Pages: 1 2 [3] 4 5 6   Go Down

Author Topic: LAB COLOR  (Read 40683 times)

laughfta

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
LAB COLOR
« Reply #40 on: November 17, 2007, 09:50:28 pm »

Scott Kelby, in The Photoshop Channels Book, says he uses Lab when he wants to add detail and vibrance to his images, especially landscapes. He adds that he uses it so much that he's created an action to apply it.

I use it for those reasons and especially when I want to make a local and color controlled change rather than a specific global adjustment.

I also use Lab channels often for creating masks.
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
LAB COLOR
« Reply #41 on: November 17, 2007, 11:22:27 pm »

Quote
Scott Kelby, in The Photoshop Channels Book, says he uses Lab when he wants to add detail and vibrance to his images, especially landscapes. He adds that he uses it so much that he's created an action to apply it.

I use it for those reasons and especially when I want to make a local and color controlled change rather than a specific global adjustment.

I also use Lab channels often for creating masks.
[a href=\"index.php?act=findpost&pid=153730\"][{POST_SNAPBACK}][/a]

Yes, indeed Scott does - Chapter 5, "Creating Vibrant Color", and it is very easy. He acknowledges the source of the technique being Dan Margulis, and he does it on a duplicate image layer so he can adjust it or discard it as he wishes. This of course was published well before CR 4.x and Lightroom became available, where (I believe from using them) much of the same kind of effects can be achieved wihout having to implement the L*a*b conversion and without needing a duplicate image layer. In my PPE presentation I did show a "canyone-type" image where I compared steepening the *a* and *b* curves on the one hand versus using Vibrance in CR 4.1 on the other, and the difference between the two results was indistinguishable - I really lost track of which image was which. Now that is one image. Maybe after a great many such comparisons one could build a set of prototypical situations where one approach would be preferred to the other - but that already tells us how diversified the options have become for achieving similar objectives. While the L*a*b approach worked fine, nowadays, personally, I would prefer the up-stream approach whenever it does what I want.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

laughfta

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
LAB COLOR
« Reply #42 on: November 18, 2007, 06:13:51 am »

Quote
He acknowledges the source of the technique being Dan Margulis, and he does it on a duplicate image layer so he can adjust it or discard it as he wishes.

Gee, that's true; I believe he mentions Dan (in his acknowledgments) as being the "bottom line when it comes to color".    (Sorry; but I sat quietly through most of this thread  )

Quote
This of course was published well before CR 4.x and Lightroom became available, where (I believe from using them) much of the same kind of effects can be achieved wihout having to implement the L*a*b conversion and without needing a duplicate image layer.

Yes, and I use Vibrance fairly often--especially when I want to prepare images for the Web. When making prints, I am much more interested in subtleties, and concentrate on enhancing local areas for adjustment. This is the primary reason knowledge of Lab is important: even Andrew said there is a role for Lab in some processing. Vibrance does a good job: it does it globally, and it has fixed settings. Lab can be used to apply local adjustments, giving the user the option to control all aspects of the adjustment--whether color, degree of effect, or sharpness.

My underlying point, as you know Mark, is  that if we are consistently herded away from understanding the effects of any technique Dan Margulis or anyone  touts or even mentions, we will be the less for it.

On the other hand, and I rarely get the chance to mention it, pointing up the flaws or alternatives in such procedures is equally  important. It's just that these points carry much less weight (indeed are often hidden altogether) when presented with obvious personal bias.

Quote
In my PPE presentation I did show a "canyone-type" image where I compared steepening the *a* and *b* curves on the one hand versus using Vibrance in CR 4.1 on the other, and the difference between the two results was indistinguishable - I really lost track of which image was which. Now that is one image. Maybe after a great many such comparisons one could build a set of prototypical situations where one approach would be preferred to the other - but that already tells us how diversified the options have become for achieving similar objectives. While the L*a*b approach worked fine, nowadays, personally, I would prefer the up-stream approach whenever it does what I want.

Well said, and happily noted, Mark. I completely agree.

after edit:

Quote
This of course was published well before CR 4.x and Lightroom became available, where (I believe from using them) much of the same kind of effects can be achieved wihout having to implement the L*a*b conversion and without needing a duplicate image layer.


