Downsampling for monitor

380 views5 repliesLast post: 7/30/2004
I´m looking for the best technique to downsample my images for monitor resolution. I usually work with drum scans on 675ppi and downsample to 96ppi monitor resolution. I use PS CS; sometimes I use bicubic and other times bicubic smoother and then sharpen but I was never really happy with the results. I'm not a big fan of sharpening- or maybe I'm not good at it. I want my small images to retain the smooth tonality and crispness of the original scans ( from MF and LF slides) How to achieve that? I've seen web images that stirred my envy!

regards,
J.
#1
"Jytzel" wrote in message
I´m looking for the best technique to downsample my images for monitor resolution. I usually work with drum scans on 675ppi and downsample to 96ppi monitor resolution. I use PS CS; sometimes I use bicubic and other times bicubic smoother and then sharpen but I was never really happy with the results. I'm not a big fan of sharpening- or maybe I'm not good at it. I want my small images to retain the smooth tonality and crispness of the original scans ( from MF and LF slides) How to achieve that? I've seen web images that stirred my envy!
regards,
J.
-------------

DON'T downsample >and< resize - just resize for the screen.

Journalist
#2
"Journalist-North" wrote in
news:L7tMc.104914$:

DON'T downsample >and< resize - just resize for the screen.

Just how in the hell do you expect to "resize" an image for the screen without it being resampled?
#3
From: (Jytzel)

I´m looking for the best technique to downsample my images for monitor resolution. I usually work with drum scans on 675ppi and downsample to 96ppi monitor resolution. I use PS CS; sometimes I use bicubic and other times bicubic smoother

According to the Photoshop Help file you should use 'bicubic SHARPER' when downsampling and smoother when upsampling (look for 'Choosing an interpolation method' in the Help files).

I usually use 'bicubic sharper' without sharpening the initial file first with MF film scans and get very good results. Often I don't even need to run USM on the final small file, or if I do it's a very small amount, so the designers did a good job with this new feature. If you're downsampling by a significant amount you might try doing it in steps of 50% instead of in one giant step.

Bill
#5