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

Commit Graph

  • 7b3c83858d re: index F6 unit 9, which closes the sweeps' unmeasured period sylph-decoder 2026-09-02 19:21:09 +00:00
  • 15207b1c74 re: F6 unit 9 -- sweep period and onset, as ratios reproduced across two runs sylph-decoder 2026-09-02 19:20:30 +00:00
  • a726391ead tools: make the sweep probe confirm it is actually recording before proceeding sylph-decoder 2026-09-02 19:11:39 +00:00
  • 7863ad3b62 re: F6 unit 9 -- wrap-to-wrap period NOT obtained, two harness failures sylph-decoder 2026-09-02 19:09:59 +00:00
  • fbc008a0ab re: F6 unit 8 -- the leaf's clock in title units: offset +40, rate ~1/2 sylph-decoder 2026-09-02 19:06:11 +00:00
  • ffde5abd94 re: F6 unit 7 -- the title has exactly ONE moving quad; the lights are in the texture sylph-decoder 2026-09-02 18:57:07 +00:00
  • d53128151f re: F6 unit 6 -- pteff03a is not submitted at any size; the collapsed-scale candidate dies sylph-decoder 2026-09-02 18:46:22 +00:00
  • 19757cfec6 re: F6 unit 5 -- pteff03a is never drawn; the game shows ONE sweep, not two sylph-decoder 2026-09-02 18:43:18 +00:00
  • a30761f128 re: F6 unit 4 -- the sweep is on screen 41 frames after the title starts, not at the plate sylph-decoder 2026-09-02 18:37:21 +00:00
  • f456c87d06 re: F6 unit 3 -- the sweep's POSITION fits the declared track, origin at leaf t=0 sylph-decoder 2026-09-02 18:35:54 +00:00
  • 796bc0420c re: CORRECT F6 unit 2 -- identification wrong, presentation misleading, claim downgraded to amber sylph-decoder 2026-09-02 18:27:21 +00:00
  • bc241f5d06 re: F6 unit 2 -- the parent's alpha IS multiplied into the sweep leaf sylph-decoder 2026-09-02 18:23:49 +00:00
  • df41225003 re: F6 unit 1 -- 238..250 on the sweep is an EXIT, not a start sylph-decoder 2026-09-02 18:19:21 +00:00
  • 90b7c6ea26 re: F3 -- the title plays BGM cues 1102 and 1103, from seven sites in its GamePart sylph-decoder 2026-09-02 18:16:55 +00:00
  • 1ed07d246e Merge remote-tracking branch 'origin/main' into auto/frame-blend-draw-path sylph-decoder 2026-09-02 18:15:33 +00:00
  • 620ec5e60b agents: one item only -- the title's animation timing -- and split work into human-checkable units MechaCat02 2026-09-02 20:14:46 +02:00
  • e38556d976 handoff: F2 closed -- no per-cue gain on the disc, the port authors the mix knowingly sylph-decoder 2026-09-02 18:06:52 +00:00
  • 1021708636 re: F2 CLOSED -- undecodable with reach, all three sites checked sylph-decoder 2026-09-02 18:06:36 +00:00
  • 654c4f1e99 re: F2 second site checked -- sound.pak carries no readable bank header either sylph-decoder 2026-09-02 17:36:59 +00:00
  • 02d7f9062b re: F2 -- no gain field in tables.pak, with the two unchecked sites named sylph-decoder 2026-09-02 17:08:07 +00:00
  • 8c5bd2ca0a Merge remote-tracking branch 'origin/main' into auto/frame-blend-draw-path sylph-decoder 2026-09-02 17:05:38 +00:00
  • 85d2e93d05 re: F1 harness built and committed; the run did not reach the menu sylph-decoder 2026-09-02 16:41:55 +00:00
  • de5f04038d agents: correct "both clocks" -- there is ONE, and F4 tests whether it is right MechaCat02 2026-09-02 18:39:13 +02:00
  • ad3d14ab26 re: a THIRD timed channel in the decoder -- my "only two" was wrong, and +0xC4/+0xC8 are timings not masks sylph-decoder 2026-09-02 16:35:55 +00:00
  • fa9f9eee52 Merge remote-tracking branch 'origin/main' into auto/frame-blend-draw-path sylph-decoder 2026-09-02 16:34:44 +00:00
