From 1b3a9e9086fd7f4c7b410477c2dfc71050f1f18e Mon Sep 17 00:00:00 2001 From: sylph-decoder Date: Sun, 30 Aug 2026 18:21:20 +0000 Subject: [PATCH] re: a third title->menu replicate, from a run that ran the wrong experiment Set out to measure menu -> EXTRAS, the reverse of the pair measured at a 2-frame gap, to test whether the black gap is a property of the screen pair or of the direction. The run did not do that. A three-part patch to fade_decompose.sh asserted two of its three replacements and left the third -- the branch condition -- unchecked. It silently failed, so WHERE=menu2extras fell through to the `title` branch. The capture is well-formed and is of a different transition than intended, which is the build-ordinal error's shape again: right-looking output for the wrong object. What caught it was the log LACKING the navigation lines the intended branch prints; the data itself looked entirely fine. Salvaged, because the accidental transition is one already measured twice: run outgoing ramp black incoming decay 1 67-70: 63,127,191,255 3 73-77 2 64-67: 63,127,191,255 3 70-74 3 92-95: 63,127,191,255 3 98-103 Three independent runs, gap = 3 frames every time, outgoing ramp byte-identical in all three. That takes "the black gap is not a load" from two replicates to three, and makes the 4-frame outgoing ramp as solid as anything measured here. menu -> EXTRAS remains open; the condition is fixed (with an assertion this time) and the run has not been taken. METHOD gains two entries. Assert every edit, not most of them -- and have each branch announce itself in the log, so a run that took the wrong path says so before its numbers are read. And: "appears nowhere in crates/" is a claim about a TREE. sylpheed-port found SYLPHEED_KF_TIME_SHIFT live at ui_layout.rs:497 on their branch, which carries the stale era; both statements are true of different trees. With main 145 commits behind and each agent on a topic branch, any claim about what the code contains needs its ref attached. Co-Authored-By: Claude Opus 5 (1M context) Claude-Session: https://claude.ai/code/session_01Wuu56cE8vJGTBtn1ppsk8v --- docs/re/METHOD.md | 20 ++++++++++++++++++ docs/re/data/fade-four-transitions.txt | 23 ++++++++++++++++++++ tools/re-capture/fade_decompose.sh | 29 +++++++++++++++++++++++--- 3 files changed, 69 insertions(+), 3 deletions(-) diff --git a/docs/re/METHOD.md b/docs/re/METHOD.md index 2b10a61f..295d516b 100644 --- a/docs/re/METHOD.md +++ b/docs/re/METHOD.md @@ -532,6 +532,26 @@ agent's loop prompt, i.e. nowhere durable. See [`README.md`](README.md) for the rule vocabulary. **That is a sample, not a sweep**, and older headings are the likelier to be stale for having had more chances to be overturned. +* 🔴 **Assert EVERY edit, not most of them.** A three-part patch to a capture + script asserted two replacements and left the third unchecked. The third + silently failed, so `WHERE=menu2extras` fell through to the `title` branch and + the run produced a **well-formed capture of a different transition** — which I + came close to analysing as the intended one. Same family as the build-ordinal + error: right-looking output for the wrong object. ⚠️ **What caught it was the + instrument's own log lacking lines the intended branch prints**, not the data + looking wrong — the data looked fine. So: assert every replacement, and **have + each branch announce itself in the log**, so a run that took the wrong path says + so before its numbers are read. + +* ⚠️ **"Appears nowhere in `crates/`" is a claim about a TREE, and I stated it + without one.** I reported `SYLPHEED_KF_TIME_SHIFT` as removed and absent from the + code; `sylpheed-port` found it **live at `ui_layout.rs:497` on their branch**, + which carries the stale era. Both true, of different trees. On a project where + `main` is 145 commits behind and each agent works from a topic branch, *any* + statement about what the code contains needs its ref attached — the same + discipline as "state what the number is a number of", applied to scope rather + than to units. + ## Runtime / emulator * **Look at the PNG** — and check its dimensions. diff --git a/docs/re/data/fade-four-transitions.txt b/docs/re/data/fade-four-transitions.txt index a408a315..4d502420 100644 --- a/docs/re/data/fade-four-transitions.txt +++ b/docs/re/data/fade-four-transitions.txt @@ -110,3 +110,26 @@ frame untextured full-screen textured (distinct) draws tex 78 [64] [8, 96, 246, 255] 8 6 79 [64] [9, 192, 245, 255] 8 6 80 [64] [10, 244, 255] 8 6 + +################################################################################ +# A THIRD title -> menu REPLICATE, 2026-08-30 -- from a run that executed the +# WRONG EXPERIMENT. The script was invoked as WHERE=menu2extras but a condition +# edit had silently failed to apply, so it took the `title` branch instead. The +# capture is well-formed and is of a different transition than intended -- the +# same shape as the build-ordinal error, caught only because the log lacked the +# navigation lines the intended branch prints. +# +# run outgoing ramp black frames incoming decay +# 1 (fadecap2) 67-70: 63,127,191,255 70,71,72 = 3 73-77 +# 2 (fadecap4) 64-67: 63,127,191,255 67,68,69 = 3 70-74 +# 3 (m2e) 92-95: 63,127,191,255 95,96,97 = 3 98-103 +# +# ✅ THREE INDEPENDENT RUNS, gap = 3 frames every time, and the outgoing ramp is +# byte-identical in all three (63, 127, 191, 255 -- steps of exactly 64). That +# strengthens "the black gap is not a load" from two replicates to three, and the +# ramp's exactness across runs makes the 4-frame outgoing duration as solid as +# anything measured here. +# +# 🟡 STILL OPEN: menu -> EXTRAS, the reverse of the pair measured at a 2-frame +# gap. The instrument is fixed (WHERE=menu2extras now reaches its branch) and the +# run has not been taken. diff --git a/tools/re-capture/fade_decompose.sh b/tools/re-capture/fade_decompose.sh index cea4cace..fa63f37e 100755 --- a/tools/re-capture/fade_decompose.sh +++ b/tools/re-capture/fade_decompose.sh @@ -24,6 +24,10 @@ # WHERE=menu (default) arms on the main menu and presses (B) -> menu -> title. # WHERE=title arms on the boot title and presses (A) -> title -> menu. # WHERE=extras navigates to EXTRAS, arms there, presses (B) -> menu. +# WHERE=menu2extras navigates to EXTRAS, (B)s back to the menu with focus +# restored, arms there and presses (A) -> EXTRAS. The +# REVERSE pair, to test whether the black gap is a property +# of the screen pair or of the direction. # Tests whether "(B) has no black interval" is a rule or # one screen pair -- sylpheed-port's BLOCKED.md ask #1. # The two transitions have DIFFERENT declared fade-ins for the incoming screen -- @@ -71,7 +75,7 @@ if [ "$WHERE" = "menu" ]; then done [ "$s" = "menu" ] || { echo "NO MENU (screen=$s)"; exit 2; } BTN=B -elif [ "$WHERE" = "extras" ]; then +elif [ "$WHERE" = "extras" ] || [ "$WHERE" = "menu2extras" ]; then python3 "$SD/pad.py" tap A 0.5 for _ in 1 2 3 4 5 6; do sleep 4; s="$(screen)"; echo " after A: $s" @@ -90,12 +94,31 @@ elif [ "$WHERE" = "extras" ]; then shot "$OUT/try$try.png" w="$(python3 "$SD/which_title_screen.py" "$OUT/try$try.png")" echo " try $try: $w" - case "$w" in extras*) found=1; break;; esac + # ⚠️ REQUIRE A MARGIN. The first version accepted any line starting + # "extras", but which_title_screen.py returns a nearest-match even for a + # screen outside GP_TITLE entirely -- where the margin collapses to ~0.1. + # A bare prefix test would then accept a wrong screen on a coin flip. The + # control puts a true match at ~10-11; require > 5. + m=$(echo "$w" | sed -n 's/.*margin \([0-9.]*\).*/\1/p') + ok=$(awk -v m="${m:-0}" 'BEGIN{print (m>5)?1:0}') + case "$w" in extras*) [ "$ok" = 1 ] && { found=1; break; } + echo " (rejected: margin ${m:-?} <= 5)";; esac python3 "$SD/pad.py" tap B 0.5; sleep 5 python3 "$SD/pad.py" dpad UP 0.3; sleep 2 done [ -n "$found" ] || { echo "NEVER REACHED EXTRAS"; exit 3; } - BTN=B + if [ "$WHERE" = "menu2extras" ]; then + # We are ON extras. (B) returns to the menu with focus RESTORED on EXTRAS + # (measured 4/4, menu-navigation-semantics.md), so arming there and pressing + # (A) gives the menu -> EXTRAS transition: the exact reverse of the pair + # already measured at a 2-frame gap. + python3 "$SD/pad.py" tap B 0.5; sleep 6 + shot "$OUT/back-on-menu.png" + echo " back on menu: $(python3 "$SD/which_title_screen.py" "$OUT/back-on-menu.png")" + BTN=A + else + BTN=B + fi else # arm on the title itself and press (A): the incoming screen is then build 5, # whose declared fade-in is 12 units where build 4's is 16.