Does CRI matter for scans of Black & White negatives?

[photonaro]

There's a lot said about CRI for camera-scans of color films - what about B&W films? Does the CRI of the light source matter here?

For B&W films, is a 150$ light table produced by a renowned photography company really any better than a 40$ light box for tracing with 3 light temperatures and different light intensities?

Info much appreciated

 

[thinkbrown]

In my experience, not in the slightest

 

[MattKing]

Just so you know, CRI isn't necessarily particularly useful for colour scanning either. The specifications and ratings that might matter are only used for industries like commercial motion picture lighting, and the equipment that comes with those ratings is a lot more expensive than the equipment that hobbyist photographers use.
CRI is oriented more toward visual appearance under reflected light, and is better suited to deciding upon fabric and paint and the painting above your couch than it is for reproducing digitally the colour dyes in your film. That being said, if CRI is all you can access economically, then go with that for colour work, if you are going to do that in the future.
My only concern with choosing a light source for black and white only is that if you do decide to do colour as well, you end up needing two light sources.

 

[TheFlyingCamera]

I would think (but I could be entirely wrong) that it would be beneficial to have as accurate a light source as possible when scanning via DSLR/Mirrorless regardless of the source material (b/w, color neg, color transparency) because the sensor doing the capturing is a color sensor, and so it could be misled by the transmitted light. I would think this would be even more helpful when scanning non-traditionally developed b/w materials, like film developed with a staining developer where a portion of the density of the negative is derived from the stain. If the light transmitted through that stain is not accurate and truly full-spectrum, it could have a deleterious impact on the final image capture.

That said, there's a law of diminishing returns here - is the Nth percent improvement worth the nx increase in cost to achieve it? if your goal is to have a version of your negative to share on the internet, then probably not. If your goal is to make 40x50 inch prints to sell in a gallery for thousands of dollars apiece, then probably yes.

Given the cost differential between a sub-90 CRI light and a 95 CRI light, I would absolutely spring for the extra to get the 95. If you're going to also do critical color work, then I would seriously consider the 99 CRI lamp. But the relatively small gain in accuracy comes at an exponential increase in price, so you'd need to be doing a LOT of it to justify it. Or you want to offer it as a service to others and can use the improved lamp as a marketing tool.

 

[koraks]

CRI is totally irrelevant for scanning b&w.

I would think (but I could be entirely wrong) that it would be beneficial to have as accurate a light source as possible when scanning via DSLR/Mirrorless regardless of the source material (b/w, color neg, color transparency) because the sensor doing the capturing is a color sensor, and so it could be misled by the transmitted light

Sorry, no. For b&w it doesn't matter one bit and for color it's a whole different ballgame where multi-channel capture is arguably (and to an extent, demonstrably) best.
Most of what's published about high-CRI white light sources for scanning is either methodologically dubious (if not downright simplistically dogmatic along the lines of "surely, more must be better"), or shows marginal differences with lower (e.g. 85-90) light sources.
Also, the camera sensor doesn't care what materials besides air the light has traveled through.

 

[wiltw]

When light is used to illuminate an object, and we want that oblect to be seen or photographed with the most faithful reproduction of all the hues, the CRI does matter. Folks in the textile industry can be very picky about the light that their goods are being viewed under, and they can been extremely picky about the reproduction of hues faithfully on film and on the printed page.
This article's first illustrations make it immediately apparent the importance of CRI on perceptions of color in real life.

Understanding CRI & Why Its Important with LED Lights

Understand the CRI rating of an LED light source and how it affects the environment around it.

www.takethreelighting.com

Many years ago, early in my own use of digital photography, I shot the MacBeth Color Checker under an assortment of illumination types, to see how it affected the appearance of the Color Checker. Unfortunately, I cannot find my shooting notes for this illustration, but de facto one can visualize real difference in hues being reproduced, even ignoring issues of CRI!