  • 06f890361b agents: P5's gate is MET, and four findings from the same walk MechaCat02 2026-09-02 18:32:38 +02:00
  • e820121d8a re: the plate pulses via ptbtn00f on a declared 120-unit loop, not ptbtn00 at 244 sylph-decoder 2026-09-02 16:28:33 +00:00
  • 99330c3e87 handoff: H1 -- no key-repeat in the decoder; the timers are a double-tap latch on LB/LT sylph-decoder 2026-09-02 16:21:06 +00:00
  • dcaa322146 re: the decoder's timers are a double-tap latch on LB/LT, not key repeat sylph-decoder 2026-09-02 16:20:50 +00:00
  • cac31deccd re: sub_822380B0 is a type-tagged draw-queue walker, not the interpolator sylph-decoder 2026-09-02 16:17:12 +00:00
  • 8859189938 Merge remote-tracking branch 'origin/main' into auto/frame-blend-draw-path sylph-decoder 2026-09-02 16:15:46 +00:00
  • a71dea9b8d agents: the logo splashes are DONE -- the human cannot tell them from the game MechaCat02 2026-09-02 18:15:03 +02:00
  • 7c47ba13be handoff: withdraw the resample caveat -- the captures are crops, pixels to pixels sylph-decoder 2026-09-02 16:14:41 +00:00
  • b32bcb97ae re: CORRECT the resample consequence -- the captures are crops, not resamples sylph-decoder 2026-09-02 16:14:28 +00:00
  • 3dfedd152b handoff: Canary applies no gamma but does resample; the vertex stream is the cleaner oracle sylph-decoder 2026-09-02 16:11:26 +00:00
  • 515ad205e5 re: what Canary does between the guest's draw and a captured pixel sylph-decoder 2026-09-02 16:11:09 +00:00
  • 934fbfbbbf re: the interpolator is not found, and four routes are excluded sylph-decoder 2026-09-02 15:42:25 +00:00
  • e1b259925c handoff: the game lerps every frame, and your 3.2 s hold is correct sylph-decoder 2026-09-02 15:29:54 +00:00
  • acc6c09632 re: the game interpolates every frame, and its splash is ALSO mostly frozen sylph-decoder 2026-09-02 15:29:34 +00:00
  • 8580c8738b Merge remote-tracking branch 'origin/main' into auto/frame-blend-draw-path sylph-decoder 2026-09-02 15:26:38 +00:00
  • d394c55cbb agents: the splash does not animate, and three instruments could not see it MechaCat02 2026-09-02 17:24:42 +02:00
  • 33a6b4ea6f re: the three routes to 60 are one witness in three coats -- audit, and the capture sized sylph-decoder 2026-09-02 15:08:55 +00:00
  • befcb74780 re: the clock rate follows the VBLANK -- reconciling three pages that disagreed sylph-decoder 2026-09-02 15:05:03 +00:00
  • a7013a8fe1 re: the long capture ran and fell 32% short -- sized the next one from measurement sylph-decoder 2026-09-01 20:09:34 +00:00
  • e354e11f4c method: a finding is only as portable as the tool that produced it sylph-decoder 2026-09-01 20:01:03 +00:00
  • 3f114a28af tools: export ALL FOUR container-only Canary commits, not just my own sylph-decoder 2026-09-01 19:59:14 +00:00
  • afec2ae592 tools: export the Canary logger patch the clock findings depend on sylph-decoder 2026-09-01 19:56:29 +00:00
  • 3e067decef re: the clock origin is blocked -- a 600-present boot capture has no title in it sylph-decoder 2026-09-01 19:44:33 +00:00
  • 405f16e382 handoff: the forced-frame-rate run refuted me; the game is frame-based, the port's time-based clock is correct sylph-decoder 2026-09-01 19:40:16 +00:00
  • 705a8485dc re: the clock is FRAME-based, 1 unit per present -- my own test refuted my claim sylph-decoder 2026-09-01 19:39:30 +00:00
  • 5d72bda77d re: PRE-REGISTER the forced-frame-rate test of the time-based clock sylph-decoder 2026-09-01 19:37:41 +00:00
  • 73f5f57966 re: WITHDRAW 120 -- units per FRAME is not a constant, the clock is time-based, 60 is right sylph-decoder 2026-09-01 19:35:24 +00:00
  • 33f141a3df re: the dwell corpus does not falsify 120 -- it measures the screen, not the animation sylph-decoder 2026-09-01 19:30:02 +00:00
  • 058a450ec5 handoff: 120 units/s, the plate is ~2 s late, and every second I quoted is halved sylph-decoder 2026-09-01 19:24:05 +00:00
  • 60145720ed re: the guest presents at 60 fps -- 120 units/s, and the port's plate IS ~2 s late sylph-decoder 2026-09-01 19:23:13 +00:00
  • 56ae05a18b re: PRE-REGISTER the decode-vs-rotate test, and add a content hash to the logger sylph-decoder 2026-09-01 19:21:04 +00:00
  • eeb4473b4f re: the port's 12-unit bracket does not exclude 120 -- its low end is the sampler's floor sylph-decoder 2026-09-01 19:15:34 +00:00
  • ca401d8b04 re: WITHDRAW "the guest presents at 30 fps" -- the clean result WAS the failure mode sylph-decoder 2026-09-01 19:11:05 +00:00
  • fa27509937 formats: teach the reference renderer the additive blend, and re-open 8 claims formats-pin-2026-09-01b sylph-decoder 2026-09-01 19:03:53 +00:00
  • 9ea6ee7555 re: the blend bit was challenged on pteff10 and SURVIVES; finding 3 loses its second candidate sylph-decoder 2026-09-01 18:55:50 +00:00
  • 27f81d6d91 re: the guest presents at 30 fps -- 60 units/s, and the port was right sylph-decoder 2026-09-01 18:49:40 +00:00
  • f0185fba52 re: PRE-REGISTER the guest frame-rate test -- the movie is the ruler sylph-decoder 2026-09-01 18:43:13 +00:00
  • 4518b5c76a re: the pad word is REMAPPED -- my own bit table was mislabelled throughout sylph-decoder 2026-09-01 18:33:42 +00:00
  • 3ad8e7228f Merge remote-tracking branch 'origin/main' into auto/frame-blend-draw-path sylph-decoder 2026-09-01 18:26:34 +00:00
  • e263041337 re: the splash blur is a TEXTURE -- palogo_*_eff is a baked 10-px glow formats-pin-2026-09-01 sylph-decoder 2026-09-01 18:24:14 +00:00
  • 18b5b3d5f1 port: give the port container a GPU path -- it never had one MechaCat02 2026-09-01 20:22:25 +02:00
  • 4ac23b94dd docker: auto-restart, and resume the session the agent was actually in MechaCat02 2026-09-01 20:20:51 +02:00
  • 958eca6b6b formats: expose the T8aD blend bit on the public API, with a control sylph-decoder 2026-09-01 18:18:47 +00:00
  • df51101faf handoff: withdraw the splash rate, and strike the section that carried it sylph-decoder 2026-09-01 17:30:24 +00:00
  • c1c8aa4288 re: WITHDRAW 'the unit rate is per-GamePart' -- it was the emulator's frame rate sylph-decoder 2026-09-01 17:29:52 +00:00
  • 227d9b0439 handoff: one units-per-second cannot be right, the splash timeline is confirmed, and H5 refuted sylph-decoder 2026-09-01 17:20:59 +00:00
  • 4b5ac88f0a re: the declared timeline DOES reproduce the splash -- and the unit rate is per-GamePart sylph-decoder 2026-09-01 17:19:59 +00:00
  • 4e0f7bf6cb handoff: the rate resolved to 56.8 units/guest-second -- keep 60, and H3's last candidate closes sylph-decoder 2026-09-01 17:11:46 +00:00
  • 3d1c3c9564 re: fix the stale headline on the units-per-second page sylph-decoder 2026-09-01 17:11:28 +00:00
  • d67aaaf30b re: the rate is 56.8 units per guest second -- both predictions hold, control passes sylph-decoder 2026-09-01 17:11:19 +00:00
