick something while ignoring the general result, here are some examples from the same collections of pictures you used.
Z5 on left, Note5 on right, over sharpened and exposed.
original.jpg

original.jpg

original.jpg

original.jpg
 
Last edited:

King p1n

Senior Member
Jul 26, 2013
1,635
819
Your looking at early September firmware...

But yeah the Z5 photos aren't as good as Sony and DXOMark are making it sound.

 
  • Like
Reactions: dehnhaide

Top Liked Posts

  • There are no posts matching your filters.
  • 15
    Dear all,

    Like they say "the devil's in the details" and "the beauty is in the eye of the beholder"... Nonetheless this is not meant to inflame anyone but rather to open your eyes for what is to be seen.
    It's not about fanboys wars but about waking up Sony until it's not too late. The idea for the pictures below is simple: they represent 100% zooms in comparisons of some of the currently available "heads on" camera tests between Sony Z5 and other models.
    So far I will save my comments for later on. I should add I am a pro photographer for more than 15 year and - aside from the artistic misconceptions I might have - at least I properly understand what digital image quality about.

    Source of images used in this comparison:
    http://www.eprice.com.tw/mobile/talk...ompact-review/
    AND
    http://www.phonearena.com/news/First...arison_id73408

    Provided other comparisons will show-up on the web I will take my time to add more of such 100% match-up. Enjoy and... spread the love! :highfive:

    >>>>
    proof_1.jpg

    >>>>
    proof_2.jpg

    >>>>
    proof_3.jpg

    >>>>
    proof_4.jpg

    >>>> Correction: the caption on the right should read Galaxy S6 not Note5. My bad! :(
    proof_5.jpg

    >>>>
    proof_6.jpg

    >>>>
    proof_7.jpg

    >>>>
    proof_8.jpg

    >>>>
    proof_9.jpg

    >>>>
    proof_10.jpg

    >>>>
    proof_11.jpg

    >>>>
    proof_12.jpg

    >>>>
    proof_13.jpg
    11
    As a professional photographer I can say that this phone so far has a very impressive camera.
    The dynamic range is insane, I haven't encountered a situation where the highlights has been blown out or the shadows completely dark yet.
    The exposure metering system in the camera does a very good job of finding the right exposure.
    The color reproduction is very very good, far better than any other smartphone I've seen.
    Yeah the colors are "duller" and less "contrasty" than other phones, but this is because it is neutral. And everyone has got to love neutral.
    Neutral means that it is much easier for anyone to edit the photo to get it where you want it.

    I see that some people complain about detail and sharpness. And yes, there is some lack of sharpness and some "smudgyness" in the corners. But for **** sake, my $1300 nikon lens is also less sharp and "smudgy" in the corners. The center sharpness of this camera is very good and the capture of details is also good. There is no chromatic aberration at all, not even at F/2. And I've tested this in the corners in very high contrast situations. Even my Nikon D800 with the $1300 24-120 gets chromatic aberration on the widest aperture.

    Ok, the 23 mp photos isn't perfect when you zoom in 100%. But who the **** uses their smartphone as a professional camera anyway? Most of people will not print photos bigger than A4 and at 23mp this will mean that the original photo will be downscaled and the photo appear sharper after print. And who prints phone photos anyways? Let's face it guys, most people take photos for snapshots or to post on social media. And for social media you don't really need photos bigger than 1500 on the short side. Again this will mean that the 23mp photos of the Z5 will be downscaled and appear sharper.

    It might not be perfect in a lab or testing it to the max zoomed 100% in. But in real life use, this phone has a camera that is so good that I will actually start leaving the D800 at home some times. And just to put it that way, my D800 with 24-120 is not perfect in a lab either.
    Remember that Sony makes sensors for Nikon and have some awesome cameras themselves. I knew I wouldn't be disappointed even though there are so many haters.

    Here you can find some test shots done today. The test shots were done in Manual mode 23mp without any tweaking of the settings.

