Retract the pulse-length conclusion; show the minimum in the drill table

The fifth ladder, the first with a minimum pulse, moved the floor down a
full rung: only 5 percent failed to mark, and 5 percent now strikes. The
trace carries eight current segments where the run before it had seven,
with fire beginning at 6.2 s exactly at rung 1 and the usual flat
saturated final segment anchoring the count from the other end.

That retracts what the previous entry concluded. Pulse length is not
irrelevant: 10 percent moved from no mark at F100, with three times the
dose per millimeter, to a mark at F300 at the same density, the only
change being its pulses growing from 36-71 us stubs to 106 us. The
matched-pairs argument was sound but drawn entirely from comparisons at
or above 20 percent density, where every pulse length in play was already
long enough - it generalized from the one regime where pulse length does
not bite. Above ~100 us dose governs; below it pulse length does; below
~36 us the supply does not strike. The factory's 100 us quantum sits on
that boundary.

dladder now reads laser_pulse_min_ticks and prints what is actually
emitted. Without that its table reports the pulse density alone would
give, which is wrong wherever the minimum applies - at min 6 the bottom
four rungs all emit 213 us and vary their rate instead, and the operator
reads that table to interpret the material.

docs/BRINGUP.md

@@ -1090,22 +1090,28 @@ Open items only. Anything closed is in `CAMPAIGN-LOG.md`.
     under both models - the motion grid is identical and every density FIRE
     tick is one the analog run also fired, so the model only ever masks.
 
-    The base period is settled: four bench ladders (F300 at periods 20, 40
-    and 10, then F100 at 20) put the same six rungs on the material every
-    time, and the matched pairs across periods separate the variables - at
+    **Both dose and pulse length matter, in different regimes.** Four
+    ladders run with no minimum pulse (F300 at periods 20, 40 and 10, then
+    F100 at 20) put the same six rungs on the material every time - 20 %
+    and up - and their matched pairs looked like a clean answer: at
     identical pulse length, halving density killed the mark; at identical
-    density, varying pulse length 3x changed nothing. Feed did not move it
-    either: 10 % at F100 carries 44 % more energy per millimeter than 20 %
-    at F300, which marks, and still left nothing. So the low-end marking
-    limit is average power, not dose per length and not pulse length, and
-    the period can be chosen on other grounds. It stays at 20.
+    density, varying pulse length 3x changed nothing; and feed did not move
+    it either, 10 % at F100 carrying 44 % more energy per millimeter than a
+    marking 20 % at F300 and still leaving nothing.
 
-    What the ladders did expose is a floor of our own making, since fixed:
-    at 5 % the model emitted 36 us stubs and the `hv_current` trace shows
-    **no discharge at all** for that rung, while 10 % drew current for its
-    full 15 s and simply marked nothing. The factory never emits below one
-    100 us tick, and reaches low density by skipping windows instead -
-    which is now what `laser_pulse_min_ticks` does.
+    That reading was too broad. Every one of those comparisons sat at or
+    above 20 % density, where the pulses in play were already long enough.
+    A fifth ladder with a 3-tick minimum moved 10 % from nothing to a mark
+    at the **same density and less dose**, purely by lengthening its pulses
+    from 36-71 us to 106 us. So above ~100 us the outcome follows dose;
+    below it pulse length dominates - too short and the energy does
+    nothing, shorter still (36 us, one tick) and the supply does not strike
+    at all, which the `hv_current` trace showed as a rung with no discharge
+    for its full 15 s. That is what the factory's 100 us quantum protects,
+    and what `laser_pulse_min_ticks` now protects.
+
+    With a minimum in place the low end is decoupled from the base period,
+    so the period is free to be chosen on other grounds. It stays at 20.
 
     Owed: the user-facing scale. The factory maps its whole 1-100 power
     scale onto density 18.9-79.5 % (fit from the three captures; Full Power

docs/CAMPAIGN-LOG.md

@@ -2949,10 +2949,47 @@ only what it must. Rule 15 in the stream harness holds both halves: no burst
 below the minimum (excepting one clipped by fire going off mid-burst), and
 the rendered density still exact.
 
