Leica and DxOMark {MERGED}

[A Sfeir]

Advertisement (gone after registration)

"Not everything that counts is countable, and not everything that is countable counts"

Albert Einstein

[Advertisement]

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

[Lindolfi]

In Dutch there is an expression "Meten is Weten" or "To Measure is to Know", which is used by many people and it is completely misleading in my opinion.

 

To measure is to obtain data. Data can never be knowledge.

 

Knowledge is obtained by going through many empirical cycles in which understanding is slowly increased. Measurements can (but do not need to) have a place in those cycles, but are not knowledge in themselves.

 

But is is fun to do measurements (at least I think it is fun) and if it helps to understand experience (for instance why do I experience images from a certain camera to be muddy), that's great.

[newyorkone]

DxOMark, that's Ken Rockwell in a laboratory.

 

LOL. So true.

[colonel]

LOL. So true.

 

In fact they are complete opposites

 

Ken Rockwell is opinion, DXOMARK is measurement

[kdriceman]

In fact they are complete opposites

 

Ken Rockwell is opinion, DXOMARK is measurement

 

Not really. By applying a numerical ranking using a mostly arbitrary weighting of their measured data, they are expressing an opinion. But they are more misleading than KR because their ranking is based on measured data so the reader thinks it is an objective rating. We all know KR's reviews are subjective. I find some value in their data, but no value in their ratings.

[CaptZoom]

Disclosure: I am not a DXO Mark user. Last time I seriously checked any of their results was several years ago when I was planning my first major digital camera investment (5Dmkii), and ended up deciding the results did not provide me with anything I found useful.

-- --

DXO Mark's advantage is consistency in their methodology and ranking. Within the context of their parameters, the results are useful and the ranking meaningful (the only variable that changes is the sensor). If I remember correctly, DXO is very clear that the results of their measurements are not to be confused with a camera review or to be used as the ultimate criteria for one camera's superiority over another. I believe they include a similar disclosure with every sensor test.

 

Clearly, there are many here who find little, if any use, for DXO Mark's results. And, there are many who do in fact find these results to be meaningful. What I don't understand is all the negativity that surrounds any discussion about DXO Mark's results. Like any other resource, DXO Mark is available for those that are interested in that sort of thing. And even though I do not make use of this resource, I appreciate that it's available. It is good to have resources available, especially those that try to minimize subjectivity and enhance consistency.

[tashley]

Advertisement (gone after registration)

My 2 cents.

 

Firstly, you absolutely need to understand how DXO tests and what meaning they claim for their results.

 

Secondly, I have never had a camera whose IQ threw up a surprising DXO score. Great IQ cameras have great DXO scores and poor IQ cameras have poor IQ scores. The M9 has great IQ as judged from certain very specific angles only: it has no AA filter and is paired with great glass so its low ISO files look very sparkly. That look is increased by the fact that it has relatively ho-hum DR, which gives the files a superficial 'pop'. Some people, me included, like its colour but no one would ever claim that it has enough colour depth to be accurate rather than attractive. It has poor ISO performance by any modern standards.

 

When I first saw M files, really saw and played with them, and then shot a bunch of my own, my first thought was 'these are notably less good than a Sony RX-1 or D600" and I mentally gave it a provisional DXO of low to mid 80's.

 

Bottom line, if you use a lot of gear, then I think it rapidly becomes clear that DXO measurements are a very good correlation with what one sees on one's own screen or prints.

 

Regarding the M score: it is 'good enough'. Crucially, it has the same DR as a Phase One P65+ and for my money, DR is the most important aspect of a modern sensor as long as the colour depth and ISO performance are OK. It might not be a D800 or an IQ180 or an RX-1 but it is still capable of serious, serious images and if you want a proper rangefinder, it is the best there is. And if you want to get the best from much of the best glass there is, it is the only game in town.

[dwbell]

As there is sloppy manual focusing there is also sloppy auto focussing.

 

My 2c, modern SLR/DSLRs are capable cameras but getting best out of AF system requires sometimes bit more than selecting auto everything and point & shoot. I managed many sharp & well focussed images from 85mm f1.4 AFD fully open so it is possible. You could call it automated manual focusing - using single point AF instead of multi pattern focusing.

 

Indeed, as do I. Although with the canon version 85mm f/1.2

 

You are right and Jono is right. I shoot Canon AF a great deal and with static subjects this is true. In a dynamic environment I prefer the RF focus because if I'm focusing through a narrow opening in say, moving people, I know I've got what I want with the RF even if something moves in front of my subject. Or, often I will pre focus on a subject so I can wait for the right moment and then put the camera down (or redirect) so I'm not standing there with a camera pointed at a subject, waiting. Then when you get that right moment, and quickly recompose, you don't want to risk the AF picking up something behind or in front of the subject. There are ways to handle it with an AF system, (I do it all the time when shooting sports) but I find it is just a little more inconvenient or adds an additional step than the very simple and straightforward RF.

 

Sure, agreed. Back button focus covers most of what you describe. Lock's it, leaves it there, doesn't change when you re-raise the camera to your eye.

 

But both of this suggests that these, mostly seasoned, photographers are comparing range finder use to ALL AUTO ENGAGED DSLR use? With all focus points lit up, in green box mode, with AF and AEL on the same button etc? Surely that can't be right? And if it is it's about as useless a negative criticism of AF as there can be?

 

Someone throw me a bone here? I still don't get it? How does AF on a modern DSLR mean you, the photographer, are not selecting what the camera focuses on?

[jaapv]

I've just revisited DxO, and can anyone explain to me how colour depth can be measured in decimals of bits? What is 0.7 or 0.3 bit??

[Gerard]

This bit, that bit... to be frank; I don't give a 'bit' about DxO shenanigans.

 

[mjh]

I've just revisited DxO, and can anyone explain to me how colour depth can be measured in decimals of bits? What is 0.7 or 0.3 bit??

Based on their noise measurements they are calculating how many distinct hues can be resolved (which is limited by noise) and express this number as a binary logarithm, i.e. a logarithm to base 2. That is, 2 ^ colour depth gives the number of hues that can be distinguished.

[Peter Branch]

I've just revisited DxO, and can anyone explain to me how colour depth can be measured in decimals of bits? What is 0.7 or 0.3 bit??

 

Perhaps in the same way that Leica M lenses are ~7.5 bit coded. (6 bit X 3).

[kdriceman]

Indeed, as do I. Although with the canon version 85mm f/1.2

 

 

 

Sure, agreed. Back button focus covers most of what you describe. Lock's it, leaves it there, doesn't change when you re-raise the camera to your eye.

 

But both of this suggests that these, mostly seasoned, photographers are comparing range finder use to ALL AUTO ENGAGED DSLR use? With all focus points lit up, in green box mode, with AF and AEL on the same button etc? Surely that can't be right? And if it is it's about as useless a negative criticism of AF as there can be?

 

Someone throw me a bone here? I still don't get it? How does AF on a modern DSLR mean you, the photographer, are not selecting what the camera focuses on?

 

 

 

I can't throw you a bone here dwbell. I have to agree with you. AF on modern DSLRs is pretty advanced and it takes some effort to learn and use it effectively. Maybe some don't go to the effort and have a misunderstanding of how to make your AF do what you want. I do find it more awkward in a lot of cases than MF with a rangefinder, but I can still do whatever I want with AF.

[pgk]

Someone throw me a bone here? I still don't get it? How does AF on a modern DSLR mean you, the photographer, are not selecting what the camera focuses on?

It doesn't. Both manual dRF and AF dSLR cameras are precision focusing instruments. Used intelligently they are both very capable - although the 85/1.2 Canon can be a bit hit & miss in my experience as its very unforgiving, and unless you have the luxury of using 10x in Live View, the precise, absolute point of focus can be hard to determine.

[dwbell]

It doesn't. Both manual dRF and AF dSLR cameras are precision focusing instruments. Used intelligently they are both very capable - although the 85/1.2 Canon can be a bit hit & miss in my experience as its very unforgiving, and unless you have the luxury of using 10x in Live View, the precise, absolute point of focus can be hard to determine.

 

Indeed. At f/1.2 I'm tripod mounted in available, or using flash, and I don't focus recompose, just because of the limited DoF. Instead I shoot to crop in post - in order to avoid the bullseye composition. But up at f/2 - /f2.8 territory she's a tameable (if heavy and huge) beast.

