Question about B&W processing in SEP 2

[bnnrcn]

Do you process your images in ACR or LR before going to SEP 2 and why? I can understand that pre-processing is necessary for cropping, cloning and lens corrections etc. because they can't be done in SEP 2 , but why do I have to set the other sliders like exposure,contrast,clarity etc. before going to SEP2 , is it a must and what is the point?

My book about SEP 2 suggests that beginners should do their edits only in SEP 2 for learning purposes (except cropping,cloning,lens corrections) but when I watched some videos about SEP 2 conversions, the images were first edited in ACR or LR nicely and then they were sent to SEP 2.

I really don't understand why do I have to use (for instance) 'clarity' and /or 'contrast' sliders in ACR beforehand , when there is much more powerful 'structure' and 'contrast' sliders in SEP2.

So, I would like to hear your opinions about the matter. Thanks in advance

[Mike Buckley]

The short answer is that it shouldn't matter.

I imagine that the suggestion to do as much as possible using a RAW converter is rooted in two concepts: One concept is that doing so is probably ideal when working solely in color, so it seems reasonable to maintain the same workflow when working with monochromes. The other concept is that developing the file in color as much as possible before converting to monochrome maintains the underlying RGB data in a manner that results in a better monochrome version. Vincent Versace explains that concept very thoroughly in his book From Oz to Kansas.

As to that second concept, Manfred and I got in an informative discussion a couple months ago when he explained that he immediately post-processes monochromes in monochrome rather than first using the typical color workflow.

Notice that the Google Nik Collection web site at least implies that the first step in the workflow is to convert from color to monochrome and then to make the other adjustments afterward. There is even this video of making global adjustments they produced that explains workflow and it doesn't mention anything about working the color version first.

[Manfred M]

Binnur - Let me continue along the vein of what Mike hinted at and perhaps give you my reasons for my workflow, which differs from his. I started off as a B&W film photographer when I was in my teens; mostly because colour film and processing were to high for my limited budget. In B&W film work, colour filters were used to enhance and modify certain tone values; just like a polarizing filter can be used to darken up the sky, a yellow filter would be used in B&W work to give the clouds more "pop". If you want to give them even more pop, a red filter would be used.

B&W is all about tonality, whereas colour photography is all about colour. When I decide to convert a B&W image, that is my very first step; convert B&W and while doing so, use the six colour sliders (red, yellow, green, cyan, blue and magenta) to get the image looking the way I want in B&W. I play with the tones in the image, much like I would have shot B&W film with a filter (except I have more filters and more variability as to how I chose to play with them). I can either brighten or darken the leaves or grass, to get the image looking the way I want. I can do the same thing with the clouds and the sky or the water. As I am thinking in monochrome, I don't have to worry about how this would look in colour, because it doesn't matter. I just assemble tones that make for a good looking image.

This also simplifies any post-processing, because I am working on the final tones I'm looking for. I don't have to worry about colour channels and colour casts.

Is there a downside? Perhaps a philosophical one where I don't like throwing away any data until the very last step (which is where Versace is coming from, I think. Normally I tend to think that way as well, but not in B&W conversion work. I've read his book, and I can respect him for his opinions, but that doesn't mean I feel that he is correct (in fact I think is wrong)..

I often do my own B&W conversions and bypass SE2 completely. I sometimes get lazy and throw a colour file in and do things the "normal" way and sometimes I take my B&W conversions (after all, the colour channels have been preserved) and run things through SE2 to see if I like the image better.

There is no better way than to try things yourself. There is no right or wrong answer, use what works best for you.

[Mike Buckley]

Is there a downside? Perhaps a philosophical one where I don't like throwing away any data until the very last step (which is where Versace is coming from, I think.

No, that's not quite his reasoning. The RGB data is never thrown away because the monochrome image is still displayed in an RGB color space.

In fact, Versace explains that using the RAW data to manipulate monochrome tones is "certainly acceptable in many cases, but it will never produce optimal results." His reason for delaying converting from color to monochrome as long as possible in the workflow is "because the better defined your image structure, the better your final image will be. Therefore, it is best to retouch a face in a full color image rather than in a black-and-white one. This is true of landscapes and still life images as well."

