LL perf (Avatar/animation): Every rigged LOD change or alpha dirty marks the whole avatar's complexity stale #337

New issue

Closed

opened 2026-10-04 14:09:23 +00:00 by sky_line · 1 comment

sky_line commented

2026-10-04 14:09:23 +00:00

Owner

From the LL-origin inefficiency audit (finding A16). File/line references are from commit 0229f8872c; re-locate them before working.

Every rigged LOD change or alpha dirty marks the whole avatar's complexity stale

Where: llvovolume.cpp:2061-2064, llface.cpp:345-350. The recompute is "~5 ms" per its own comment (llvoavatar.cpp:2796).
Also: an empty main-loop lambda post (:11742-11748).
Likely. S-M.

Verify the cost (profile or timer) before and after changing it.

From the LL-origin inefficiency audit (finding A16). File/line references are from commit 0229f8872c; re-locate them before working. **Every rigged LOD change or alpha dirty marks the whole avatar's complexity stale** - **Where:** `llvovolume.cpp:2061-2064`, `llface.cpp:345-350`. The recompute is "~5 ms" per its own comment (`llvoavatar.cpp:2796`). - **Also:** an empty main-loop lambda post (`:11742-11748`). - Likely. S-M. Verify the cost (profile or timer) before and after changing it.

sky_line added the

perf

label

2026-10-04 14:09:23 +00:00

sky_line added this to the Slipstream project

2026-10-06 15:04:14 +00:00

sky_line added the

status

in-progress

label

2026-10-08 02:07:16 +00:00

sky_line referenced this issue from a commit