Luminance in LAB vs. RGB?

My point, as you know, was not whether he makes a strong or weak argument but that his views are controversial.

It remains that many think in practice the difference is negligible and not worth changing color modes.

I notice you ignore his 8-bit argument, which is even more at odds with the consensus.
 
I could have but that was the real purpose of the post. The explanation of "the claw" was a mere ruse to take a swipe at Kelby's unfunny jokes.
 
Page 133. To be accurate he says open in 16-bit but then convert to 8-bit at your convenience.

This guy, who wrote the classic PS Channel Chops book, apparently finds it "productive" to debate it.
In fact he goes on a rant against this convert to 8-bit theory, here
http://attentionphotoshoppers.libsyn.com/
He specifically mentions the LAB book.

While I found the LAB book interesting I think it is one of those books many will read and very few will put into practice because the techniques are of questionable utility in light of easier and more efficient alternatives.
 
Page 133. To be accurate he says open in 16-bit but then convert to
8-bit at your convenience.
This guy, who wrote the classic PS Channel Chops book, apparently
finds it "productive" to debate it.
In fact he goes on a rant against this convert to 8-bit theory, here
http://attentionphotoshoppers.libsyn.com/
He specifically mentions the LAB book.
While I found the LAB book interesting I think it is one of those
books many will read and very few will put into practice because
the techniques are of questionable utility in light of easier and
more efficient alternatives.
I hadn't seen that reference on that page. My point is that the book is not "about" 8-bit vs. 16-bit. Make your own choice and then read the book for everything else it can teach you. Apparently Dan chooses to argue this point elsewhere (which doesn't matter to me one way or the other), but that argument is not what the book is about.

Whether or not you find LAB mode useful (I find it incredibly useful on at least 50% of my images), this is the best book I've seen (I own at least 10 other books on PS) to teach you about channel blends, blend modes, curves, blend-if settings, color correction, color modes, contrast enhancement, separating color from contrast, sophisticated retouching without masks or selections, etc....

While I use and appreciate what I've learned about the LAB mode, I'd hazard a guess that more than 80% of what I've learned is really generic to any mode and can be used in RGB or CMYK too.

I don't need to argue this with you, just thought I'd share my experience. Your mileage may vary.

--
John
Gallery: http://jfriend.smugmug.com/portfolio

Smugmug discount coupon '3elo1xh75JSiI' to save $5
 
Oh, I missed the Ch. 14 summary since it isn't in the Big Listing
of Chapter Summaries ( http://www.dgrin.com/showthread.php?t=18203 ).
Thanks for that. I guess I'll just search for the summaries that
aren't in that list so I can read them.
I posted the chapter 14 summary just yesterday and it hasn't been added to the main page yet by the owner of that page. It's on his todo list (I emailed him about it yesterday). I think there are two more chapter summaries coming soon, but I don't think there are others that aren't in the main list so no need to go searching.
--
John
Gallery: http://jfriend.smugmug.com/portfolio

Smugmug discount coupon '3elo1xh75JSiI' to save $5
 
I notice you ignore his 8-bit argument, which is even more at odds
with the consensus.
I find it interesting how people like to discredit everything someone says because they disagree with one point they make. One can go through life that way, but that closes off a lot of sources of good information.

I happen to disagree with several points that Dan makes in his book. I even say so in my chapter 14 summary that I posted yesterday on one of the points he makes in that chapter. But, on the whole he knows a ton more about how to get the most out of PS than I do and I learn a lot from studying his writings. Read, understand, learn, make your own decisions about what is useful to you.
--
John
Gallery: http://jfriend.smugmug.com/portfolio

Smugmug discount coupon '3elo1xh75JSiI' to save $5
 
... and just
saturate the L channel. I really liked the effect this had on the
water and waves, and I was curious how this effect could be
achieved in RGB.
What does that mean? Saturation applies only to colors so it is impossible to saturate the L channel. It only contains luminence information.

--
Ed C.
 
I find it interesting how people like to discredit everything someone says because they disagree with one point they make.
Obviously there is some kind of projection going on here. To state the book has controversial ideas (it does) and to point to two specific examples, from memory, well how one then reaches the conclusion that that constitutes "people" who "like to discredit everything someone says because they disagree with one point". Like I said, serious projection.
One can go through life that way, but that closes off a lot of sources of good information.
Again, obviously some tendency to want to make philosophical statements about "life" rather then addressing the specific issues.
I happen to disagree with several points that Dan makes in his book.
Of no relevance to whether his book makes controversial points or not.
I even say so in my chapter 14 summary that I posted yesterday on one of the points he makes in that chapter.
As you brought it up, why you find the need to "summarize" his content when the primary source material is readily available is a mystery. Are you his translator?
But, on the whole he knows a ton more about how to get the most out of PS than I do and I learn a lot from studying his writings.
Again, irrelevant to the point made re whether it is controversial or not.
Read, understand, learn, make your own decisions about what is useful to you.
 
