DXO Mark Leica M10 Score

[M11 for me]

Advertisement (gone after registration)

I mean that Leica enthousiast appreciate their gear, whether from the latest generation or not.

Nikon/Canon people tend to think they can only make nice pics with the latest stuff.

As a Leica enthousiast I appreciate all my gear. But if I had to sell all my gear for a reason or another I would try to keep my 5D MkIV with all my lenses ma flashes and filters.

[jonatdonuts]

We live in a world today where no one is making a bad camera or lens. 

 

If you were to take the same picture with a Sony, Canon, Nikon, Fuji, Leica with similar focal length at the same settings, edit it the same way, and make16x24 prints, most of us would have great difficulty discerning which picture was taken with which camera from our naked eyes. 

 

Image quality has been 'more than good enough' for sometime now that DXO and technical reviewers are now measuring with very, very fine margins that have very little difference on how we view an image once its printed. I mean, how many of us really need 15 stops of dynamic range at base ISO?

 

These days, we are spoilt for image quality, and the added benefits of technology nowadays have practically zero difference as to what makes a good photograph in terms of light, composition, emotion. The art of photography seems to have been forgotten in this discussion. 

 

The only real difference between cameras now is the build quality, ergonomics, and overall usability. And for most of us old soul, manual focus masochists, the Leica M10 gives you great image quality in a small package which is difficult to put down. 

[Advertisement]

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

[pgh]

Everything you say is entirely reasonable. Heck, my A7R2 sits in a closet alongside all of my Leicas when I want big prints or shoot landscapes. 35mm format seems to me to be a jack of all trades and master of none, except maybe the thin DOF wars. However, there's no denying that the files are great.

 

However, as a counter I would argue that if you can't get a decent shot from the M10, it isn't the camera. Most people (including me 90% of the time) don't need more.

There was never a question of being able to make a decent shot with the M10. This is a subjective thing anyways. But yea, it's just as easy/hard as it always has been with any M. 

My comment was in regards to resolution in terms of final print quality - which, at least for me, is pretty important. The difference shows up - I can see it. I wish it were a better sensor. It would be so much nicer if, in return for my $7k purchase, I could at least have a file that's as good/robust as the D800 I got in 2013 or whatever. 

[FlashGordonPhotography]

There was never a question of being able to make a decent shot with the M10. This is a subjective thing anyways. But yea, it's just as easy/hard as it always has been with any M. 

My comment was in regards to resolution in terms of final print quality - which, at least for me, is pretty important. The difference shows up - I can see it. I wish it were a better sensor. It would be so much nicer if, in return for my $7k purchase, I could at least have a file that's as good/robust as the D800 I got in 2013 or whatever. 

 

But a Nikon D5 costs only $400 less than an M10 and it doesn't have the same sensor IQ as your 2013 camera. Surely as Nikon's "flagship" it should have the highest image quality? How does this supposedly "professional" camera have such a lousy sensor? The D5 IQ gets thoroughly trounced by the D850 which is getting close to half the price. What's going on there?

 

Or put another way......

 

Nikon D5. Costs $6500. Scores 88

Leica M10. Costs $6900 Scores 86

Canon 1Dx2 Costs $6000 Scores 88.

 

Three brands. Three "flagship cameras". Same price bracket. Similar resolution. Almost the same scores. Yet only one of them gets constantly bashed because it doesn't compete with the sensor of a D8xx series camera. Well guess what. NO 24mp camera competes with the D850 sensor.

 

None of these score like the D850 because none of them were designed to do what the D850 does. Nor does the D850 do what these cameras do. Apples to apples.

 

Gordon

[M11 for me]

We should NOT look at the overall DxO score as this one is not relevant. We should just look at the noise (SRN 18%) and the dynamic range. Further the graphs should be set to „Print“ rather than „Screen“ as this latter shows the pixel level. What DxO says is that a top APS-C sensor compares to the M10 sensor. This is no bashing just a fact. Note that the big Nikons D5 or Canon are designed for speed. That is another market.

 

Here sensors are compared. This is a very valid discussion. The M10 makes it all up with its wonderful lenses. To my experience an ooc image looks better with the M10 compared to one out of my MkIV. I have to work on an M10 picture much less in LR to get the result that I expect. So I see a difference.

