Blender Git Loki
January 27, 2020, 15:39 (GMT) |
Avoid own global variable, use existing DRWManager struct |
January 27, 2020, 15:13 (GMT) |
Only cache framebuffers for layers that aren't always redrawn |
January 27, 2020, 14:14 (GMT) |
Get draw-layers to work with multiple viewports |
January 27, 2020, 12:08 (GMT) |
Use hash, not list to store and lookup DRWLayer items |
January 27, 2020, 11:46 (GMT) |
Fix memory leak when only drawing gizmos |
January 27, 2020, 11:24 (GMT) |
Merge branch 'master' into temp-gizmo-decoupled-redraws |
January 24, 2020, 16:42 (GMT) |
Draw-layers: Get it all to work! Create the framebuffers for caching viewport rendering results, and "composite" all layers together once each is drawn. Compositing is a bit of a big word for this, since we basically just draw all layers on top of each other with alpha blending. After talking with Jeroen, we decided to keep things simple for now, we can add more features (like per-layer color-management) and do small improvements later. |
January 24, 2020, 10:37 (GMT) |
Merge branch 'master' into temp-gizmo-decoupled-redraws |
January 20, 2020, 15:23 (GMT) |
Initial introduction of DRWLayer abstraction Used to wrap different drawing steps, i.e. the main scene drawing, editor overlays and debug text. Later overlays should get their own layer. Right now only the drawing commands are wrapped into per layer callbacks. The idea is to let each draw into different framebuffers and to composit them together at the end, so that we can avoid redrawing all layers if only one needs updating. |
January 17, 2020, 16:47 (GMT) |
Add separate region redraw tag for UI overlays (gizmos) |