"The complete color gamut?" What's that?Originally Posted by Sirius Glass
"The complete color gamut?" What's that?
Brian Ellis
Before you criticize someone, walk a mile in their shoes. That way when you do criticize them you'll be
a mile away and you'll have their shoes.
This is a complicated question.
When you scan, you are adding a step which has a certain resolution attached to it, You can translate from ppi or dpi (pixels or dots per inch) to lp/mm (line pairs per mm) by dividing by two. (Each line pair requires two piixels.) But every other step in the process also has an associated resolution. Combining different steps results in a LOWER resolution than any individual step.
To calculate the combined resolution, you have to compose MTF functions, which is hard to do. Most people use one of two rules of thumb. The most commonly used rule is to sum the reciprocals of the individual resolutions to get the reciprocal of the combination. Another rule used by some people is to sum the reciprocals of the squares of the individual resolutions to get the reciprocal of the square of the combination.
When you take a picture, the lens has a certain resolution and the film also has a resolution. The film resolution is determined in part by the coarseness of the grains, but other factors also play a role. The net result for modern large format lenses on typical films is unlikely to be greater than 60 lpm. When you scan, even with the best scanner, you are going to reduce that further. If you then print, that will reduce the resolution further.
You have to compare that to enlarging and printing or contact printing, which also has a resolution.
Since there are so many factors to compare, one is unlikely to be able outside a laboratory to be able to measure all these factors and get an accurate result for the final resolution. So it is best to judge the results you get by eye to see if you can see any difference. But if you do so, keep in mind that each print should be viewed at the proper distance, depending on its size. Looking at a 16 x 20 print from a few inches away is going to show defects no matter how you go about it.
My experience is that I can make up to 16 x 20 prints by scanning, which viewed from about two feet, are indistinguishable from prints I made years ago with an enlarger.
As to color, my experience is that scanning and massaging the image in a photoeditor is much superior to using an enlarger and printing on color paper. I even had a color head and an expensive color analyzer, and I never got anything that I found completely satisfactory. In particular the state of the solutions used in processing the print plays a role. A photo processing lab can do a reasonable job keeping those constant, but typically there is some variation. If you are willing to spend enough, you can get good results, but rou tine printing is not up to the same standards. Printing on a good inkjet printer, in my experience, is more likely to produce consistent results. In addition modern inkjet printers and printing papers resist fading better than conventional printing papers.
One thing to keep in mind is that transparency films have a limited latitude. One is usually much better off using color negative films if one wants a large range of colors.
The complete color gamut is what the eye sees. See reference: http://en.wikipedia.org/wiki/Gamut
Digital camera and digital scanning of the color gamut:
http://dx.sheridan.com/advisor/cmyk_color.html
http://www.google.com/imgres?imgurl=...ed=0CEcQ9QEwBA
Once the information is lost, no printing process can bring it back. Example: take a black & white photograph on film. The color information is lost. Printing it on color paper or using a digital photographic printer will will not bring back the colors.
The point is that color information is lost by using a digital camera or scanning film and it at cannot be brought back once lost. The OP asked:The answer is that there are quantifiable losses. The question is Are these losses acceptable?I haven't studied this one much, but given one has a top flight scan of a well exposed neg, how much of that neg gets taken away from the scanning/processing of it into a file? In other words, say we look at the neg on light table...how much of the color/tones/shadows/information/etc. gets lost in the scanning process "and" does the scan actually make the neg look more "digitized" and not as "realistic" as the neg does?
Steve
Nothing beats a great piece of glass!
I leave the digital work for the urologists and proctologists.
Steve,
I am flabbergasted by your complete lack of understanding on this topic - and your propensity to post repeatedly about it. I went to all your references. They often referenced unrelated topics like the CMYK space, I had to work to find any reference to scanning. None of them showed related information about the color gamut of PMT's.
The answer is that the loss is in the category of "almost none". It is not like photographing a b&w print. PMT's can suck the marrow out of a piece of film. My clients constantly send me thank you notes and emails.
You really ought to get out and see some good prints.
Lenny
EigerStudios
Museum Quality Drum Scanning and Printing
As Leonard Evens has intimated quantifying any comparison between digital scanning and optical printing could be a bit of a mathematical Tour de Force. But thinking about it a bit does help.
Both flatbed scanners and optical enlargers can transfer image information via a lens or in the case of a PMT based scanner by a beam of illumination that can sample a potentially very small section of the film.
The flatbed scanner and optical enlarger can be essentially identical, visually, if the optics of both are of comparable resolution (typically in the 5 to 10 um airy disc resolution limit for white light. There is obviously some quantization with a flatbed due to the discrete nature of the sensors and optical distortion associated with each individual pixel as replicated in the digital data stream. But the nature of the distortion is not so materially different than occurs within the resolution limit of a broad area enlarging lens of equivalent resolving power. Both will produce a sort of Gaussian intensity profile within the resolution limit but the intensity profile is lost in the flatbed in going from sensor through the A to D conversion process.
A PMT (Photo Multiplier Tube) based scanner is potentially a different beastie. The very best of these are capable of very small spot sizes (say about 3 um diameter) as a sampling area. Such a small aperture can sample almost the smallest grains and even yield information about the edge of grains. High DMax values are possible due to the high electron multiplication in the PMT (gains in excess of 100,000 can be had which leads to high contrast at the micro scale as well as high contrast at larger apertures. The crispness of a PMT based scan, well executed, can be a visually significant departure from the average flatbed or optical enlarger based gear.
Nate Potter, Austin TX.
If he is true to his principles, Steve (Sirius Glass) never prints, he only looks at negatives and transparencies.....otherwise, some "information is lost".
Come to think of it.....I doubt that ANY photographic process can capture the full gamut of visual impressions that a properly functioning human eye can see, so perhaps Steve should just dispense with the camera all together and leave it to photographers who live in the real world, whether digital or analog.
I don't think I've ever read more idiotic statements than these Steve: "Example: take a black & white photograph on film. The color information is lost. Printing it on color paper or using a digital photographic printer will will not bring back the colors. The point is that color information is lost by using a digital camera or scanning film and it at cannot be brought back once lost."
No wonder I was having so much trouble making a color print from my B&W negatives! Who would have guessed that the digital images displayed on my monitor, printed on my printer and which by all accounts APPEARED to be color, were in fact something else......good grief.
Gee, that happened to me once. Real bummer -- was looking for a nice green tone and all I got was Zone V gray tone. I realize now that the real problem was that my optical system could not capture detail at the molecular level.
And now I have having the same problem with scanners. They just all seem incapable of giving a good RGB scan from a B&W negative.
Sandy
For discussion and information about carbon transfer please visit the carbon group at groups.io
[url]https://groups.io/g/carbon
No, not true. I print color and black & white in my darkroom up to 16"x20". For any larger that there are several all optical photo labs in Los Angeles that I used. I rarely take slides anymore thought.
Steve
Nothing beats a great piece of glass!
I leave the digital work for the urologists and proctologists.
Tyler Boley's article should shed some light on the subject. Look at the first row of images. Compare the film scan to the silver contact... lots of lost detail.
http://www.custom-digital.com/2008/09/bw-print-quality/
John V.
ScanHi-End Moderator
That was a very fine comparison Tyler did. I really appreciated the creaminess of the silver print example along with more nuanced detail.
The silver detail can be tweaked considerably however by different enlarging lenses and even much much more by going from a diffusion source to a condenser large aperture source then to a small aperture source and finally approaching a point source using a variable aperture in the light source. Essentially that is altering the f/no of the light source. As that source f/no. increases the contrast in the projected image increases as does the acuity of detail, but the inter-grain modulation that yields creaminess decreases.
How one chooses to treat the projected analogue image is just a function of ones' artistic intent; as is the scanning conditions and subsequent digital manipulation.
I was curious about the parameters used for the silver image he showed - enlarger lens, and lamphouse particularly.
Nate Potter, Austin TX.
Posting Permissions
Reply With Quote
Bookmarks