diff --git a/docs/re/f6-unit9-sweep-period-and-onset.md b/docs/re/f6-unit9-sweep-period-and-onset.md
index 090eb314..1dcd8583 100644
--- a/docs/re/f6-unit9-sweep-period-and-onset.md
+++ b/docs/re/f6-unit9-sweep-period-and-onset.md
@@ -38,19 +38,36 @@ and gave an 8% spread where this gives 0.35%.
 The cycle boundary is unambiguous: the sweep enters at `x=-1.540` and wraps from
 `x=+1.840` back to `x=-1.540`, so first-appearance→wrap is one whole period.
 
-Calibrating the baseline with the plate's declared 12-unit ramp (238→250) puts
-it at ~48–52 title units, so in title units: **period ≈ 670–720**, **the sweep
-leads the plate by ≈ 40**.
+🔴 **Do not convert these to title units.** Calibrating via the plate's declared
+12-unit ramp gives a title clock that puts `title start → plate` at 75 units
+against a declared 238 — see below. The ratios stand alone; the conversion does
+not.
 
 ## What this says about F6
 
-**The sweep does not start at title `t=0`, and it does not start exactly with the
-plate.** It starts a short, reproducible interval *before* the plate — ~0.8 of
-the baseline, ≈40 title units, well under a second.
+**The sweep starts about HALFWAY between the title's first drawn element and the
+plate's appearance.** Clock-free, and the quantity to implement against:
 
-So the human's report is approximately right and the port's behaviour is not: at
-`t=0` the port starts it roughly 200 units early. The human could not have seen a
-40-unit lead as separate from the plate, and did not claim to.
+| | f6 | f6b | agreement |
+|---|---|---|---|
+| **(onset − title start) / (plate − title start)** | **0.489** | **0.507** | **3.7%** |
+
+So it starts neither at `t=0` (the port's behaviour) nor just before the plate.
+
+🔴 **This corrects an earlier version of this page**, which said "≈40 title
+units, well under a second" and "at `t=0` the port is ~200 units early". Both
+came from converting the ratio into title units through the plate's declared
+12-unit ramp — the same calibration this page already flags as conflicted. The
+conflict is worse than stated: `title start → plate` measures **75 title units**
+through that calibration, where the declared data puts the plate at `t=238`.
+**3.2× out, not 1.75×.**
+
+**Every title-unit figure on this page is withdrawn.** The ratios never needed a
+clock and are unaffected.
+
+⚠️ Note the two ratios describe the same fact and do not conflict: the baseline
+(`-0.740` → plate) is only ~64% of `title start → plate`, so 0.79 baselines and
+0.51 title-spans are the same interval measured against different rulers.
 
 ## 🟡 A conflict I did not smooth