LEICA M9 vs Phase one

[ErikKaffehr]

Hi,

I have it from a book: Science for the Curious Photographer (page 138)

....
I end this section with some estimates of the dynamic ranges associated with various imaging
systems. These numbers have large errors because estimates of noise and visual quality are subjective and
technology continues to improve. As before, one stop represents a factor of two and n stops means a factor
of 2n :
Sensors (DSLR): 10 - 14 stops (RAW readout)
8 – 9 stops (JPG readout)
Color print film: 7 stops
Color slide film: 5 stops
Human eye: 5 - 7 stops (fixed retina -without iris change)
...

It may depend a bit on the interpretation of dynamic range. Also print film can be overexposed a lot it saturates smoothly, and that may be interpreted as extended DR.

Best regards
Erik

I always thought negative film had about 11 or 12 stops of DR, compared to about 6 stops for positive film.

[Slough]

Hi,

I have it from a book: Science for the Curious Photographer (page 138)

....
I end this section with some estimates of the dynamic ranges associated with various imaging
systems. These numbers have large errors because estimates of noise and visual quality are subjective and
technology continues to improve. As before, one stop represents a factor of two and n stops means a factor
of 2n :
Sensors (DSLR): 10 - 14 stops (RAW readout)
8 – 9 stops (JPG readout)
Color print film: 7 stops
Color slide film: 5 stops
Human eye: 5 - 7 stops (fixed retina -without iris change)
...

It may depend a bit on the interpretation of dynamic range. Also print film can be overexposed a lot it saturates smoothly, and that may be interpreted as extended DR.

Best regards
Erik

It is very hard to find reliable sources e.g. from Kodak, or some other source which is likely to be written by trained engineers, rather than cribbed from an unknown source. There are many conflicting opinions online. Here is another quote from a book which supports my statement:

http://books.google.co.uk/books?id=byROCPd...nge&f=false

It looks as if your 7 stops for colour print film refers to the DR that can be captured by traditional wet printing. Many authors say that print film has more latitude, and I think what they mean is that because the DR is so great, you can underexpose/overexpose the print by many stops and still get a usable image. In other words, when you create the print, the DR you get is limited by the print, but you can select from the wide range of DR available in the film. But if you scan with a high quality scanner, you can get more of the DR. That is consistent with tests I have seen:

http://www.janrik.net/MiscSubj/2007/FilmVs...SLR_Images.html

The above looks to have been done carefully and methodically. So I stand by the statement that colour print film has significantly more DR than colour slide film, and a DSLR. The Fuji camera based on the D200 was said to get closer to colour print film.

I don't argue that a DSLR has significantly more DR than colour slide film. I found the latter often useless due to limited DR. But I do think many people overestimate the DR of a DSLR, given that the measures DR depends strongly on how you define acceptable. Maybe 9 or 10 stops, or 11 for the D3/D3x. (No I've not done tests, so take that with a truck load of salt.)

[Anders_HK]

Interesting, I really like the viewfinder of the M8, though I've never shot with it. Virtually all of my work is at about 5 to 15 feet away from the subject and with a wide angle lens so I think the M9 could be ideal. I hear the M9 files are very similar to the M8 files. Any way you could tell me about the different file characteristics of an M8 file when printed at say 10 by 13 inches compared to an Aptus 65 file. I know it's hard.
thanks
-a

Although I too like the viewfinder of M8, that and the camera is small, good for walk around carry. Files from M8 and Aptus 65 are much different. Being a Kodak sensor the M8 sensor renders more like Phase-One, but it is not MFDB rendering, perhaps... simplified, yet beautiful are best words to describe. Compared to Aptus 65, the Aptus provides far more details, dimension, DR etc. Colors are pleasing from both, though I prefer the Aptus for colors, sort of like different films... The Aptus can be print much larger than M8, obvious. Simply different tools.

M9 ?? - This sort of sums it up - http://www.l-camera-forum.com/leica-forum/...three-days.html

The M8 is a great smaller companion to a MFDB. Now if they made a DRF bigger (sensor, body, viewfinders, lenses) then we could talk compare to MFDBs.

Rgds
A


P.S. Anyone looking at M9 thought, check prices of Leica lenses. With M8 I use Leica M8 2.8 (not too expensive) and two Voigtlanders. If only M9 for serious landscape... might be cheaper picking up used MFDB and MF lenses...

[ErikKaffehr]

Hi Slough,

Thanks for the articles! They seem to be well researched.

I was surprised myself at those seven steps. There is a short note at the end of

http://www.janrik.net/MiscSubj/2007/FilmVs...SLR_Images.html

that may explain how the 7 steps DR for print film is arrived at.

I have very little experience with print film tried it, didn't like it end of story. On the other hand I have a lot of experience of BW-film.

What I have seen on digital is that there is a very large amount of data that can be extracted from the shadows, but it clips overexposed highlights. I don't think that negative film is very tolerant for underexposure but very tolerant for overexposure.

This link has some other interesting findings: http://www.clarkvision.com/imagedetail/dyn...nge2/index.html
 
Best regards
Erik

It is very hard to find reliable sources e.g. from Kodak, or some other source which is likely to be written by trained engineers, rather than cribbed from an unknown source. There are many conflicting opinions online. Here is another quote from a book which supports my statement:

http://books.google.co.uk/books?id=byROCPd...nge&f=false

It looks as if your 7 stops for colour print film refers to the DR that can be captured by traditional wet printing. Many authors say that print film has more latitude, and I think what they mean is that because the DR is so great, you can underexpose/overexpose the print by many stops and still get a usable image. In other words, when you create the print, the DR you get is limited by the print, but you can select from the wide range of DR available in the film. But if you scan with a high quality scanner, you can get more of the DR. That is consistent with tests I have seen:

http://www.janrik.net/MiscSubj/2007/FilmVs...SLR_Images.html

The above looks to have been done carefully and methodically. So I stand by the statement that colour print film has significantly more DR than colour slide film, and a DSLR. The Fuji camera based on the D200 was said to get closer to colour print film.

I don't argue that a DSLR has significantly more DR than colour slide film. I found the latter often useless due to limited DR. But I do think many people overestimate the DR of a DSLR, given that the measures DR depends strongly on how you define acceptable. Maybe 9 or 10 stops, or 11 for the D3/D3x. (No I've not done tests, so take that with a truck load of salt.)

[Slough]

Hello Erik

I too have little experience with print film, so this is akin to two blind men discussing the colours and shades in a Turner painting. Anyway, the second link you give used a Sprintscan device to scan the film. This is I think the same as the Microtek device which I once owned. My experience is that it clips highlights. So that might explain why he thinks that colour print film clips highlights, whereas some other people give much larger values of DR. Anyway, I acknowledge that this is somewhat hand waving, as I do not have hard facts about the scanner. Perhaps the only way to solve this is to expose some film with a test image having a colour scale with a large DR, then print the film on paper with a range of exposures, and see how much of the test image is reproduced. I think that wouild be definitive, but tedious to do, and not within my abilities.

Leif

Hi Slough,

Thanks for the articles! They seem to be well researched.

I was surprised myself at those seven steps. There is a short note at the end of

http://www.janrik.net/MiscSubj/2007/FilmVs...SLR_Images.html

that may explain how the 7 steps DR for print film is arrived at.

I have very little experience with print film tried it, didn't like it end of story. On the other hand I have a lot of experience of BW-film.

What I have seen on digital is that there is a very large amount of data that can be extracted from the shadows, but it clips overexposed highlights. I don't think that negative film is very tolerant for underexposure but very tolerant for overexposure.

This link has some other interesting findings: http://www.clarkvision.com/imagedetail/dyn...nge2/index.html
 
Best regards
Erik

[ErikKaffehr]

Hello Leif,

"I too have little experience with print film, so this is akin to two blind men discussing the colours and shades in a Turner painting." This was a good one! :-)

So we dig up a lot of partly contradicting evidence from what seems to be well executed serious research. Hopefully one of scientifically minded friends takes up the gauntlet and tries to sort out things.

I checked on your home page, as your "Leif" sound Scandinavian to me, but it seems that we are no countrymen. Happened to see your film vs. D200 evaluation. Which is quite impressive. I'm in progress on doing something similar on Velvia 67 vs. 24 MP digital.

It's here: http://83.177.178.241/ekr/index.php/photoa...-sony-alpha-900

My impression right now is pretty consistent with yours, but I need to look into more detail and make additional tests to sort out some things.
Best regards
Erik

Hello Erik

I too have little experience with print film, so this is akin to two blind men discussing the colours and shades in a Turner painting. Anyway, the second link you give used a Sprintscan device to scan the film. This is I think the same as the Microtek device which I once owned. My experience is that it clips highlights. So that might explain why he thinks that colour print film clips highlights, whereas some other people give much larger values of DR. Anyway, I acknowledge that this is somewhat hand waving, as I do not have hard facts about the scanner. Perhaps the only way to solve this is to expose some film with a test image having a colour scale with a large DR, then print the film on paper with a range of exposures, and see how much of the test image is reproduced. I think that wouild be definitive, but tedious to do, and not within my abilities.

Leif

[Slough]

Hello Leif,

"I too have little experience with print film, so this is akin to two blind men discussing the colours and shades in a Turner painting." This was a good one! :-)

So we dig up a lot of partly contradicting evidence from what seems to be well executed serious research. Hopefully one of scientifically minded friends takes up the gauntlet and tries to sort out things.

I checked on your home page, as your "Leif" sound Scandinavian to me, but it seems that we are no countrymen. Happened to see your film vs. D200 evaluation. Which is quite impressive. I'm in progress on doing something similar on Velvia 67 vs. 24 MP digital.

It's here: http://83.177.178.241/ekr/index.php/photoa...-sony-alpha-900

My impression right now is pretty consistent with yours, but I need to look into more detail and make additional tests to sort out some things.
Best regards
Erik

I am no more Scandinavian than most English people, which means that if you go back 1,000 years, you will find lots of Norwegian and Danish ancestors. I think the whole film versus digital thing was done and dusted years ago, but I got sick of people continually coming out with the same old misinformation. That idiot Ken Rockwell is still spreading such nonsense.

[BJL]

... in a few years time 20MP+ FF sensors will be standard issue, maybe even in modest consumer grade cameras. And the 'only' barrier to achieving top rate images will be imagination, and hard work.

Imagination and hard work and good lenses. My hope is that sensor resolution will eventually be good enough that lenses are the only significant limit on image resolution, sharpness, etc., since ever better sensors of the same format do not cost much more to make. Maybe 35mm format is getting close with 24MP and zoom lenses.