Yes, its global. How easy it will be to mask, if necessary, will depend on the shot. There are some shots where this is probably the simplest and best way to approach this sort of issue.
There are also some shots where the easiest, and perhaps best, solution is a set of RGB curves, one for each channel. In general, I tend to prefer this approach to one that involves masking and painting on the mask. There are other ways to mask too, and there's always the blending options.
I've got one minor nit to pick about something Andrew said. He said that the only way to know a neutral is a neutral is in a well behaved RBG color space. You can also tell neutrality from 0 readings in the AB channels in LAB. And you don't have to convert the file to LAB to get the LAB numbers. All you need to do is set the right hand info palette numbers to LAB, and voila. There are other uses for LAB numbers as well, beyond finding neutrals. If you get fur that is showing negative in the B channel, that should raise a red flag. So should grass that is B negative. And there are a host of other things that LAB numbers can help you with.
When it comes to using my eyes, I'm still in the "Reagan" school: Trust, but verify. For me, becoming sensitive to casts and color problems is definitely an ongoing learning process. If others get great results just from using their eyes, then more power to them. I'm not there yet, and doubt that I ever will be.
As one can see from the animated GIF image of the Photoshop Color window/palette - the same skintone has a similar pleasing ratio in both RGB and CMY(K) and that the Lab colour reading is also very intuitive from the slider positions, as a visual indicator of the colour.
www.curvemeister.com has updated it's custom curves plug, and now has "watches" where the watch hands can be used to evauluate the relationship of colour in a similar but totally different fashion.
Just like with setting a neutral, one has to use care and judgement and careful evaluation when setting skintones.
The basic guideline/suggested "ratio" or "formula" of R to G to B, or C to M to Y or A to B in Lab space (not forgetting L) is, as mentioned a guideline and it will depend on many factors on whether it is appropriate for the particular image at hand (and faces and children can be extra challenging). It may not be appropriate for the lighting or subject to force a "bronzed tan" colour found in a good lighting with no shadow etc (when it comes to memory colours, most people prefer a skintone that is *not* true to life, just as they may prefer the sky to be more saturated than gray, or grass more greener etc).
P.S. The colour samples below were taken from a colour managed BetterLight scanback camera image file, they are unedited values which fall within the basic formula/ratio.
Yes, I've seen that optical illusion and I'll be sure to keep it in mind when
I'm shooting a chessboard with a cylinder on it. But the fact is, that if you
get an image to look and print in such a way that you are pleased with the
final result, then no amount of theory, men from Mars or channel ratios
is going to make that image "incorrect". Of course I'm not saying one should
discount all these methods. I use them all the time. I'm just saying that
sometimes people forget that the actual reason for using all of these fancy
tools is to get an image that we like. Nothing more.
How about when shooting a frame with a shadow across it also?
Bernard,
I fully agree that the end result from all the talk about color and numbers is to create an image, or more importantly for some of us, a print, that is pleasing to the eye of the beholder - especially when examined under a good, full spectrum mimic of daylight.
It's got to be lots more complicated than that doesn't it?
I can point you into a direction whereby an author will instruct you to convert to Lab, blend all kinds of channels, convert to CMYK, jump through some more hoops and the net result is you've thrown away more useful bits and time. And you'll feel real macho after the 23 steps and some of your fellow Photoshop users might be impressed.
Again, KISS, render pixels as best you can BEFORE you get to Photoshop and remember Bruce's quote above. Life's too short to fall in love with difficult Photoshop techniques for the sake of the technique, not the final results.
Yes, I've seen that optical illusion and I'll be sure to keep it in mind when
I'm shooting a chessboard with a cylinder on it.
I use that URL a lot to illustrate NOT that you can't trust your eyes when looking at colors in context but that you can't trust your eyes when looking at solid colors in an attempt to calibrate a display visually. A Colorimeter would measure both grays as the same color of course. Which is what a Colorimeter needs to do to profile a display, read solid colors of known color values.