-Owed next: the user-facing scale. Under this model `$35` and `$36` are a
-density floor and ceiling, so mapping S onto the usable band is a settings
-choice rather than new code — but the floor's value wants a finer ladder than
-the 10 %→20 % step these four runs give.
+### The fifth ladder, and a conclusion retracted
+
+Same ladder, period 20, F300, with the 3-tick minimum in place. **The floor
+moved down a full rung: only 5 % failed to mark, and 5 % now strikes.**
+
+The trace carries eight current segments where the F100 run had seven.
+Segmenting by time rather than by zeros — at 5 % density the sampled current
+aliases, so isolated zeros appear mid-rung and cannot serve as boundaries —
+the rung period is 5.25 s and lines up end to end: fire begins at 6.2 s,
+exactly at rung 1's start, boundaries fall at 11.4, 16.6, 21.9, 27.3, 32.6,
+38.1 and 43.3 s, and the span is 42.0 s against 41.4 s for eight rungs. The
+final segment reads 937–981 flat and saturated, which can only be full
+density. Rung 1 shows peaks of 291, 286 and 204 where the F100 run held a
+flat zero for the rung's entire fifteen seconds.
+
+**This retracts the conclusion in the entry above.** Pulse length is not
+irrelevant: 10 % moved from no mark at F100 — with three times the dose per
+millimeter — to a mark at F300, at the same density, the only change being
+its pulses growing from 36–71 µs stubs to 106 µs. The matched-pairs argument
+was sound but drawn entirely from comparisons at or above 20 % density, where
+every pulse length in play was already sufficient; it generalized from the one
+regime where pulse length does not bite. Above ~100 µs dose governs, below it
+pulse length does, and below ~36 µs the supply does not strike at all. The
+factory's 100 µs quantum sits exactly on that boundary.
+
+Not read into: the low-rung current means (76 and 82 raw for 5 % and 10 %).
+At those duties a 3.3 Hz point sample of a pulsed current carries
+presence-versus-absence and nothing more. Noted as a confound, though it cuts
+against the result rather than for it — this ladder started at MPos 0,0 after
+the controller restart, so it may be on different material than the stacked
+Y=0/24/48/72 runs.
+
+Owed next: whether a longer minimum reaches further down (`min_ticks = 6`,
+213 µs, is set on the bench for the next ladder), and then the user-facing
+scale. Under this model `$35` and `$36` are a density floor and ceiling, so
+mapping S onto the usable band is a settings choice rather than new code, but
+the floor's value wants a finer ladder than the 10 %→20 % step. The trick has
+a ceiling of its own: a longer minimum at fixed dose means longer gaps, and
+once gap × feed approaches the beam spot a line dots. At 5 mm/s a 4.5 ms gap
+is 22 µm against a ~200 µm spot; at 2000 mm/min it is 150 µm, where dotting
+would start to show.
 
 ## Superseded status notes
 

scripts/bench/live_fire_drills.py

@@ -521,6 +521,8 @@ DLADDER_PITCH = 3.0                     # mm between rungs
 STREAM_RATE_HZ = 28160                  # machine tick (GFSINK_RATE default)
 PWM_PERIOD = 127                        # 7-bit power byte against PWMSAR
 CONF = os.environ.get('GFHOME_CONF') or '/data/forgefirm.conf'
+PULSE_MIN_KEY = 'laser_pulse_min_ticks'
+PULSE_MIN_DEFAULT = 3                   # glowforge_laser.c PULSE_MIN_TICKS_DEFAULT
 
 
 def conf_get(key):
@@ -621,9 +623,16 @@ def drill_dladder(g):
 
     # What each rung actually emits. The on-count is dithered between
     # adjacent integers, so the burst below is the mean.
+    min_ticks = int(conf_get(PULSE_MIN_KEY) or PULSE_MIN_DEFAULT)
+    if min_ticks < 1:
+        min_ticks = 1
     print('period %d ticks = %.0f us at %d Hz -> %.0f Hz pulse rate'
           % (period, period * 1e6 / STREAM_RATE_HZ, STREAM_RATE_HZ,
              STREAM_RATE_HZ / float(period)))
+    print('minimum pulse %d ticks = %.0f us (%s): below it the model skips'
+          % (min_ticks, min_ticks * 1e6 / STREAM_RATE_HZ,
+             PULSE_MIN_KEY if conf_get(PULSE_MIN_KEY) else 'driver default'))
+    print('periods and carries the debt, so the pulse never falls under it.')
     print('rungs (drawn in order, alternating direction, +Y between):')
     levels = []
     for pct in DLADDER_PCT:
@@ -638,8 +647,12 @@ def drill_dladder(g):
         # skipped instead - that is the short end this drill is for.
         lo = int(on)
         tick_us = 1e6 / STREAM_RATE_HZ
-        if lo == 0:
-            burst = '1 tick (%.0f us) on ~%.0f%% of periods' % (tick_us, on * 100)
+        if on < min_ticks:
+            # Below the minimum the model skips periods and carries the
+            # debt, so the pulse holds at the minimum and the rate drops.
+            burst = '%d ticks (%.0f us) every %.1f periods (%.0f Hz)' % (
+                min_ticks, min_ticks * tick_us, min_ticks / on,
+                STREAM_RATE_HZ / float(period) * on / min_ticks)
         elif on == lo:
             burst = '%d ticks (%.0f us)' % (lo, lo * tick_us)
         else: