Sensor in the M-E

If quantitative tests were useless, Leica and Zeiss would not publish MTF and distortion charts. But they do. Full stop.
 
I thought DxO revenues came from DxO Optics Pro raw processing software and a series of specialzed film simultion plug ins. If I remember correctly the DxOMark web site is a means to promote the technical excellence of DxO Optics Pro.

The camera evaluations they publish are essentially derivative measurements thay already make for the R&D conducted in support of DxO Optics Pro. They also have data for lenses.

A few years ago I discovered that DxO Optics Pro outperformed ACR for the DSLR/lens combination I used professionally. In my case the improvements were not practical because my workflow demanded overnight image delivery. Speed and convenience are more important to my clients than a 5-10% improvement in IQ.

I am not motivated to defend DxO nor am I interested in annoying Leica owners. At the same time it is relevant to state that every single method DxO uses is completely documented in mathematical detail. I personally can't see how DxO's results are not objective.

While I usually detest automobile/camera analogies, all DxOMark's data measures is the signal-to-noise ratio of the sensor to raw file data flow. Of course S/N is critically related to dynamic range and bit depth. This is similar to only evaluating the overall performance of sports cars' drivetrains. Of course these tests don't completely characterize any camera or lens. They just objectively measure one critical component of the complicated mixture of objective and subjective factors that comprise the whole of the camera.

If I used digital Leica Ms, I would point of that careful optimization of exposure is typically more important than the technical potential of the camera's S/N. The large number of cameras with higher S/N than the Leica CCD sensors rarely function at their full potential because sub-optimal exposure is the norm Convenience trumps evaluation and thinking about exposure. In general photographers tend to place more faith in the camera automation than their own experience and ability to evaluate and optimize exposure.

Finally, DxO will soon test the new Leica CMOS sensor. Let's suppose this system has the best S/N of any camera evaluated by DxO. Does this mean these results would be silly, unimportant and unauthentic? Should camera buyers ignore those results too?

Brilliant post. Agree with every single word.
 
Most sensors of similar size and generation have similar performance in the field. ISO has indisputably improved, and the effects of antialiasing filters are, for real-world use, exaggerated -- often wildly so.

Far more important than using a specific sensor or lens are careful focus and exposure. A Pentax K5 (first generation) or Nikon D7000 can destroy the IQ of an M9 with a $4000 lens under the right conditions. Under other conditions, the M9 will be better.

The trick is to learn your gear and learn how to exploit its strengths. Always has been, always will be.

The magic bullet is not equipment. It is technique.
 
Most sensors of similar size and generation have similar performance in the field. ISO has indisputably improved, and the effects of antialiasing filters are, for real-world use, exaggerated -- often wildly so.

Far more important than using a specific sensor or lens are careful focus and exposure. A Pentax K5 (first generation) or Nikon D7000 can destroy the IQ of an M9 with a $4000 lens under the right conditions. Under other conditions, the M9 will be better.

The trick is to learn your gear and learn how to exploit its strengths. Always has been, always will be.

The magic bullet is not equipment. It is technique.

Thank You!

Not everything is Lost.
 
A few points after reading this thread:

1. DXO will not invalidate your photography if your camera gets a low score.

2. "Image Quality" has absolutely nothing to do with the quality of a photograph.

3. Everybody gets all worked up when Leica get spoken of negatively, however, again, this will not affect your photography.

4. Leica and especially (some) Leica users go around touting their brand and their lenses as the "best" so I mean really what do you expect? Best is a concept that doesn't exist.

5. Leica is kind of a snobby brand so people love to see them bashed. I'm not saying you're all snobs, but lets be honest, when the product videos involve white gloves and Hermes leather, there is an element of luxury which is a little cringe worthy. I'm sorry that this must be associated with the brand you love, but this is the reality of the situation. However like I said in point 1, this will not affect your photography.

6. STOP ACTING LIKE PEOPLE WHO USE DSLRs ARE INFERIOR PHOTOGRAPHERS! Everybody has a different workflow that they like. Leica only makes one type of camera without a Program mode, and I really don't see what the big deal is. An F100 is absolutely capable of being operated as manually as an MP.

Point 7 is a lot like point 1. Tests like these, and camera reviews, and internet forum posts, or the opinion of anybody in the world have very little to do with your photography. Grab a camera you like and make pictures you like. The only thing that really matters in art and photography is taste. Get killer taste and it will take you further than any lens, sensor, or technique.

:)
 
5. Leica is kind of a snobby brand so people love to see them bashed.

6. STOP ACTING LIKE PEOPLE WHO USE DSLRs ARE INFERIOR PHOTOGRAPHERS!

Self-esteem is not self-respect. While self-respect demands that one be treated as an equal human by another person, self-esteem is based upon the evaluation of one’s achievement by others--and hence enters into a hierarchy. When I hold myself in high esteem, somebody else must be held in low esteem.
 
