Question - I’m using write nodes (awesome) to render my shots to my server - I am working remotely- I am also exporting the same clip as a viewable quicktime to upload to shotgun - when the clip auto reimports it loses the shotname meta data, i am using a set of tokens to make it easy to just right click export with my preset into the shot name_version folder - but i’m having to manually write in the shot name back into the auto reimported clip each time (boring).
Is there an easy way to keep that metadata? Am I doing something wrong? Or Is there a way or a python script to also export a quicktime from the exr sequence i’ve written out - that would be ideal
OpenClips would keep their [shot name]. You could render and then just version up your openClip. Then you wouldn’t need to keep entering the [shot name] every time
Ah yes open clip… i’m not running the timeline tho only working on shots and i’ve already have all the shot as exr’s so it would seem silly to republish them - is there way of making the shot an openclip without going through all the republishing etc?
You could “fake” an openclip AFTER the render is completed with the Pattern Browsing, if your export structure is consistent.
This is an exemple: My render is named “sh0010_v001” and this is its export path:
shots/sh0010/flame_renders/v001/sh0010_v001.[0001-0100].exr
Now I open the MediaHub and go to the “shots” folder (the root folder for all my shots).
In the Pattern Browsing tab, I create this new preset using the tokens:
{shotName}/flame_renders/v{version}/{shotName}_v{version}.{frame}.{extension}
I save it, click on the “Pattern Browsing” button (next to the “Scan Subdirectories”), and now a pseudo openclip appears in my root folder because Flame detected a clip that could match my rules.
Not only this clip keeps all the metadatas i set via my pattern (in this case, the shot name and its version), but if i make a new version of my render (v002, v003, …), I can update it directly in the “format options” ribbon or the “Extended” menu, the same way you would do with openclips. This works because you explicitly expose what the version folder is in your pattern.
This is my workaround when our timeline is in Flame but we also work with Nuke artists.
I just need to import the Nuke renders once, and from there I can update the versions in my timeline the same way I would do with openclips
@knhn I am using the python script @PlaceYourBetts mentioned. That auto imports the render from the write node. All I want is the auto imported render to keep the metadata - I see that making it an openclip would make it easier. I’ll figure something out. Thanks for all the tips everyone
@mikeysmith This should work out of the box if you use the Add to Workspace option available in the Write File node as of Flame 2024. If you want the script to work you will need to ask its author (@MikeV) to update it .
Sorry, but I really don’t see the point of the add to workspace functionality. Surely most people would render first to make sure your shot is ok before exporting?
Or is that just me.
It works for all situations, when you render a denoised plate, a pre render, you would want those added back in. Its also the way of working unmanaged, all on the server instead of locally. Always check the render before submitting it.
