I know, Nik, that you want me to play with your image. I'm going to do that, but, sorry, I'm not going to promise to improve it. The reason is that I don't think the channel-based curve technique is the best technique for every image. In fact, as I keep saying, I think that Dan made a kind of mistake by introducing it before blending a basic technique for improving contrast for different elements of an image independently based on their colors.
Nevertheless channel based curves have their place. So first let's agree about what exactly we are talking about. Dan contends that:
Using separate curves on a luminosity blended curves layer gives better control over different areas of the image in some cases. Note that because the curves are luminosity blended, this is a contrast move, not a color move.
Using separate curves on a color blended curves layer gives better control over color.
2 should be uncontroversial. It's the first step of the portrait recipe and I use it all the time to correct the yellow-magenta balance in flesh. I hope you do that, too. If not, I'm glad to supply some examples.
So we are really talking about how effective 1. is. Before we start with it, let's remember an important fact: Photoshop's luminosity formula:
.3R + .6G + .1B
Why is this relevent here? It means that changes to the green channel curve have 6x more impact on the luminosity of the image than changes to the blude curve and 2x more impact than the red curve. This imbalance is the reason I think this technique is difficult to use and should be taught as a secondary technique to channel blending (chapters 6 and 7.)
Nevertheless, let's see what channel based curves can do in RGB. I'll start by opening up the image with a false profile, same as Duffy did. I'm doing this to open up the shadows and to give the curves more room to work. Here is the image after I assigned it to a 1.0 gamma sRGB profile and then converted it back to real sRGB:
Washed out, but except for the sky, a better starting point for our pedagogical purposes than the original. If I were illustrating blending instead of curves, at this point the first thing I'd do is use the red channel as a luminosity blend for the sky to recover some more detail there, or better in this case, keep the original before the false profile and use that for the sky. In fact, I think I'll do just that:
With this starting point, I wrote a composite curve to try to reestablish a good dark point, get some depth in the faces and get better contrast in that flag and sky. (Remember I'm using a luminosity blended curves layer here.)
Can we improve on this by using the separate channel curves? Yes.
(Again, remember this is a luminosity blended curves layer.)
Why is this better? Because I was able to keep better shadow detail while establishing a dark point. I was able to do this by making the dark end of the red (the weak channel for the soldiers' pants) steep. I got better depth in the faces, sky, and reds of the flag by making the green channel steeper in the highlights and midtones where these things live. What about the blue channel? Forget it. It's only 10% of the luminosity calculation, so it doesn't do much and I ignored it.
Is this the best I can do with this image? No. It's meant only as an illustration of what can be done with RGB channel based curves vs the composite curve.
Maybe I'll redo this image using all the tools and see how I can do. But that isn't the point is it? You wanted to know what channel based curves were good for and I have tried to illustrate.
When you do the "once for color/once for contrast" approach to curves, the point seems pretty obvious. The more ways that you can segregate contrast before curving, the finer control you can have. By the same token, thats why curving in CMYK gives even more control. Instead of three areas to work with, you have four, and as a bonus, one of those areas is a natural selection of shadows.
Dan, however, shows curving the separate channels in a Normal blend. This is a much trickier thing to do. I've been practicing it quite a bit over the past few months, and am still surprised (sometimes unpleasantly) by the results. I think the resistance to PP5's approach in the early chapters comes from the union of color and contrast in these curves.
The main trouble I have with the "once for color/once for contrast" approach is that a big change in contrast does change the color. In LAB terms, brown and red are the same thing. But the commonsense approach says that they are different colors. That said, more and more I'm coming around to seeing the usefulness of the two step approach to color and contrast.
Duffy
(Edit: I suppose I should have called it "once for contrast/once for color". Because color changes when you change contrast, it makes sense to me that you should get your contrast the way you want, or close, before doing the fine color correction.)
Duffy, I've also been thinking about the reaction to those early chapters, especially in comparison to the success of the LAB book. I've even talked to Dan about it. Of course, these days, he's experimenting with something totally different: false profile to decrease gamma a lot, convert to CMYK. Steepen K, keep copy of K channel somewhere, convert to LAB. Steepen A+B through a layer mask containing that saved K channel, HIRALOAM all three channels. Of course, his results look great, mine, not so much yet. Dan demoed this at MIT on Wednesday.
In the last advanced class I took, Dan really stressed luminosity blending for contrast enhancement. I'm just beginning to get my mind around this.