[furcafe]

I think the frustration comes when someone chooses to rely on the pattern/matrix focusing, where the camera can choose to focus on something other than what the photographer intended. Like matrix metering, it is only as good as the databases that are programmed into the camera & while it may be good for many, even most, situations, it can fail miserably in tricky conditions. Yes, of course, you can put the camera into single-point focus mode, but then the speed advantage over manual RF focusing, at least for those of us who are good at it, is rather minimal, especially when compared to the size/weight penalty of the dSLR body. A similar reasoning may apply to metering & setting exposures; yes, you can go manual, but if the automation isn't helping you, it can all seem a bit superfluous.

But both of this suggests that these, mostly seasoned, photographers are comparing range finder use to ALL AUTO ENGAGED DSLR use? With all focus points lit up, in green box mode, with AF and AEL on the same button etc? Surely that can't be right? And if it is it's about as useless a negative criticism of AF as there can be?

 

Someone throw me a bone here? I still don't get it? How does AF on a modern DSLR mean you, the photographer, are not selecting what the camera focuses on?

[dwbell]

I think the frustration comes when someone chooses to rely on the pattern/matrix focusing, where the camera can choose to focus on something other than what the photographer intended. Like matrix metering, it is only as good as the databases that are programmed into the camera & while it may be good for many, even most, situations, it can fail miserably in tricky conditions. Yes, of course, you can put the camera into single-point focus mode, but then the speed advantage over manual RF focusing, at least for those of us who are good at it, is rather minimal, especially when compared to the size/weight penalty of the dSLR body. A similar reasoning may apply to metering & setting exposures; yes, you can go manual, but if the automation isn't helping you, it can all seem a bit superfluous.

 

I'm surprised but it seems to be as I suggested above. That, comparing all DSLR focus points on to manual RF centre patch then? Yeah. RF wins every time. Poor use of one system is worse than correct use of the other? No surprise though, for me. As to the other opinions of speed, weight, size, etc I wasn't addressing those and don't feel they relate my question.

 

Ok, that's pretty much cleared up my confusion on that opinon. Many thanks and sorry to veer a little OT.

[furcafe]

Well, with all due respect it's not all about "[p]oor use of one system" v. "correct use of the other." Speaking from personal experience, I shoot a lot of music shows, parties, etc. in bad lighting & ended up getting a D700 because of its better high ISO performance compared to the M9, not because of its other capabilities. I had no problem focusing the M9 in bad light, but wanted the option of shooting at ISO 3200-6400. So speed, weight, & size are relevant if you're using a dSLR, correctly, like a RF or old-school manual focus SLR. This is why some of us are more excited about the improved high ISO performance of the M 240 than any of the other new features (&, to bring things back on topic, why the DxO ratings have some relevance to real world use of cameras).

 

I'm surprised but it seems to be as I suggested above. That, comparing all DSLR focus points on to manual RF centre patch then? Yeah. RF wins every time. Poor use of one system is worse than correct use of the other? No surprise though, for me. As to the other opinions of speed, weight, size, etc I wasn't addressing those and don't feel they relate my question.

 

Ok, that's pretty much cleared up my confusion on that opinon. Many thanks and sorry to veer a little OT.

[Advertisement]

Hallo,

Look here: Leica M

[dwbell]

Furcafe - we're doing that Internet thing of talking past each other and ignoring each others points. My specific question was how users find a modern DSLR somehow autonomously decides "for itself" what to focus on. Specifically where a RF camera focuses on "what I want." This being completely contrary to my experiences using both systems in all manner of conditions.

 

As I've learned, it doesn't. Unless you allow it to. So to me, the question is answered with user skill / knowledge.

 

To this limited comparison on autofocus 'overriding' the photographer and 'choosing what it wants to focus on', size, weight, speed, ISO etc are irrelevant.

 

In case it helps I'm on the list for the M240 and own one M9 and one M9P with 4 lenses.

[furcafe]

Yes, we are in agreement as to the proper use of autofocus. And I strongly suspect we are also in agreement that some systems are better than others for certain conditions/jobs.

 

Furcafe - we're doing that Internet thing of talking past each other and ignoring each others points. My specific question was how users find a modern DSLR somehow autonomously decides "for itself" what to focus on. Specifically where a RF camera focuses on "what I want." This being completely contrary to my experiences using both systems in all manner of conditions.

 

As I've learned, it doesn't. Unless you allow it to. So to me, the question is answered with user skill / knowledge.

 

To this limited comparison on autofocus 'overriding' the photographer and 'choosing what it wants to focus on', size, weight, speed, ISO etc are irrelevant.

 

In case it helps I'm on the list for the M240 and own one M9 and one M9P with 4 lenses.