An image of a human is hundreds, thousands or millions of solid colors but in context, it looks like a human. The original image shown here and the corrections are prefect examples. Once you do calibrate and profile the display, the degree of such optical illusions is pretty low. They can exist but the point is, the optical illusion specifies a very well defined deficiencies in human perception that I don't think illustrates well that you can't now use your calibrated and profiled display to visually edit a complex image.
BTW, if you print the optical illusion (or the image in question), the same holds true. We have to look at the image in context, either on the display or print. If it looks good to us, even due to some optical illusion, then that's OK. If the optical illusion fools us into setting a device in a consistent way, that's bad.
I've got one minor nit to pick about something Andrew said. He said that the only way to know a neutral is a neutral is in a well behaved RBG color space. You can also tell neutrality from 0 readings in the AB channels in LAB.
Yes of course, true. I was referring to RGB and CMYK (neither of which guarantee neutrality numerically unless you've got a well behaved working space). But you I'd certainly hope anyone working in RGB would be in such a color space, that being the case, no need to look at Lab numbers, a rather non intuitive scale to work in.
Lab has been greatly over sold considering its original design. Lab was designed as a tool for matching solid colors. It was not designed as a tool for matching overall image appearance. It has all kinds of issues.
Lab contains 16.7 million colors (as does any 3-channel, 8-bit-per-channel
space). But because the Lab primaries are at the limits of human vision, you
end up with a color space that contains 16.7 million colors, of which perhaps 5 or 6 million are actually perceptible, let alone reproducible.
One of the best posts on the subject dates way back to 1999, long before it was so popular to spoon feed users on Lab books, by the great Bruce Fraser. Its as pertinent today as it was then, maybe more so considering what we can now accomplish using a Raw rendering workflow that doesn't have any use for Lab.
Let me make it clear that I'm not adamantly opposed to Lab workflows. If
they work for you, that's great, and you should continue to use them.
My concern is that Lab has been oversold, and that naive users attribute to
it an objective correctness that it does not deserve. (AR: he said that in 1999, its more over sold today!).
Even if we discount the issue of quantization errors going from device space
to Lab and vice versa, which could be solved by capturing some larger number
of bits than we commonly do now, (though probably more than 48 bits would be
required), it's important to realise that CIE colorimetry in general, and
Lab in particular, have significant limitations as tools for managing color
appearance, particularly in complex situations like photographic imagery.
CIE colorimetry is a reliable tool for predicting whether two given solid
colors will match when viewed in very precisely defined conditions. It is
not, and was never intended to be, a tool for predicting how those two
colors will actually appear to the observer. Rather, the express design goal
for CIELab was to provide a color space for the specification of color
differences. Anyone who has really compared color appearances under
controlled viewing conditions with delta-e values will tell you that it
works better in some areas of hue space than others.
When we deal with imagery, rather than matching plastics or paint swatches,
a whole host of perceptual phenomena come into play that Lab simply ignores.
Simultaneous contrast (AR: that optical illusion above), for example, is a cluster of
phenomena that cause the same color under the same illuminant to appear differently
depending on the background color against which it is viewed. When we're working with
color-critical imagery like fashion or cosmetics, we have to address this
phenomenon if we want the image to produce the desired result -- a sale --
and Lab can't help us with that.
Lab assumes that hue and luminance can be treated separately -- it assumes
that hue can be specified by a wavelength of monochromatic light -- but
numerous experimental results indicate that this is not the case. For
example, Purdy's 1931 experiments indicate that to match the hue of 650nm
monochromatic light at a given luminance would require a 620nm light at
one-tenth of that luminance. Lab can't help us with that. (This phenomenon
is known as the Bezold-Brucke effect.)
Lab assumes that hue and chroma can be treated separately, but again,
numerous experimental results indicate that our perception of hue varies
with color purity. Mixing white light with a monochromatic light does not
produce a constant hue, but Lab assumes it does -- this is particularly
noticable in Lab modelling of blues, and is the source of the blue-purple
shift.
