diff --git a/docs/re/captures/f101-index-shift-before-after.png b/docs/re/captures/f101-index-shift-before-after.png
new file mode 100644
index 0000000..1317e68
Binary files /dev/null and b/docs/re/captures/f101-index-shift-before-after.png differ
diff --git a/docs/re/structures/xbg7-mesh.md b/docs/re/structures/xbg7-mesh.md
index cb58ed7..244fc1c 100644
--- a/docs/re/structures/xbg7-mesh.md
+++ b/docs/re/structures/xbg7-mesh.md
@@ -1448,3 +1448,13 @@ the grouped `.dat` break composites in `ptc_pack` (`f102`/`f104`/`e107`, whose
 marker lists are documented not to map onto the stored blocks), `e201_bdy_03_m`
 (2 containers) and `_rou_f402_dead` (9) — each already suspect on other grounds,
 and now the concrete next targets.
+
+**And it is plainly visible** — the check the numbers kept failing at.
+`ship_render --static` on `f101` (the ACROPOLIS, whose 25 448-vertex hull carried
+2 156 degenerate triangles) both ways:
+[before / after](../captures/f101-index-shift-before-after.png). The old decode is
+a torn mess of shards; the new one is a coherent capital ship with flat decks,
+panel lines, bridge tower and funnels. Same 31 991 triangles, same 4 placements,
+same bounds — only the wiring differs. Third time this project has learned it:
+**render the output**; a metric that cannot see a 1 600-unit slab could not see
+this either.