[SilentShutter]

How the heck did they test it without any lens? I click Lenses tested and it says that no lens has been mounted on the camera?!

 

The lens can be a negativ factor in the calculation so it makes sens to test it without it right ?

[M11 for me]

Advertisement (gone after registration)

What strikes me is the inaccuracy of the ISOs (manufacturer ISO vs. measured ISO). I know no other camera with such a high deviation of almost a full stop. As a matter of fact an ISO setting of 3200 on the M10 is only 1600. For a precision instrument as the M10 should be this is very bad news. All cameras have a certain deviation. For marketing reasons this is quite usefull. But almost a full LV is too much. No wonder I said somewhere else in this forum that the M10 is capable for very high ISOs. I believed that the value was correct with a Leica.

 

Maybe DxO had a pre series camera???

[FlashGordonPhotography]

We should NOT look at the overall DxO score as this one is not relevant. We should just look at the noise (SRN 18%) and the dynamic range. Further the graphs should be set to „Print“ rather than „Screen“ as this latter shows the pixel level. What DxO says is that a top APS-C sensor compares to the M10 sensor. This is no bashing just a fact. Note that the big Nikons D5 or Canon are designed for speed. That is another market.

 

Here sensors are compared. This is a very valid discussion. The M10 makes it all up with its wonderful lenses. To my experience an ooc image looks better with the M10 compared to one out of my MkIV. I have to work on an M10 picture much less in LR to get the result that I expect. So I see a difference.

 

My bolding...

 

So what you're essentially saying is that I can compare one 24MP sensor to the D850/A7R3/APSC but not if it's in a D5 body? Interesting......

 

I would argue that the market for the M10 may just be the most specialised of all in modern photography.

 

The M10 is designed with specialised microlenses to perform best with M lenses. DXO doesn't take the micro lenses into account. The M10 is designed as a highly portable camera that works with a set of small prime lenses usually off a tripod. That is also another market.Leica stste they chose 24MP because the camera is used mostly hand held. Heat generation were also considerations to make the newer smaller body.

 

If DXO is simply a measurement of the raw sensor, without any influence from any part of the camera design excluding the D5 and 1Dx2 is BS. In that case the Nikon flagship also compares to a top APSC camera. This is a fact. You can't just discard the D5 because you want to. We're either comparing just the sensors, in which case the M10 is competitive or we aren't, in which case the M10 isn't a competitor to the high res cameras. It's a specialty tool to work with M lenses. If that's the case since it works the best with M lenses it's sensor is actually superior to any other. Is there any digital camera made that performs better with the entire range of current M lenses? See? You don't get to pick and choose your arguments and present them as the whole story.

 

If, as you state, the M10 is judged to be in the same class as a top APSC sensor, so are the vast majority of ALL 24MP 35mm format sensors. How do you justify excluding the D5 and 1Dx2. Because they're big? Maybe because they don't come in silver? We're either comparing sensors alone or in concert with what the cameras were designed to do. DXO compares sensors alone. The D5 and 1Dx2 are comparable to the top APSC sensors.

 

Who, in their right mind considers the M10 and D850/A7R3 as competitors? Manual focus/auto focus. High res/ medium res. large lenses/small lenses. Lots of menu options/super simplified. Video/No video. DSLR/EVF/Rangefinder. Zooms/No zooms. super teles/135APO. Electronic Interface/Analogue controls. $3-4K/$7K. OCF/No OCF. etc. etc. etc.

 

They're just not related but yet the comparisons continue because DXO put the D850 on the web page with the M10 review.

 

Gordon

[FlashGordonPhotography]

The lens can be a negativ factor in the calculation so it makes sens to test it without it right ?

 

Or in the case of a sensor specifically designed to have specialised micro lenses that optimise IQ with it's native lenses maybe the opposite.

