DXO Mark Leica M10 Score

[Chaemono]

Advertisement (gone after registration)

I meant ‘Andy’, of course.

[Chaemono]

First shots in low light M10 vs. Sony α7R III just to warm up. Love how the Sony feels and shoots. It's so responsive. They must have done something to that sensor, though. M10 seems to have better dynamic range  . I love the M10 output. The Sony raws uncompressed are about 85 MB. The Sony lens used is the 50/1.4 Zeiss Planar FE. It's kind of crap wide open but an excellent performer stopped down a bit as it should be. At f/2.8 and f/4.0 used here it's sharp and contrasty, so good enough to bring out the best from that Sony sensor. Raw files can be provided.

 

Uncompressed JPEGs here: https://www.smugmug.com/gallery/n-Jfdr66/

 

α7R III + 50/1.4 Zeiss Planar FE AWB by the camera, Highlights -11, zero NR, not touched otherwise in LR.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 3200 f/2.8 @1/60 sec.

 

 

M10 + Noctilux AWB by the camera, Highlights -11, zero NR, not touched otherwise in LR

ISO 3200 f/2.8 @1/60 sec.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Chaemono]

And two more. Raw files can be provided.

 

Uncompressed JPEGs here: https://www.smugmug.com/gallery/n-Jfdr66/

 

Sony α7R III + 50/1.4 Zeiss Planar FE AWB by the camera, Exposure +2, Highlights -42, AND Shadows +60, zero NR

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 3200 f/2.8 @1/250 sec.

 

 

M10 + Noctilux AWB by the camera, Exposure +1.8, Highlights -42, zero NR

ISO 3200 f/2.8 @1/250 sec.

[M11 for me]

Thank you very much for your comparisons. You conclude that the M10 DR is better. Indeed we can see that at 3200 it is not worse. I am not fully sure but I wonder if we can not see that the M10 puctures are a bit lighter than the Sony pics? I want to put the attention onto the differences in the meaning of 3200 with Sony and Leica. DxO shows a difference of a bit more than ⅓ LV between the 2 cams (at ISO 3200 the „real“ value is approx 2200 resp 1600 with Sony resp with M10). For your camera settings you should then leave the exposure time and the aperture as they are but adapt the ISOs accordingly.

[Chaemono]

Thanks for your feedback. I’ll try to take it into consideration. I’m not concluding anything, yet. I’m just getting started. I’ll look for details in the highlights and the shadows with equal adjustments made to both or none at all in LR. But the pictures above are not ideal for that. They were just the best quick and dirty shots I could do last night after unpacking the α7R III and charging the battery a bit. Hopefully, lots more to follow .

[anickpick]

Thanks for the comparisons.

 

It is in line with my comparisons (Leica SL and Sony A7rii at that time).

 

In my comparisons, I chose same shutter speeds and aperture, so slightly different ISO values resulted, but it did not make a relevant difference in my findings. The cameras have different metering systems too, so I simply pushed the files a bit in post so they were about equal. I used the 50 Apo Summicron on both cameras. While noise levels were not so different, I found the noise patterns in the SL files much nicer and homogeneous.

 

In your comparisons, I also prefer the output of the M10 at ISO 3200, even if pushed Exp. +2 in post.

 

The Sony A7riii will clearly pull ahead if you push the files even more and lift shadows up to +100.

For my uses, such a scenario is not relevant, so I prefer the M10 with its sensor.

However, for some landscape shooters who expose for the highlights and in post push Exp. +4 and lift shadows +100, the Sony is the better choice.

Edited December 28, 2017 by anickpick

[Chaemono]

Advertisement (gone after registration)

Two more, M10 vs. α7R III now with higher ISO for the Sony but probably still not high enough.

 

Uncompressed JPEGs here: https://www.smugmug.com/gallery/n-Jfdr66/

 

α7R III + 50/1.4 Zeiss Planar FE, WB adjusted to try to match M10, exported from ARW in LR

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 3200 f/4.0 @1/125 sec.

 

 

M10 + Noctilux, WB adjusted to try to remove tint, exported from DNG in LR

ISO 2500 f/4.0 @1/125 sec.

[Chaemono]

And now lifting the shadows to +100 for both. M10 in the uncompressed JPEGs of the link upscaled to 37.5 MP. Raw files can be provided.

 

Uncompressed JPEGs here: https://www.smugmug.com/gallery/n-Jfdr66/

 

α7R III, Highlights -15, Shadows +100, NR +30, LR default sharpening

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 3200 f/4.0 @1/125 sec.

 

 

M10, Highlights -25, Shadows +100, NR +30, LR default sharpening

ISO 2500 f/4.0 @1/125 sec.

[Chaemono]

Again, slightly higher ISO for the Sony. As far as I can tell, there is no issue with DR with the M10. In fact, with the Noctilux stopped down a bit, it kicks butt the way it captures the tones and the light.

 

Uncompressed JPEGs here: https://www.smugmug.com/gallery/n-Jfdr66/

 

α7R III, ARW opened in LR and WB adjusted to try to match M10, then exported as JPEG

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 1000 f/2.8 @1/125 sec.

 

 

M10, DNG opened in LR and WB adjusted a bit, then exported as JPEG

ISO 800 f/2.8 @1/125 sec.

Edited December 29, 2017 by Chaemono

[Chaemono]

And now with the Shadows lifted by +100

 

Uncompressed JPEGs here: https://www.smugmug....llery/n-Jfdr66/

 

α7R III, Exposure +0.75, Highlights -15, Shadows +100, NR +30, LR default sharpening

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 1000 f/2.8 @1/125 sec.