After posting this I came upon some additional information about Kelby's workflow in which a poster said Kelby's new book Seven Points includes several chapters in which he uses Lab for his results. I only relay this so that those who haven't yet experimented with the Lab color space will know that it is not obsolete, though it may well be less necessary.
« Last Edit: November 18, 2007, 07:50:27 am by laughfta »
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
LAB COLOR
« Reply #43 on: November 18, 2007, 08:53:58 am »

Quote
When making prints, I am much more interested in subtleties, and concentrate on enhancing local areas for adjustment. This is the primary reason knowledge of Lab is important: .................. Vibrance does a good job: it does it globally, and it has fixed settings. Lab can be used to apply local adjustments, giving the user the option to control all aspects of the adjustment--whether color, degree of effect, or sharpness.

My underlying point, as you know Mark, is  that if we are consistently herded away from understanding the effects of any technique Dan Margulis or anyone  touts or even mentions, we will be the less for it.

................................

[a href=\"index.php?act=findpost&pid=153790\"][{POST_SNAPBACK}][/a]

Gloria, forgetting all the names - Dan, Andrew, Scott, me, you, but just focusing on the tools themselves   , let us recall that within CR4.x /LR, one is not confined to global adjustments of Saturation with the Vibrance tool (which itself is less global than it may appear) or the Saturation control, but there is also the HSL tab with eight colour groups each of which is amenable to three sets of controls: saturation, hue and luminance. The Adobe folks moved off the conventional RGBCMY grouping into a configuration which research and experimentation indicated to them would be more useful to photographers. This is innovative and I have read some positive feedback about it. I too find it works well. So with these added 24 controls, CR/LR is capable of much more targeted work than used to be the case only within the past year.

I don't think there's any issue of being herded away from techniques - it's just a discussion about alternatives and their pros and cons. No-one needs to take anyone's word for it - at least I don't and neither do you - everyone with serious  interest in it should be encouraged to experiment with the options and come to their own conclusions.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

laughfta

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
LAB COLOR
« Reply #44 on: November 18, 2007, 09:20:31 am »

Quote
Gloria, forgetting all the names - Dan, Andrew, Scott, me, you, but just focusing on the tools themselves 


Once more, you bring us back to the issue at hand, Mark  

Quote
within CR4.x /LR, one is not confined to global adjustments of Saturation with the Vibrance tool (which itself is less global than it may appear) or the Saturation control, but there is also the HSL tab with eight colour groups each of which is amenable to three sets of controls: saturation, hue and luminance. The Adobe folks moved off the conventional RGBCMY grouping into a configuration which research and experimentation indicated to them would be more useful to photographers.


In my experience with these controls, they are wonderful and innovative and highly useful. The downside is that they can't be limited beyond their predetermined effects. By which I mean, for example, if you choose a color slider to work on, all of that color in your image will be effected. Much of the time, I want this. When I don't, I need to tweak in PS, and that happens a lot. I need other techniques at that point, and Lab offers much in that regard.

I am not saying that those sliders are not easy to use and highly effective, I'm just saying when they aren't giving me the result I want, where I want it, I need to turn to other techniques. And those other techniques are much harder to learn and understand, but the results enable me to make the subtle but extremely important changes to create an image I intend to print and hang.

Quote
I don't think there's any issue of being herded away from techniques - it's just a discussion about alternatives and their pros and cons. No-one needs to take anyone's word for it - at least I don't and neither do you - everyone with serious interest in it should be encouraged to experiment with the options and come to their own conclusions.


I don't think it's that simple, Mark. But it should be.
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
LAB COLOR
« Reply #45 on: November 18, 2007, 09:46:57 am »

Gloria, yes, basically what we don't have in CR/LR are conventional channel ops, blend modes, layer styles, masking, overlays, etc. And for sure, when one needs those tools it is necessary to work in PS. I don't think anyone suggests that CR/LR are replacements for PS. They were never meant in that way - rather they are complementary and will most likely remain so for quite a while regardless of the progress to come in raw conversion technology. What's happening, and I see this continuing, is that as programs such as CR/LR keep advancing and adding capability, they will make it incrementally less necessary to deploy advanced techniques in PS.

I see at the final statements in both of our previous posts there is the word "should". Well, what to say, "should" is "should", "eh"?  
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Panopeeper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1805
LAB COLOR
« Reply #46 on: November 22, 2007, 09:09:08 pm »

Quote
was wondering if, when and why anyone employs LAB color in their workflow? I applied some of the methods to some of my landscape photos and was very impressed with the outcome.

