Practical Info
This is a weekly video chat meeting for planning and discussion of Blender Eevee/viewport module development. Any contributor (developer, UI/UX designer, writer, …) working on Eevee/viewport in Blender is welcome to join.
For users and other interested parties, we ask to read the meeting notes instead so that the meeting can remain focused.
- Google Meet
- Next Meeting: June 26, 2023, 11:30 AM to 12:00 PM Amsterdam Time (Your local time: 2023-06-26T09:30:00Z → 2023-06-26T10:00:00Z)
Attendees:
- Jeroen
- Michael
- Miguel
Eevee-next
- Subsurface scattering has landed in main. [#107407 - EEVEE Next: Subsurface Scattering - blender - Blender Projects]
- Viewport HDRI selector is being worked on. Initial version would not support blurring.
- Modified the way how textures are bound to shaders. Eevee-next used to bind internal textures and buffers to slots upto 15, leaving the lower slot numbers for user textures. This has been switched; internal textures and buffers are now occupying lower slots, and user textures the higher slots. Freeing more slots for user textures. Slot 0 and 1 are reserved still for Grease pencil [#108984 - EEVEE Next: Improved sampler indices - blender - Blender Projects]
- Improved validation messages so it is easier for developers to track the reason that raises the messages.
Overlay-next
- Porting extra passes and shapes is in progress. Contains a refactor how render passes are organized. [#109059 - WIP: Overlay Next: Shapes and Extras - blender - Blender Projects]
Metal backend
- Memory freeing issue in Blender 3.6 has the highest priority. There has been a first round of feedback. This week BCon4 is scheduled hence this issue will have the top priority. [#108940 - Fix #108792: Ensure Metal buffers correctly freed on exit - blender - Blender Projects]
- Continued with the EDR/HDR change. [#105662 - MacOS: Enable support for EDR rendering - blender - Blender Projects]
Vulkan backend
- Added support for compute indirect. [#108879 - Vulkan: Indirect Compute - blender - Blender Projects]
- Some experiments have been done to for device selection. Current outcome is that it is to early to add support for these configurations. The backend should mature more, before this could be tackled. It has been investigated to speed up platform support by installing GPUs of multiple vendors in a single case. Currently we switch GPUs multiple times a day.