Originally posted by Ikarus I took some shots of my computer screen that illustrate the typical behavior of my copy. The top crop is f/4, the second f/2, the third the same with contrast enhancement. F/4 is crips and sharp, but note the line doubling at f/2 - the letters have a shadow with a shift of 45 degrees by about 1 pixel. The focusing was done in live view.
I thought this looked very interesting and so I decided to perform the same test with my own copies. However, I discovered that my LCD screen was creating a sort of color registration offset effect(somewhat similar to your own examples) which didn't seem to exist with a reflective surface. And so I moved the test to a paper target instead.
And so to help with consistency, would it be possible for you to use a paper target with text or similar to conduct your test shots on as well? That way we could compare notes. Because I'd be very interested in seeing how the results compare between them(different models).
Also in testing, I discovered that the best way to nail focus was to fine tune the lens at
f/5.6 and then drop to f/2 and take the shot afterward without touching anything. Since the J-9 can be very tricky to nail focus with wide open(and by tricky, I mean veeeeery tricky
).
And to help illustrate, here's an example of what I found on paper using a 1987 MC black version:
fr. RAW 100% (untouched): fr. RAW 200% (untouched): fr. RAW 200% w/RawTherapee: