[theora] NHW codec - improvement of precision (Daniel Hendrycks)

[Raphael Canut]

Hello,

Yes, that's right that my codec is "closed" source and undocumented for the
moment, but this can be easily corrected.

(As I said in my first post), I have contacted Monty Montgomery first.I had
few but great discussions with him, and Monty also encouraged me to present
my work on the channel.But that's right that I don't know if he was ok that
I post that often at each update (will have to slow down)? Maybe the best
would be to ask directly to Monty and Xiph members if it is a polite way to
post on the Theora channel, or definitely not?

I have also sent to Monty all the source code of my codec and algorithm
detail (he also very kindly helped me to improve my codec).The problem is
that my codec is absolutely not finalized (my algorithm is definitely not
good on the 5% of rather compressed and "very blurred" images... and that's
a major drawback...), the source code would also really need to be cleaned
up...

About all this, I think the best is maybe to have an answer from Monty
and/or Xiph core members?

Many thanks again for your time,
Raphael


2012/6/2 David Kuehling <dvdkhlng at gmx.de>

> >>>>> "Raphael" == Raphael Canut <nhwcodec at gmail.com> writes:
>
> > Hi again,   If it could be of interest, I have added on my demo page
> > the 3 test images compressed with a new version that I have just
> > made... I have improved precision on the UV chroma components (and
> > corrected a very little bug).    Now, it is directly possible to
> > compare the results of the first version with the last version, and
> > see the precision improvements...
>
> Am I overlooking something or is this a *closed* source, undocumented
> image codec that you keep posting about?  What makes you think that
> theora at xiph.org is the appropriate channel to use for distributing your
> project's updates?
>
> David
> --
> GnuPG public key: http://dvdkhlng.users.sourceforge.net/dk.gpg
> Fingerprint: B17A DC95 D293 657B 4205  D016 7DEF 5323 C174 7D40
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.xiph.org/pipermail/theora/attachments/20120602/62fce97f/attachment.htm