If I were Dan (or his editor), I'd unify PP and the LAB book. I'd start with the first four chapters of the LAB book. I'd do impossible colors, including RGB channel curves for color. I'd teach blending. I'd teach sharpening including HIRALOAM and how to blend HIRALOAM and USM. Only then would I get into RGB channel curves for contrast.
But then who am I? I can't reproduce Dan's results even after they have been clearly demonstrated any number of times...
Curves in luminosity blend.
Heck, I didn't get that from Dan's book. However, as I indicated in the other thread, I have recently started to use curves (and levels) in Luminosity mode almost exclusively and even it was only a master curve, I was quite satisfied with this approach. Now, after reading your post, I feel kinda stupid, since in L-mode I could go to individual channels and get finer controls of some colors, yet still not screwing up a whole lot... So, if nothing else (which is not the case), your post was a great help/revelation!
Separate steps for contrast/color.
Again, great approach, makes total sense. It is much easier to control this beast this way!
Chanel curves in normal mode for the color shift fix.
Since I shoot RAW, it's not an issue. Pretty much all my portrait work was done in the controlled light environment, which means one shot with the gray card, one click in ACR - and whatever color cast was present is gone for good. I do understand it's possible to do via channels, but to me it would be a huge overcomplification.
Thing I didn't get from the book and I'm still not getting - false profile. Why is it a better starting point? How one decides on which gamma value to choose? Why not use middle Levels slider in PS or Exposure in ACR, which are, to a degree, doing the same thing, and are far easier to control?
It would be great if you can elaborate on this...
[Chanel curves in normal mode for the color shift fix.
Since I shoot RAW, it's not an issue. Pretty much all my portrait work was done in the controlled light environment, which means one shot with the gray card, one click in ACR - and whatever color cast was present is gone for good. I do understand it's possible to do via channels, but to me it would be a huge overcomplification.
I shot this in raw under mixed light (window, florescent). I tried various things in ACR, but there was no escaping that mixed cast. In the end I converted with AUTO white balance:
See how the window light makes a blue cast in the highlights (and the artificial light makes a yellow cast in the darker parts of his face?
Easily fixed with one blue curve:
I just command-clicked on the parts of the image I didn't want to change, then command clicked on the part with the blue cast and moved that point toward darkness, to get the blue out of the highlights without changing anything else. Because this is the blue curve, it doesn't have much effect on contrast, but I could have set the blending mode of the curve layer to "color" if I were being paranoid.
This move is a prelude to the rest of the workflow. It's a first order color balance correction that targets more narrowly than color balance in ACR or a LAB curve correction. Sometimes a blend can do better, but this is my workhorse correction for mixed cast portraits (including ballet under theater lights.)
Thing I didn't get from the book and I'm still not getting - false profile. Why is it a better starting point? How one decides on which gamma value to choose? Why not use middle Levels slider in PS or Exposure in ACR, which are, to a degree, doing the same thing, and are far easier to control?
I'm also just getting the hang of false profiles. It takes some setup, but I've watched Dan do amazing things with it over and over. It's like having a huge curve library at your command. Dan likes to tell the story of a photojournalist who uses only false profiles as a post processing technique. But it's a very powerful way to make the entire image lighter, darker, more colorful, etc with a single move. In the case of your civil war image, it really opens up the shadows, a good preparation for curves, blending, etc. But as I said, I'm still getting the hang of this. Hope the reading group gets to it.
I shot this in raw under mixed light (window, florescent).
I see... Makes sense.
Also this whole "locking the parts you don't want to change" idea became MUCH clearer after you guys started to do your magic in this thread.
As to the false profiling... I guess you must have this library and this amount experience first to be able to use it effectively. I also need to re-read the book about how it's done, since during the first read it was way over my head...
Curving in separate channels vs. composite
I'm new to the forum. I've been reading the posts on this thread and seen
much discussion/question about the benefits of using curves in separate
channels vs. composite curves. I think a recent post demonstrated the former
reasonably well.
The only thing I'd add, based on my own experience (including taking one
of Dan's classes) is that this seems to be especially effective in CMYK,
and much trickier in RBG. I don't completely understand the science behind
it, but it seems that you can make bigger moves in CMYK curves that bring
out the details (read contrast) without screwing up the colors noticeably.
I can post a picture or two if anyone's interested.