 

Gordon

[FlashGordonPhotography]

What strikes me is the inaccuracy of the ISOs (manufacturer ISO vs. measured ISO). I know no other camera with such a high deviation of almost a full stop. As a matter of fact an ISO setting of 3200 on the M10 is only 1600. For a precision instrument as the M10 should be this is very bad news. All cameras have a certain deviation. For marketing reasons this is quite usefull. But almost a full LV is too much. No wonder I said somewhere else in this forum that the M10 is capable for very high ISOs. I believed that the value was correct with a Leica.

 

Maybe DxO had a pre series camera???

 

If DXO could test Fuji sensors you'd see Leica aren't alone. Canon's used to use digital gain between whole stops as a command in the EXIF. we only found that one out because early versions of independent raw converters didn't know that and the files looked darker than in Canon's software. Back in the film days i had two Canon newF1N's. Same model and the meter on one was out by 2/3 of a stop from the other. I just did my own testing and moved on. I do the same today. I don't rely on a website to tell me where noise becomes unacceptable. i think of ISO rating being like VW's emissions reporting.

 

i agree with you but short of having all manufacturers send a camera for independent testing and certification it'll never happen.

 

Gordon

[pgk]

Whatever happened to cascaded MTF graphs? Each factor in the imaging process can be assigned data (lens and sensor) and the cascade indicates the result. They still only give purely technical, theoretical informations and a camera/lens/sensor is a package which offers much more. These discussions are much like comparing a Landrover with a Porsche by assigning them with an overall performance number - hardly relevant in the slightest.

[M11 for me]

Hi Gorden

 

This is all a point of view. And depending on that point of view you can be absolutely right. I just took the big Nikons and Canons out as they are designed for speed and sport. You can now agree that this is a special segment or not. The M10 for many of us in the full frame segment is the alternative of a D8xx or 5D Mk IV. That is my view and it might be personal and some might then not agree.

 

At the end I believe that Leica should be on the top with the others. Of course we like the M10 and with its lenses it make stunning photographs. And its small and elegant and and and. That is why I bought it. I do a lot of landscape with the M10 and it is clear that dynamic range is essential for this type of photography. Lets then put it that way: Our images would be even better with a 15LV sensor.

[M11 for me]

If DXO could test Fuji sensors you'd see Leica aren't alone. Canon's used to use digital gain between whole stops as a command in the EXIF. we only found that one out because early versions of independent raw converters didn't know that and the files looked darker than in Canon's software. Back in the film days i had two Canon newF1N's. Same model and the meter on one was out by 2/3 of a stop from the other. I just did my own testing and moved on. I do the same today. I don't rely on a website to tell me where noise becomes unacceptable. i think of ISO rating being like VW's emissions reporting.

 

i agree with you but short of having all manufacturers send a camera for independent testing and certification it'll never happen.

 

Gordon

Of course with Fuji you are right. This is well known indeed.

[pgk]

At the end I believe that Leica should be on the top with the others. Of course we like the M10 and with its lenses it make stunning photographs. And its small and elegant and and and. That is why I bought it. I do a lot of landscape with the M10 and it is clear that dynamic range is essential for this type of photography. Lets then put it that way: Our images would be even better with a 15LV sensor.

 

Currently I run a photo gallery and there are landscape and other images on display from a large variety of cameras including consumer dSLRs (APS) high MPixel 'pro' dSLRs, & my M9s. By far the most important aspect of any image is its content. Would any images be better shot on better sensors? Perhaps, maybe, possibly - I don't know. Bottom line is that all are 'fit for purpose' and adding more resolution, dynamic range, etc., might satisfy us as photographers seeking perfection but by far the most important thing is to get out and be inspired to take great photos - technical specifications rarely help with this. I'm not suggesting that Leica shouldn't utilise better sensors, but what I am saying is that the fact that they don't really isn't a big deal most of the time. Its far more about the rangefinder package that having a state-of-the-art sensor.

[Steve McGarrett]

So one can't make good landscapes with slides? Landscape photography is not only "point towards the sun and then push shadows by 5EV"... that is a kind of landscape photography (extremely widespread today on the Internet, by the way), but surely not the only one.

Edited December 18, 2017 by Steve McGarrett

[pgh]

But a Nikon D5 costs only $400 less than an M10 and it doesn't have the same sensor IQ as your 2013 camera. Surely as Nikon's "flagship" it should have the highest image quality? How does this supposedly "professional" camera have such a lousy sensor? The D5 IQ gets thoroughly trounced by the D850 which is getting close to half the price. What's going on there?

