But I'm going to probably break some internet rule and cross-post a message. As someone who is also looking at the S3, I found this short commentary from one of the pro groups' boards I subscribe too (the ones I keep hyping here), very interesting. I'm sure many users here will too. I'm not posting names to protect the author:
"Funny you should mention this. I was at an all-day digital workflow seminar
yesterday sponsored by Fuji. The speaker was Will Crockett and he was
using, and hyping a bit (as expected), an S3. He had a series of four
shots, all in-camera jpegs, that had been blown up by a pro lab to 30"x40"
prints. All were the highest quality jpegs, the "fine" setting, but each
was a different native file size -- ~38MB, ~18MB, ~10MB, and ~4MB ( approx.
12, 8, 4 and 1MB's respectively as jpegs?). I absolutely could not believe
the quality of the 30x40 print created from the 1MB jpeg! Yes, there was a
little bit of aliasing in the fine detail, but that had mostly gone away by
the print from the next size up, and was completely gone in the print from
the 8MB jpeg. Areas of continuous color were smooth and accurate, there
appeared to be little or no loss of either shadow or highlight detail, and
there were no "mushy" midtones. The little bit of aliasing was all I could
see to tell the difference.
His points about accurate exposure within small tolerances, not messing with
the files too much in post, and using a lab that's prepared to rez up
something that much were well taken, not always practical for all shooters,
and do tend to limit one's options down the line. Nevertheless, judging
only by the quality of a print that large from a jpeg that small, I was
pretty darn impressed."
Sounds good to me...
Regards,
Andy Pearlman
Andy Pearlman Studio 
Google
IP: 216.67.130.57
Linear Mode