I'm an RGB guy. The first day of Dan's class, we had to do everything in
CMYK. Based on that, I now do almost all of my curving in CMYK, and I'm
very happy with the results. Occasionally, I'll try it in RGB. Depending on
the picture, it can work well. But mostly, I find it easier (and quicker) to
do in CMYK.
The black point/white point setting is entirely by the numbers. Contrast
setting is done by moving the cursor over the interesting part of the
picture, and seeing where it lands on the curve (an earlier post discussed
this - I also didn't get this from the book, but saw Dan doing it in person).
The only thing I'd add, based on my own experience (including taking one
of Dan's classes) is that this seems to be especially effective in CMYK,
and much trickier in RBG. I don't completely understand the science behind
it, but it seems that you can make bigger moves in CMYK curves that bring
out the details (read contrast) without screwing up the colors noticeably.
I can post a picture or two if anyone's interested.
I'm an RGB guy. The first day of Dan's class, we had to do ever
The ease of writing curves for contrast without altering colors in CMYK makes sense. There are four channels, not three.
The individual color curves have lesser effect due to the addition of the black channel where much of the contrast and shadow detail lie. This is why curves in CMYK can be easier, if your brain is used to thinking in CMYK as pre press color editors are used to.
Unfortunately, my brain tends to work in RGB for some reason, hence writing CMYK feels like drawing with my non-dominant hand
"Every file has 10 channels"
With all due respect, I don't believe that any given single color space is a panacea. Different images and different tasks require different treatments, different tools and different spaces. Mastering one space is good, but why stop there?
Unfortunately, my brain tends to work in RGB for some reason, hence writing CMYK feels like drawing with my non-dominant hand
Just keep saying to yourself "CMYK is RGB". C = R, M = G, Y = B. More cyan is less red, more magenta is less green, more yellow is more blue.
Make a new blank document in Photoshop. Try RGB curves. Move the light endpoint of the red curve toward darkness. What happens? Less red and the image becomes cyan. Why? Because green and blue light is what's left.
Now cancel your curve and convert that plain white document to CMYK. Get a curves dialog and pull the light end of the cyan curve toward darkness. What happens? It become cyan, not surprisingly. You've added cyan ink, making the image darker. But how? By adding the ink that reflects green and blue but not red.
But you knew that. Mostly the very same curves translated from RGB to CMYK with R=C, G=M, B=Y will do the same thing. Except in the shadows. And where the shadows start is a function of the definition you use of CMYK. See PP5E, ch 5. So use CMYK when you are trying to conquer the shadows or when you are preparing for a CMYK press.
I guess it's hard for people who lack the knowledge/experience of the abstract topologies/ "spaces". For me personally all these colorspace shenanigans fade compared to an extremely simple, nice, easy to use and oh-so-forgiving L(2,oo)
The hardest thing for me coming from RGB was understanding neutrals.
R=G=B is pretty straightforward, but CMYK isn't at first. It turns out
not to be that big a thing, but it was an incredible mental block for me.
Getting past that was quite rewarding, allowing me to solve
problems I couldn't in RGB (bringing detail out in shadows, as you
mention, but also in highlights with custom conversions).
Another practical issue is setting up the curves plot to be consistent
in both spaces (and understanding that less is more, so to speak -
more C ink is less R glow).
In retrospect, I'm amazed that it seemed so hard at first.
The hardest thing for me coming from RGB was understanding neutrals.
These days I almost always curve twice, once for color and once for contrast, setting the blending mode of the curves layer with color and [/i]luminosity[/i]. Or I blend for contrast and curve for color. Or I work on casts in RGB and then go to CMYK for work on the shadows with the K curve. I fuss with the "cycan" curve for flesh after LAB corrections by using the red curve in RGB; only when there is some reason this won't work do I do it in CMYK. In any case, I try to avoid needing to neutralize in CMYK.
Another practical issue is setting up the curves plot to be consistent
in both spaces (and understanding that less is more, so to speak -
more C ink is less R glow).
One of Dan's best innovations, little appreciated, but very important for seamless 10-channel work.
I just came across this thread and am trying to go through Dan's book (2002 PS7 version but I'm sure the concepts are just about the same). Ok, I'm only in chapter 2 but I don't know where he is coming from with some of the stuff he says. I often feel like he's talking down to me. It would help if he would take just a bit more space to explain the rationale behind some of the decisions and assumptions he makes.
I'm not a complete newbie but am entirely self-taught from my camera through PS and have plenty of gaps to fill in. But I'm not sure this book is where I should first be going to do this. Perhaps something more of a 'Post processing finishing school 101' course would be better before I decide (if I decide) to get into Dan's ideas.
