What does OPTIMIZATION do for the IrfanView JPEG lossless transformation plugin?

What does the IrfanView lossless JPEG rotation plugin "optimization"
actually do?

When I rotate a JPG digital photograph using the IrfanView "Options - JPEG
Lossless Operations" plugin, one of the options is to "Optimize JPG file".

The IrfanView 3.98 help doesn't explain what it means to "Optimize the JPEG
file".

Does anyone know what this "optimization" setting actually does?
Should we leave it checked or unchecked?

Joy wrote:

What does the IrfanView lossless JPEG rotation plugin
"optimization" actually do?

It uses custom Huffman tables.

http://www.impulseadventure.com/phot...ized-jpeg.html

Should we leave it checked or unchecked?

Depends on your paranoia-level. It is said that some decoders doesn't
handle it, but I haven't run across any yet. You can always convert
back to non-optimized form without quality loss, so it's not that
dangerous.

(Follow-up set to alt.comp.freeware)
--
Toke Eskildsen - http://ekot.dk/

Joy wrote:
What does the IrfanView lossless JPEG rotation plugin "optimization"
actually do?

When I rotate a JPG digital photograph using the IrfanView "Options - JPEG
Lossless Operations" plugin, one of the options is to "Optimize JPG file".

The IrfanView 3.98 help doesn't explain what it means to "Optimize the JPEG
file".

Does anyone know what this "optimization" setting actually does?
Should we leave it checked or unchecked?

I can't answer your question, but recommend upgrading to the current
(4.0) version of Irfanview. Don't forget to upgrade the plugins at the
same time!

On May 18, 10:04 am, Joy wrote:
What does the IrfanView lossless JPEG rotation plugin "optimization"
actually do?

When I rotate a JPG digital photograph using the IrfanView "Options - JPEG
Lossless Operations" plugin, one of the options is to "Optimize JPG file".

The IrfanView 3.98 help doesn't explain what it means to "Optimize the JPEG
file".

Does anyone know what this "optimization" setting actually does?

Jpeg is a two step algorithm where first step transforms the image
into a more easily compressible form and drops visually less important
information. Second step (huffman) encodes the remaining information
in as few bits as it can.

In Jpeg, the Huffman part can either use a predefined set of tables or
try to come up with custom tables based on what will work best for
that image. First option is obviously easier to implement and uses
less processing power so most implementations just do this due to
laziness and cameras do this to save battery.

Irfan view's 'optimization' setting makes it calculate and use custom
tables which means better compression at no additional loss of
quality.

Should we leave it checked or unchecked?

I think you can use it safely. Some old or badly written applications
might not support Jpegs with custom tables but this should be rare.

Sachin Garg [India]
www.sachingarg.com | www.c10n.info