Sean Reid's M9/MM/M240 Comparison

[Bob]

Funny weird stuff from M8 with IR filters?
Or maybe just a bad case of boot oder

-bob

 

[douglasf13]

Ha! In that case, I think the green blob adds to the pic. Great, Bob!

Anyway, sorry to get so off topic.

 

[Jeff S]

I never said it didn't happen, just that it hasn't happened to me. And even if I had gotten a case of 'green boots,' I'd have to weigh the frequency against the otherwise obvious benefits. Each photographer needs to make judgments based on his/her own use and experience; hence my recommendation to Douglas to try it and see.

Jeff

PS At least we know her jacket wasn't black.

 

[jstaben]

If one is inclined to upgrade to new "M" then those folks will likely believe and continue to feel that CMOS and CCD are at parity.

If one is inclined to keep their M9/M8 etc then those folks will likely believe that CCD is better.

blind tests on web are meaningless. You could show two images and have a hard time telling Monochrom vs. m43 camera B&W and we all know there is a difference in performance there, for example.

My feeling is that there are definitely characteristics of different sensors. One is not better than the other and that's why choices are good. I have an M9P and D800E and although D800E is technically a better sensor I often find the images on the M9P more pleasing. Also using primes/Zeiss etc on Nikon.

 

[Double Negative]

Sooner or later, every M8* user will get a green blob. Nature of the UV/IR filters and both Leica's and Heliopan's filters do it. Typically when there's a strong specular light. In these situations you have four choices:

1) Suck it up
2) Clone it out
3) Pull the filter
4) Get an M9

 

[Bob]

Shoot an MP

 

[Jeff S]

In these situations you have four choices:

1) Suck it up
2) Clone it out
3) Pull the filter
4) Get an M9

5) Convert to b/w.
6) Don't shoot in those conditions.
7) Keep it and post it on the web for all to see

Jeff

 

[edwardkaraa]

Inspired by the CCD vs CMOS discussion, I revisited my archives and browsed many files, all shot with Carl Zeiss lenses, on Canon 5D and 1Ds2, Sony A900. What strikes me the most is the almost organic quality of the M9 files and the "roundness" of shapes, while I found the shapes by all other cameras quite flat looking. Flat not in terms of contrast (well, that too), but flat in that they lack the roundness (not to be confused with 3D) of the M9. Very interesting discovery for me.

 

[CharlesK]

I am fascinated about the CCD vs CMOS discussion also, I think there is a good case for now at least to keep the M9 where possible, while having the improved ergonomics and functionality of the M240. At base ISO's, I feel the M9 will become the classic, sought after look. With the improved firmware and profiles the new M240 will look similar, and in reality maybe very hard to pick the difference on the computer screen.

Having the RX1, FF Sony's latest CMOS sensor, the IQ, DR and low light ability have amazed me. At base ISO's, the M9 has the definite clean, rich and full look that the RX1 files seems to lack, even with extensive PP'ing.

 

[algrove]

That makes me think the Phase One MF digital backs might be sought after for many years with the P45+ being the long exposure leader. It's needed when shooting ISO 50. I've never even tried ISO 100 with it.

 

[edwardkaraa]

Fully agree, Charles.

I am fascinated about the CCD vs CMOS discussion also, I think there is a good case for now at least to keep the M9 where possible, while having the improved ergonomics and functionality of the M240. At base ISO's, I feel the M9 will become the classic, sought after look. With the improved firmware and profiles the new M240 will look similar, and in reality maybe very hard to pick the difference on the computer screen.

Having the RX1, FF Sony's latest CMOS sensor, the IQ, DR and low light ability have amazed me. At base ISO's, the M9 has the definite clean, rich and full look that the RX1 files seems to lack, even with extensive PP'ing.

 

[doobooloo]

This is a very interesting discussion. I have never really known or thought about the CCD vs CMOS factor until recently with the M240's release, but what I have known very intimately is whether I like the look out of a certain camera or not. I don't need to know *why* one looks better versus the other... And the why applies to both the "supply" side (sensor, lens, etc.) and the result side (technical differences in output files).

But it's fascinating that without thinking about the "why" many of us are emotionally more responsive to outputs from one camera versus the other. And perhaps that's all that matters.

Despite the perceived technical superiorities, I never really fell in love with Nikon D3/D3S files, and the same with Canon 5D Mk2. However I felt a lingering attraction to Sony's A850's outputs and much more so with those from the M9. Not sure why, and pixel peeping and stretching the limits of RAW files in Lightroom I did not feel that one was superior in versus the other in terms of amount of information contained in the files, but the way they're presented... Despite the horrid auto WB on the M9, I often still preferred the look versus similar shots out of the Nikons and Canons.

I admit, this could all be in my head, since Leica as a brand and the pretty shape of the M9 command a significant emotional premium to start. And I've seen plenty of photos from others that I can form a strong and lasting emotional attachment with that were shot with Nikons and Canons. So, I dunno. What I do know is that I look forward to opening up the files from my M9 much more so than I did with any other camera, and I guess it doesn't really matter why.