When light is simply a source to expose an image being recorded, the color temperature can affect the exposure of the material, since the material is (or is not sensitized) with certain assumptions. Shooting tungsten balanced film in the cool early morning light illustrates what can happen with wrong source temp vs. assumed response of the material being exposed.
And variable contrast print papers illustrate that the color of the illuminating light itself can affect the paper's response with more or lessened contrast characteristics.

 

[koraks]

When light is used to illuminate an object, and we want that oblect to be seen or photographed with the most faithful reproduction of all the hues, the CRI does matter.

Sure, but we're talking about scanning film now. Not how the film is being exposed. From your argument about original capture/photography no conclusions can be drawn about the digitization of film using a camera or what light source is (or isn't) optimal for this. The only possible (but not even definitive) exception I'd make is photographing slide/positive film, in particular color.

 

[MsLing]

High CRI source is not always significant necessary while scanning B&W even color film, IMO. For example, narrowband tricolor source like Status M, can accordance dyes in emulsion and bayer filter well and decreases crosstalk, that means cleaner and purer color. And visible monochromatic light source, especially short wavelength, may provide higher resolution if you reproduce film which was developed with normal formular. Pyro developers may stain your emulsion and display various densities with different wavelength. But I think normal full-spectrum light can cover films well. BTW, Pyros are not what I prefer and I don't know much about them, so I quote Sandy King's example and you can read something on his site.

 

[koraks]

@MsLing I agree with your summary in which IMO you addressed the key points.

 

[wiltw]

Sure, but we're talking about scanning film now. Not how the film is being exposed. From your argument about original capture/photography no conclusions can be drawn about the digitization of film using a camera or what light source is (or isn't) optimal for this. The only possible (but not even definitive) exception I'd make is photographing slide/positive film, in particular color.

I was merely commenting about the use of light to illuminate objects reflectively being photographed (first part with illustration), vs. the use of light to project thru neg/slide for copy purposes, and the fact that even color balance alone could affect color reproduction, even without consideration of CRI. I meant to include (and I remember keying in...it somehow got lost?!...my error, probably during edit to add the illustration), citing color of light affecting multicontrast printing paper contrast under an enlarger, but otherwise not relevant (fluorescent source vs incandesent bulb vs LED heads) with single contrast papers. My discusion was merely about the relevevance of color balance, and not addressing CRI effect itself. I quoted an article about the relevance of CRI, to complement what I said about color balance.

 

[photonaro]

Just so you know, CRI isn't necessarily particularly useful for colour scanning either.

That being said, if CRI is all you can access economically, then go with that for colour work, if you are going to do that in the future.
My only concern with choosing a light source for black and white only is that if you do decide to do colour as well, you end up needing two light sources.

Thanks for your reply!
I'm a bit confused - from your point of view is there a difference between a light source for B&W and a light source for color?

 

[photonaro]

Many years ago, early in my own use of digital photography, I shot the MacBeth Color Checker under an assortment of illumination types, to see how it affected the appearance of the Color Checker. Unfortunately, I cannot find my shooting notes for this illustration, but de facto one can visualize real difference in hues being reproduced, even ignoring issues of CRI!

Thanks for your input! In these examples it's reflected light, though, right? I wonder whether these differences would be noticeable in the black-gray-whites if it were film with light going through them rather than reflected.

When light is simply a source to expose an image being recorded, the color temperature can affect the exposure of the material, since the material is (or is not sensitized) with certain assumptions. Shooting tungsten balanced film in the cool early morning light illustrates what can happen with wrong source temp vs. assumed response of the material being exposed.

And variable contrast print papers illustrate that the color of the illuminating light itself can affect the paper's response with more or lessened contrast characteristics.

I believe as well that the color temperature affects the result of the contrast also in the DSLR scanning of b&w negatives, at least that's what my experiments so far have shown. Since I don't know the CRIs of the light sources I've used, I am still left with question marks about that...

 

[photonaro]

CRI is totally irrelevant for scanning b&w.