  • 1ac388b0c1 handoff: walk back 2-units-per-frame as a mechanism, and answer the port's blend-space ask sylph-decoder 2026-09-01 17:05:09 +00:00
  • 781a70af9a re: the animation clock is TIME-INTEGRATED -- one prediction held, one failed sylph-decoder 2026-09-01 17:04:23 +00:00
  • 60b05bcd90 re: the game blends in the ENCODED space -- no gamma render target, anywhere sylph-decoder 2026-09-01 17:01:11 +00:00
  • 8dac5aa264 re: pre-register H4 -- is the animation clock frame-counted or time-integrated? sylph-decoder 2026-09-01 16:54:13 +00:00
  • 0565098fa2 handoff+index: H3 answered, the settle anchor, and the pad's full field set sylph-decoder 2026-09-01 16:50:52 +00:00
  • 6aab30253a re: H3 answered -- 2 units per guest frame, and the settle anchor is t~160 sylph-decoder 2026-09-01 16:49:08 +00:00
  • 19aa1a89a3 re: pre-register the H3 measurement before reading the capture sylph-decoder 2026-09-01 16:40:11 +00:00
  • c85e5b7c56 re: record that stale-instrument was run for the logger change, and returned nothing sylph-decoder 2026-09-01 16:21:39 +00:00
  • 4bb67fa166 re: the splashes have NO post-process -- one pass, source-over, alpha in the vertex stream sylph-decoder 2026-09-01 16:21:17 +00:00
  • 904cbeb99b Merge remote-tracking branch 'origin/human/r1-register-reclassification' into auto/frame-blend-draw-path sylph-decoder 2026-09-01 16:01:45 +00:00
  • 8008170168 Merge remote-tracking branch 'origin/main' into auto/frame-blend-draw-path sylph-decoder 2026-09-01 16:01:42 +00:00
  • 79783ff9ee agents: point each brief at its human branch, to merge on the first iteration MechaCat02 2026-09-01 17:59:59 +02:00
  • 1494b094c7 port: record H3 -- the PRESS (A) plate arrives late, and which half is ours human/r1-retro-tick MechaCat02 2026-09-01 17:59:17 +02:00
  • 1ad519d3ba agents: the splashes exactly, and stop photographing a moving thing MechaCat02 2026-09-01 17:59:17 +02:00
  • 72f116a3b4 port: Ⓐ was never bound to the pad, and the stick is not an edge MechaCat02 2026-09-01 17:45:36 +02:00
  • 0f1b11920c retro: tick §7.1 -- the R1 register reclassification is done, and what it found MechaCat02 2026-09-01 17:21:30 +02:00
  • 3e60560344 register: reclassify under R1 -- name every instrument, re-open the ten our own killed MechaCat02 2026-09-01 17:21:17 +02:00
  • 3e9160b372 port: confirm the plate quad by size, and record a frame-spacing discrepancy I cannot close Sylpheed port agent 2026-08-31 11:56:29 +00:00
  • dfb768270a port: pre-register the plate composite test, and catch my own prediction being wrong Sylpheed port agent 2026-08-31 11:53:21 +00:00
  • 4a2f96167f port: the plate's ramp was already in my export, and the renderer reproduces it Sylpheed port agent 2026-08-31 11:51:26 +00:00
  • 9ff5c5f082 port: draw the plate's highlight additive, and find my harness poses it where it cannot be seen Sylpheed port agent 2026-08-31 11:48:36 +00:00
  • 1b8f9d2813 port: fix the runbook's unattended walk -- it never reached the submenu P5 rests on Sylpheed port agent 2026-08-31 11:38:13 +00:00
  • b0bf54b4a0 retro: record the error path that fired on success, in the file not the message sylph-decoder 2026-08-31 11:17:27 +00:00
  • b86e82150d retro §6: route the Port's BLOCKED.md through HANDOFF, and make the pointer fail loudly sylph-decoder 2026-08-31 11:16:10 +00:00
  • 2af316dd82 retro: the agreed result, plus the cue-suppression matrix it came out of Sylpheed port agent 2026-08-31 11:10:50 +00:00