That said, of the two photos above, I immediately and instinctively preferred the first one and was hoping that was the M9 file and was relieved to find out it was so.

 

[BANKER1]

Excellent points. My first toe into the waters of MFD was with the Hassy CFV on my 500 C/M. If the files were opened in Photoshop they were converted to DNGs. The DNG files were awful. They were dull, dark, and the color was horrid. In Flexcolor the files were absolutely stunning. So, it seems, Hasselblad had worked hard and provided "out of the box" images that needed only slight alterations. If you think about it, the 35 mm camera makers give you "out of the box" images that enhanced the positive attributes of the 35 mm systems. Conversely, MFD manufacturers provided "out of the box" images that brought photographers to the Medium Format systems in the first place. So, I'm not sure there is a CCD versus CMOS "look". Now, I also know, that there are inherent differences in the sensors that allow the manufacturers to only go so far, and camera manufacturers have chosen the sensors that allow them to take advantage of their individual systems.

As I understand it, the Leia sensor is a European version that should contain the most recent technical innovations. So, if they are presumably working with a clean slate, they may just be able to bring us the most useful advantages of the RF system, using top notch lenses, into the new M.

Greg

 

[kuau]

Well looking at the Buy/Sell forum, all I see are M9's for sale now.....
Am I to assume that everyone is going for the new "M"

 

[jstaben]

Well looking at the Buy/Sell forum, all I see are M9's for sale now.....
Am I to assume that everyone is going for the new "M"

Well you have thousands of people wanting an M, many are M9 owners. You will see a lot on the forums. Now is a good time to buy an M9! Long term who knows, maybe they will rise in price!

 

[Amin]

I think the point was that its the temptation to calibrate a CMOS sensor to have fantastic high ISO which compromises it rather than the actual sensor technology. AFAIK there are no modern CMOS sensors which do not do this. The new CMOSIS sensor is much less ambitious (as was the sensor in the Sony A900).

I understand your point about sensor tuning beyond intrinsic sensor technology itself, but the chosen base ISO of 200 in this camera represents a decision to prioritize high ISO performance at the expense of low ISO performance. Don't get me wrong - I'm sure the low ISO performance is, for all practical purposes, wonderful. I also don't doubt that it will be better than the M9 at all ISO values. However, my understanding is that - all other things being equal - a sensor designed with a lower base ISO will have better image quality at base ISO than a sensor designed with a higher base ISO.

I was told by a sensor expert that a 4/3 sensor designed with a base ISO of 25 would have the same dynamic and tonal range and same signal/noise at that ISO as a 35mm format sensor with a base ISO of 100. If paired with high quality prime lenses that excelled at f/2.8, such a camera could be an incredible compact and lightweight tool for landscapers. However, such sensors do not exist, presumably because they would sacrifice low light, high ISO image quality.

 

[georgl]

The actual sensitivity of a sensor is defined by sensel area (they "collect" photons during the exposure - 4x area -> 4x more photons) and quantum efficiency (basically translating light into electric current) - the signal can then be amplified (in an analogue stage or digitally) to the desired higher "sensitivity".

The quality of implementation (electric interference etc.) and the amplifier/ADCs varies, but generally speaking, an efficient sensor with high basic sensitivity will offer superior IQ (regarding noise and to a certain degree DR) at any ISO setting.
The rather different amplification/conversion-technology in the CCD-world (external, 1-4x amps/ADCs) might lead to a faster degradation of IQ compared to modern CMOS (on-chip, column-based).

What can be compromised in favour of higher sensitivity is the density of the CFA, using more "transparent" filters which reduce "color precision". Other than that, a low-noise sensor (+ additional circuitry) will ALWAYS be preferable.

 

[MirekE]

What can be compromised in favour of higher sensitivity is the density of the CFA, using more "transparent" filters which reduce "color precision".

Is there any way to see or verify the higher "color precision" of such sensors? I have seen several indications that Leica (M9, specifically) uses more selective filters and I wonder how it would be reflected in actual images. I don't want to open a can of worms here, but one of the DxO numbers is about color depth and the collected numbers are more or less the same as competition, which rises my curiosity.

 

[douglasf13]

Is there any way to see or verify the higher "color precision" of such sensors? I have seen several indications that Leica (M9, specifically) uses more selective filters and I wonder how it would be reflected in actual images. I don't want to open a can of worms here, but one of the DxO numbers is about color depth and the collected numbers are more or less the same as competition, which rises my curiosity.

Here are some great posts from Joakim (thesuede,) who works in the industry:

a850 vs a900, 1 stop better noise performance? - FM Forums

 

[MirekE]

Thanks for taking the time and finding this in old posts on FM for me. TheSuede has also been a great source of information and I was not aware of these particular posts. Very helpful!