Professional Photoshop 5th Edition by D Margullis is a challenging book. I don't think any of us who participated in that book review would say it was all easy sailing. But it is a very worthwhile journey, that many of us felt the need to take.
I found that I had to work through each example - Kind of like a math textbook:D Only very few folks can just read their calculus textbook, and say "oh yeah, that was obvious".
If this is your first endeavor into Photoshop self education be prepared to be challenged along the way. There are simpler, more direct texts around - Scott Kelby writes a number of good ones. Many of us had already read them, and wanted a deeper understanding of what was happening when we edit an image.
Dan did not title his book PROFESSIONAL Photoshop for no reason. It is addressed to pre press professional and would be serious photographers. It is not a tome for folks without a serious desire to learn. Dan also assumes a more than average understanding of Photoshop usage as well.
If you have specific questions feel free to drop in here and we'll see if some one can't round up an answer for you. It will be good review for those of us who reviewed chapters in this book.
I just came across this thread and am trying to go through Dan's book (2002 PS7 version but I'm sure the concepts are just about the same).
While what Jim (Pathfinder) said above is very true, I sincerely recommend to cough out some $$ (it's New Year!) and get 5th edition (you got 4th), as well as his LAB book. Yes, concepts are the same, but he himself changed alot over the course of time. You will find his latest books no less challenging, but more adequate for the digital age... I still own and hold dear a copy of 4th edition (as wel as the other two), but if I had to choose, I'd start with the 5th... Or maybe with LAB - it's kinda easier...
Just my 0.0000002 of the f/stop.
Thanks for the encouragement, guys. I've ordered the Kelby CS2 book and will let that fill in any gaps before proceeding with Dan's books. I don't mind taking extra time to make sure my foundation is solid. I'm sure I'll be redoing many of the photos on my site!
Comments
Done.
http://www.dgrin.com/showthread.php?t=58191
TIA!
Nevertheless channel based curves have their place. So first let's agree about what exactly we are talking about. Dan contends that:
- Using separate curves on a luminosity blended curves layer gives better control over different areas of the image in some cases. Note that because the curves are luminosity blended, this is a contrast move, not a color move.
- Using separate curves on a color blended curves layer gives better control over color.
2 should be uncontroversial. It's the first step of the portrait recipe and I use it all the time to correct the yellow-magenta balance in flesh. I hope you do that, too. If not, I'm glad to supply some examples.So we are really talking about how effective 1. is. Before we start with it, let's remember an important fact: Photoshop's luminosity formula:
Nevertheless, let's see what channel based curves can do in RGB. I'll start by opening up the image with a false profile, same as Duffy did. I'm doing this to open up the shadows and to give the curves more room to work. Here is the image after I assigned it to a 1.0 gamma sRGB profile and then converted it back to real sRGB:
Washed out, but except for the sky, a better starting point for our pedagogical purposes than the original. If I were illustrating blending instead of curves, at this point the first thing I'd do is use the red channel as a luminosity blend for the sky to recover some more detail there, or better in this case, keep the original before the false profile and use that for the sky. In fact, I think I'll do just that:
With this starting point, I wrote a composite curve to try to reestablish a good dark point, get some depth in the faces and get better contrast in that flag and sky. (Remember I'm using a luminosity blended curves layer here.)
Can we improve on this by using the separate channel curves? Yes.
(Again, remember this is a luminosity blended curves layer.)
Why is this better? Because I was able to keep better shadow detail while establishing a dark point. I was able to do this by making the dark end of the red (the weak channel for the soldiers' pants) steep. I got better depth in the faces, sky, and reds of the flag by making the green channel steeper in the highlights and midtones where these things live. What about the blue channel? Forget it. It's only 10% of the luminosity calculation, so it doesn't do much and I ignored it.
Is this the best I can do with this image? No. It's meant only as an illustration of what can be done with RGB channel based curves vs the composite curve.
Maybe I'll redo this image using all the tools and see how I can do. But that isn't the point is it? You wanted to know what channel based curves were good for and I have tried to illustrate.
When you do the "once for color/once for contrast" approach to curves, the point seems pretty obvious. The more ways that you can segregate contrast before curving, the finer control you can have. By the same token, thats why curving in CMYK gives even more control. Instead of three areas to work with, you have four, and as a bonus, one of those areas is a natural selection of shadows.