There are a whole slew of other perceptual effects that Lab ignores, but
that those of us who work with imagery have to grapple with every day if our
work is to produce the desired results.
So while Lab is useful for predicting the degree to which two sets of
tristimulus values will match under very precisely defined conditions that
never occur in natural images, it is not anywhere close to being an adequate
model of human color perception. It works reasonably well as a reference
space for colorimetrically defining device spaces, but as a space for image
editing, it has some important shortcomings.
One of the properties of LCH that you tout as an advantage -- that it avoids
hue shifts when changing lightness -- is actually at odds with the way our
eyes really work. Hues shift with both lightness and chroma in our
perception, but not in LCH.(AR: We have something like this in Lightroom).
None of this is to say that working in Lab or editing in LCH is inherently
bad. But given the many shortcomings of Lab, and given the limited bit depth
we generally have available, Lab is no better than, and in many cases can be
worse than, a colorimetrically-specified device space, or a colorimetrically
defined abstract space based on real or imaginary primaries.
For archival work, you will always want to preserve the original capture
data, along with the best definition you can muster of the space of the
device that did the capturing. Saving the data as Lab will inevitably
degrade it with any capture device that is currently available. For some
applications, the advantages of working in Lab, with or without an LCH
interface, will outweigh the disadvantages, but for a great many
applications, they will not. Any time you attempt to render image data on a
device, you need to perform a conversion, whether you're displaying Lab on
an RGB monitor, printing Lab to a CMYK press, displaying scanner RGB on an
RGB monitor, displaying CMYK on an RGB monitor, printing scanner RGB to a
CMYK press, etc.
Generally speaking, you'll need to do at least one conversion, from input
space to output space. If you use Lab, you need to do at least two
conversions, one from input space to Lab, one from Lab to output space. In
practice, we often end up doing two conversions anyway, because device
spaces have their own shortcomings as editing spaces since they're generally
non-linear.
The only real advantage Lab offers over tagged RGB is that you don't need to
send a profile with the image. (You do, however, need to know whether it's
D50 or D65 or some other illuminant, and you need to realise that Lab (LH)
isn't the same thing as Lab.) In some workflows, that may be a key
advantage. In many, though, it's a wash.
One thing is certain. When you work in tagged high-bit RGB, you know that
you're working with all the data your capture device could produce. When you
work in Lab, you know that you've already discarded some of that data.
I don't think any of the color spaces are more or less intuitive than the others. To a certain extent they are all learned, and which one seems intuitive is probably more a function of which one a person learned first and/or uses most.
Thanks for posting the post by Bruce Fraser. Very interesting stuff.
I don't think any of the color spaces are more or less intuitive than the others.
I don't find positive and negative a and b star values inherently intuitive. I don't know I'd start a new user with such a scale, especially since there are no capture or output color spaces based on Lab and there are few reasons to assign color using this scale for the tasks at hand.
And if we look at modern image capture and processing, everything is going to start from Raw into some RGB working space so might as well start there. Defining neutrals, highlight and clipping values (0-255 or 0-100%) is pretty easy to teach and grasp.
Getting back to macho image processing, if you've got the time and desire, sure, work in Lab, XYZ, CMYK, whatever.
Bernard,
I fully agree that the end result from all the talk about color and numbers is to create an image, or more importantly for some of us, a print, that is pleasing to the eye of the beholder - especially when examined under a good, full spectrum mimic of daylight.
Shooting a frame of film ( an image) with a shadow across it - like a side lighted cylinder on a chess field, or maybe a sunlit grain bin, next to a harvested cornfield. I live in the MidWest, remember!!??
I can point you into a direction whereby an author will instruct you to convert to Lab, blend all kinds of channels, convert to CMYK, jump through some more hoops and the net result is you've thrown away more useful bits and time. And you'll feel real macho after the 23 steps and some of your fellow Photoshop users might be impressed.
