Dear Roger,
Dear Jan,
Except, of course, that sharpness and fine grain are to some extent mutually exclusive: you will never get both the maximum sharpness of which a film is capable, and the finest grain of which it is capable, at the same time from the same developer.
with all respect, but in this generalization that is wrong.
I've tested dozens of different developers in the last decades. And there are several which deliver both excellent sharpness
and very fine grain (see below).
Especially in the last 15 years there have been quite a lot R&D by innovative smaller chemistry manufacturers (the big ones like Kodak and Ilford have not cared for chemistry for a long time; they don't even make it anymore by themselves: Tetenal is producing both the old Ilford and Kodak formulas for them).
Try for example Spur HRX, Adox ADX, Rollei RLS, Moersch Finol, Moersch Eco (for slow and medium speed films), Moersch Tanol, Prescysol EF, Spur Dokuspeed, Spur Nanospeed (for HR films) etc.
That excellent sharpness and very fine grain exclude each other to a certain extent is correct for the old (outdated) Kodak, Ilford, Agfa and Tetenal developer formulas.
But it is not true in general. And especially not true for several new, recent formulas.
Also, the concept of an "ideal characteristic curve" is dubious:a lot will depend on where you place the minimum exposure (shadow detail) on the characteristic curve. The late Geoffrey Crawley believed in keeping the straight line portion of the characteristic curve as straight as possible, but you're still going to have a toe and a shoulder...
Cheers,
R.
For the best tonality an even, consistent increase in density from Zone I / II upward is optimal = linear characteristic curve.
Toe:
If you have a (very) good film-developer combination the toe is below Zone II / I. In that area it is irrelevant for the picture.
Shoulder:
Depends on the film-developer combination. Not everyone has a shoulder.
For example both Delta 100 and FP4+ developed in Ilford DD-X have an absolutely straight linear curve without any shoulder up to Zone X. Perfect linear characteristic curve.
Cheers, Jan