Re: filmscanners: File format

[Arthur Entlich <artistic@ampsc.com>]

I decided to put my money where my mouth is ;-) and I just took an 
image, which mainly consisted of some color areas and some typeface, and 
ran it through Photoshop's jpeging.  With one set, I used level 4 and 
the other set I applied level 2.  In each case, I did a series of ten 
jpegs applications, each time starting with opening the previously 
created file and applying the same jpeg level, and then saving that 
file, then using the saved file for the next application.  Below are the 
file sizes for each application

    LEVEL 4    LEVEL 2
1  170,700    122,197
2  168,292    120,947
3  167,927    120,761
4  167,865    120,702
5  167,862    120,692
6  167,802    120,698
7  167,801    120,698
8  167,787    120,698
9  167,785    120,698
10 167,789    120,698

So, what does this show?  Well, first of all, obviously something is 
changing in these files, and I don't believe it is just Photoshop 
tagging or thumbnails (which I admit, I should have probably turned off 
before doing this, oh well).  BUT, more importantly, there appears to be 
a point where JPEG reaches a saturation point in terms of its ability to 
use the algorithms to compress.  In fact, it begins to very slightly 
enlarge the files after that point.

Now, I'll admit, being Photoshop, perhaps it is recording something 
beyond image info within the files, like history or who knows, but the 
mechanism seems to be that the more drastic the jpeg compression, the 
sooner it reaches maximum compressibility.  Also, I can't say that just 
because the file sizes become stagnant that the pixels are identical, 
either.

Anyway, that's as far as I'm going with this.  I suspect that if I had 
used a smaller jpeg compression, the file would have taken more 
applications to be reduced to the minimum size.

Art

Laurie Solomon wrote:

>> Each time there would be some generational loss.
> 
> 
> Not necessarily true. If you open and close ( or resave) the compressed file
> without changing the compression from one quality level to another in the
> case of .jog or without resampling the image prior to closing or resaving
> the file, there will be no more degradation than opening and closing or
> resaving a raw uncompressed file.
> 
> When you open a compressed file you have uncompressed it, so resaving it
> with the same compression as before or without engaging in any resampling
> prior to resaving the file  should not result in any additional losses in
> data or quality.
>