[theora-dev] one-line fix reduces distortion

[Arc Riley]

On Sun, Feb 08, 2004 at 04:59:24PM -0500, Timothy B. Terriberry wrote:
> 
> That line was actually correct until you broke it in your most recent 
> commit (encode.c v1.16), which I, at the least, was under the impression 
> was just libogg2 changes. Please try to keep commits for different 
> things separate in the future. It's also helpful if you maybe specify 
> bug report IDs or provide URLs, so some of us have a chance of figuring 
> out what you're talking about.

You should subscribe to the CVS list.  It contains diffs for every CVS 
commit.  I looked over the related code for a few hours and, as far as I 
could tell, this was a mistake.  In the sorting routine which it is part 
of why is the previous value not reset?

Before committing I verified that it did not break bitstream 
compatability (if it changed the format it would not be able to 
playback) and didn't do anything obviously horrendous.  If the routine 
in question is not suppost to be reset, I'd be interested to know 
why.  For a sorting routine this seems extremely atypical behavior.

libogg2 changes, with exception for the examples, has been finished for 
about two months now.  Apparently I was the one who was mailed the 
patch because I was the last to modify encode.c.

I'm encoding for a side by side compairison now to verify that the 
change reduces intraframe distortion @ same bitrate, which just by 
eyeing the results of the post-change it appears to.
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'theora-dev-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.