Originally Posted by redcrown
A lot of people guess that it's just a "midtone contrast" adjustment. It does not take much testing to see that it's much more than that. It adjusts all tones, yet does protect shadows and highlights.
Then it can't affect all tones <g>. I agree with you, it's not just Midtone now that I've looked further. The image I supplied was Lab inside the LR.'s Develop module so that might play a role. Running Mac Holbert's routine I see a big difference but even that is odd when viewing the numbers, probably due to gamma and blending of the layers*. In LR I do see very dark and very light tone's having no numeric differences in all the settings. And George replied to me via email about trying PV2012 vs. 2012. I didn't see much difference on the Lab image but the one he sent was by default more aggressive.
*The blend if are set for values to begin the effect but when I run a single pixel sample over a 0-255 gradient, this isn't correlating.
In the example of Andrew's step wedge, that's why measuring in the middle of the wedges shows so little change.
Nor on Mac's technique. But the differences in when the numbers change and by how much is quite different, at least on this end, using Lightroom or Photoshop. In the end, I guess what visually looks best counts.
I'd prefer to do this work parametrically in LR/ACR if I can get the look I want.