Sean Reid's M9/MM/M240 Comparison

[edwardkaraa]

Afaik, CCD and CMOS are like slide and negative film. They work in opposite ways, and curiously their output does resemble slide film (CCD) and negative film (CMOS). CCD images to my eyes have a certain density and depth only visible with slide film, while CMOS produces light and superficial color very much like negative film. Both are good and produce excellent results, but are fundamentally different.

Now what differentiates color response of similar sensors in different cameras are the CFA design and demosaicing algorithms. The density and the hue of the color filters have a direct influence on the color output and separation characteristics. Of course how the raw converter decides to interprete this information plays also an important role.

So is it possible to replicate the CCD look of the M9 with the M CMOS sensor by controlling the CFA characteristics and the demosaicing algorithms? IMHO, not really, but I'm no expert. Obviously there will be varying opinions, but I guess we will know soon enough when the first batches hit the shelves.

 

[ptomsu]

I think all this discussion about CCD versus CMOS is something happening in the minds of people coming from different camps (different sides).

Truth is that both use digitized information resulting from light falling onto different technology photo sites and then being processed and calculated for the final image afterwards.

As long as all information on color, DR, brightness, etc. is contained coming from either sensor, it is only a matter of the final software (which is stored in FW in case of cameras) what final look an image will have and what reserves are still in the resulting OOC image to compensate for any weakness (brights, darks, etc.)

And then there is of course also post processing which allows for further adoption to the final style one likes, but actually the images should already come OOC in a way so that there is almost no further PP needed to achieve the optimum look as the scene really was if all in camera (FW and handling of photography parameters by photographer) are handled in the right way.

Having said that - the only reason that CCD might look different than CMOS is because of the SW/FW is not mature enough.

 

[Double Negative]

I'm actually looking into this CCD vs. CMOS issue right now. Pretty much since the beginning, I've noticed that my Canon DSLR files look different than my Leica files. Different how? Less contrast, saturation - pop. What Ed said above has a bit of a ring to it.

Both cameras are set to produce flat files (RAW/DNG anyway) and I do everything in post. While my PP recipe has evolved, it hasn't changed much. It's not the camera resolution despite the Canon being 8.2MP, the M8 10.3MP and the M9 18MP. I do chalk it up to:

Evolution of ACR (Canon on older versions, Leica the latest)
Lack of an AA filter on the Leica bodies
CCD vs. CMOS, perhaps?
Leica prime vs. Canon zoom

But the one thing that stands out to me is ACR. It's easy enough to test and that's exactly what I'm going to do. Re-process older Canon files in the latest ACR and according to how I PP my files today.

 

[algrove]

IMHO, Ming Thein seems to have extracted his old CCD look out of the M with CMOS.

Perhaps ACR helps as DN suggests, but Ming has an uncanny way that brings out the best of most sensors from what I can tell.

 

[Jeff S]

Describing the way the sensors work really doesn't have any bearing on these positions. It all hinges around the intent of the processing of the information and whether this is more important than the data from the sensor.

And there are even more variables, e.g., the presence or lack of AA filter, the actual design of the CCD or CMOS sensor (not all are created equally), including use of micro-lenses, etc, etc.

Frankly I don't know or care too much about most of this, anymore than I care about how my car engine and other internals work to achieve its performance. The articles written about it, by reputable sources, are interesting, but at the end of the day, I do my own test drives and evaluate accordingly. And I can't make that judgment until the car (or camera) is actually available for sale.

Kudos to Erwin Puts (here) for reserving judgment on the M's performance until the final version is complete and relevant RAW developers are available.

Jeff

 

[Vivek]

That (Puts) review is one that will read and look forward to.

It does not even need a subscription! :thumbs:

 

[Double Negative]

...It does not even need a subscription! :thumbs:

Yeah, I'm so done with Reid Reviews. I subscribed for two years... But that stupid Flash interface has got to go. Since it didn't - I did.

 

[fotografz]

IMHO, Ming Thein seems to have extracted his old CCD look out of the M with CMOS.

Perhaps ACR helps as DN suggests, but Ming has an uncanny way that brings out the best of most sensors from what I can tell.

I think he's a good photographer, and there is some nice sense of rich color to the shots he posted ... different than M9 files, but nice in their own right.

However, I also feel many of the M shots looked to crispy and slightly over-processed. Perhaps a difference in tastes, but I personally have a slightly uneasy feeling of "trying to hard" in post.

I still tend to gravitate to Sean's more neutral or native approach when initially viewing potential, since we all have different goals.

The caveat is that the M isn't done yet ... so it's all a bit premature ... fun, but premature.

-Marc

 

[Guy Mancuso]

My opinion but nothing should be done in post when doing any review. WB maybe a touch of sharpening but always on a neutral playing field.