Again, KISS, render pixels as best you can BEFORE you get to Photoshop and remember Bruce's quote above. Life's too short to fall in love with difficult Photoshop techniques for the sake of the technique, not the final results.
Andrew, I am sure you know that many of us, here on the Finishing thread, have read, and studied at some length, both of Dan's books. Many of us feel that we have learned new skills as a result. His techniques are not always intuitive, or simple to master, and I, for one, have much, more to learn.
If there are faster and easier ways to accomplish the same results, I am always a Big Fan of easier and faster, if the results are of the same level of quality. I like KISS.
In that vein, I do like the new Adobe Raw convertor 4.3 - I find that to be a real valuable asset in my workflow.
Comments
Easy enough to mask out in PS though.
Moderator of the Technique Forum and Finishing School on Dgrin
There are also some shots where the easiest, and perhaps best, solution is a set of RGB curves, one for each channel. In general, I tend to prefer this approach to one that involves masking and painting on the mask. There are other ways to mask too, and there's always the blending options.
I've got one minor nit to pick about something Andrew said. He said that the only way to know a neutral is a neutral is in a well behaved RBG color space. You can also tell neutrality from 0 readings in the AB channels in LAB. And you don't have to convert the file to LAB to get the LAB numbers. All you need to do is set the right hand info palette numbers to LAB, and voila. There are other uses for LAB numbers as well, beyond finding neutrals. If you get fur that is showing negative in the B channel, that should raise a red flag. So should grass that is B negative. And there are a host of other things that LAB numbers can help you with.
When it comes to using my eyes, I'm still in the "Reagan" school: Trust, but verify. For me, becoming sensitive to casts and color problems is definitely an ongoing learning process. If others get great results just from using their eyes, then more power to them. I'm not there yet, and doubt that I ever will be.
Duffy
www.curvemeister.com has updated it's custom curves plug, and now has "watches" where the watch hands can be used to evauluate the relationship of colour in a similar but totally different fashion.
Just like with setting a neutral, one has to use care and judgement and careful evaluation when setting skintones.
The basic guideline/suggested "ratio" or "formula" of R to G to B, or C to M to Y or A to B in Lab space (not forgetting L) is, as mentioned a guideline and it will depend on many factors on whether it is appropriate for the particular image at hand (and faces and children can be extra challenging). It may not be appropriate for the lighting or subject to force a "bronzed tan" colour found in a good lighting with no shadow etc (when it comes to memory colours, most people prefer a skintone that is *not* true to life, just as they may prefer the sky to be more saturated than gray, or grass more greener etc).
http://www.panix.com/%7Erbean/color/color4.txt
http://www.panix.com/%7Erbean/color/
P.S. The colour samples below were taken from a colour managed BetterLight scanback camera image file, they are unedited values which fall within the basic formula/ratio.
Stephen Marsh.
http://members.ozemail.com.au/~binaryfx/
http://members.ozemail.com.au/~binaryfx/
http://prepression.blogspot.com/
Yes, I've seen that optical illusion and I'll be sure to keep it in mind when
I'm shooting a chessboard with a cylinder on it.
get an image to look and print in such a way that you are pleased with the
final result, then no amount of theory, men from Mars or channel ratios
is going to make that image "incorrect". Of course I'm not saying one should
discount all these methods. I use them all the time. I'm just saying that
sometimes people forget that the actual reason for using all of these fancy
tools is to get an image that we like. Nothing more.
http://bertold.zenfolio.com
Bernard,
I fully agree that the end result from all the talk about color and numbers is to create an image, or more importantly for some of us, a print, that is pleasing to the eye of the beholder - especially when examined under a good, full spectrum mimic of daylight.
Moderator of the Technique Forum and Finishing School on Dgrin
I can point you into a direction whereby an author will instruct you to convert to Lab, blend all kinds of channels, convert to CMYK, jump through some more hoops and the net result is you've thrown away more useful bits and time. And you'll feel real macho after the 23 steps and some of your fellow Photoshop users might be impressed.
