Thank you for the recommendation. I looked at few pages available in the net and purchased the book.
Thank you. I had a feeling that the integrated contrast should be close to linear, as we see it. Your formula makes it perfectly clear.
As far as I understand the more contrasty is the paper the shorter is the input range, the usable range as you marked it on your graph. But even with a steepest curve we still have the toe and shoulder I think, is it right? From printing the tablet I can see that the lengths of the toe and shoulder change somehow, but not by very much. Does it sound right?
I can also see that on a polycontrast paper the curve "breaks", is it because it is a mixture of two (three?) emulsions that start to be "density misaligned" in development if one goes to extreme contrast with filters?