> Now open b and c in [image editor of choice]. Copy b to
> clipboard, paste on top of c, set merge mode to 'subtract'.
> Resulting image is totally black -- ie, b and c are identical.
JPEG is a LOSSY bit-reduction method. No two ways around it.
However, where it has an advantage is in the fixed nature of the
algorithm. The greatest amount of data loss occurs in the first
compression. Any further saves will line up with the previously
saved files nearly identically.
However...
Remove ONE row of pixels from the edge of the file! Now the 8x8
block of the previous save will not line up with the new save. The
algorithm becomes lossy all over again.
This is very similar to what we had to deal with in audio bit-rate
reduction for broadcast applications. DOLBY AC-x had a fixed
algorithm whereas MPEG was adaptive. On a first-pass, MPEG would have
a better audio quality, but when reprocessed multiple times, the
Dolby schemes had no further degradation.
JPEG is an interesting creature. It holds up very well to repeated
processing like you have done in your test, but in real-world where
editing occurs (including color and luminance changes, cropping,
sharpening, etc., you will get onother whole set of losses.
AG
____________________________________________________________________________________
The fish are biting.
Get more visitors on your site using Yahoo! Search Marketing.
http://searchmarketing.yahoo.com/arp/sponsoredsearch_v2.php
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|