DxOMark trashes the Leica M9 sensor

[Hobby Shooter]

Good cameras do not become bad bcoz DxO "trashes" them ...

That is a tautology.

When buying a (new) camera, I dont have access to complete information about their "goodness". I might read subjective tests, measurements, or hire the cameras for a weekend or more. I still won't be certain which is the right for me, or which will allow me to take the better pictures for the next 4-5 years.

I dont have the skills, patience or equipment to do what DXO does. Certainly not for every camera on the market. Therefore I use their excellent data to aid me in choosing a camera.

I'm not challenging or disproving (nor am I capable of) what DxO claims, but I've never bought a camera or lens based on what DoX says. Their tests/claims are irrelevant to me.

How can you be so certain? Do you ever see noise in your pictures?

-h

That is not a tautology

[CarlTN]

The low score of the Leica CCD doesn't surprise me.  But, I do think DXOMark's test results have always been skewed (as do most of us Canon people). 

And I still don't understand why so much emphasis is placed on dynamic range, color fidelity, and low noise...at low ISO levels.  On any modern full frame camera (regardless of brand...well maybe except for Leica, haha)...the noise is so low below ISO 800, that it's extremely silly to worry about it.  Even the noise on my 5 year old XXD body is so low below ISO 640, that it's almost absurd to think about.  Autofocus performance and lens sharpness are really the only things left to think about, at the low ISO's.  I get basically no detail loss at all, due to any obsessive NR I might dial in, in this range.

I mean, if we're going to worry about the noise and dynamic range of a sensor at the lower ISO range, then we're really talking about small, point-and-shoot sensors (and to some extent 4/3 sensors)...but definitely not huge full frame sensors.

What counts most of all for me, is the noise performance, and how I can maximize it, above ISO 1000 (since I enjoy doing nature and wildlife shooting, and don't use strobes very often).  And starting around this range or a little above it...all of Canon's full frame sensors, pull ahead of all Nikon's sensors save one, the D4 (or the older and "better" D3s with its bigger pixels).

Perhaps sensors will see a revolution of improved performance soon...or perhaps not.  I'm still mostly happy with what I've got now, as humble as it is. 

[fussy III]

[/quote]


If a camera is allready asserted to be "good", then claiming that some measurement cannot change that assertion is, in my understanding, a tautology.

-h
[/quote]

There is often a discrepancy in language between logic and meaning. Personally I prefer to subordinate logic to meaning. If we do this in the case treated here, we must first step outside of the inner logic of words that have been combined. If we do this, the sentence treated might read as follows and does not represent a fitting example of a tautology:

"A camera to date regarded by a wide usership as a good camera will not be regarded as a bad camera (by whomever) from now on just because DXO ...  "

The art of language often lies in brevity from the speaker's side and in the ability or willingness of a recipient to make correct preassumptions. And sometimes it is just a matter of well-meaning. 

[Maui5150]

Further proof of the irrelevance of DxOMark scores. 

[lol]

Wow, so much hatred for DxO? They're telling us nothing that hasn't already been widely reported throughout the years. The M9's sensor wasn't state of the art when it was new, and that was a lot of years ago now.

I think it does show for many people, you don't need to always be on the cutting edge. Once things get good enough, you don't look to upgrade unless something useful comes along with it. So even if the M9 sensor isn't the best, it does the job well enough for most.

As for DxO sensor scores as a whole, I do not think it fair to outright say their tests are wrong. They are only testing some factors and weighing them a certain way. Unless your needs exactly match those, then their overall scores will be of limited value. I believe their sub tests do show useful data but I wouldn't normally look at the overall score.

[Alrik89]

Who gives a s**t about DXO results?

[David Hull]

Interestingly, the low marks didn't seem to have affected the brand's prestige or sales.  The same might be said of Canon's marks vs Nikon. It would seem most people view DxO Mark as little more than a scientific curiosity.  Not something you'd base a serious purchasing decision on.

Most people, even most photographers, probably does not even know about DXO mark. I believe that primarily people reading forums like this knows about DXO mark.

The DXO measurements (and similar data produces by others) affect my equipment choices. But they are far from my only source, since they cover only a part of what makes a camera "good" to me.

I think that DXO have seriously shifted landscape photographers towards Nikon. Those who chose a 5Dmk2 in the previous generation might have chosen a D800(e) today partially motivated by DXO. Unless their lense needs (or investement) keeps them back.

-h

True for newbies, perhaps but anyone else especially someone with a bit of understanding and some experience in digital photography knows that the differences between what the two brands are offering are relatively minor and easy to work around.

[7enderbender]

This is not the first time that such claims from this dxo company seem unbelievable. I don't have the physics background to dive into this. But this seems rather a very singular approach to what is clearly a sum of multiple parts. I don't own a Leica (unfortunately) and I don't own a Micro 4/3rd either. But the real life results that I see are pretty obvious in favor of the Leica system. And I'm just talking about the various test shots that I've seen and should not include the "Leica buyers are more serious about photography" factor which may or may not be true.

Maybe the Leica sensors are technically "inferior" in this or that way but I still like looking at the results some folks are getting. And I have to admit that I wish I could afford the M system. It would work well with my preferences.

[Rienzphotoz]

Good cameras do not become bad bcoz DxO "trashes" them ...

I'm not challenging or disproving (nor am I capable of) what DxO claims, but I've never bought a camera or lens based on what DoX says. Their tests/claims are irrelevant to me.

How can you be so certain?
-h

Yes, I am absolutely certain bcoz I never bought a camera or lens based on what DoX or some other xxx company says or claims and their tests/claims are irrelevant to me.