Dan, however, shows curving the separate channels in a Normal blend. This is a much trickier thing to do. I've been practicing it quite a bit over the past few months, and am still surprised (sometimes unpleasantly) by the results. I think the resistance to PP5's approach in the early chapters comes from the union of color and contrast in these curves.
The main trouble I have with the "once for color/once for contrast" approach is that a big change in contrast does change the color. In LAB terms, brown and red are the same thing. But the commonsense approach says that they are different colors. That said, more and more I'm coming around to seeing the usefulness of the two step approach to color and contrast.
Duffy
(Edit: I suppose I should have called it "once for contrast/once for color". Because color changes when you change contrast, it makes sense to me that you should get your contrast the way you want, or close, before doing the fine color correction.)
In the last advanced class I took, Dan really stressed luminosity blending for contrast enhancement. I'm just beginning to get my mind around this.
If I were Dan (or his editor), I'd unify PP and the LAB book. I'd start with the first four chapters of the LAB book. I'd do impossible colors, including RGB channel curves for color. I'd teach blending. I'd teach sharpening including HIRALOAM and how to blend HIRALOAM and USM. Only then would I get into RGB channel curves for contrast.
But then who am I? I can't reproduce Dan's results even after they have been clearly demonstrated any number of times...
Thank you so much for your time!
This is exactly was I was looking for :-)
Now, to the things I got (and not...
- Curves in luminosity blend.
- Separate steps for contrast/color.
- Chanel curves in normal mode for the color shift fix.
- Thing I didn't get from the book and I'm still not getting - false profile.
Why is it a better starting point? How one decides on which gamma value to choose? Why not use middle Levels slider in PS or Exposure in ACR, which are, to a degree, doing the same thing, and are far easier to control? 
Once again, I do appreciate your sharing!Heck, I didn't get that from Dan's book. However, as I indicated in the other thread, I have recently started to use curves (and levels) in Luminosity mode almost exclusively and even it was only a master curve, I was quite satisfied with this approach. Now, after reading your post, I feel kinda stupid, since in L-mode I could go to individual channels and get finer controls of some colors, yet still not screwing up a whole lot... So, if nothing else (which is not the case), your post was a great help/revelation!
Again, great approach, makes total sense. It is much easier to control this beast this way!
Since I shoot RAW, it's not an issue. Pretty much all my portrait work was done in the controlled light environment, which means one shot with the gray card, one click in ACR - and whatever color cast was present is gone for good. I do understand it's possible to do via channels, but to me it would be a huge overcomplification.
It would be great if you can elaborate on this...
I shot this in raw under mixed light (window, florescent). I tried various things in ACR, but there was no escaping that mixed cast. In the end I converted with AUTO white balance:
See how the window light makes a blue cast in the highlights (and the artificial light makes a yellow cast in the darker parts of his face?
Easily fixed with one blue curve:
I just command-clicked on the parts of the image I didn't want to change, then command clicked on the part with the blue cast and moved that point toward darkness, to get the blue out of the highlights without changing anything else. Because this is the blue curve, it doesn't have much effect on contrast, but I could have set the blending mode of the curve layer to "color" if I were being paranoid.
This move is a prelude to the rest of the workflow. It's a first order color balance correction that targets more narrowly than color balance in ACR or a LAB curve correction. Sometimes a blend can do better, but this is my workhorse correction for mixed cast portraits (including ballet under theater lights.)
I'm also just getting the hang of false profiles. It takes some setup, but I've watched Dan do amazing things with it over and over. It's like having a huge curve library at your command. Dan likes to tell the story of a photojournalist who uses only false profiles as a post processing technique. But it's a very powerful way to make the entire image lighter, darker, more colorful, etc with a single move. In the case of your civil war image, it really opens up the shadows, a good preparation for curves, blending, etc. But as I said, I'm still getting the hang of this. Hope the reading group gets to it.
I see... Makes sense.
Also this whole "locking the parts you don't want to change" idea became MUCH clearer after you guys started to do your magic in this thread.
As to the false profiling... I guess you must have this library and this amount experience first to be able to use it effectively. I also need to re-read the book about how it's done, since during the first read it was way over my head...
I'm new to the forum. I've been reading the posts on this thread and seen
much discussion/question about the benefits of using curves in separate
channels vs. composite curves. I think a recent post demonstrated the former
reasonably well.
The only thing I'd add, based on my own experience (including taking one
of Dan's classes) is that this seems to be especially effective in CMYK,
and much trickier in RBG. I don't completely understand the science behind
it, but it seems that you can make bigger moves in CMYK curves that bring
out the details (read contrast) without screwing up the colors noticeably.
I can post a picture or two if anyone's interested.
I'm an RGB guy. The first day of Dan's class, we had to do everything in
CMYK. Based on that, I now do almost all of my curving in CMYK, and I'm
very happy with the results. Occasionally, I'll try it in RGB. Depending on
the picture, it can work well. But mostly, I find it easier (and quicker) to
do in CMYK.
The black point/white point setting is entirely by the numbers. Contrast
setting is done by moving the cursor over the interesting part of the
picture, and seeing where it lands on the curve (an earlier post discussed
this - I also didn't get this from the book, but saw Dan doing it in person).
John
Please do. Thanks.
The individual color curves have lesser effect due to the addition of the black channel where much of the contrast and shadow detail lie. This is why curves in CMYK can be easier, if your brain is used to thinking in CMYK as pre press color editors are used to.
Unfortunately, my brain tends to work in RGB for some reason, hence writing CMYK feels like drawing with my non-dominant hand
Moderator of the Technique Forum and Finishing School on Dgrin
With all due respect, I don't believe that any given single color space is a panacea. Different images and different tasks require different treatments, different tools and different spaces. Mastering one space is good, but why stop there?
Just keep saying to yourself "CMYK is RGB". C = R, M = G, Y = B. More cyan is less red, more magenta is less green, more yellow is more blue.
Make a new blank document in Photoshop. Try RGB curves. Move the light endpoint of the red curve toward darkness. What happens? Less red and the image becomes cyan. Why? Because green and blue light is what's left.
Now cancel your curve and convert that plain white document to CMYK. Get a curves dialog and pull the light end of the cyan curve toward darkness. What happens? It become cyan, not surprisingly. You've added cyan ink, making the image darker. But how? By adding the ink that reflects green and blue but not red.
But you knew that. Mostly the very same curves translated from RGB to CMYK with R=C, G=M, B=Y will do the same thing. Except in the shadows. And where the shadows start is a function of the definition you use of CMYK. See PP5E, ch 5. So use CMYK when you are trying to conquer the shadows or when you are preparing for a CMYK press.
So what's hard about CMYK?
The hardest thing for me coming from RGB was understanding neutrals.
R=G=B is pretty straightforward, but CMYK isn't at first. It turns out
not to be that big a thing, but it was an incredible mental block for me.
Getting past that was quite rewarding, allowing me to solve
problems I couldn't in RGB (bringing detail out in shadows, as you
mention, but also in highlights with custom conversions).
Another practical issue is setting up the curves plot to be consistent
in both spaces (and understanding that less is more, so to speak -
more C ink is less R glow).
In retrospect, I'm amazed that it seemed so hard at first.
One of Dan's best innovations, little appreciated, but very important for seamless 10-channel work.
I'm not a complete newbie but am entirely self-taught from my camera through PS and have plenty of gaps to fill in. But I'm not sure this book is where I should first be going to do this. Perhaps something more of a 'Post processing finishing school 101' course would be better before I decide (if I decide) to get into Dan's ideas.
Professional Photoshop 5th Edition by D Margullis is a challenging book. I don't think any of us who participated in that book review would say it was all easy sailing. But it is a very worthwhile journey, that many of us felt the need to take.
I found that I had to work through each example - Kind of like a math textbook:D Only very few folks can just read their calculus textbook, and say "oh yeah, that was obvious".
If this is your first endeavor into Photoshop self education be prepared to be challenged along the way. There are simpler, more direct texts around - Scott Kelby writes a number of good ones. Many of us had already read them, and wanted a deeper understanding of what was happening when we edit an image.
Dan did not title his book PROFESSIONAL Photoshop for no reason. It is addressed to pre press professional and would be serious photographers. It is not a tome for folks without a serious desire to learn. Dan also assumes a more than average understanding of Photoshop usage as well.
If you have specific questions feel free to drop in here and we'll see if some one can't round up an answer for you. It will be good review for those of us who reviewed chapters in this book.
Moderator of the Technique Forum and Finishing School on Dgrin
Just my 0.0000002 of the f/stop.
I agree, the 5th edition is really much more relevant for the digital photographer today. I suspect even the author would agree.
Moderator of the Technique Forum and Finishing School on Dgrin