M10 micro contrast

[PaulJohn]

Advertisement (gone after registration)

I’ve just read on another forum that an owner of the M10 and M240 thought that the micro contrast of the M10 was not as good as the M240.

 

Has anyone else observed this?

 

I have wondered if pushing sensor technology to cater for dynamic range and high iso capabilities has a consequence on other unmeasurable attributes such as micro contrast and tonality.

 

There is a mantra in business that what gets measured gets done. Perhaps this is to the detriment of getting the most from our lenses?

 

 

Sent from my iPhone using Tapatalk

[Jeff S]

That's all fixed with the exposure and contrast controls (and import presets) in PP with the M10. The files straight out of the M 10 seem to lack 'punch', looking flatter and lacking contrast, but that's due to default out-of-camera settings. There's a lot of recoverable data in those files.

 

Jeff

[adan]

I wouldn't go by "what someone thought" - I'd want to see some controlled samples that show a difference (or not).

 

(It's always - comforting - to come up with reasons not to shell out $7000/£5600 )

 

When the M9 came out, many people were certain that it was not as sharp (per pixel) as the M8. Never actually demonstrated.

 

When the M240 came out - with more DR and thus lower contrast between neighboring tones, some M9 users thought the M240 was "not as sharp, per pixel." Also never actually demonstrated.

 

It is inherent with higher DR, that by squeezing in, say, 13 "stops" of tones in place of 10, the individual tonal steps are not as "different" from their neighbors - a 7.7% difference vs. a 10% difference. And are thus a bit less distinguishable.

 

As a former M9 user, at first blush, the flatter, less contrasty M10 files (like M240 files) appear less sharp - per pixel - than the M9. But appearances are deceiving, and by adding back a little global contrast in the mid-tones to M10 images (overriding a bit of the DR), the micro-contrasts perk up as well.

 

Here's an example - from a Leica demo M10 before I acquired my own. First version, straight from the camera. Second version, with auto Adobe "exposure settings" applied, which clipped the blacks significantly (-42). Which reduced the DR and increased contrast - and the eyelashes appear much sharper. No Sharpening or Clarity was applied, BTW.

 

I made this trial precisely because I had heard the same "thoughts" about the M10's resolution and edge contrasts.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Given that the M10 is resolving fine hairs down to a level of 1 to 1.5 pixels (which is about all one can expect from a demosaiced Bayer color image), I doubt the M240 does any better. If I ever get an M240 long enough to do a similar comparison, I will.

Edited December 23, 2017 by adan

[PaulJohn]

I wouldn't go by "what someone thought" - I'd want to see some controlled samples that show a difference (or not).

 

(It's always - comforting - to come up with reasons not to shell out $7000/£5600 )

 

When the M9 came out, many people were certain that it was not as sharp (per pixel) as the M8. Never actually demonstrated.

 

When the M240 came out - with more DR and thus lower contrast between neighboring tones, some M9 users thought the M240 was "not as sharp, per pixel." Also never actually demonstrated.

 

It is inherent with higher DR, that by squeezing in, say, 13 "stops" of tones in place of 10, the individual tonal steps are not as "different" from their neighbors - a 7.7% difference vs. a 10% difference. And are thus a bit less distinguishable.

 

As a former M9 user, at first blush, the flatter, less contrasty M10 files (like M240 files) appear less sharp - per pixel - than the M9. But appearances are deceiving, and by adding back a little global contrast in the mid-tones to M10 images (overriding a bit of the DR), the micro-contrasts perk up as well.

 

Here's an example - from a Leica demo M10 before I acquired my own. First version, straight from the camera. Second version, with auto Adobe "exposure settings" applied, which clipped the blacks significantly (-42). Which reduced the DR and increased contrast - and the eyelashes appear much sharper. No Sharpening or Clarity was applied, BTW.

 

I made this trial precisely because I had heard the same "thoughts" about the M10's resolution and edge contrasts.

 

M10conDRedges.jpg

 

Given that the M10 is resolving fine hairs down to a level of 1 to 1.5 pixels (which is about all one can expect from a demosaiced Bayer color image), I doubt the M240 does any better. If I ever get an M240 long enough to do a similar comparison, I will.

Thank you Adan that is really helpful and very interesting. I will have a play with contrast as by default I tend to reduce it for people shots partly due to reducing saturation. It’s interesting to see how previous editing methods can wrong on new cameras.

 

 

 

Sent from my iPhone using Tapatalk

[Tailwagger]

It is inherent with higher DR, that by squeezing in, say, 13 "stops" of tones in place of 10, the individual tonal steps are not as "different" from their neighbors - a 7.7% difference vs. a 10% difference. And are thus a bit less distinguishable.

 

In my limited experience,  I think this is a key point, particularly when claiming the files of the newer model look flatter, duller, less pop, etc than the previous one.  Increased DR is a somewhat of a double edged sword. It certainly makes capturing more difficult scenes, less so. But the more range you have on the front end, the more work is required to reconcile compressing it to accommodate the limited ability of the backend to display it. No clue how the standard profiles work their magic, or lack thereof, but they seem rather linear, which I assume is intentional as it provides the best starting point for tweaking rather than generically imposing someone else's belief about how the final result should appear.     

[albertknappmd]

Customization is the key point that was so ably brought up in the prior threads.

I always optimize at every iteration (from M8 to 9 to 240 and now.. 10...

Albert