Walker,
I'm sorry for any miscommunication. We are saying the same thing. I maybe didn't word my statement well and therefore your confusion. First, the mid-point between 13 and 97 is (13+97)/2 is 55, not 61.5 as you said. So an L*50 in photoshop or the profile connection space will be, as I said, 55 in the print with that paper. So far so good. What you didn't understand from my wording was the whole thing with the patches. Instead of patches let's take a test chart in gray gamma 2.2. We open the test chart in photoshop, make sure we're in gray gamma 2.2, put the colorpicker in grayscale mode over the 50% patch and it reads K- 50%. Good. Now, what is the ideal L value going to be for that patch in a print on the previously mentioned paper? Continuing on, let's print a different test chart created in dot gain 20, making sure we are in dot gain 20, we then grab the color picker in grayscale mode, put it over the 50% patch and it reads K-50%. What is the ideal L value going to be for that patch in a print on that paper? Repeat again with a test chart in gray gamma 1, which we agree is the same as L. What the ideal L value for the 50% patch?
So now let's looks at how you interpreted what I was asking, and the real values of the patches. You read the L values for wherever L*50 was in the photoshop file. In our hypothetical printing situation, how, I'm not sure since L*50 in gamma 2.2 lands somewhere around the 54% patch. So unless you are printing a 100 step tablet, it's going to be hard to read without extrapolating between patches. Although we know it's going to always be 55 for that paper, so there's no need to read anything. This brings me to my next point- you said
"Long story short: keep your target in the grayscale space of your images. If you images are Dot Gain 20, your target should be. If you images are gray gamma 1.0, your target should be. We, as an industry, have settled on gray gamma 2.2."
I disagree. One can convert a file to gray gamma 2.2 or dot gain 40 or gray gamma 2.789 and the L values stay the same. As long as you can keep the underlying values straight by not confusing convert and apply profiles, and remember to embed a profile, you'll be fine. As you said, the L values are usually device independent. I can print a photo in gray gamma 2.2 then convert it to dot gain 20 and they will come out the exact same since the underlying lab values in the pcs have not changed.
One more thing- You said
"Trouble happens when the target is gray gamma 2.2 and images are dot gain 20"
I guess that's sort of true. More correctly, the problem is a misunderstanding between apply profile and convert to profile. I can easily print a gray gamma 2.2 target in either gray gamma 2.2 or dot gain 20, or dot gain 40 and get the same results if I convert between profiles before printing. Applying a different profile will cause trouble. When I get photos from clients, the only trouble could happen when there is no embedded profile. If a client sends me a photo without an embedded profile, I can guess at what profile it was in. But let's say the client's photo was in dot gain 20, and photoshop opens it up and applies the gray gamma 2.2 profile. The same trouble happens, but in reverse of what you said. The wrong profile was applied, and the meaning of the lab values have changed.
Honestly I was just trying to help Sandy King with a calibration issue. I'm not trying to figure out anything or trying to do anything. I don't even use qtr. In the end we are saying the same thing, you know your stuff, but communication in forums is always awful.
Best,