Hi Jeremy, Just letting you know I'm back from Holidays now, I'll try and go back through and review what I have missed. And we should line up a chat maybe next week if you are around. .malcolm

Nuke's OCIOLogConvert node was being applied in the wrong direction. (What was listed as log->lin was actually the

CineonLogToLin was always occupying a sort of transform netherworld. It's too specific to count as 'general math', but not specific enough to be useful in many practical cases for linearizing film

These add a new transform type, AllocationTransform, which is useful in mapping HDR color ranges into LDR code spaces. Two options are supported: a uniform allocation, and a log-2 allocation, and

Merging this into trunk, no objections. -- Jeremy

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

I've removed the TODO file internal to the codebase, in favor of the github issue tracking. (Which is really nice!) Please feel free to add any issues (bugs or features) to this list as you encounter

No one has objected, so I'll roll this into the master branch. -- Jeremy

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

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

In the absence of any objections, I will check this into master. -- Jeremy

FileTransform will attempt to load the specified lut formats, if the initial read fails. I also had to make some of the 3D lut loaders use stricter parsing as they now get called (in error

That's fair to add it as an option. I'll add it as an official TODO, and will implement it once we have a few more examples of this type. I'm not sure if this will be a transform specific option, or

* Simplified the .3dl loading code * Full support for .3dl shaper

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

Yup, sending the map directly to the yaml stream works great. Thanks. -- Jeremy <malcolmh...@...> wrote:

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

LGTM I'm pretty sure yaml-cpp supports std::map serialization does the following work? sorry I can't test this myself. --snip-- // Roles if(m_impl->roles_.size() > 0)

LGTM .malcolm

+1 I was tempted to always have a lut that would match in the searchpath, say a global identity lut. As this is possible to circumvent this way I would fix this later. The identity lut option should