Obviously there is some kind of projection going on here. To state the book has controversial ideas (it does) and to point to two specific examples, from memory, well how one then reaches the conclusion that that constitutes "people" who "like to discredit everything someone says because they disagree with one point". Like I said, serious projection.
One can go through life that way, but that closes off a lot of
sources of good information.
Again, obviously some tendency to want to make philosophical statements about "life" rather then addressing the specific issues.
I happen to disagree with several points that Dan makes in his book.
Of no relevance to whether his book makes controversial points or not.
I even say so in my chapter 14 summary that I posted yesterday on
one of the points he makes in that chapter.
As you brought it up, why you find the need to "summarize" his content when the primary source material is readily available is a mystery. Are you his translator?
But, on the whole he knows a ton more about how to get the most out
of PS than I do and I learn a lot from studying his writings.
Again, irrelevant to the point made re whether it is controversial or not.
Read, understand, learn, make your own decisions about what is
useful to you.
I no longer know what we're arguing about. In my opinion (you are free to have a different opinion), this book has more concrete, non-controversial teachings about layers, blend modes, channels, blend-if settings, contrast enhancement, color correction, color enhancement, color modes and channel blends than any book I've seen.

Since you asked, I wrote a chapter summary as part of an online reading group ( http://www.dgrin.com/showthread.php?t=18203 ) that has summarized most of the chapters in the book for the purposes of discussing what people have learned from the book.

I'll be signing off from this part of the discussion now.

--
John
Gallery: http://jfriend.smugmug.com/portfolio

Smugmug discount coupon '3elo1xh75JSiI' to save $5
 
Thank you for allowing me an opinion. Like I said initially, it's an interesting book. Some good, some controversial, some gibberish. But interesting.
 
I posted the chapter 14 summary just yesterday and it hasn't been
added to the main page yet by the owner of that page. It's on his
todo list (I emailed him about it yesterday). I think there are
two more chapter summaries coming soon, but I don't think there are
others that aren't in the main list so no need to go searching.
That would explain it. Thanks for the info.

--
Patrick O'Leary
http://patrick.greentaperacing.us/blog/gallery/
 
All I know is that in this script there are selections for each of the channels, to either saturate or have no change. There's also a slider for saturation percentage. I'm guessing that whichever channels are selected with the "saturate" option, the curves are adjusted in the same way based on the percentage selected.

I've looked at the script, but I really can't make heads or tails out of how it works. If you know anything about script-fu, I'll be glad to send you a link to where you can download the script and see it for yourself.
 
It appears, from my limited understanding of Scheme, and the reference for gimp-curves-spline in the database browser, that the LAB Color Punch script-fu is applying a sharp S-curve to each channel it is instructed to "saturate". The control points are: [(0, 0), (curvestr, 0), (255-curvestr, 255), (255, 255)]. "curvestr" is the strength of the curve, which it appears is set in the script-fu dialog box. This is similar to (not identical to; this will spline instead of be straight) the basic curves applied to the A and B channels in Margulis Ch. 1. However, it is not typically a good idea to apply this same curve to the L channel--depending on how much of the histogram of the L channel is used, you might plug some shadows or blow out highlights by this technique.

The script is (1) a shortcut for Margulis Ch. 1 curves on A and B, and (2) a really bad way of adjusting the L curve. I would recommend just learning to use the Curves on the LAB image directly, rather than through the script. At least, use the script to adjust A and B, but don't let it touch the L; do it yourself to enhance contrast in the areas of the image that need it.

Good luck learning this stuff; it's technical but it's worth it in the end.

--
Patrick O'Leary
http://patrick.greentaperacing.us/blog/gallery/
 
Thanks, Patrick.

I don't know how Photoshop handles working in LAB, but in GIMP, when an image is decomposed into LAB, you get three grayscale representations, one for each channel. It is very difficult for me to adjust curves and be able to visualize through the grayscale representations what the effect will be on the final image. While the script does not allow for very specific control of the curves, it at least lets me try something and see the results much more quickly than having to do the decomposing and recomposing myself.
 
Right, I forgot that the GIMP decomposes into separate documents. In that case, following from what I said before, go ahead and run the script-fu on the A and B channels. The L channel looks like a greyscale image, and though it's difficult to visualize, if you apply a curve to make it look like a better greyscale image, expanding contrast where needed and compressing it where you don't, you'll notice it when you recomposite into RGB. Or apply the script to A and B and then adjust the curve on the RGB composite, which is not exactly the same thing but close enough.

It's been a couple of years since I last used the GIMP seriously--it's a great program at a great price, but I don't know how I'd live now without adjustment layers.

--
Patrick O'Leary
http://patrick.greentaperacing.us/blog/gallery/
 

Keyboard shortcuts

Back
Top