Skip to content

Rendering optimization #78 (#81)#88

Merged
bridgesign merged 1 commit into
claude/memory-engine-occlusion-4T55mfrom
dev
May 15, 2026
Merged

Rendering optimization #78 (#81)#88
bridgesign merged 1 commit into
claude/memory-engine-occlusion-4T55mfrom
dev

Conversation

@bridgesign
Copy link
Copy Markdown
Collaborator

  • Replace per-layer surfaces with single render surface + per-artist RGBA cache

  • Reuse graph cache across camera moves via offset blit and zoom stretch

  • Substitute short edges and skip sub-pixel edges and nodes in graph rendering

  • Clean up docstring formatting in default_drawers

  • Move cached artist rendering behind a RenderManager callback

  • Merge get_culling_bounds and world_to_screen_scale into get_viewport

  • Skip edge.linestring access for short or cached edges in graph rendering

  • Move constants to init

  • Revert graph render cache to lazy edge_line_points dict

  • Apply skip/short edge optimization to render_map_sensor

  • Avoid strip re-render and bilinear cost on zoom-out

  • Clarify cached artist handler docstring

  • render agent optimization

  • get_edges in a box

  • edge caching in waiting_simulation

  • edge finding optimizations

  • Fix projection updates on window resize

  • Remove will_draw from artist interfaces and use ArtistType for static cache routing

  • Replace per-artist pixel caches with per-layer surfaces

  • Guard render projection against zero screen size

  • Change layer cache artist_names from set to tuple

  • cache mismatch bug fix

  • Remove unused _static_layers from RenderManager

  • Clamp render dimensions on resize


* Replace per-layer surfaces with single render surface + per-artist RGBA cache

* Reuse graph cache across camera moves via offset blit and zoom stretch

* Substitute short edges and skip sub-pixel edges and nodes in graph rendering

* Clean up docstring formatting in default_drawers

* Move cached artist rendering behind a RenderManager callback

* Merge get_culling_bounds and world_to_screen_scale into get_viewport

* Skip edge.linestring access for short or cached edges in graph rendering

* Move constants to init

* Revert graph render cache to lazy edge_line_points dict

* Apply skip/short edge optimization to render_map_sensor

* Avoid strip re-render and bilinear cost on zoom-out

* Clarify cached artist handler docstring

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

* render agent optimization

Co-authored-by: Copilot <copilot@github.com>

* get_edges in a box

Co-authored-by: Copilot <copilot@github.com>

* edge caching in waiting_simulation

* edge finding optimizations

* Fix projection updates on window resize

* Remove will_draw from artist interfaces and use ArtistType for static cache routing

* Replace per-artist pixel caches with per-layer surfaces

* Guard render projection against zero screen size

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

* Change layer cache artist_names from set to tuple

* cache mismatch bug fix

* Remove unused _static_layers from RenderManager

* Clamp render dimensions on resize

---------

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Mehul Sinha <mehulsinha73@gmail.com>
Co-authored-by: Copilot <copilot@github.com>
Co-authored-by: Mehul Sinha <72033137+mehulsinha73@users.noreply.github.com>
@bridgesign bridgesign merged commit 6483148 into claude/memory-engine-occlusion-4T55m May 15, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants