Leica M Monochrom - saving my bad exposures

It is not just the interpolation artefacts; the optical aberrations caused by the color transitions in the Bayer filter play an important role as well.
borge said:
You base your posts on your own assumptions and not facts.
I have several 5D3 and X-Pro1 files that I have worked with extensively previously, which is some of the latest and greatest cameras on the market with modern CMOS sensors. One of the sensors (X-Pro1) isn't a full frame sensor, but that doesn't really matter at all.

By pushing the files from these sensors you get a whole lot of interpolation artifacts and other artifacts as well as very visible banding and noise even at base ISO (100 for the 5D3 and 200 for the X-Pro1). Compared to the files from the MM the effective useable quality for print and high quality presentations on the MM files are far better. The useable latitude is far greater.

I'm not comparing sensors either. The sensors themself doesn't interest me at all. The fact that the files from the MM doesn't need to be interpolated in the same way as every other camera has to be (due to CFA's) is probably the main reason for the difference.

I base my articles on my own experience, and the freedom and useable latitude that I have experienced with the MM so far is incredible and really surprising as well. I don't care if it's CMOS, CCD, old sensor, new sensor, old camera, new camera, or whatever. The result is what matters, and that is the point of my articles. So far the files from the MM has surprised me in regards to how far you can push the files in post-processing compared to other cameras that I have taken several tens of thousands of exposures with (X-Pro1, 5D3, 5D2, 60D, 40D).
Click to expand...
 
Sigh. Any of these cameras exceed the needs of 99.9 percent of us. Come on. ISO 40,000? My 1DMkIV can do that, but I've never had any need, even in some serious darkness.
 
The need to record color information is a huge handicap when modeling analog signals via digital means. The color models certainly work, but they have to represent continuous frequency information along with amplitude and spatial information. Color models have more variables and more complexity. But this has little to do with the signal-to-noise ratios of shadow regions.

Authentic shadow portrayal depends primarily on the signal-to-noise ratio in the shadow regions. The SNR is primarily determined by optimum exposure. Put another way, the SNR difference between camera systems can be small compared to signal loss due to underexposure error. Throwing away half the available signal has a profound effect. Unneeded underexposure will make any camera look worse than it needs to. Letting uninteresting, non-essential highlights clip is the key to maximum shadow quality. This approach insures the best possible shadow SNR. When every highlight must be accurately digitized to produce an aesthetically acceptable image, then the system noise becomes critical because the shadow-region signal can not be increased. Signal loss due to the Bayer filter array affects both the highlights and shadows equally, so removing the CFA only increases the total exposure and does not selectively increase the shadow SNR.

ISO is important to provide flexibility in the choice of aperture and shutter speed, but ISO does not inherently determine exposure. ISO does not determine the signal level for the shadow regions... exposure does. The signal gain from removing the CFA increases the flexibility of shutter and aperture selection. This is valuable since a higher shutter speed (or more DOF) is possible. Exposure flexibility is a good thing

I would be shocked of the OP would not be equally impressed by the shadow regions in a well-exposed image from a D800, among others. I say this not to disparage the M9M. The M9M's images speak for themselves. Instead, my point is we now enjoy stunning performance from all high-performance camera systems. No doubt there are subjective, aesthetic shadow differences that depend on both the camera system and how the image is rendered. Stiil, excellent shadow-region quality is widely available provided the photographer properly exposes the frame.
 
willie_901 said:
The need to record color information is a huge handicap when modeling analog signals via digital means. The color models certainly work, but they have to represent continuous frequency information along with amplitude and spatial information. Color models have more variables and more complexity. But this has little to do with the signal-to-noise ratios of shadow regions.

Authentic shadow portrayal depends primarily on the signal-to-noise ratio in the shadow regions. The SNR is primarily determined by optimum exposure. Put another way, the SNR difference between camera systems can be small compared to signal loss due to underexposure error. Throwing away half the available signal has a profound effect. Unneeded underexposure will make any camera look worse than it needs to. Letting uninteresting, non-essential highlights clip is the key to maximum shadow quality. This approach insures the best possible shadow SNR. When every highlight must be accurately digitized to produce an aesthetically acceptable image, then the system noise becomes critical because the shadow-region signal can not be increased. Signal loss due to the Bayer filter array affects both the highlights and shadows equally, so removing the CFA only increases the total exposure and does not selectively increase the shadow SNR.

ISO is important to provide flexibility in the choice of aperture and shutter speed, but ISO does not inherently determine exposure. ISO does not determine the signal level for the shadow regions... exposure does. The signal gain from removing the CFA increases the flexibility of shutter and aperture selection. This is valuable since a higher shutter speed (or more DOF) is possible. Exposure flexibility is a good thing

I would be shocked of the OP would not be equally impressed by the shadow regions in a well-exposed image from a D800, among others. I say this not to disparage the M9M. The M9M's images speak for themselves. Instead, my point is we now enjoy stunning performance from all high-performance camera systems. No doubt there are subjective, aesthetic shadow differences that depend on both the camera system and how the image is rendered. Stiil, excellent shadow-region quality is widely available provided the photographer properly exposes the frame.
Click to expand...

Willie,

Thanks for the carefully written thoughtful post.

Cal
 
willie_901 said:
I would be shocked of the OP would not be equally impressed by the shadow regions in a well-exposed image from a D800, among others. I say this not to disparage the M9M. The M9M's images speak for themselves. Instead, my point is we now enjoy stunning performance from all high-performance camera systems. No doubt there are subjective, aesthetic shadow differences that depend on both the camera system and how the image is rendered. Stiil, excellent shadow-region quality is widely available provided the photographer properly exposes the frame.
Click to expand...

I never said I wouldn't be impressed by a D800 file. My comparisons and what I say is only based on my experiences, and I have never had a chance to try a D800, so I really can't say anything about it other than that I've read that it has a great sensor.
 
brbo said:
In your opinion, how much better is DR in Monocrom compared to Nikon D800?
Click to expand...

I don't know, I have never tried a D800, or any recent Nikon at all for that matter (I think the last one was the D300 or something, like 4-5 years ago).
 
Bo,

I didn't mean to imply you had any opinion of the D800. I was trying to say your shaow/underexposure recovery experience with the M9M is one many of us have had with numerous cameras. I had just read a post elsewhere by an interior photographer who was shocked at how much shadow detail he could recover from raw files from a new D800 while keeping the windows just barely overexposed.

I apologize if my anbiguous writing bothered you.
 
jaapv said:
It is not just the interpolation artefacts; the optical aberrations caused by the color transitions in the Bayer filter play an important role as well.
Click to expand...

Jaap, I think you made a similar point over on the LUF recently. Could you elaborate a bit more on this as I'm not yet convinced your statement isn't just voodoo dressed up in a bit of technical language.
 
Ian, it was discussed at the introduction of the MM. It appears more logical than voodoo to me. If you pass light through a piece of glass made up of mini kaleidoscopic parts it would be more surprising if there were no adverse effects.
 
Pickett Wilson said:
Sigh. Any of these cameras exceed the needs of 99.9 percent of us. Come on. ISO 40,000? My 1DMkIV can do that, but I've never had any need, even in some serious darkness.
Click to expand...

That's only true if you use digital cameras like the way you use a film camera e.g when instincts kick in to use 1/15 f2.8 at ISO 1600 on streetlights... but if you need creativity, you can maybe try shooting at somewhere dim at F22 at ISO 40,000
 
jaapv said:
Ian, it was discussed at the introduction of the MM. It appears more logical than voodoo to me. If you pass light through a piece of glass made up of mini kaleidoscopic parts it would be more surprising if there were no adverse effects.
Click to expand...

Ok. I thought you were suggesting that the CFA was exaggerating already existing lens related optical aberrations such as chromatic aberration.
 
You must log in or register to reply here.
Back
Top