Sorry, no. For b&w it doesn't matter one bit and for color it's a whole different ballgame where multi-channel capture is arguably (and to an extent, demonstrably) best.
Most of what's published about high-CRI white light sources for scanning is either methodologically dubious (if not downright simplistically dogmatic along the lines of "surely, more must be better"), or shows marginal differences with lower (e.g. 85-90) light sources.
Also, the camera sensor doesn't care what materials besides air the light has traveled through.

Thanks koraks! Are you saying that CRI in itself doesn't matter for color films either, but that it's multi-channel capture that determines the resemblance to real life colors?

 

[MattKing]

Thanks for your reply!
I'm a bit confused - from your point of view is there a difference between a light source for B&W and a light source for color?

If the spectral distribution of your source is really spiky, and the spikes are at inconvenient places relative to the colour sensitivities of your sensor, then the source may average out in a way that gives you a good result for your eyes and the CRI measurement, but a relatively poor result when scanning the dyes in a colour negative.
In that case, the source may be fine for black and white, but at the very least add to your work load when scanning colour.
The LED sources are more likely to be better - i.e. less inconveniently spiky - then they once were.
But if you have ever wondered why pro level LED light sources for professional video/film work are as expensive as they are, it helps to understand about how relatively well matched their output is to the colour sensitivities of the sensors/film.
The same applies to viewing transparencies or colour prints - its just that our visual systems differ a bit from sensors, in particular.

 

[photonaro]

If the spectral distribution of your source is really spiky, and the spikes are at inconvenient places relative to the colour sensitivities of your sensor, then the source may average out in a way that gives you a good result for your eyes and the CRI measurement, but a relatively poor result when scanning the dyes in a colour negative.
In that case, the source may be fine for black and white, but at the very least add to your work load when scanning colour.
The LED sources are more likely to be better - i.e. less inconveniently spiky - then they once were.
But if you have ever wondered why pro level LED light sources for professional video/film work are as expensive as they are, it helps to understand about how relatively well matched their output is to the colour sensitivities of the sensors/film.
The same applies to viewing transparencies or colour prints - its just that our visual systems differ a bit from sensors, in particular.

Ok, got it. I think . Thanks for the explanation!

 

[wiltw]

Thanks for your input! In these examples it's reflected light, though, right? I wonder whether these differences would be noticeable in the black-gray-whites if it were film with light going through them rather than reflected.



I believe as well that the color temperature affects the result of the contrast also in the DSLR scanning of b&w negatives, at least that's what my experiments so far have shown. Since I don't know the CRIs of the light sources I've used, I am still left with question marks about that...

Largely the CRI and/or color balance would not really matter with B&W materials reproducing B&W source material, unless the B&W reprodution emulsion was DESIGNED to be sensititive to specific light color -- like variable contrast printing paper for the darkroom.

As for being a light source under which a dSLR copies are made, it should not matter because the dSLR sensor is designed to be virtually uniform across the visible spectrum; if your camera was set to B&W JPEG capture, especially! I might understand some variation if you were using one of the profiles from the factory, which did some de-emphasis, like a more pastel capture JPEG, that is intended to reduce contrast with de-emphasis of the more saturated tones.

 

[reddesert]

I have used a cheap USB-powered light box for tracing as mentioned by the OP (it was even less than $40) as a light source for DSLR scanning of black and white negatives, held down with a plate of glass. It works fine. I wasn't doing a super precise job, just making some "contact sheets" and images for drugstore commercial B&W paper prints. You have to readjust the color balance of the scanned negative, but IMO you would have to do that with any DSLR B&W scan no matter what light source, because the camera white balance won't get it exactly right and you need to turn it into your preferred shade of B&W anyway.

Again IMO, you will /should spend more time on worrying about getting the exposure and dynamic range of the scan correct. Debating whether the CRI will throw off the camera sensor, for B&W work, is spending time on unimportant considerations. I do think that such a USB lightbox would be too blue for color work, and make balancing the scan a nuisance, but I haven't tried it.