 

 

M10, Exposure +0.65, Highlights -25, Shadows +100, NR +30, LR default sharpening

ISO 800 f/2.8 @1/125 sec.

Edited December 29, 2017 by Chaemono

[Chaemono]

And now crops

 

Uncompressed JPEGs here: https://www.smugmug....llery/n-Jfdr66/

 

α7R III

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

M10

Edited December 29, 2017 by Chaemono

[Chaemono]

And two more. How is this for dynamic range? 

 

Uncompressed JPEGs here: https://www.smugmug....llery/n-Jfdr66/

 

α7R III

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 1000 f/2.8 @1/125 sec.

 

M10

ISO 800 f/2.8 @1/125 sec.

[rodf]

Thanks to Chaemono for all of his excellent comparisons and diagnosis. To the unscientific mind and eye the M10/Noctilux scores 95 and the Sony /zeiss 85 on a qualitative basis.  I read somewhere, and it sounds perfectly logical to me, that if you sqeeze 40 megapixels into a 35 X 24 sensor each pixel will be a lot smaller than if you squeeze 24 megapixels into the same area.  Therefore the amount of light that each pixel accepts is less for a 40 megapixel sensor and this may explain the different quality of the images as displayed by Chaemono. I can understand why Leica have decided that 24 Mps is the optimum and I detect no reduction in sharpness in the M10 images as compared to the Sony, in fact the M10 looked a bit sharper to my maybe biased eye, maybe the Noctilux lense has something to do with that but a camera body is not much use without a lense! 

[jmahto]

So what is the conclusion? M10 good enough or equally good?

[anickpick]

Conclusion: The M10 is slightly preferable as long as you stay below ISO 10'000 (including adjustments in post). Both cameras are easily good enough for most uses. So better forget about the DXO scores. There are much more relevant factors when choosing a camera.

Edited December 30, 2017 by anickpick

[Chaemono]

So what is the conclusion?

Why so impatient? I’m not done, yet.

[M11 for me]

Thank you so much for this excellent work and the sharing with us. What you presented up to now is very interesting as i can not see any advantage of Sony up to now (Chaemono wants to go on :-))

 

But still I have a question: We very often read about the very special Leica "somewhat" (rendering etc.). When we look now at the examples above we just come to the "conclusion" that it is just a question of working a bit on the rulers in Lightroom to get results that look very much alike and one can not distinguish the cameras from each other when looking at those pictures (the ones above). This is nothing against or for one brand as the experience of working with Leica M10 is still not comparable with anything else. But still I allow myself to make that point: Are the results of different top of the range cameras just equal at the end? It might be more difficult in Lightroom to get a satisfying result fast enough with one or another cam. My personal M10 experience (10 months by now) is that I am faster with Leica than with Canon: It takes me longer with Canon to get the beautiful result of the Leica. And still when we look at the examples above both cameras render a very, very similar result just ooc resp. with the same LR adjustments.

[M11 for me]

. . . .  I read somewhere, and it sounds perfectly logical to me, that if you sqeeze 40 megapixels into a 35 X 24 sensor each pixel will be a lot smaller than if you squeeze 24 megapixels into the same area.  Therefore the amount of light that each pixel accepts is less for a 40 megapixel sensor and this may explain the different quality of the images as displayed by Chaemono. I can understand why Leica have decided that 24 Mps is the optimum and I detect no reduction in sharpness in the M10 images as compared to the Sony, in fact the M10 looked a bit sharper to my maybe biased eye, maybe the Noctilux lense has something to do with that but a camera body is not much use without a lense! 

I think that your conclusions are not correct without further explanation. Note that you can set your measurements in DxOMark to "Screen" or to "Print". That takes into account exactly that point: It is true though that a smaller pixel captures less light than a bigger one (besides being logical this is a simple fact). Its exactly the same when comparing big or small sensors). And therefore the smaller pixel creates more noise when it has to render the same brightness as the larger pixel and when both get the same amount of light. BUT: We never look at a single pixel but at the whole picture. And all pixels together do not create more noise. On the contrary: As a whole they render more detail (more pixels on the same sensor size) without more noise (DxO "Print" setting). The DxO measurements show this too: The recent full frame sensors with 25 to over 50 Mpixs do not produce more noise than sensors with less Mpix. 

 

Here too: I am completely happy with the 24 from the M10. We learn that high Mpix sensors are more affected by shaking which is an effect that is not suitable for a mostly hand held camera as the M10 is.

[Chaemono]

DxOMark:

“While the M10’s overall score and sub-scores are good, they fall behind the best full-frame image quality we’ve found with the Sony A7R III and the Nikon D850...In fact, the Leica M10’s sensor scores are more on par with the best APS-C chips we’ve tested in mirrorless cameras, such as the Samsung NX500 and the Sony A6300.”

 

Why do I get the sense that we’ll be looking at more pictures, a lot more pictures.

[Advertisement]

Hallo,

Look here: Leica M

[M11 for me]

DxOMark:

“While the M10’s overall score and sub-scores are good, they fall behind the best full-frame image quality we’ve found with the Sony A7R III and the Nikon D850...In fact, the Leica M10’s sensor scores are more on par with the best APS-C chips we’ve tested in mirrorless cameras, such as the Samsung NX500 and the Sony A6300.”

Why do I get the sense that we’ll be looking at more pictures, a lot more pictures.

And then lets publish our new LUF score. DxO is useless.