I often use Lab, always for a single reason. This reason may be irrelevant for other users.

I often "enrich" the images with pseudo-HDR, i.e. developing two TIFF versions from the same raw file. I do this almost always for the sake of the sky.

This way I end up with at least two layers: the sky and the everything else. However, the tonality of the sky is often "wrong", i.e. not what I want in that setting. In other cases the tonality is good, but I am still experimenting with different lightnesses.

Changing the luminousity in RGB changes the color as well (there is no color integrity). Changing the color is a recipe for frustration. Therefor I switch to Lab space, adjust the lightness and color of the sky and convert it back to sRGB. I do this usually in an own document, for I don't want to convert the entire image back and forth with the associated losses.
Logged
Gabor

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
LAB COLOR
« Reply #47 on: November 22, 2007, 09:40:57 pm »

Quote
I often use Lab, always for a single reason. This reason may be irrelevant for other users.

I often "enrich" the images with pseudo-HDR, i.e. developing two TIFF versions from the same raw file. I do this almost always for the sake of the sky.

This way I end up with at least two layers: the sky and the everything else. However, the tonality of the sky is often "wrong", i.e. not what I want in that setting. In other cases the tonality is good, but I am still experimenting with different lightnesses.

Changing the luminousity in RGB changes the color as well (there is no color integrity). Changing the color is a recipe for frustration. Therefor I switch to Lab space, adjust the lightness and color of the sky and convert it back to sRGB. I do this usually in an own document, for I don't want to convert the entire image back and forth with the associated losses.
[a href=\"index.php?act=findpost&pid=155099\"][{POST_SNAPBACK}][/a]

I don't understand the workflow here. How do you isolate the sky on one of the layers? What's the real purpose of the Lab conversion? Do you start your workflow from a raw file? If so you can create two versions, one having the sky the way you like it, the other everything else the way you like it, render each into Photoshop and blend them using a mask perhaps with a gradient, without converting to Lab. You can change the luminosity of a layer without affecting its colour by clipping a Curves Adjustsment layer to it in Luminosity blend mode. Have you compared these options with the Lab approach? I think compared with trips to Lab these options may result in more non-destructive editing flexibility and smaller file size. But I emphasize this is a tentative suggestion because your workflow isn't entirely clear (to me).
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Panopeeper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1805
LAB COLOR
« Reply #48 on: November 22, 2007, 11:03:54 pm »

Quote
How do you isolate the sky on one of the layers?

This is the most difficult part, but it has no relevance in this question. It is sometimes painstaking, combining several methods, though there are cases, when it is quite simple.

Quote
What's the real purpose of the Lab conversion?

To be able to adjust the color and the lightness independently of each other. In RGB mode if you adjust the lightness, the color too changes.

Quote
Do you start your workflow from a raw file? If so you can create two versions, one having the sky the way you like it, the other everything else the way you like it, render each into Photoshop and blend them using a mask perhaps with a gradient, without converting to Lab

As I posted it, this is exactly what I am doing. However, it is difficult to judge the "ideal" lightness and tonality of the sky, while one does not see the other areas. Usually the sky is too bright with the "normal" lightness of the rest. When I reduce the lightness, the sky gets darker, usually more blue, but the rest becomes worthless, one does not see them "side by side".

Plus, color adjustment in ACR is by far not as clear as in Lab. In fact, it can be disastrous.

Quote
You can change the luminosity of a layer without affecting its colour by clipping a Curves Adjustsment layer to it in Luminosity blend mode

Yes, you can. That's "poor man's choise".

Quote
I think compared with trips to Lab these options may result in more non-destructive editing flexibility and smaller file size

There is nothing destructive in this setting (I don't lose any colors in the sky, at least I have never noticed that).

The file size has nothing to do with the subject, as I convert the sky back in RGB and insert it in the original image in most cases.

I did not want to elaborate on this, because it really depends on the actual circumstances, but sometimes I keep everything in Lab (I archive the file this way), and I convert it in RGB only when creating a specific version for Web or for printing.

Anyway, the file size is not a consideration. My files are usually between 100 and 800 megabytes, a few more or less megabytes don't make any difference.