 

Or put another way......

 

Nikon D5. Costs $6500. Scores 88

Leica M10. Costs $6900 Scores 86

Canon 1Dx2 Costs $6000 Scores 88.

 

Three brands. Three "flagship cameras". Same price bracket. Similar resolution. Almost the same scores. Yet only one of them gets constantly bashed because it doesn't compete with the sensor of a D8xx series camera. Well guess what. NO 24mp camera competes with the D850 sensor.

 

None of these score like the D850 because none of them were designed to do what the D850 does. Nor does the D850 do what these cameras do. Apples to apples.

 

Gordon

Apples to apples would not be comparing sports oriented speed demons with a camera that is not. The most likely replacement/subsitute for Leica M would be smaller body camera with a prime lens that is oriented towards travel, street, portability. Not the "flagship" - Leica makes nothing that compares in any way to the full body D5 or 1dx2 in terms of intended use. Those cameras intentionally sacrifice a bit in image quality to speed. They are also cameras specifically and usually used by photojournalists, who usually see images printed across a magazine spread at largest output. They are plenty good for that, as is the Leica M. With those cameras you get a lot by sacrificing a bit of IQ. ... with the Leica, it's a camera with basically no automation - what is the legitimate reason for not putting in a better sensor? We don't need more speed out of it. The only reason I can think of is that they don't feel like their lenses actually hold up to higher res sensors (maybe why they refer to it as the 'sweet spot') - which is another problem altogether given that less expensive Zeiss lenses render quite well across 42 mp, or they don't want spend the money it would cost to do so, which, at $6800 for a camera seems a little frustrating. Might as well make the camera 7k even and give us a better sensor, at that point the difference in cost isn't that significant. 

 

If you want to go apples to apples I think you take similar size/use cameras - this is an A7r/II/III, an X pro 2, an Rx1rII, a Ricoh GR, an Olympus Pen F, etc - and with those cameras the IQ varies widely. However, at 2x the price of cameras used for similar purposes, the IQ should at least do better than occupy a middling realm imo. 

 

I don't care about the Dx0 score, but I just really think that for what they charge, Leica could offer better IQ. It used to be a big reason people bought in to Leica. For 35mm, you couldn't get a camera that made a better negative (from 28-90mm at least). 

[pgk]

The only reason I can think of is that they don't feel like their lenses actually hold up to higher res sensors .....

 

If you read through many posts and threads on the forum which discuss the advantages of mechanical manual focus lenses together with the inherent disadvantage they have compared with electronically integrated lenses you will realise that things aren't as simple as they were in the days of film cameras. Trying to wring image 'quality' out of lenses and hi res sensors is not as simple as putting the lens in front of the sensor and looking at the image before it has had some processing, and its this processing which shifts the goalposts whether we like it or not. Any 'sweetspot' is determined by a whole load of factors and this isn't a simple scenario anymore.

[FlashGordonPhotography]

Apples to apples would not be comparing sports oriented speed demons with a camera that is not. The most likely replacement/subsitute for Leica M would be smaller body camera with a prime lens that is oriented towards travel, street, portability. Not the "flagship" - Leica makes nothing that compares in any way to the full body D5 or 1dx2 in terms of intended use. Those cameras intentionally sacrifice a bit in image quality to speed. They are also cameras specifically and usually used by photojournalists, who usually see images printed across a magazine spread at largest output. They are plenty good for that, as is the Leica M. With those cameras you get a lot by sacrificing a bit of IQ. ... with the Leica, it's a camera with basically no automation - what is the legitimate reason for not putting in a better sensor? We don't need more speed out of it. The only reason I can think of is that they don't feel like their lenses actually hold up to higher res sensors (maybe why they refer to it as the 'sweet spot') - which is another problem altogether given that less expensive Zeiss lenses render quite well across 42 mp, or they don't want spend the money it would cost to do so, which, at $6800 for a camera seems a little frustrating. Might as well make the camera 7k even and give us a better sensor, at that point the difference in cost isn't that significant. 