[Brev00]

I do not claim any special expertise but I have always foliowed the procedure of processing first in ACR until full satisfied that I have the range of tones and an accurate display of colors. If I wish to prepare the image for black and white I might adjust accordingly--adding more blacks and contrast. I simply learned to do as much as possible in ACR and follow that procedure. I don't always use Silver Efex but when I do, I enjoy the zone indicator which helps to give me a precise array of tones. I don't find that the structure in Silver Efex duplicates clarity or any other feature in my ACR (Elements, which is limited). It is more similar to the tonal contrast feature in Color Efex, at least in terms of the cloud texture it can give. So, I would say that a full raw conversion is part of my workflow but Silver Efex might eliminate my need to use Color Efex for those sort of fine contrast adjustments. I think the raw conversion turns an amorphous blob into a nicely formed sphere and then Silver Efex creates the final, etched globe. But, as I said, I am not a dedicated black and white photographer with film history. Just someone who likes black and white a lot.

[Manfred M]

No, that's not quite his reasoning. The RGB data is never thrown away because the monochrome image is still displayed in an RGB color space.

If you do the conversion non-destructively, then this is true. If the conversion is destructive, then the original RGB values are replaced with alternate ones that are mapped to monochrome values. The original data has been thrown away and replaced with what are effectively gray scale values. A "red" with a RGB value of (124,50,23) maps to (66,66,66) on conversion. So yes, effectively we still have an RGB colour model to drive our displays, but the three colour values are identical.

"because the better defined your image structure, the better your final image will be. Therefore, it is best to retouch a face in a full color image rather than in a black-and-white one. This is true of landscapes and still life images as well."

I think where I find fault with Versace's argument is that he provides no backup to his statement other than the proverbial "trust me". If one converts up from, one can see exactly how each and every edit will impact the final monochrome image. This ongoing feedback will be missing when one edits the coloured version. I would suggest you try both methods to see if you find any quality issues or get a "better" or "worse" image.

I treat colour and B&W totally differently when I do my shot planning and shooting. I can't see why I wouldn't do the same thing in post. In B&W I'm working in tonal values, in colour I'm not. I prefer to see the tones I'm editing and not hoping things go right when I convert at the end.

[bnnrcn]

Thank you Mike, Manfred and Larry. I really appreciate your help. I think everybody has their own way of editing B&W images and I have to choose my way too.

I think I will convert to monochrome at the beginning of my work flow because I want to learn SEP 2 thoroughly. I also agree with Manfred about working in tonal values for a B&W image. I prefer to see tones instead of colors in front of me while editing. I don't want to guess how my B&W image wil look by working in color and converting it later and I don't want to get frustrated. In addition , I belive that some images suit only B&W treatment , so why do I have to bother myself spending time to edit in color first.

Larry 'clarity' and 'structure' sliders are not the same but they are similar in some ways. If you apply them on top of each other your image might look overdone and weird. That's why I find it strange to apply them both and prefer only one of them . If I had to choose I would choose 'structure' slider because it has seperate choices for highlights,shadows and midtones as well as global 'structure' and 'fine structure' sliders.

I'm happy to be here in CinC, because I know that there are always people here to help and share their knowledge. Thanks again

[Mike Buckley]

If you do the conversion non-destructively, then this is true.

That's what we're discussing. Versace explains when it's fine to do it destructively and why. As an example, I almost always change my photos of glass to monochrome destructively because the decision making is faster and with equally good results as changing non-destructively.

I think where I find fault with Versace's argument is that he provides no backup to his statement other than the proverbial "trust me".

It's only an opinion. It's not a statement of physical properties that can be supported. You and I are saying it's okay -- trust us -- to work the color or the black-and-white image during post-processing. If it's okay for us to do that, it should be okay for him too.

I would suggest you try both methods to see if you find any quality issues or get a "better" or "worse" image.

If you're looking for empirical evidence produced by trying both methods, I don't think that's possible to happen. That's because you're not going to make the exact same decisions when tweaking the color image as when tweaking the monochrome image.

By the way, I still haven't seen anyone recommending the reasons to work the monochrome image instead of the color image with the detail that Versace and Freeman lay out to work the color image. I browsed the Barnbaum book that you like and it's worth the price and effort just to be able to view his photos. However, I didn't find anything about his digital workflow other than one idle comment that he works in black-and-white. If I missed something more definitive, please point me in the right direction.

[Manfred M]

Mike - I guess I need to understand why one solution works better than another and that is missing from Versace's book, From Oz to Kansas. I don't like empirical solutions, I need to understand the when and why, not just the "how to". That's the techie part coming out; with a background in math and sciences, it's just the way my brain is wired. I guess that's why I feel quite comfortable with Barnbaum, with his math and physics background. Now granted, he does spend a lot of time on film, and with my long history with film, both shooting and in the wet darkroom, I find much of this material more interesting than the average digital photographer might. To me the real reason I like his books is that he gets me thinking about my own approach when composing, shooting and post processing.