An interesting assertion, but it is both unsupported and tangential to the present discussion.

Your use of the word unsupported is unfamiliar to me. If you are asking for citational sources, the distinction comes from Luke Gibbons reading Avishai Margalit. Avishai Margalit, The Decent Society, Naomi Goldblum trans. Cambridge & London, Harvard University Press, 1996. Luke Gibbons, “Guests of the Nation: Ireland, Immigration, and Post-Colonial Solidarity” in Traces II - “Race” Panic and the Memory of Migration. Meaghan Morris and Brett de Bary ed. Hong Kong, Hong Kong University Press, 2001: 93 passim.

The distinction between self-respect and self-esteem helps us understand how to avoid points 5 and 6 in the post by SaveKodak. Most of the bashing that goes on here is motivated by self-esteem.

Hope this helps.
 
It does indeed help.

Indeed, that looks like a very interesting book. Gibbons and Margalit seem have drawn a somewhat specialized distinction between self-respect and self-esteem for the purpose of drawing out a specific line of argument. That's fine, and it may even be an argument to which i am sympathetic; but's certainly not a distinction that was (to me, at least) obviously embedded in the plain meaning of the terms.

As an academic, I run up against this problem reasonably often. I find it useful to say up front when I'm using plain language in a technical sense, and I also prefer, where possible, to say whose meaning I'm using.

Anyway, thanks for clarifying.
 
It does indeed help.
thanks for clarifying.

You're welcome :)

As an academic, I run up against this problem reasonably often. I find it useful to say up front when I'm using plain language in a technical sense.

I'm an academic, too!

I have taught in four languages (English, Chinese, French and Japanese) on three continents for an extended length of time, and issues related to language, translation and community are areas of specialization. "Plain language", like "common sense", is a baldly ideological construction, and research reveals (perhaps surprisingly) that it is most often related to various forms of exclusionary politics, cultural nationalism, and/or the division of labor (of which technocratism is a part).

I'm used to speaking to audiences that are linguistically, culturally, etc., heterogeneous, and I've found that what works better than assumptions about "plain language" is rather an attitude of openness to the multiplicity of vocabularies, accents, and mixtures in every situation.
 
Of course they have an agenda, they want to drive web traffic to their site that sells of DxO Pro raw software and film plug-ins.

It is also fair to say DxO Mark focuses on the S/N ratio of the data written to the in-camera raw file. So their agenda is to measure and compare S/N ratios. The sensor S/N is a major factor in determining the dynamic range.

Not even DxO claims their analyses represent all you need to know to make purchase decisions.

Unless I am mistaken, you are not thinking their results are inauthentic or intentionally misleading.
 
. . . 6. STOP ACTING LIKE PEOPLE WHO USE DSLRs ARE INFERIOR PHOTOGRAPHERS! Everybody has a different workflow that they like. Leica only makes one type of camera without a Program mode, and I really don't see what the big deal is. An F100 is absolutely capable of being operated as manually as an MP.

Point 7 is a lot like point 1. Tests like these, and camera reviews, and internet forum posts, or the opinion of anybody in the world have very little to do with your photography. Grab a camera you like and make pictures you like. The only thing that really matters in art and photography is taste. Get killer taste and it will take you further than any lens, sensor, or technique.
Highlight 1: Why? A lot of them are. Otherwise they'd use Leicas, 11x14 inch Canhams, or whatever other camera is best for the job, rather than whatever's expensive, idiot-proof and heavily advertised.

Highlight 2: Only with difficulty.

Seriously, I agree with the second paragraph completely. I'd substitute 'skill' for 'taste', but as this is too easy to confuse with 'technical skill', maybe your wording is better. Which makes you wonder why any of us spends this much time trying to persuade people of this.

My excuse is that people are well served by reading suggestions that don't fit the 'received wisdom'. Even if they decide to follow the herd, at least they may have thought about why they are following the herd -- possibly because what suits the herd suits most people, which is why there's a herd.

Cheers,

R.
 
Since this discussion has fully derailed into sensor wars, I'll give my two cents.

There are certain shooting conditions that favor cameras or lenses with certain technical qualities. Low light photography needs a high S/N ratio for ISO 3200+ and lenses with large apertures, wild life photographers need a long lens for many animals, architectural photographers want low distortion and high sharpness, etc.

If you know you want to do one of these things, you should evaluate potential gear with that role in mind. I quite like aspects of the M9, but as much of my photography is in low / available light, it's not a camera I would buy (at current prices). The M240 has a much better high ISO performance, and as such I'm thinking about getting it (very much down the road). Before M9 owners jump on me and mention all the nice portraits they've taken in low light, I should specify that this is my subjective judgement. Others think the M9 is good enough, but I do not.

