mercenariesengineering / openexrid Goto Github PK

Cannot get OFX plugin to run on After Effects.

The openfx plug-in is not running on fusion.

Error: fr.mercenariesengineering.openexrid returned kOfxStatFailed for OfxImageEffectActionRender
OpenEXRId1 failed at time 0

Hello!

I realize enabling "false color" doesn't work if there is no RGB channel in source.

Here is a simple graph:

My deepread doesn’t have any RGB channel.

If I use it directly, like DeepOpenEXRId2 and set false color ON, and put my viewer on it, nothing appear, it’s black.

My only solution is to "force" a RGB channel to exists in the Deep graph merging two deep inputs (DeepOpenEXRId1).

If I enable false color on DeepOpenEXRId1, I have proper color.

I suspect the problem appear here:

https://github.com/MercenariesEngineering/openexrid/blob/master/nuke/DeepOpenEXRId.cpp#L320

The else statement is not reached if no RGB channel exists.

I suggest to add something like:

Channel redChannel = DD::Image::getChannel ("rgba.red");
Channel greenChannel = DD::Image::getChannel ("rgba.green");
Channel blueChannel = DD::Image::getChannel ("rgba.blue");

On top of doDeepEngine(), then, after needed += Mask_Alpha;, add:

if (_colors)
{
    needed += redChannel;
    needed += greenChannel;
    needed += blueChannel;
}

It should do the job but more test is obviously needed, maybe creating channel is done another way. The idea is simply to ensure RBGA is available when "false color" is enabled.

Thanks in advance, keep the good job!

CMake build fails with missing package FindOpenEXR.cmake:

CMake Error at openexrid/CMakeLists.txt:1 (FIND_PACKAGE):
  By not providing "FindOpenEXR.cmake" in CMAKE_MODULE_PATH this project has
  asked CMake to find a package configuration file provided by "OpenEXR", but
  CMake did not find one.

  Could not find a package configuration file provided by "OpenEXR" with any
  of the following names:

    OpenEXRConfig.cmake
    openexr-config.cmake

  Add the installation prefix of "OpenEXR" to CMAKE_PREFIX_PATH or set
  "OpenEXR_DIR" to a directory containing one of the above files.  If
  "OpenEXR" provides a separate development package or SDK, be sure it has
  been installed.

I've the feeling that either OpenEXRConfig.cmake or openexr-config.cmake is installed locally on Mercenaries workstations, but not released in openexrid. What OpenEXRConfig.cmake are you using ? Is that the one from USD ?

Hello There,

This is Aghiles, from the 3Delight team. Questions:

1. Why do we just have one ID per EXR file ? It is interesting to have different IDs (materials, namespaces, etc ...) in one file (Like in CrytoMatte). Maybe I missing somethings ?
2. Why not consider any uint32 slice as an ID deep space ? Or jut list the deep slices in metadata. This will fix problem 1 above.

Let me know.

--
Aghiles

hi,
We are using natron to check the exr output from our renderFarm.
In the last tests were we using the v1.0-beta.19 and had display problems, but now on v1.0-beta.25 Natron won't read anymore exrID.

(Sorry for filing an issue just to ask a question, but GitHub doesn't give anything else. It allows commenting on code, but it doesn't have any way to comment on a project...)

Being able to do per-object color correction like this is definitely a useful goal, but can't you do this with standard deep EXR files and an ID channel (and optionally some attributes to assign names to IDs)? If I render that out of Arnold in Maya 2017, I can apply color adjustments to single objects by selecting samples with that ID. File sizes aren't affected too badly, just a moderate increase due to the extra channel and having more samples where more than one object ID affects a pixel.

Bounding box per id would allow fast RoD computation, and is fairly easy to add in Id map

If a DeepOpenExr/Id node in nuke is unconnected to a viewer and to a source, and if a user pick in the viewer: Nuke just crash ( Not Gently )

Add a version file, a version tag in the open-fx plugin, a change log.