B&W filter--to be or not to be

Are there any benefits to using a B&W lens filter, such as the LEE 23A Red to enhance digital B&W images. I'm defining "enhance" here as a mild infrared look, not the full out Hoya R72 look. People keep saying that all I need is to play around with the color channels in PS and that lens filters are pointless for the look I want.

Is this true? Or does anybody combine B&W lens filters with Photoshop RGB channel mixing?

IMO, yes. PS doesn't add two new glass surfaces or reduce exposure.

I'm with tzalman here, a simple colour filter over the lens adds nothing that you can't easily do in PS (apart from an extra glass surface to cause reflections, flare etc).

To add to that, if you add the wrong color physical lens filter for the effect you are after, you've effectively made it more difficult to to get the desired effect later.

I've heard of people using them for the challenge, or the fun of it. But everything I've read agrees with the other posts above... color filters for BW are one type that are no longer needed with digital.

In fact, when I first started shooting with my dSLR, I thought it was pretty cool how easy it was to instantly simulate the effect of different filters on my RAW file in Canon's DPP software. Most other software like Lightroom and PS can do this also (and with even more fine-tuned control.) You may even find presets to get you 'close' to they Hoya look, and then tweak further from there.

If using a digital sensor with a Bayer CFA infront of the sensor (or Fuji Foveon) I would not want to use a B&W colour correction filter on the lens. The big advantage of converting to monochrome using a full colour image (especially in the RAW converter) is that you have full control over all of the avilable colour channels (8 in LR/ACR for example). Adding a colour correction filter will simply remove some of the light spectrum ultimatly reducing the effectivness of the channel mixing tools.

The Bayer FCA actually acts as either a red blue or green (×2) colour correction filter split on per sensel locations. Since we can then mix any colour we want from this you can see why additional filters are unnecessary.

The only current production digital camera that I can think of that could benefit from the use of B&W colour correction filters is the Lieca Monochrome M. As this camera completely lacks a colour filter array.

Alan

texshooter wrote in post #16963888
People keep saying that all I need is to play around with the color channels in PS and that lens filters are pointless for the look I want.

Is this true? Or does anybody combine B&W lens filters with Photoshop RGB channel mixing?

It's true enough for most people's purposes. IMO, if you have a specific need to use a lens filter instead of digital adjustment, you'll know it, and you won't need other people to tell you that you need a filter.

I like B&W conversions in Lightroom/ACR instead of Photoshop RGB channels.

Filters were sledgehammers, a good digital editor is a scalpel.

It depends. You can use filters to adjust, selectively, the exposure during capture.

http://www.libraw.org …a-filters-on-digicam.html

Kirk

kirkt wrote in post #16965448
It depends. You can use filters to adjust, selectively, the exposure during capture.

http://www.libraw.org …a-filters-on-digicam.html

Kirk

That's actually an interesting idea. So why isn't this more widely used - or is it widely used by landscapers? PITA? Filter quality? The fact you'd often have a CPL and this would add an extra filter and exposure time?

texshooter wrote in post #16963888
Are there any benefits to using a B&W lens filter, such as the LEE 23A Red to enhance digital B&W images?

Using color filters makes your lens sharper. The blue light that gives you a purple fringe get filtered out and there's no purple fringe.

I suspect the difference is mostly theoretical.

ejenner wrote in post #16966649
That's actually an interesting idea. So why isn't this more widely used - or is it widely used by landscapers? PITA? Filter quality? The fact you'd often have a CPL and this would add an extra filter and exposure time?

I do not know the answer to your questions.

That being said ( ) I would guess that most digital photographers shooting color images are not aware of color filters or see no point in using them, because if they are being used for some "artistic" purpose, that can be simulated with software.

Using filters for digital, raw acquisition may be of benefit if you have a good understanding of what it is you are trying to achieve and understand how filtering can produce the desired results.

In the case of a daylit scene with the sky present in the image, after WB, folks might find that the sky contains noise that they find objectionable. Why is there noise in the sky? Or how about in the shadows, which are essentially lit not by the direct light of the sun, but by the fill light of the sky?