Again, KISS, render pixels as best you can BEFORE you get to Photoshop and remember Bruce's quote above. Life's too short to fall in love with difficult Photoshop techniques for the sake of the technique, not the final results.
Author "Color Management for Photographers"
http://www.digitaldog.net/
I use that URL a lot to illustrate NOT that you can't trust your eyes when looking at colors in context but that you can't trust your eyes when looking at solid colors in an attempt to calibrate a display visually. A Colorimeter would measure both grays as the same color of course. Which is what a Colorimeter needs to do to profile a display, read solid colors of known color values.
An image of a human is hundreds, thousands or millions of solid colors but in context, it looks like a human. The original image shown here and the corrections are prefect examples. Once you do calibrate and profile the display, the degree of such optical illusions is pretty low. They can exist but the point is, the optical illusion specifies a very well defined deficiencies in human perception that I don't think illustrates well that you can't now use your calibrated and profiled display to visually edit a complex image.
BTW, if you print the optical illusion (or the image in question), the same holds true. We have to look at the image in context, either on the display or print. If it looks good to us, even due to some optical illusion, then that's OK. If the optical illusion fools us into setting a device in a consistent way, that's bad.
Author "Color Management for Photographers"
http://www.digitaldog.net/
Yes of course, true. I was referring to RGB and CMYK (neither of which guarantee neutrality numerically unless you've got a well behaved working space). But you I'd certainly hope anyone working in RGB would be in such a color space, that being the case, no need to look at Lab numbers, a rather non intuitive scale to work in.
Lab has been greatly over sold considering its original design. Lab was designed as a tool for matching solid colors. It was not designed as a tool for matching overall image appearance. It has all kinds of issues.
Lab contains 16.7 million colors (as does any 3-channel, 8-bit-per-channel
space). But because the Lab primaries are at the limits of human vision, you
end up with a color space that contains 16.7 million colors, of which perhaps 5 or 6 million are actually perceptible, let alone reproducible.
One of the best posts on the subject dates way back to 1999, long before it was so popular to spoon feed users on Lab books, by the great Bruce Fraser. Its as pertinent today as it was then, maybe more so considering what we can now accomplish using a Raw rendering workflow that doesn't have any use for Lab.
Author "Color Management for Photographers"
http://www.digitaldog.net/
Thanks for posting the post by Bruce Fraser. Very interesting stuff.
Duffy
I don't find positive and negative a and b star values inherently intuitive. I don't know I'd start a new user with such a scale, especially since there are no capture or output color spaces based on Lab and there are few reasons to assign color using this scale for the tasks at hand.
And if we look at modern image capture and processing, everything is going to start from Raw into some RGB working space so might as well start there. Defining neutrals, highlight and clipping values (0-255 or 0-100%) is pretty easy to teach and grasp.
Getting back to macho image processing, if you've got the time and desire, sure, work in Lab, XYZ, CMYK, whatever.
Author "Color Management for Photographers"
http://www.digitaldog.net/
I'm not sure what this is in reference to.
Right on.
http://bertold.zenfolio.com
Shooting a frame of film ( an image) with a shadow across it - like a side lighted cylinder on a chess field, or maybe a sunlit grain bin, next to a harvested cornfield. I live in the MidWest, remember!!??
Moderator of the Technique Forum and Finishing School on Dgrin
Andrew, I am sure you know that many of us, here on the Finishing thread, have read, and studied at some length, both of Dan's books. Many of us feel that we have learned new skills as a result. His techniques are not always intuitive, or simple to master, and I, for one, have much, more to learn.
If there are faster and easier ways to accomplish the same results, I am always a Big Fan of easier and faster, if the results are of the same level of quality. I like KISS.
In that vein, I do like the new Adobe Raw convertor 4.3 - I find that to be a real valuable asset in my workflow.
Moderator of the Technique Forum and Finishing School on Dgrin