 

If you want to go apples to apples I think you take similar size/use cameras - this is an A7r/II/III, an X pro 2, an Rx1rII, a Ricoh GR, an Olympus Pen F, etc - and with those cameras the IQ varies widely. However, at 2x the price of cameras used for similar purposes, the IQ should at least do better than occupy a middling realm imo. 

 

I don't care about the Dx0 score, but I just really think that for what they charge, Leica could offer better IQ. It used to be a big reason people bought in to Leica. For 35mm, you couldn't get a camera that made a better negative (from 28-90mm at least). 

 

I suppose like most people I am influenced by my personal expectations. In the way I'd use an M10 it exceeds, quite easily, the IQ I need. So I think it's brilliant. I shoot my M10 handheld, mostly for spontaneous photography and so the resolution works fine for me. I wouldn't print bigger and I never need more DR when I shoot the M10. In the places I use an M10 I don't run out of resolution or DR. Heck I rarely do that with my Pen F or CL or XPro2. The M10 out performs all of them.

 

If i want more DR or resolution, it'll be when I'm shooting landscapes. And I'd never use an M10 for that because of other limitations that aren't sensor related. In those cases Ill put an X1D on a tripod. Where I use the M10 the IQ is more than fine. If I were wanting more I wouldn't have bought it. When I want insane resolution of huge dynamic range it's for landscapes and the M10 is a crap landscape camera, regardless of the sensor. If it had the D850 sensor it'd still be a crap landscape camera. So I don't get why we'd compare the D850 or A7R3 to an M10.

 

I also have issues comparing most cameras, except maybe a D5 to an EOS 1DX2 or a mid range DSLR to another. I'd never even consider a D850 because it's not mirrorless. So I don't compare it to a Sony A7R3 because I'll never but a D850 regardless of the sensor. I bought a S007 and only use it occasionally because it's a DSLR. Love the lenses. The IQ is fantastic. Still a DSLR. After that I decided never to even consider an OVF again as I know I won't use it. So for me it's just not comparable to a Sony A7R3. I'd not compare an A7R3 to an M10 because the fast Sony lenses are huge and AF. 

 

So, for me, I tend to look at cameras in isolation. Is the M10 going to perform as I want it to under the conditions I intend to use it. Yes. I buy it. No. I don't. I am not cost sensitive with cameras. If it were half the price I wouldn't treat it differently. I have also stopped trying to have a single camera that does it all. I but gear to fill a specific purpose. If I need something different I pick up a different camera.

 

If Leica release a 40MP SL tomorrow I'll be at the head of the list. If they do the same for an M10 I probably won't.

 

Gordon

[pgk]

So one can't make good landscapes with slides? Landscape photography is not only "point towards the sun and then push shadows by 5EV"... that is a kind of landscape photography (extremely widespread today on the Internet, by the way), but surely not the only one.

 

Of course you can, and from negatives! In fact this year I've just 're-'printed a landscape from a Kodachrome slide that I took in 1980 (on a Nikon with a borrowed and damaged 28/3.5). I digitally copied the slide in 4 sections, put them together and finally had it printed better than I've ever been able to before. In the Gallery we represent one photographer who (as far as I am aware) has never owned a digital camera - he works on 6x9 and 35mm. Camera and medium need to support the vision not the other way around.

[Advertisement]

Hallo,

Look here: Leica M

[Steve McGarrett]

Of course you can, and from negatives! In fact this year I've just 're-'printed a landscape from a Kodachrome slide that I took in 1980 (on a Nikon with a borrowed and damaged 28/3.5). I digitally copied the slide in 4 sections, put them together and finally had it printed better than I've ever been able to before. In the Gallery we represent one photographer who (as far as I am aware) has never owned a digital camera - he works on 6x9 and 35mm. Camera and medium need to support the vision not the other way around.

 

Of course, but my sentence was just a (rethorical) way to say that you don't need the absolutely best DR around to shoot good landscape photography... and that there are more things to evaluate in rating a sensor than just DR and ISO (filter stack thickness and microlenses technology, for example... less astigmatism and better sharpness across the frame usually comes in handy for landscapes...)

Edited December 19, 2017 by Steve McGarrett