Hakan SevenandClaude Opus 4.8 aea1fd2caa perf(render): arena on by default; trim per-edit resident + text work ...

Three cuts to the per-edit cost, on top of the incremental wire arena:

- Enable OCS_WIRE_GPU_PATCH by default (kill-switch OCS_WIRE_GPU_PATCH=0), so the
  in-place / append-tombstone arena — validated on the reference drawing (wire
  block ~300ms + 265MB upload -> ~2ms) — is on for everyone; a bail always falls
  back to the identical batched path.

- Resident patch: an all-Modified edit never reorders, so reassemble by walking
  the recorded assembled order (~10k handles) instead of scanning the whole
  document (~800k) to re-derive draw order. Add/Remove still scan (positions
  change). ~14ms -> <1ms on a move.

- SDF text: a new content id misses the glyph cache every edit, but if no
  text-bearing entity (Text/MText/Dimension/MLeader/Leader/Table/attribute/INSERT)
  changed since the last build the glyphs are identical, so reuse them. ~23ms -> 0
  on a plain-geometry edit.

A move/edit on this 800k-entity drawing now spends ~50ms (dominated by the undo
document clone) instead of ~85ms, and the wire GPU upload is gone.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-07-15 10:14:44 +03:00

..

camera.rs

fix(camera): orbit target in f64 to avoid extent-scale jitter

2026-07-05 03:26:07 +03:00

dispatch.rs

feat(properties): editable per-object leader dimension overrides

2026-07-13 00:28:42 +03:00

mod.rs

refactor(scene): group modules into role-based subfolders

2026-06-16 11:33:46 +03:00

render.rs

perf(render): arena on by default; trim per-edit resident + text work

2026-07-15 10:14:44 +03:00

transform.rs

refactor(precision): widen command transform scalars and defaults to f64 (#311)

2026-07-12 00:34:38 +03:00

viewport_pane.rs

feat(viewport): hide ViewCube then render bar as the viewport narrows

2026-07-10 14:58:48 +03:00