Personally, I find Versace's explanations on B&W conversion weak and at best anecdotal. I do have his book and find it very difficult to get through, mostly because of his writing style; something half way between a cookbook showing us how to do things and a philosophy textbook. I don't quite understand why he needs to put all those quotes in that have little or no connection to the subject at hand or why he needs 100 words to explain as concept that should only take 20.

Let me take another kick at why I've come to my conclusions. As I recall, you're not a Photoshop user, so let me go through the three "global" approaches that a Photoshop / Adobe Camera RAW (ACR) user could do a colour to B&W conversion.

1. Convert in ACR - This is the same approach Lightroom users would be able to use. As ACR is a parametric editor, what happens in the background is that the original colour image data is unchanged, but the software creates instructions as to how to convert every pixel into a gray scale tone. Importing this into Photoshop effectively means you are editing a colour image, but seeing the impact on a gray scale representation.

Eight individual sliders allow you to tweak the actual colour values so that they look the way you want as gray scale values.

Once imported into Photoshop, you are essentially editing a full colour image, but this detail is totally hidden from you. This is probably my least preferred method of doing B&W. I

2. Destructive conversion in Photoshop - use the <Image> <Adjustments> <Black & White> menu / drop down box to convert. Six colour channels plus hue and saturation adjustments possible. The tone (sepia, blue, etc.) can be adjusted here too.

This is my preferred method, after all, I'm working in B&W and all of the colour tools still work. I don't have to worry about colour casts or any other adjustments that impact a colour image. Just like the old B&W film days, I work with a 100% B&W image. Any cloning or content away functions only care about tone matching and I don't have to worry about what the conversion will look like. It's already done.

3. Non-Destructive conversion using an adjustment layer - This is a method that lets you go either way; early conversion or late conversion. Just add a B&W adjustment layer in the appropriate location in the layer stack <Layer> <New Adjustment Layer> <Black & White>. The colour sliders can be used any time in the cycle.

For early conversion, place the adjustment layer just above the image being edited and for late conversion., but the adjustment layer as the top layer. The main reason I don't use it is that it adds an extra layer of complexity, where it is not needed. File size is a touch larger as well.

[Mike Buckley]

Personally, I find Versace's explanations on B&W conversion weak and at best anecdotal.

I'll point out for the last time so I don't risk being rude that Barnbaum's book that you pointed my way offers no explanation of a recommended workflow for producing a digital monochrome image. So, I can't compare the two authors' explanations.

I can only conclude that Barnbaum either felt that no explanation is necessary or that he became sloppy and forgot to include it. If he felt that no explanation is necessary, he was sloppy for failing to mention that. It's especially sloppy in my mind considering that his book displays a lot more monochrome images than color images and that it devotes an entire chapter to digital photography.

[bnnrcn]

I just want to say that I'm marking your comments as helpful in order to let you know that I'm following the conversation. No other intentions.

[Mike Buckley]

Binnur,

It just now came to my mind that I seem to remember that you have Freeman's The Complete Guide to Black & White Digital Photography. He explains that "There is plenty of conflicting advice on the workflow sequence, but in reality, there are no truly compelling reasons for doing any particular procedure first, or last, or wherever." However, he then contradicts that statement immediately to the right in a sidebar that lists the essential steps that must be taken. One of them is to set the hue adjustments. It's not possible to set the hue adjustments if your first step is to convert to monochrome and if your workflow never reverts to the color version.

By the way, he also devotes two pages to the workflow of using SEP. At the time, it was the first version, not the current version. He literally explains the workflow recommended by the developer and doesn't explain whether he likes or dislikes anything in that workflow.

He devotes two pages to the use of SEP 2.

[Mike Buckley]

The more I think about why I work the color version first, the more I think I know why: The first three books I read on the subject by Odell, Versace and Freeman (I think in that order) explained to use that workflow. I've never seen any book or heard of any book recommending otherwise. So, I simply got used to using that workflow.

[bnnrcn]

Mike, I have read Michael Freeman's book and a book about the Nik Software at the same time very recently and I will look at it the Freeman's statement again. But as far as I understand from all arguments , from the books and videos, there is no certain answer to my question and everybody has their own way. Also, even if you convert to monochrome as first step, it is possible to adjust the effect of the colors in the B&W image by using 'Color Sensitivity' sliders in SEP2. You can even see the original color image and the B&W image with the adjusted color effects together in the 'Split Preview' option. So, because I'm new into SEP2, I was a bit confused about which way to go and started this thread.