That in no way proves that their claims are irrelevant to you. It only perhaps indicates that you are ignorant of their results. Not the same thing.

Do you ever see noise in your images?

-h

You seem to be under some misguided impression that you know more than me, about what is relevant to my needs.

[David Hull]

True for newbies, perhaps but anyone else especially someone with a bit of understanding and some experience in digital photography knows that the differences between what the two brands are offering are relatively minor and easy to work around.

I think that I have a some understanding and experience in digital photography. While I agree that for general photography, the differences tends to be minor, when the shot depends on a large capture DR, I find the work-arounds to be clumsy, time-consuming and possible shot-ruining.

Not something that newbies or professionals would want to do if they could avoid it.

-h

I think that I have a some understanding and experience in digital photography.

As do I and I have never found the so-called DR issues of any of the Canon products to be a significant limitation -- nor have countless thousands of others who use the stuff on a day-to-day basis for both professional and amateur work.
 
The particular value of this feature (like any other) is clearly dependent on what you shoot and how you shoot it (as you seem to allude).  Over the years of discussing this stuff, and having significant personal experience with the gear in question, it has been my observation that most of these differences are significantly over hyped by individuals who seem to have specific agendas to promote.  This is probably not the case with you, but there has been plenty of it about.

[David Hull]

This is not the first time that such claims from this dxo company seem unbelievable. I don't have the physics background to dive into this. But this seems rather a very singular approach to what is clearly a sum of multiple parts. I don't own a Leica (unfortunately) and I don't own a Micro 4/3rd either. But the real life results that I see are pretty obvious in favor of the Leica system. And I'm just talking about the various test shots that I've seen and should not include the "Leica buyers are more serious about photography" factor which may or may not be true.

Maybe the Leica sensors are technically "inferior" in this or that way but I still like looking at the results some folks are getting. And I have to admit that I wish I could afford the M system. It would work well with my preferences.

You seem to be doing the error of believing that "if camera A has better DR than camera B, then camera A must produce better images. If it does not, then DR must be wrongfully measured".

This is not so. People make wonderful images using low-DR capturing.

Just like a 36 MP camera does not in itself make better pictures than an 18 MP camera. The photographer makes pictures, the camera facilitates. A high-resolution camera can enable some images that would be hard to do on a 18 MP camera.

-h
-h

That sums it up right there IMO, there seem to be way too much focus on minutia, IMO.

[RS2021]

Dxo and Ken Rockwell make photography what it is..."entertainment". Don't complain.
I have always loved the form factor of the classic Leica...low key and classy but good performers at the same time.

[Rienzphotoz]

You seem to be under some misguided impression that you know more than me, about what is relevant to my needs.

I don't know, neither do I claim to know what is relevant to you.

If you choose to be ignorant, neither do you.

-h

Just bcoz I do not worship DxO or share your interest does not make me ignorant. But I do agree with you that you "don't know" what is relevant to me.

[CarlTN]

People are using exposure bracketing to extend the Dynamic Range. This indicates to me that quite a few people think about DR at low ISO, and would like to have more of it without having to resort to exposure bracketing.

Both are relevant aspects of a camera for most users. Lens sharpness is thoroughly reviewed on the net. Autofocus is hard to review objectively, and sadly, we are often only fed the subjective opinion of some reviewer.
-h

I'm not sure why you're questioning my own thoughts about what I'm personally concerned with, and implying they're irrelevant.  When I mentioned lens sharpness and autofocus performance, I was saying that these are the only aspects of performance I personally am left with when shooting at low ISO, as noise or dynamic range don't concern me down there, because it's already very good.  Perhaps read what I said again...No need to nitpick, is there?

Regarding dynamic range at low ISO...I disagree with you.  I assume you are heavily into HDR work?  People will always exposure bracket to produce HDR images, no matter how big a dynamic range the sensor can achieve.  Why?  Because they like doing it.  They will always want to maximize the number of bits they get in a final HDR image, to ease their own mind, and to essentially use as a creative toy later.  But at the end of the day, what they achieve merely becomes an 8 bit "painterly" cartoon before it gets printed or displayed anywhere as a jpeg.  It's still not preserving dynamic range in reality as the eye/brain saw it (or in the future, as some very superior sensor saw it.)  It's artistically impressionistic of reality.  In the future, if there are display devices and software that can accurately display 32 or 48 bits of dynamic range (and they are widely adopted...and people actually enjoy viewing pictures/video on them)...ONLY THEN will maximum sensor dynamic range (at low ISO) be truely a valid issue...in my opinion.  This would require a contrast ratio orders of magnitude beyond even the claimed "10 million plus" of OLED displays, etc.  It would also require the display device's ability to reproduce luminosity equal to that of the sun, while also being able to reproduce a black that is literally zero light output.  I won't hold my breath on this...nor will I quibble over the difference between 16 or 13 bits of dynamic range in RAW files...because mostly I produce prints.

Even if future sensors do achieve 18, 20, 24, 32, or 48 bits of dynamic range...most people will probably be paying $1k on photo software plugins that will squeeze the dynamic range back down so it gets represented on a print.  It will still look cartoonish...and the more elements in the picture that actually stretched the dynamic boundaries, the more cartoonish it will look.  Sure it can sometimes look great and artistic...but it's still not representing what your eye/brain saw, the way it saw it.

[Rienzphotoz]

But I do agree with you that you "don't know" what is relevant to me.

I have never claimed to know, so this is just you fighting strawmen.

-h

Since you think I'm "fighting strawmen" why are you replying