Of course, none of those technical qualities will result in better photos. They simply remove barriers to the photos you might create. We've all had photos "lost" because we didn't focus fast enough, or because our film was too slow (or fast!), or due to other technical quirks. A poor craftsman blames his tools, but an even poorer one pretends they don't matter.
 
Highlight 1: Why? A lot of them are. Otherwise they'd use Leicas, 11x14 inch Canhams, or whatever other camera is best for the job, rather than whatever's expensive, idiot-proof and heavily advertised.

Highlight 2: Only with difficulty.

R.

Cheaper cameras do not necessarily go hand in hand with cheaper personal vision. If there was a camera that was truly idiot proof then everyone would be exceptional. This is not the case. Some talented people can only afford to use the camera they have, or have time to use on that particular assignment.

Highlight 2: If you don't know how to use an F100 that's okay. Putting it in manual is as easy as taking the M8.2 off of 'snapshot' mode.

I wish this forum was a little more supportive.
 
You realize that an el-cheapo Nikkor 50 1.8 surpasses any old Leica lens which are known for their "magic" while the Nikkor has "aberations".

There's a limit to how "magical" the Leica lenses can be. In this digital age, the sensor really holds it important place.

My 1.8 AiS long nose is pretty sweet, but I prefer my f/2 Ai's aberrations when wide open. ;)

All these lens/camera tests attempt to quantify ever smaller differences between products in a manner that will cause consumers to go out and buy the latest unit, regardless of what he/she is currently using. Many reviewers feel compelled to much hyperbole, in part, I believe, to keep the supply of new product to review coming from manufacturers who know well how to play the reviewing landscape.

For many this means quantifiable stuff in exchange for unit dollars. You start talking about un-quantifiable stuff and value per dollar gets pretty cloudy, even though anybody who has ever owned a camera for very long knows that a good deal of the attraction for cameras and the photographs they produce involves un-quantifiable stuff.

Leica is certainly facing challenges in today's marketplace: Price of labor, their scale of production, their insane build quality goals. It ends up being very expensive to produce a very tiny package of superb quality, with the price padded (certainly) to establish some sort of exclusivity. Big Deal. The big two do this as well. If you're old enough to remember, Nikon once sat on their arrogant haunches too, but only until Canon came up and stole about a third of their customers, and they have been at each other's throats ever since. It has resulted in some pretty amazing devices. Certainly not very good 'investments', not even the lenses, by older measures. But there's a declining rate of return going on if you look closely, responded to with feature creep. CMOSIS has a 70MP sensor out, now. Why aren't Nikon and Canon using it? Because it's a supply chain chess game, that's why. Besides, I will submit, the importance of the sensor is only for the pixel count: Everything else about its 'signature' can be corrected/created in post. Eventually, in-camera.

All of this chatter re leica's quality and price, mine included, is pointlessly academic. Buy and/or use what you want, and move on.

s-a

PS - I could see Leica producing a digital MP, entirely mechanical except for a replaceable sensor. You would send the camera in to Solms for a sensor upgrade and CLA because the rest of the camera is nonpareil.
 
Last edited:
To me its matter of choice... In this case here, there is no other choice other than the M9/M-E platform until perhaps early next year when the new M is more widely available.

I want to shoot a rangefinder but digitally - film is just too much of an inconvenience. My style of shooting and anticipating moments require a rangefinder. I tried using x-mount bodies and sony focus peaking but the dynamics are all wrong. You can live view all you want but I *need* the rangefinder module.

They can talk all about DSLRs, MFs, digital backs providing better quality but that is not how I shoot.
 
I thought DxO revenues came from DxO Optics Pro raw processing software and a series of specialzed film simultion plug ins. If I remember correctly the DxOMark web site is a means to promote the technical excellence of DxO Optics Pro.

I am not motivated to defend DxO nor am I interested in annoying Leica owners. At the same time it is relevant to state that every single method DxO uses is completely documented in mathematical detail. I personally can't see how DxO's results are not objective.

DXO makes its real money from licensing image optimization algorithms to cameraphone makers, i.e. how to compensate for the lousy optics and sensors used in those cameras. The RAW processing software is a sideshow.

Disclaimer: their former VP of Engineering is a classmate. That said, the DXO staff is mostly composed of maths PhDs from a country (France) where the best students gravitate towards maths, and they really know their imaging algorithms, to a far higher degree than even camera manufacturers. Adobe is probably the only other firm with that level of expertise.

The DXO score is a weighted average of scores in each dimension of image quality they evaluate. The priorities implicit in the weighting may not match what you and I weigh as most important. It doesn't help that some of what is considered sharpness, acutance, is a subjective factor.
 
Back
Top Bottom