I have finished watching the two series of LAB videos on kelbytraining.com and I tried an interesting experiment. At the end of the series are some images from others he applies his techniques to. After seeing his work on the first one I decided to try working on the image on my own (they supply many of the images). I stayed in RGB, did an Image/Adjustments/Auto Colour to the image, then used Akvis Enhancer and voila, I had a better result in two steps than Margolis had in 20 or so. Might be luck, so I did the same thing with the second image after only watching the beginning of his process. This time I had to do a bit of saturation adjustment in addition to the above three steps, but still, we're looking at 3 vs. 20 steps. For the last image I did my adjustments before watching the video and the results were very similar - 3 steps for me, 20 for him.
So, does this mean that L.A.B. is a crock? Hardly, but it's like I say to my patients, if you see a gall bladder surgeon with abdominal pain - there's a good chance he's going to recommend surgery - it's the only tool he has, and he's either going to make that tool work, or admit he can't help you. To a hammer, everything looks like a nail, even a screw. Well, seems to me that by embracing L.A.B. you run the risk of being the carpenter who only has a hammer. Great tool but sometimes a screwdriver is what's really called for.
I think this means that we need to learn where L.A.B. excels, where it can do things not possible other ways and Margulis certainly shows that - his blurring of a and b to reduce noise was nothing short of miraculous. Some of the manipulations which only apply to the problem areas of an image without messing with the rest of it through 'blend if' and 'apply image' are very impressive. His showing me how to down play the damage to light tones in my sharpening routine is really important to me. Other techniques, well just maybe a screwdriver is sometimes the better tool.