Thank you very much for your contribution to the thread. You and Manfred have certainly been very informative about the matter.

Binnur,

It just now came to my mind that I seem to remember that you have Freeman's The Complete Guide to Black & White Digital Photography. He explains that "There is plenty of conflicting advice on the workflow sequence, but in reality, there are no truly compelling reasons for doing any particular procedure first, or last, or wherever." However, he then contradicts that statement immediately to the right in a sidebar that lists the essential steps that must be taken. One of them is to set the hue adjustments. It's not possible to set the hue adjustments if your first step is to convert to monochrome and if your workflow never reverts to the color version.

By the way, he also devotes two pages to the workflow of using SEP. At the time, it was the first version, not the current version. He literally explains the workflow recommended by the developer and doesn't explain whether he likes or dislikes anything in that workflow.

He devotes two pages to the use of SEP 2.

[bnnrcn]

Manfred, thank you very much for your contribution to the thread and being very informative about the matter.

[Mike Buckley]

Also, even if you convert to monochrome as first step, it is possible to adjust the effect of the colors in the B&W image by using 'Color Sensitivity' sliders in SEP2.

That's great to know! Fortunately, I haven't had the need to delve deeply into the capabilities of SEP2 and had not come upon that one.

[Manfred M]

I'll point out for the last time so I don't risk being rude that Barnbaum's book that you pointed my way offers no explanation of a recommended workflow for producing a digital monochrome image. So, I can't compare the two authors' explanations.

Mike - Barnbaum does all of his B&W photographs in film and only does colour with digital. He mentions that in one of his books, but it may be in his latest book, The Essence of Photography: Seeing and Creativity, so the lack of colour to B&W conversion simply a reflection that he doesn't do any. I find it far more interesting in how he looks at his subjects and how this guides his technique for B&W work. To me this is the critical aspect of shooting in B&W, not the details of the conversion process.

I also find that certain authors seem to make a lot more out of the colour to B&W conversion than is really necessary. The cynic in me suggests it may have more to do with selling books than any real mystery in the process. There are really only two variables that one can control in a monochrome image; tonal range and contrast (and this is true for both film and digital work). By changing the colour distribution in the original image, we can enhance or subdue certain colours, much as we did in the film days with filters, but with much more flexibility (filters had set colours and densities). Honestly, it's that simple.

By working at an appropriate bit rate and colour space, we can also apply all of the functionality our editing tool(s) have built in, even though we are working on a grayscale image.

[Brev00]

The colors that you effect with the color sensitivity sliders in Silver Efex are from a range of colors you are importing from another place, whether your camera directly or ACR with full processing or ACR immediate, default transmission. If, for example, your white balance is set to 4000 instead of 7000 in your conversion, you will have a different set of colors to work with in Silver Efex. At least this is my understanding. You can fiddle with the red slider all day, but you it will not have the same result with the 4000 K file as the 7000 K file. So, that is why I like to develop the colors completely the way I want before moving to Silver Efex. One can purposely process an image in a 'distorted' way to get the color range one wants. I may be wrong about this but I do believe one has the greatest set of possibilities in the raw state and every choice from there creates further limitations. Since Silver Efex is not a raw processor, one has to feed it.

[Mike Buckley]

Mike - Barnbaum does all of his B&W photographs in film and only does colour with digital...the lack of colour to B&W conversion simply a reflection that he doesn't do any.

Not true, though I have no idea how prevalently he makes digital monochromes.

First, I became aware of his book because you wrote that you were pleasantly surprised to see that he post-processes his black-and-white images by working with the monochrome image. Second, I found the statement in his book that verified that, though I didn't mark it and am not going to take the time to look it up. Third, on page 233 as part of his explanation of a digital process, he explains why he converted a particular image from color to monochrome. That monochrome image is displayed. That process also confirms that he works on the monochrome image.

I also find that certain authors seem to make a lot more out of the colour to B&W conversion than is really necessary. The cynic in me suggests it may have more to do with selling books than any real mystery in the process.

The cynic in me agrees with the cynic in you.

[Manfred M]

I'll have to go back to re-read.

One thing that I find with a number of authors is that they change their minds over time. I have a book by Freeman where he spends perhaps three paragraphs on B&W,

I'm going through Dan Margulis's first book and last book right now (both are about colour correction), and he certainly contradicts himself, but does suggest that we should pay attention to his last book because it covers off his evolution and learnings over time.