Finally, a note: the selective color adjustments of PS CS3 are really appealling at first sight, but they can be devastating. They are no contest for Lab (I do use this feature often, but warily).
Logged
Gabor

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
LAB COLOR
« Reply #49 on: November 23, 2007, 09:57:59 am »

Quote
Plus, color adjustment in ACR is by far not as clear as in Lab. In fact, it can be disastrous.

You need more practice <g>. In the right hands, the ACR adjustments can be a much finer controlled tool.

Quote
Yes, you can. That's "poor man's choise".

What makes it poor?

Quote
There is nothing destructive in this setting (I don't lose any colors in the sky, at least I have never noticed that).

It is destructive no way around it using Lab. At least do it in 16-bit.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
LAB COLOR
« Reply #50 on: November 23, 2007, 09:59:56 am »

Quote
This is the most difficult part, but it has no relevance in this question. It is sometimes painstaking, combining several methods, though there are cases, when it is quite simple.
To be able to adjust the color and the lightness independently of each other. In RGB mode if you adjust the lightness, the color too changes.
As I posted it, this is exactly what I am doing. However, it is difficult to judge the "ideal" lightness and tonality of the sky, while one does not see the other areas. Usually the sky is too bright with the "normal" lightness of the rest. When I reduce the lightness, the sky gets darker, usually more blue, but the rest becomes worthless, one does not see them "side by side".

Plus, color adjustment in ACR is by far not as clear as in Lab. In fact, it can be disastrous.
Yes, you can. That's "poor man's choise".
There is nothing destructive in this setting (I don't lose any colors in the sky, at least I have never noticed that).

The file size has nothing to do with the subject, as I convert the sky back in RGB and insert it in the original image in most cases.

I did not want to elaborate on this, because it really depends on the actual circumstances, but sometimes I keep everything in Lab (I archive the file this way), and I convert it in RGB only when creating a specific version for Web or for printing.

Anyway, the file size is not a consideration. My files are usually between 100 and 800 megabytes, a few more or less megabytes don't make any difference.

Finally, a note: the selective color adjustments of PS CS3 are really appealling at first sight, but they can be devastating. They are no contest for Lab (I do use this feature often, but warily).
[a href=\"index.php?act=findpost&pid=155120\"][{POST_SNAPBACK}][/a]

Well, I have some news for you:

The way in which you capture the sky IS relevant to this discussion, because you can often do it using the colour groups in Camera raw and avoid all those difficult procedures you allude to above.

You are wrong about not being able to envision the final results by doing a luminosity blend of two separately developed raw captures. This is a very well-known technique which many photographers employ frequently and successfully. You do not need perfection at the raw development stage. All you need is two images that replicatre approximately the two luminosity scales you want in the final result, bring them into PS, layer them, layer mask them, at which point you see the whole thing together, give each its own clipped Curves Adjustment layer, and you are off to the races.

You are wrong about a Curve in RGB Luminosity Blenad Mode being a poor man's choice relative to using the L curve in L*a*b - in fact it is the other way around. You won't find this in Dan Margulis' Canyon Conundrum book, but the factual  conundrum is that the L channel in Lab mode has only fair chroma consistency when shifting the L Curve while holding the *a* and the *b* constant. I've tried both and couldn't tell the difference - LaCie 321 display and Epson 4800 printer. So no - if I want *relatively* pure luminosity adjustments, AND colour adjustments, and I want to keep them separated, I stay in RGB and use two Curves Adjustment Layers - one for luminosity and the other for colour. One often sees this recommended as "best practice".

Selective Color Adjustment Layers are only devastating in the hands of people who devastate images with them. They are not if you use them properly. They are far more intuitive and manageable than manipulating the *a* and *b* curves of L*a*b. Of course they are different tools and may therefore be used to achieve different effects. One shouldn't pre-suppose they are necessarily substitutes. "Selective Color" is one of the more unsung heros in the Photoshop arsenal. Very interesting and useful things can be done with it.

