To answer your first question it is worthwhile to store them in 48-bit form.
When potentially making adjustments in levels and curves in the future you
will be less likely to have banding or posterization in the image. See
Bruce Fraser's most enlightening article on this issue at
As to converting from 48 to 24 and back to 48 bit, don't even bother:
there is no quality loss but there is no quality gain whatsoever either.
The benefits of 48 bit, more colors available to fill the gaps when
compressing the color or tone gamut, are just not there as you merely end up
with a multiple of the 24-bit colors when you increase the bit number.
As to publishing on the web, there is no quality loss to my knowledge in
reducing the size in PS.
Maris
----- Original Message -----
From: "Preben Kristensen" <krille@tin.it>
To: <filmscanners@halftone.co.uk>
Sent: Wednesday, May 16, 2001 2:14 AM
Subject: Re: filmscanners: 48-bit batchscanning
| I've just started 48-bits batchscanning (try to say this quickly a couple
of
| times) on the SS4000 in Silverfast 5.2 - following Ian Lyons' instructions
| at
|
| It works like a dream. The images are now coming out in gamma 2.2 (or 1,8
if
| you prefer) looking very good "out of the box".
|
| Is it worthwhile (storage space) to store them in 48 bit or can one use 24
| bit. I guess what I am really asking is: Is there a quality loss going
from
| 48 bits to 24 bits and back again to 48 bits? (in case I would want to
| adjust the images at a later stage)
|
| Also: I would like to publish these images on the web as large thumbnails.
| Is it OK quality-wise to reduce the size in PS 6.01 or is it better to
| rescan at a lower resolution.
|
| Preben
|
