[veysey]

To Traclo - Glad it helped. In terms of bulk processing, you should certainly be able to do it with the command line, and I think through the GUI as well (change your preferences?)

To Kovid - Yes, I was somewhat puzzled too. I am guessing that it may be in part related to the encoding of the original image. To fully make clear what I'm talking about, here are the details.

I begin with one cbz file, and use the following command to convert:

ebook-convert /home/Media/Print/Comics/Toles/Toles.cbz Toles_v2.epub --output-profile sony --no-sort --cover /home/Media/Print/Comics/Toles/Toles.jpg --title "Tom Toles Cartoons" --authors "Tom Toles" --author-sort "Toles, Tom" --publisher "Washington Post" --tags "News, Comics, Humor"

When I add the "--no-process" flag, I get a different sized file (Toles.epub below):

#1: 1.1M Toles.cbz
#2: 5.3M Toles_v2.epub
#3 (--no-process): 1.1M Toles.epub

Breaking it down by unziping the files, it is solely due to images. A representative file from all three:

#1: 50K Toles-09.12.2009.jpg
#2: 128K 0_1.png
#3: 50K Toles-09.12.2009.jpg

Since I'm using the command line, I assume it's a matter of the calibre internal defaults. I'm 99% sure I haven't change anything in my preferences related to cbz conversion. I will check.

Finally, to anyone else who's curious, the example is drawn from a script that turns Tom Toles' daily cartoon into an ebook. Rather nice, since it's a 1-panel black and white cartoon. Unfortunately, the script is a horrible bash frankenstein. I'm going to learn python and try to make it something worth sharing.

Cheers,

John