As for file sizes - I agree with you - I work on files in the 100~300 range, typically, and a few MBs more or less is perhaps no big deal, but if you can deploy techniques that consistently economize on file saize without compromising image quality, all those little bits add-up to save storage and make saving, printing and retrieval just that bit more efficient.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Brian Gilkes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 443
    • http://www.briangilkes.com.au
LAB COLOR
« Reply #51 on: November 23, 2007, 04:00:33 pm »

I hesitate to enter this one , not having yet read The Canyon Conundrum, so I will just make a couple of comments from experience. I avoid moving to and from L*a*b, partly because of losses (even in 16 bit) and partly because of some distortions in the translation process. I do use it in a duplicate image , taking the L information and placing it as a Luminosity blended layer in the original image layer stack. The purpose of this is to alter the luminosity rendering to something more approximating human brightness perception. This move is used in combination with  other luminosity and colour edits.
The use of two curves layers, one for colour (HSL) and and one for tonal control , blended accordingly , is mandatory if you want any independant control of colour and tone.
I guess the book in question  does this with the *a and *b channels . In the brief time I have played with them , this adjustment seems very touchy, and so far makes no sense in jumping out of RGB.
There still remains a problem with the RGB dual blending approach, which is an inability of the Luminosity and Color blending modes to work at sufficient levels of accuracy for exacting applications. That is , the strategy definitely helps, but color still drifts when tone is changed and vv. even on fixed points. I suspect this is similar for L*a*b space, bur have not checked it out.
There is a lot on this at www.freegamma.com which proposes, as a solution, the program (actually a very complex PS droplet) Lobster.
This was mentioned in another thread some time ago. At that time Lobster was only available for Mac, but it now seem  a Win version is available.
Lobster creates large files but solves the problem . I have no interest in the product apart from knowing one of the developers and using it when appropriate. It works.
HTH
I'm going off to play with Selective Color!
Brian
www.pharoseditions.com.au
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
LAB COLOR
« Reply #52 on: November 23, 2007, 04:27:59 pm »

Quote
..............

There still remains a problem with the RGB dual blending approach, which is an inability of the Luminosity and Color blending modes to work at sufficient levels of accuracy for exacting applications. That is , the strategy definitely helps, but color still drifts when tone is changed and vv. even on fixed points. I suspect this is similar for L*a*b space, bur have not checked it out.
There is a lot on this at www.freegamma.com which proposes, as a solution, the program (actually a very complex PS droplet) Lobster.
This was mentioned in another thread some time ago. At that time Lobster was only available for Mac, but it now seem  a Win version is available.
Lobster creates large files but solves the problem . ...............

Brian
www.pharoseditions.com.au
[a href=\"index.php?act=findpost&pid=155278\"][{POST_SNAPBACK}][/a]

Brian,

Not having gone to "freegamma.com yet, which I shall, meanwhile could you elaborate a bit what you have in mind by "sufficient levels of accuracy" and relative to what kind of exacting applications? Also, do you understand how Lobster solves whatever this problem may be - I intend to visit their site too - last time I went there, it was Mac only which ruled me out so I lost interest.

Cheers,

Mark
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

laughfta

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
LAB COLOR
« Reply #53 on: November 23, 2007, 04:56:21 pm »

Quote
...but color still drifts when tone is changed...


Are you sure this is the case in PS? I remember Guillermo did some tests in Mark's 'Your Curves' thread in which he found that hue was preserved exactly in PS Luminosity blending mode, though saturation changed. Thus was not the case in CR, or of course PS normal blending mode. Guillermo's test --posts #28 and #39
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
LAB COLOR
« Reply #54 on: November 23, 2007, 06:16:47 pm »

Quote
Are you sure this is the case in PS? I remember Guillermo did some tests in Mark's 'Your Curves' thread in which he found that hue was preserved exactly in PS Luminosity blending mode, though saturation changed. Thus was not the case in CR, or of course PS normal blending mode. Guillermo's test --posts #28 and #39
[a href=\"index.php?act=findpost&pid=155300\"][{POST_SNAPBACK}][/a]

Gloria, hue is preserved in CR, but not saturation.

Mark
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

laughfta

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
LAB COLOR
« Reply #55 on: November 23, 2007, 06:55:06 pm »

Quote
Gloria, hue is preserved in CR, but not saturation.


I was looking at Guillermo's results on that one, Mark. I certainly came away with the understanding that according to Guillermo's tests, hue was not exactly preserved (though it was close) in an CR curve adjustment:

Guillermo:
Quote
ACR curves don't preserve Hue, and in addition to this don't work exactly as a PS Normal blending mode RGB curve. Only PS Luminance blending mode seems to preserve Hue, producing however an apparent saturation loss (or at least a feeling that this happens).


