OpenVAT

Changelog

• Updated tagline to better reflect the use of this self-contained tool
• Code cleanup and commenting
• Nodes cleanup and commenting
• Improved Normals rendering process
  • Added 'Packed' Vertex Normals method option
  • this option allows for a position and normal map in one export map (still includes functionality to not include normals, or render on a separate map)
• Improved menu layout, tooltips, and user interface
• Caught and removed one unnecessary def
• Stripped and un-grouped enclosed node_groups, so appending does not bring additional node groups - major pass at node restructuring and organization
• Stripped all unnecessary data from resources
• Added delete/purge functionality in cleanup (does not accumulate data when cleanup is enabled)

Changelog

• Updated default path to prompt the user to enter a path
• Updated JSON export to include number of total frames to better communicate with engine shader tools

Changelog

OpenVAT - VAT Encoder Initial Extension Release

-Unity URP shadergraph methods and examples can be found at https://github.com/sharpen3d/openvat.

Full support documentation, breakdowns, walkthrough, Unreal integration and demo files are in process and will be available here. https://github.com/sharpen3d/openvat/blob/0a165933838f013e2b0ef8dc856f68f99b5e7505/README.md

Release Notes:

• Mesh to encode must maintain consistent domain size (vertex, edge, face count and ordering)

• Range finding is rounded for ease of manual engine-shader implementation - may update for better range accuracy, but should not impact quality of encoding.

• VAT and VNRM exports are 16 bit PNG exports and have uncapped size. It is recommended to render at a resolution of 4096 or lower - resolution is calculated within the tool when an object is selected.

• Linear PNGs can become heavy if not monitored because of the lossless quality. These can be managed after the fact with careful compression (the built-in geometry node decoder can help determine quality loss with compression by re-loading in the compressed image after running your compression of choice.)

• Not all animation situations have been tested, but any position/normal information that is accessible in the dependency graph (Geometry Nodes Spreadsheet data) is able to be recorded.