If one shoots such an image and examines the R, G and B channels (which you cannot do in LR, for example) you will find that red channel is underexposed in the sky and noise is present in this channel. The noise in the composite RGB sky originates mostly in the R channel - this would imply that if you were able to expose the red channel more, you could decrease the noise in that channel and have cleaner tones in the sky. If noise is bad enough, it can start affecting color, not just produce annoying speckles. So, addressing noise in a single channel might be of benefit.

The use of a magenta filter, as outlined in the LibRaw article, essentially permits one to expose for the red channel (with some crosstalk into the other channels) and suppress the light falling on the green channel - the green channel is typically what dominates these daylit scenes and will blow first - if you base your exposure on the green channel you can only ETTR so much and the red channel will still be underexposed. By passing red and suppressing green, you can bring both channels up to their clip points via ETTR without sacrificing one over the other. Then you can white balance and go about your merry way, with better signal in the red channel. At least, that is what I gather from the exercise demonstrated in the LibRaw article.

This probably makes less of a difference in an image the is destined for black and white, as the noise in the original color image is less likely to affect the "look" of a black and white that you convert using whatever method you want. "Grain" is nice in such an image and any color effects from the noise will vanish in conversion. You can apply filters virtually to a color image in many black and white conversions to, for instance, darken the sky and lighten the complementary color (a red or yellow virtual filter).

I would file the use of color filters for raw image acquisition under "Nerd Experiment" - that is, something that I have wanted to do for a while and have not gotten around to investing the money in filters yet. Whether or not there is an actual real benefit to the exercise, I do not know. I would have to try it for my shooting style. I would think that any scene with highly saturated dominant color that would not produce an "average gray" scene would be a situation where filtering to control that color might help your overall exposure.

Note in the comments that follow the article there are some interesting discussions about the topic. One in particular is the question about biasing WB in camera to Magenta to simulate the effect. As the author of the article points out, this does not address the fundamental requirement of the problem's solution - making more light fall onto what will be the red channel. To make this happen, you need to modify the light entering the camera.

The fact that most raw converters give you no picture of the R,G and B components of your image (or the white balance coefficients used to scale the linear data in your file to achieve the selected balance) would indicate to me that most people using raw converters to make images do not care about such details - hence it is not surprising that most people do not really think about, or find interest in, filtering during acquisition.

kirk

Here is a link to PDF about a Lee Filter system:

http://www.leefilters.​com …s/pdf/100mm-System-02.pdf

- if you read through it, the various filter sets that are described also have a description of their application. It is interesting to read about the sets, what they are meant to do, and how they change exposure. The color filters (for example, the CCMagenta) are described on pg. 47. Understanding color theory and how it applies to the recording to light in a camera is a helpful knowledge base to understand the relationship between the color of a filter and the resulting effect of the pass/reject colors.

kirk

Kirk, that's interesting stuff, I just don't think it will be something I'll set out to put into practice!

I mean, heck, I don't even use polarizing filters in practice! I have a slew of them, just never put them on -- for one thing I like to work the colors in my Raw processor, and then as far as reflections go, well, it just hasn't been an issue that occupies my mind...!

Tony, you made my point splendidly! It is interesting, but not something most people would go through the effort to put into practice on a regular, or even as maybe needed, basis.

kirk

tonylong wrote in post #16968145
Kirk, that's interesting stuff, I just don't think it will be something I'll set out to put into practice!
\

I was going to say the same thing in my post, but then decided I should not be so rash.

The point about skies came to mind immediately. Even at ISO100 you can generate quite a bit of noise, especially once you start cranking the local contrast to bring out cloud details and the like. Even though one does this selectively, it still can make the sky noisy.

Now OTOH, I've never found that objectionable in print, or full-sized on a monitor (even a 4k). At higher ISO yes, definitely, but not at base ISO.

OTOH we (I certainly) spend enough on gear and travel to get sharp, high IQ shots that not even considering adding an extra filter to improve aspects of the image seems rather contradictory.