Colorspace for scans of B&W negatives

[kdriceman]

Advertisement (gone after registration)

I have been scanning my B&W negatives as 16bit grayscale tif images, but some posts by Jaapv and others in a concurrent thread M Monochrom thread piqued my curiosity about this approach. I came across this article and am now wondering what others in the forum think of scanning BW negatives into RGB color files. http://midwestmigrant.blogspot.com/2012/02/digital-darkroom-scanning-black-white.html

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[andybarton]

I have tried it, but have never seen the point, given that there's no colour information in the negative.

 

If you can see an improvement, or a difference in your scans, then why not?

[kdriceman]

I suppose theoretically you get more tones at 24 bit RGB than 16bit Grayscale. 16 million vs 65,000? But if you can't see the difference then what is the point of pushing around bigger files?

[250swb]

There is no colour information in the negative, but there have always been hues, from warm to cool, in traditional photographic papers. The emulsion could be warm, giving a brownish colour in papers like Agfa Portiga, or cool in papers like Agfa Brovira. Photographers would choose the tone of the paper in the same way they chose the contrast. In addition some papers had a warm paper base, and some neutral.

 

So colour of some sort has been an aspect of 'B&W' photography since the dawn of photography, much longer than the recent digital desire to throw it out, and do away with hue as part of the expressive possibilities in a monochrome image. An awful lot of the famous images people are familiar with would look distinctly odd without the reproduction of that base colour, but it is something that is often subtle and overlooked.

 

Making a B&W image RGB allows the digital photographer do take advantage of subtle colour shifts, warm to cool, making digital B&W images more like film images even if they have originally been scanned in Greyscale. It needs an RGB file to add a subtle colour to the inkjet print and remove the digital look of a Greyscale image.

 

Steve

[jaapv]

Conversion to RGB and toning can be easily done just before exporting the file.

[Jeff S]

So colour of some sort has been an aspect of 'B&W' photography since the dawn of photography, much longer than the recent digital desire to throw it out, and do away with hue as part of the expressive possibilities in a monochrome image.

 

One of the wonderful aspects of newer inkjet papers IMO is the increasing range of tones and textures. I choose digital printing papers for many of the same reasons I did in my darkroom days. Some of the current 'baryta' papers, ranging from warm to white, with differing hues, are terrific. And choosing mats to complement those papers is as important as ever.

 

One should be clear, too, that Epson printers (with standard Epson inks), even in advanced b/w mode, still impart some color to the print. And software like LR provide numerous toning options, including subtle split-toning, that are more flexible and less time consuming to impart than in darkroom days.

 

Any good b/w photographer needs to understand color, then and now. Use of color filters on the MM is but one example where that knowledge is key.

 

Color within the b/w digital world is not dead by a long shot.

 

Jeff

[philipus]

Advertisement (gone after registration)

scanning BW negatives into RGB color files.

 

I've tried this with Vuescan but haven't found that it makes any difference (to my eyes at least).

[stunsworth]

I suppose theoretically you get more tones at 24 bit RGB than 16bit Grayscale. 16 million vs 65,000? But if you can't see the difference then what is the point of pushing around bigger files?

 

To get those 16 million tones you'd have to be scanning a colour image - a b&w scan may not be perfectly neutral, but there are no bright red, green or blue colours. I always scanned b&w negatives as 16 bit greyscale, which would have more information than the 8 bit per channel RGB scan that you proposed.

[pop]

I did use colors for scans of B&W negatives and positives. However, they were quite old and I used some of the color information for de-speckling and reconstruction. In some cases the result could be quite simply improved by using but one of the three color channels.

 

I used the maximum available color depth which was 3*16bits, nominally.