 

[koraks]

the dSLR sensor is designed to be virtually uniform across the visible spectrum

Digital camera sensors generally show considerable crosstalk between the color channels.

 

[wiltw]

Digital camera sensors generally show considerable crosstalk between the color channels.

One might argue that is what contributes to unformity of response?!

 

[koraks]

Hehe, well played!

 

[photonaro]

As for being a light source under which a dSLR copies are made, it should not matter because the dSLR sensor is designed to be virtually uniform across the visible spectrum; if your camera was set to B&W JPEG capture, especially! I might understand some variation if you were using one of the profiles from the factory, which did some de-emphasis, like a more pastel capture JPEG, that is intended to reduce contrast with de-emphasis of the more saturated tones.

Do I understand you correctly - are you saying that color temperature or CRI of the light source should not have an impact on the contrast of an image with DSLR copies?

 

[photonaro]

You have to readjust the color balance of the scanned negative, but IMO you would have to do that with any DSLR B&W scan no matter what light source, because the camera white balance won't get it exactly right and you need to turn it into your preferred shade of B&W anyway.

That's what I was thinking too.. Besides, b&w editing is pretty much brightness and contrast only, so that's not an awful lot of work compared to any necessary color editing.

Again IMO, you will /should spend more time on worrying about getting the exposure and dynamic range of the scan correct. Debating whether the CRI will throw off the camera sensor, for B&W work, is spending time on unimportant considerations. I do think that such a USB lightbox would be too blue for color work, and make balancing the scan a nuisance, but I haven't tried it.

A couple of tracing light boxes that I have have three color temperatures, the neutral on one of them is pretty neutral in fact. One goes purple and the other is inconsistent in color...

 

[koraks]

b&w editing is pretty much brightness and contrast only

YMMV, but it never seemed to be that simple to me. Still doesn't, really!

 

[MsLing]

Here is a typical spectral sensitivity of Sony CMOS without UV-IR cut filter and there is a set of UV-IR filter's transmission curves from Kolari Vision. Maybe they will help you understand crosstalk of a CMOS.

 

[koraks]

@MsLing thanks for posting those; the first chart is the kind of chart I had in mind when I mentioned the crosstalk issue earlier. There are more/similar charts here for a number of (mostly rather old) cameras: https://maxmax.com/spectral_response.htm
The second plot I think comes with some caveats. It seems to represent the combined response across all three RGB channels with a particular Kolari filter fitted on the lens. This filter likely has a non-linear response to compensate for certain peaks & valleys that result from crosstalk also/in particular in the case of a modified (IR filter removed) camera. So I'm not sure how representative it is in the context of this thread.

As to the former plot, note that much of what happens beyond ca. 700nm or 750nm will likely not be all that relevant for camera scanning since most LED systems don't have much emission in that part of the spectrum. This means that the more problematic part of the RGB crossover (where the curves start to align again in this particular example) is something that won't affect real-world outcomes much. Moreover, in an unmodified camera, this part of the spectrum will be lopped off by the IR filter anyway.

Note finally that the kind of curves shown here can only be constructed on the basis of measurements to which in-camera processing is already applied to normalize the signals. We cannot really know what kind of math (potentially fairly complex, even) is applied to the raw sensor data as this is manufacturer-proprietary information and occurs inside the black box of the onboard image processor. I suppose that if a datasheet of the actual sensor is available (which is probably most often not the case for consumer/prosumer cameras), it might be possible to at least crudely reverse engineer this black-box behavior to a limited extent.

Again, for B&W acquisition it's not very relevant either way. Things get more interesting when digitizing color film. In that case, we enter a world of still conflicting views with reasonable arguments on either side of the divide (narrow-bandwidth vs. full-spectral width), but AFAIK the 'state of the art' in professional film scanning (e.g. for motion picture use) has for a long time been narrow-bandwidth, separate R, G, B captures because this gives the cleanest channel separation and thus is capable of extracting the most color information from the dye image.