fabi/Sylpheed
1
0
Fork 0
Code Issues 9 Pull Requests Actions Packages Projects Releases Wiki Activity

re: refutation attempt on Q2's 'shipped twice' -- it survives, with one real caveat

Browse Source 
The doubt was my own artefact: the entry dump printed only the first two sprite
names in HashMap order, making 11 and 14 look like different studios. Full sets
are identical.

7 of 8 pairs declare identical sprite sets, control included. 4/7 does not: entry
7 carries nine sprites entry 4 lacks, including ptlogo_jp and ptlogo_jpeff, so the
Japanese title is a different element inventory rather than the same screen
localised. That matches the JP capture work from the other side.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Wuu56cE8vJGTBtn1ppsk8v
This commit is contained in:
sylph-decoder
2026-08-30 22:05:03 +00:00
parent 04e8c6d8a7
commit e35c56a5ed
3 changed files with 201 additions and 105 deletions
Download Patch File Download Diff File Expand all files Collapse all files

57
docs/re/data/gp-title-pair-check.txt Normal file
View File

@@ -0,0 +1,57 @@
# Does "GP_TITLE is 8 screens shipped twice, EN/JP" hold pair by pair?
# 2026-08-30. Refutation attempt on sylpheed-port's Q2 headline. IT SURVIVES.
#
# THE DOUBT: the entry dump showed entry 11 with `palogo_gamearts` and entry 14
# with `palogo_seta` -- different studios, not a language pair. If the halves of
# a pair declare different sprites, "shipped twice" is the wrong description.
#
# ❌ THE DOUBT WAS MY OWN ARTEFACT. That dump printed only the FIRST TWO sprite
# names per entry, in HashMap order, which is not stable. Comparing the full
# sets, 11 and 14 are an IDENTICAL SET of 6 -- one screen listing all three
# studios. A truncated listing in unspecified order is not evidence of anything.
#
# CONTROL: 2/3, the PRESS (A) plate, is a known real EN/JP pair and must come out
# matching. It does.
#
# tools: crates/sylpheed-formats/examples/gp_title_pair_check.rs
0/1 loading plain 7 shared IDENTICAL SET
2/3 PRESS (A) plate [CONTROL] 2 shared IDENTICAL SET
4/7 title art 15 shared DIFFERS
only in 4: ["pteff01.t32", "ptlogoall_eff.t32", "ptlogoall_eff2.t32"]
only in 7: ["ptlogo3a.t32", "ptlogo3b.t32", "ptlogo3c.t32", "ptlogo_all_eff.t32", "ptlogo_back1.t32", "ptlogo_eff2.t32", "ptlogo_eff3.t32", "ptlogo_jp.t32", "ptlogo_jpeff.t32"]
5/8 main menu 21 shared IDENTICAL SET
6/9 EXTRAS 20 shared IDENTICAL SET
10/13 publisher splash 2 shared IDENTICAL SET
11/14 developer splash 6 shared IDENTICAL SET
12/15 loading dressed 9 shared IDENTICAL SET
# ------------------------------------------------------------------------------
# ✅ THE HEADLINE SURVIVES: 7 of 8 pairs declare IDENTICAL sprite sets, including
# both splashes and the control.
#
# 🟡 BUT ONE PAIR IS NOT A PAIR IN THE SAME SENSE. 4/7, the title art, DIFFERS:
# only in 4: pteff01, ptlogoall_eff, ptlogoall_eff2
# only in 7: ptlogo3a, ptlogo3b, ptlogo3c, ptlogo_all_eff, ptlogo_back1,
# ptlogo_eff2, ptlogo_eff3, ptlogo_jp, ptlogo_jpeff
#
# Entry 7 carries NINE sprites entry 4 does not, including ptlogo_jp and
# ptlogo_jpeff. So the Japanese title is a different ELEMENT INVENTORY, not the
# same screen with different text -- which is the same fact the JP capture work
# already recorded from the other side ("the katakana subtitle, and a crystalline
# burst BEHIND the wordmark, which the English title lacks").
#
# ⚠️ So "8 screens shipped twice" is right as a COUNT and as a structure, and the
# port's entry map agrees. It is misleading only if read as "the two halves are
# the same screen localised", which is false for exactly one pair -- and that
# pair is the title, the screen most of this corpus's capture work is about.
#
# REACH: sprite-name sets only. Two entries could share every sprite name and
# still place them differently; this does not check placements or keyframes.