Hey everyone, I've encountered a linker issue on Ubuntu 10.10 with ocio-0.7.0. I was wondering if anyone could shed some light on whats going on? Linking CXX executable ocio_core_tests /usr/lib/gcc/x8

A new profile, 'nuke-default' has been added to the 0.7.1 config bundle. http://sites.google.com/site/opencolorio/profiles This profile emulates the 'out of the box' color processing luts that come wi

I've added a nuke python script to src/testbed that lets you export specified nuke color spaces into an OCIO profile. It also is a good example of using the python API to create an OCIO profile. commi

Trivial addition, exposed ROLE_DEFAULT to python. commit 82dd5a4666f7c5456361e9ec7d78107742e5b427 -- Jeremy

Would anyone be against us copying the shared_ptr implementation internally into OCIO? That would let us make the OCIO boost dependency optional, which would let it compile on a plain-vanilla OSX inst

* LogConvert, which is a handy node that does the colorspace transform between the SCENE_LINEAR role, and the COMPOSITING_LOG role. While this isnt strictly needed (it could be accomplished using the

Added support for Truelight .cub luts. (Both 1D shaper + 3d lut). 1D shaper lut currently uses linear interpolation, until something better comes along. Note: This checkin tested with a limited number

Added explicit control of transform direction + interpolation to the Nuke OCIOFileTransform node. Commits: 3143ccdb38fee9ec774a1be94150c8f04d9e70c5 -- Jeremy

Unit tests for envvars were failing, referencing files that weren't checked into the source tree. Disabling these specific tests for now. Commits: 0a052d9090c9d693f63cdb409260627406790d75 -- Jeremy

Upon further testing, it appears the channel swizzling was backwards during CSP lut importing. I used Nuke as the baseline in terms of lut channel order, so it's possible that I've actually broken thi

Minor optimization: the potentially expensive EnvExpand call (which expands envvars in file paths), does an early exit for paths which do not contain special characters. ($ + %) Commits: 38d53d46239ec

In the public API we were storing references to m_impl using autoptrs, which really was unnecessary. This only saved one line of code per use, and made the memory model less obvious. This doesnt affec

Added full read support for Iridas .cube luts. Verified results agree with Nuke for all luts in FrameCycler distro. commits: 1880192d63718dfba54d96d88ff396a0ac85fdee -- Jeremy

The list of defined roles within a config previously maintained order. Conceptually, this is actually a mapping type where order does not matter (and the yaml serialization dropped ordering). This swi

Looking again at 3d lut index ordering, .csp files can be loaded without any float reordering - none of the prior index mojo is required. This simplifies the code quite a bit, results are identical to

* Simplified the .3dl loading code * Full support for .3dl shaper luts. Commits: 76e11f2d894856142c6d96a6d0d317e911880993 f8fe2d5af218154189ab46830afc3d78e31c91ac 0176f8e96c24f06b0a80b8c78539087946e30

On Lut loading, 1D luts are checked to determine if the underlying transform represents 'identity'. If so, they are skipped during op generation. This is particularly useful when working with 3D luts

Version 0.7.1 (Nov 15 2010): * Additional 3d lut formats: Truelight .cub + Iridas .cube * FileTransform supports envvars and search paths * Added Nuke plugins: LogConvert + FileTransform * Improved OC

This updates the API for setting the ColorSpace GPU allocation hint. (The allocation is used to determine how a particular lattice is best sampled using a fixed domain lut on the GPU). Previously all

For those not familiar with the IIF project, the Academy is sponsoring a project to standardize a single floating-point linear workflow across the motion picture industry: http://www.oscars.org/scienc