[Openexr-announce] Pixar contributes Pxr24 compressor to OpenEXR

[Drew Hess]

Hi all,

We are pleased to announce that Pixar Animation Studios has contributed
code to OpenEXR for a new 24-bit floating-point compression method.  This
new compressor is called "Pxr24" and will be available as of the 1.1.1
development release of OpenEXR.  It will also be included in the upcoming
1.2 stable release.

Values produced by the Pxr24 compressor provide the same range as 32-bit
floating-point numbers with slightly less precision, and compress quite a 
bit better.  I'll have some more detailed numbers on the compression 
improvement when I've had a chance to run more experiments.  For UINT and 
HALF datatypes, Pxr24 is lossless and compresses about as well as the ZIP 
method.

For details on the compressor, see the documentation in the upcoming 1.1.1
development release of OpenEXR.

Thanks to Loren Carpenter and Dana Batali of Pixar, and to Florian Kainz
of ILM, for making this happen.  And thanks to both of these companies for
allowing this collaboration to happen!


-dwh-