Quote Originally Posted by Michael R 1974 View Post
I think we're kind of saying the same thing. CI works fine within a "normal" exposure range. For high contrast scenes it doesn't tell me enough about the shape of the shoulder for example, particularly when applying minus development. This is where targetting the paper range can be dangerous.
Fortunately in Rafal's case, we are sticking to N-1 at most. His current curve crosses the "paper" right at the end of the data and there isn't a shoulder on the graph. The danger in this case is going to be uneven development...

But if that is the danger, your advice is still good, to look at the "greater picture". It may be better to give normal development to N-1 and expect to print on a No. 1 paper than to risk uneven development. Or repeat a test using a compensating developer to see the possibilities. Depending on the developer and technique chosen, the shape of the curve may change.