 

[Paratom]

The funny thing is that I bet > 80% of people in this discussion will probably own a new M in 1 year from now. (I am afraid I wont be part of the 20%)

However regarding the opperational advantages-I never had a real problem with the speed of the M9, the shutter noise and it never broke due to bad weather.

 

[Bob]

CMOS vs CCD,
lets say it depends a lot on mainly four variables.
1) the spectral response curves used for the bayer filters (or whatever filters) are used
2) The way that noise is controlled in the specific implementation.
3) profiles.
4) date that the design team froze the process (they are all constantly evolving)

Anyway I am eager (Guy, please no name calling) in getting my hands on one of those 240's.
If I can get my eye surgery to early enough I definitely will try once more an un-assisted RF.
Why?
Because I liked them.
-bob

 

[jstaben]

Honestly I think there is something nice about having cameras that produce a different look out of the camera. My Nikon D800E files look a lot different than my M9P files. Much more dynamic range. Sometimes that's good, and sometimes it's a bit too much. Sure I can reduce it in post, but there's something nice about not having to go into every file and work them in post to give it that. When I use my Monochrom though it's nice to have that flexibility to push the pixels around. My opinion is those limitations of the M9 CCD sensor produce some of those unique qualities. It's nice to have choices and some will prefer one vs. the other!

 

[Jeff S]

My opinion but nothing should be done in post when doing any review. WB maybe a touch of sharpening but always on a neutral playing field.

For some aspects, yes. But it would be nice to learn in an MM review, for instance, that the files are 'robust' and 'malleable' without introducing problems in PP. One can't know this without extensive PP adjustments.

Anyway, I rely on reviews more for technical aspects than for any screen shots or IQ assessments, which: (a) may be based on some aesthetic judgments by the photographer that I might not agree with; and (b) may have no relevance to real prints, particularly ones based on my own print size, style and preferences.

Jeff

 

[MCTuomey]

I decided mid-last year to save for an MM to pair with my M9 (still wait listed on the MM). Mostly because I need 1-2 stops better high iso for some club shooting I do and plan to increase. Now on the fence with the M 240 likely to arrive about the time my ticket for the MM will be up. Do I want the better of two worlds and use two cameras to get there or do I want the lesser of two worlds and use just one camera? Or is the practical image quality difference so small that better/lesser is a non-issue?

I need my own experience with the M 240 and the MM. This is going to be expensive. Sometimes I really wish I disliked rangefinder shooting ...

 

[jstaben]

I decided mid-last year to save for an MM to pair with my M9 (still wait listed on the MM). Mostly because I need 1-2 stops better high iso for some club shooting I do and plan to increase. Now on the fence with the M 240 likely to arrive about the time my ticket for the MM will be up. Do I want the better of two worlds and use two cameras to get there or do I want the lesser of two worlds and use just one camera? Or is the practical image quality difference so small that better/lesser is a non-issue?

I need my own experience with the M 240 and the MM. This is going to be expensive. Sometimes I really wish I disliked rangefinder shooting ...

I got the option to pick up an MM and promptly cancelled my M240 preorder. Couldn't be happier. Kept my M9P. The low light performance with the MM is so nice I have no need for the M240 now. If you want color low light then the choice is obvious.

 

[woodyspedden]

Excellent discussion, and Sean's review was up to his superb standards. I came away leaning heavily toward getting the MM and keeping my M9. The MM out resolves it's siblings, has the best high iso, and I believe will produce images at low and medium iso that outshine the M because of the added malleability of it's files, especially pushing shadows. This aspect is not covered in Sean's review which only looks at out of camera rendering.

+1
Woody

 

[Guy Mancuso]

CMOS vs CCD,
lets say it depends a lot on mainly four variables.
1) the spectral response curves used for the bayer filters (or whatever filters) are used
2) The way that noise is controlled in the specific implementation.
3) profiles.
4) date that the design team froze the process (they are all constantly evolving)

Anyway I am eager (Guy, please no name calling) in getting my hands on one of those 240's.
If I can get my eye surgery to early enough I definitely will try once more an un-assisted RF.
Why?
Because I liked them.
-bob

Hell I want one also . Actually either one would be nice . I'm a sick bastard too

 

[jonoslack]

Hell I want one also . Actually either one would be nice . I'm a sick bastard too

Guy - Bob
semantic arguments about CCD/CMOS are just one thing. Lovely cameras with delicious silky shutters, lovely lenses and a rangefinder are just the other.
There really is no escape :facesmack::chug:

 

[Vivek]

Preaching to the choir, Jono? :ROTFL:

Bob, Hope your lens replacements will go smoothly.

 

[jonoslack]

Preaching to the choir, Jono? :ROTFL:

Bob, Hope your lens replacements will go smoothly.

Absolutely Vivek - except that I know you aren't a member

. . . and like you I'd like to wish Bob all the best with his eye surgery

all the best