aboutsummaryrefslogtreecommitdiff
path: root/docs/tomes
diff options
context:
space:
mode:
authorValentin Popov <valentin@popov.link>2026-07-18 10:43:33 +0300
committerValentin Popov <valentin@popov.link>2026-07-18 10:43:33 +0300
commit781241742dc650353bd354868bd497d1f65011a9 (patch)
treebfb2ce82fad375d7aace338b258692e193432bc2 /docs/tomes
parent66b224a65c76d270747d8ab4be12ea2b6d536a28 (diff)
downloadfparkan-781241742dc650353bd354868bd497d1f65011a9.tar.xz
fparkan-781241742dc650353bd354868bd497d1f65011a9.zip
feat(runtime): trace visual graph phases
Diffstat (limited to 'docs/tomes')
-rw-r--r--docs/tomes/05-render.md7
1 files changed, 7 insertions, 0 deletions
diff --git a/docs/tomes/05-render.md b/docs/tomes/05-render.md
index dc8790d..5899182 100644
--- a/docs/tomes/05-render.md
+++ b/docs/tomes/05-render.md
@@ -1094,6 +1094,13 @@ process was terminated. This gives no Part 2 GPU acceptance or pixel result and
does not attribute the delay to Vulkan; its visual-dependency corpus remains a
separate profiling and compatibility target.
+`GraphVisuals` is now split observationally into `GraphVisualWears`,
+`GraphVisualMaterials` and `GraphVisualTextures`; it preserves the existing
+graph traversal, edge order and validation semantics. Repeating the controlled
+180-second Part 2 probe ended at `GraphVisualTextures`. Thus Part 2 had passed
+WEAR and MAT0 expansion before timeout; its remaining bottleneck is TEXM
+validation/archive work, not Vulkan submission or the base prototype graph.
+
`Assets` now has four ordered diagnostic sub-checkpoints: `AssetModelMeshes`
(MSH), `AssetWearTables` (WEAR), `AssetMaterials` (MAT0) and `AssetTextures`
(diffuse TEXM and lightmaps). The callback is observational: it neither changes