Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 




      :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: filmscanners: JPEG Loss - File format

>"Close" and "Save" are usually quite a lot different operations.

Good point.  I accept that I and others may have been sloppy in our use of
the terms as well as our keeping clear the consequence that have stemmed
from each of the operations.

>If you did not change anything in the file - then why would you save it
>and not just close?

Once again fair enough.

>And still, even thought these artifacts are small and hard to notice, then
>are there nevertheless.

I concur with most of your arguments and would suggest that much of the
debate appears to turn around two things.  First, it turns around some of us
( including me) not keeping the distinctions you have made between "closing"
and "resaving" clear and distinct.  Thus has definitely muddied the water
and made the discussion ambiguous and open to false debates.  Secondly, it
turns around the fact that many of us were rally referring to significant or
relevant degradations which become visibly manifested in contract to the
absolute notion of degradation which refers to the loss of any data or
information.  Thus, the protagonists may have been talking past each other
in this discussion.

However, like most discussions that do not fall apart at the seams, this
discussion has become clearer and more informative the more we talk and
eliminate ambiguities in terms and expressions. Hence, I consider it to have
been a useful and profitable conversation.

-----Original Message-----
From: owner-filmscanners@halftone.co.uk
[mailto:owner-filmscanners@halftone.co.uk]On Behalf Of Hugo Gvert
Sent: Monday, April 02, 2001 7:04 AM
To: filmscanners@halftone.co.uk
Subject: RE: filmscanners: JPEG Loss - File format

On Fri, 30 Mar 2001, Laurie Solomon wrote:

> However, my experiences with .jpg files has not been along the lines of
> increasing artifacts with each iteration; when I have gotten artifacts in
> .jpg files that were opened and resaved without any editing or changing of
> quality levels, it has been due to some other factor than the mere opening
> and closing of the .jpg file.

"Close" and "Save" are usually quite a lot different operations. With
close, the file just closes. With save, it's resaved... and it takes time,
compared to close, which does not take any more time even for large files.
In closing the file, no bit is changed from the file. Nothing.

If you did not change anything in the file - then why would you save it
and not just close?
I don't know the compression of Jpeg that well that I could say if there
even is anything that you could do to the file, without modifying any of
the pixels information. I believe all the color commands change the pixels
information that will be saved into the jpeg as well. And as the pixels
are changed, then the whole compression must be done again - with (your
changes as well as) the artifacts that were introduced while saving and
opening the file. The artifacts are now part of the new image.

And if you did not do anything to the image, absolutely anything to the
image, just opened it and resaved (not closed) it again as .jpeg, and you
did get artifacts... then where could you have gotten them? If you did not
change the pixels, and the procedure did not introduce any artifacts, then
how do you get them?

And still, even thought these artifacts are small and hard to notice, then
are there nevertheless.

And as final note: yes, there are some things that can be done to the
jpeg-file without changing any pixels information. There are some programs
that, for instance, rotate the whole image without decoding (opening) the
comressed data. That way, the image is changed (even though no pixels have
been changed) and nothing was lost in the process. Maybe there are some
white point settings or something like that, that could be changed... but
I'm now just guessing.

Best regards,

**   Hugo Gvert                                          **
**   hugo@cc.hut.fi             http://www.hut.fi/~hugo   **
**   Sometimes I think the surest sign that intelligent   **
**   life exists elsewhere in the universe is that none   **
**   of it has tried to contact us.   -- Calvin.          **


Copyright © Lexa Software, 1996-2009.