battery: wrong-model null takes about 6 minutes of rugnux (measured)

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EizBKhTqrAYago9KJG3dA3

.gitea/workflows/battery.yml

@@ -9,7 +9,7 @@ name: Validation Battery
 #
 # A ci-tier run also runs the wrong-model null (tools/battery/nullmodel.json: five sets, each without
 # --model and with five wrong models - other open-arm entries' deposited coordinates, from the PDB
-# cache under /data/battery - 30 more rugnux runs, about 7 minutes); its pass/fail table is a section of
+# cache under /data/battery - 30 more rugnux runs, about 6 minutes); its pass/fail table is a section of
 # report.md of its own and leaves the ci-tier rows untouched.
 #
 # The job fails when a ci-tier set fails on lattice or space group, crashes or has no input - in any

tools/battery/README.md

@@ -338,7 +338,7 @@ is the same protein.
 
 The pairs go to `nullmodel_results.json` in the run directory and to a **Wrong-model null** section
 at the top of the report; the sets' own rows in `results.json` are those of the run without it. It
-adds 30 rugnux runs, about 7 minutes on the RTX 5080.
+adds 30 rugnux runs, about 6 minutes of rugnux on the RTX 5080 (measured: 6.1 min of WALL_TIME).
 
 ### Stress: hunting crashes that do not reproduce