Quote
Regarding LR, the new one:
- Seems to preserve Hue much better than PS Normal, but still modifies Hue.
-Increases saturation nearly the same as PS Normal.


Sorry, I didn't realize he recapped his results on the final page of the Your Curves thread, but these are his results. Have you found/heard something to the contrary? Or do you figure the changes are too close to count?

Post # 188

Gloria
Logged

Panopeeper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1805
LAB COLOR
« Reply #56 on: November 23, 2007, 06:55:46 pm »

Quote
You need more practice. <g> In the right hands, the ACR adjustments can be a much finer controlled tool

LOL, that's good.

Pls make following experiment:

1. Create a graduated block in 16-bit Lab mode, with fixed color, for example a=30, b=60. Not gray and not an RGB primary. The graduation should run from L=100 to L=50 (for example).

2. Make a copy of it and convert it in RGB.

3. Apply some luminosity change to the Lab version, like brightness -50, whatever.

4. Do the same on the RGB version with blending mode luminosity, and merge the layers.

5. Convert the modified RGB in Lab, and compared them. If you convert the modified Lab in RGB, you see the difference, but in Lab mode you can quantify it.

Quote
It is destructive no way around it using Lab

No, it is not. More accurately, it depends. Check out the gamuts of the RGB spaces: they are very good just in the blue corner.

Quote
At least do it in 16-bit

Of course I do - not from the color space conversions, but starting with ACR, up to the point of generating the presentation version. That is part of the reasons my files are hundreds of megabytes large.
Logged
Gabor

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
LAB COLOR
« Reply #57 on: November 23, 2007, 07:22:01 pm »

Quote
Pls make following experiment:

1. Create a graduated block in 16-bit Lab mode, with fixed color, for example a=30, b=60. Not gray and not an RGB primary. The graduation should run from L=100 to L=50 (for example).

2. Make a copy of it and convert it in RGB.

3. Apply some luminosity change to the Lab version, like brightness -50, whatever.

4. Do the same on the RGB version with blending mode luminosity, and merge the layers.

5. Convert the modified RGB in Lab, and compared them. If you convert the modified Lab in RGB, you see the difference, but in Lab mode you can quantify it.

I have absolutely no idea what you're trying to propose here and the instructions are anything but clear. What layers am I blending on the RGB copy? Do you mean fade/Luminosity?  And what does "but in Lab mode, you can quantify it" mean?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
LAB COLOR
« Reply #58 on: November 23, 2007, 07:23:20 pm »

Quote
Check out the gamuts of the RGB spaces: they are very good just in the blue corner.

Oh, I also don't understand what this sentence means.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Panopeeper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1805
LAB COLOR
« Reply #59 on: November 23, 2007, 07:32:51 pm »

Quote
Well, I have some news for you:

The way in which you capture the sky IS relevant to this discussion, because you can often do it using the colour groups in Camera raw and avoid all those difficult procedures you allude to above.

Well, I have some news for you: real life is more complicated. I usually use a combination of several methods. It depends on the actual images and on the personal level of acceptance, how laborous the process is.

Quote
You are wrong about not being able to envision the final results by doing a luminosity blend of two separately developed raw captures. This is a very well-known technique which many photographers employ frequently and successfully

You did not understand what I was writing. Of course you can compare them, but not in ACR. The above way may be right with your workflow, but that is not how I am creating panoramas.

Quote
All you need is two images that replicatre approximately the two luminosity scales you want in the final result, bring them into PS, layer them, layer mask them, at which point you see the whole thing together, give each its own clipped Curves Adjustment layer, and you are off to the races

This is exactly what I do and what I described above. The difference is only that I often convert the sky in Lab for adjustment.

Quote
You are wrong about a Curve in RGB Luminosity Blenad Mode being a poor man's choice relative to using the L curve in L*a*b - in fact it is the other way around

You too could profit from the experiment I describe in my previous post.

Quote
Of course they are different tools and may therefore be used to achieve different effects. One shouldn't pre-suppose they are necessarily substitutes

You got at least this one right. I am using selective color adjustments in PS (the adjustments in ACR are of very limited use). However, when adjusting the sky, I prefer Lab. I do agree, that color adjustment in Lab may be a pain.
Logged
Gabor
Pages: 1 2 [3] 4 5 6   Go Up