theora dev - Apr 2010 - YUV clamped or unclamped ?

Hi,
can somebody confirm whether theora uses clamped (16 - 235, 16 - 240) or
unclamped (0 - 255) YUV ?

My tests are showing that Y is always in the range 16 - 235 but the U and V
channels occasionally go above 240.


Thanks,
Salsaman.


http://lives.sourceforge.net
https://www.ohloh.net/accounts/salsaman
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/theora-dev/attachments/20100414/e8f29b53/attachment.htm

salsaman wrote:
> Hi,
> can somebody confirm whether theora uses clamped (16 - 235, 16 - 240) or
> unclamped (0 - 255) YUV ?
The specification clearly states that the former is to be used (see
Section 4.3). However, there is nothing internal to the codec that
enforces this.
> My tests are showing that Y is always in the range 16 - 235 but the U
> and V channels occasionally go above 240.
The codec itself will not clamp things to these ranges, so the whole
"lossy compression" thing can occasionally produce values slightly
outside of them. This is normal.

Ah OK, thanks for clarifying. So a chroma value of 241 - 255 should just be
treated as 240 then.


Regards,
Salsaman.

http://lives.sourceforge.net
https://www.ohloh.net/accounts/salsaman



On Wed, Apr 14, 2010 at 9:36 PM, Timothy B. Terriberry <
tterribe at email.unc.edu> wrote:
> salsaman wrote:
> > Hi,
> > can somebody confirm whether theora uses clamped (16 - 235, 16 - 240)
or
> > unclamped (0 - 255) YUV ?
>
> The specification clearly states that the former is to be used (see
> Section 4.3). However, there is nothing internal to the codec that
> enforces this.
>
> > My tests are showing that Y is always in the range 16 - 235 but the U
> > and V channels occasionally go above 240.
>
> The codec itself will not clamp things to these ranges, so the whole
> "lossy compression" thing can occasionally produce values
slightly
> outside of them. This is normal.
> _______________________________________________
> theora-dev mailing list
> theora-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/theora-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/theora-dev/attachments/20100414/1d9c639f/attachment.htm