    Test shots with the Sony Z5 in manual mode. No settings altered. by TordFuglstad, on Flickr
    9
    https://www.sendspace.com/file/e451np
    Here are some photo I've taken today with Z1c and S6, you can identify it by the folder name or file name. Z1c got both 8MP and 20MP but I've been rolled back to KitKat and 20MP mode on KK is buggy and easy soft, it should perform more better on Lolipop.

    Thanks, I'll have a look in a moment. :) Much appreciated.

    ---------- Post added at 04:57 PM ---------- Previous post was at 04:39 PM ----------

    There has been a lot said, above, about how "Awful" the Z5 is in some site's side by side comparisons with the competition. As has been pointed out by myself and others, the Z5's wider field of view has not been taken into account in these tests and the phone has NOT been used at its optimum resolution (8MP) in each case. This has heavily biased the result.

    The image below shows how this can be deeply misleading, as do the following set of images.

    The Image attached shows the difference that adjusting for the different field of view and dropping to 8MP makes: the result speaks for itself. All I have done is simulate the the level playing field that the sites should have given the Z5.

    Initial image: both images cropped identically - on the left is with the image at 20MP while standing a step away, to simulate the FOV difference. This simulates how the phone has been used in the tests referenced earlier on this thread. On the right, is an 8mp image with a single step forward to adjust for the FOV. The result is dramatic.

    Thoughts are welcome. Flickr feed demonstration https://www.flickr.com/gp/136945683@N06/Tw852X
    7
    Back to the beginning...

    So now that I'm a lot more familiar with the Z5, let's go back to the original post from dehnhaide. The crops of the pics he had posted seemed to point to the Z5 pics being worse than the Note 5. So let me go through them one at a time, as now I know a lot better what to look for.

    Here are the original links if you want to play along:
    http://www.eprice.com.tw/mobile/talk/4523/4953454/1/rv/sony-xperia-z5-compact-review/
    http://www.phonearena.com/news/Firs...vs-Z3-vs-G4-camera-samples-comparison_id73408

    Pic #1: the salad. If you look at the pics closely, the Z5 pic has more noise and the lettuce is not as detailed. The noise is because it's in 23 MP rather than 8 MP, obviously. The detail is the funny part - the Z5's focus is on the plate, rather than the lettuce. Why? That's because of the 4:3 framing and much larger FOV. If the pic was taken in default 8 MP in 16:9 and proper FOV, the Z5 would have taken a different picture. As I've maintained from the beginning, comparing dissimilar pics leads to errors, as is in this case.

    Pic #2: the building. Notice the discolouration in the middle of the Z5 pic? That's glare. You can clearly see the glare coming in at the top of the Z5 pic. Once again, the larger FOV caused a problem. With the added glare, the Z5 lost detail. That should have been caught by the reviewer and re-taken.

    Pic #3: the red car. Blur and lack of night mode because the camera was using 23 MP. If the camera was left in default 8 MP, scene detection would have kicked in.

    Pic #4, 5, 6, 7, 8: Excelsior Caffe, hotel, night pics. Similar to #3 - noise from 23 MP mode and lack of night/low light scene detection. Use 8 MP mode.

    Pic #9: trees. Glare from sunlight. Look at the Z5 pic - it covers twice as much sky as the Note 5, so suffers from much greater glare. Also, the Note 5 has a purple tinge to the sky.

    Pics #10, 11: night pics. Same story as #3, 4, 5, 6, 7, 8.

    Pic #12: tree and sign. Yup, you guessed it - the Z5 is focused on the middle part of the tree, while the S6 is focused more on the lower part and the sign. So, obviously, the sign in the Z5 pic is slightly less focused. Another bad comparison.

    Conclusion: What more needs to be said? Comparing 4:3 pics to 16:9 with different FOVs was and is a bad idea. Some of the pics weren't even focused on the same thing! If they had stuck with the default 8 MP at least, the Z5 could have saved some of the pics. Those initial reviews just didn't have a clue what to do with the Z5 camera.
    :eek: