Orange Mask of Color Negative Film

[Peter York]

Hi all,

On my last trip to the southwest I brought along a box of color negative film (Fuji 160s), and now I'm coming to grips with scanning it. I know that the orange mask varies across films, but within a specific film, does it vary by processing, SBR, etc.? If so, how does scanner software correct for this variation?

Thanks,
Peter York

[rdenney]

Hi all,

On my last trip to the southwest I brought along a box of color negative film (Fuji 160s), and now I'm coming to grips with scanning it. I know that the orange mask varies across films, but within a specific film, does it vary by processing, SBR, etc.? If so, how does scanner software correct for this variation?

Thanks,
Peter York

I have used Vuescan for years when scanning color negative film. It provides a range of correction settings that have been compiled by users, and organized by film brand and type. You just select the film you used and it will set up the basic corrections for you. Then, you adjust the color balance as usual.

Transparency film is different in that it is intended for direct display and targets realistic colors at the outset. It doesn't need a different correction for each film, unless you are trying to change the film's basic color response.

Rick "who wouldn't want to try to develop the mask removal correction from scratch" Denney

[Peter York]

Thanks Rick,

I have no intention of removing the mask from scratch. Rather, I am trying to get a better understanding of what my scanner/software is doing under the hood to neutralize the mask, and this leads me to the mask itself, which I know little about. If the orange mask is not constant within a specific film due to various factors, then how does the software account for this variation? For what it is worth, I am limited to Epson Scan on a 4990.

Peter

[Paul Kierstead]

Some people will do mask removal manually by using a bit of clear film on the roll they are using. A little tougher with sheet film; usually not much clear film around (no end of roll). But the real answer is to use the built-in removal and tweak it by eye. After all, it is the eyes making the final judgment.

[rdenney]

I have no intention of removing the mask from scratch. Rather, I am trying to get a better understanding of what my scanner/software is doing under the hood to neutralize the mask, and this leads me to the mask itself, which I know little about. If the orange mask is not constant within a specific film due to various factors, then how does the software account for this variation? For what it is worth, I am limited to Epson Scan on a 4990.

You might consider adding Vuescan to your resources, though I have no experience with Epson Scan. The price is low and the capabilities substantial. And you can test it first to see if it works for you.

I think of the orange mask as being similar to biasing a transistor. You have to apply a DC voltage to a transistor just to get it up into its operating range, and then it can do things like amplify an AC voltage that you superimpose on top of that. The orange is the color needed to bring the emulsion into it's operating range in terms of sensitivity and accuracy. Removing the orange mask requires a base blue-green filtration before reversal.

The orange mask has been in my experience quite constant for a given emulsion, with slight variations, perhaps, across lots. But even with sheet film, you could burn a sheet by processing it without exposure to use for determining a base correction for that film type. As you said, though, you don't want to remove the mask manually. With my old Acer scanner and previous software, I was able to get basically realistic color by reversing the image and then setting the film base to black. But that was a blunt instrument and required a lot of subsequent correction.

I just acquired an Epson V-750 Pro to go along with my Nikon film scanner (which only goes up to 6x9), and it will be my first experience with Silverfast. But I have a feeling that I'll still being using my trusty Vuescan.

Rick "who likes scanning color negative film" Denney

[Vlad Soare]

I know that the orange mask varies across films, but within a specific film, does it vary by processing, SBR, etc.?
I believe it does, though I haven't tested this myself. A friend of mine, who was operating a minilab (the classic type, not digital) a few years ago, could tell where I had developed my negatives just by looking at the mask. He used to say "look at that beautiful shade of the mask, you've developed this at ....., haven't you? They always use fresh chemicals."
When printing for friends, myself included, he used to set the machine on manual mode and to dial filters manually for every individual frame. The machine had no preview display; he knew what filters to choose just by looking at the negative. And he always got it right. :)

I never tried to have several identical films developed by different labs to see if the masks look different, because I wasn't too interested in the matter. But, knowing how experienced my friend was in appraising negatives by eye, I tend to believe him.

[mrladewig]

You asked first about the color negative mask. In my experience it is different between films, especially between brands, but consistent in that film regardless of development. I find that Fuji's color negative films have a somewhat red-orange base while Kodak I would describe as more brown-orange. But I've never noticed a difference between a particular film base tint (say Ektar 100) between different developers or labs.

When I first headed down the color negative path, I was using Silverfast SE then Ai. I thought it was giving me some strange watercolor sorts of effects to the film and I created a post about that here several months ago. The answer given was that I was not providing a steep enough curve across those tones, but this was not correct. The problem is something to do with silverfast, but I don't know exactly what it is.

Later I bought a Nikon for scanning 35mm film and quickly learned that the manufacturer's software provided a much different result than silverfast. No more watercolor look. I also learned pretty quickly how to control the color balance tools automatically without relying on the film settings from a software package. Now I'm scanning LF color negative film in Epson Scan and I'm much happier with the control. I use Epson's auto balance feature to get the film close, then adjust all the histograms to taste, build the curve for each color channel as needed and scan. Its hard to get everything exactly perfect in the scan software but IMO best to get it as close as possible. I usually use the selective color layer in Photoshop to polish off the color balance.

I also find color negative gives me a great deal of flexibility that cannot be matched by slide films. I still shoot alot of slide film, but increasingly I find a place for color negative.

[Richard Martel]

Hi,
After reading the above posts a thoght came to me, (they are rarely occuring phenomina though). I wonder if one could scan in, say, 48 bit color positive then read the RGB values and somehow use those values to render the mask clear. I have, in the past used white and black point to eliminate the mask. I'd play around with this thought but I'm getting ready to leave for AK and the Arctic.

regards, Richard

[thechrisproject]

I tried VueScan's built in mask removal for my film (portra 400nc) and then tried removing it manually. The manual process yielded MUCH better color. And it was pretty easy once I had this:
http://benneh.net/blog/index.php/2008/04/21/better-colour-neg-scanning-with-vuescan/

[Larry Gebhardt]

Every brand is different, and it seems development does change the mask color slightly.

With my scanner I pretty much need to remove the mask manually. I get much better results scanning as a positive because the software is so primitive. Just invert and bring the RGB black points down until the edges are at 0 using levels works to get a good starting point. I include the blank area around the frame and use this for the black point. I suspect the software does pretty much this same process as it could be automated.