+| 13.9 | **The WM1A volume curve from a table the user brings** — new; the reference device had it copied in over adb, not by an install. **PARTIAL 2026-09-14** (reference device, Wampy installed): the install step, run as root, refused a junk `/contents/ov_127x.tbl` and installed all three tables from Wampy's copies with matching hashes; `cinder-voltable wm1a` then applied from Cinder's directory and `--volcurve` read vol 40 → 0x54 (84), vol 60 → 0x7C (124), monotonic. **The boot check failed first time**: `mkdir -p` under root's umask 077 made both directories 0700, so the uid-100 launcher logged `needs Sony's ov_127x.tbl` with the file present — fixed (explicit `chmod 755`); the next boot logged `volume curve: wm1a applied` at 25 s. (a) and (c) pass by the install step run over adb; still to do: the same through a real install, and (b), a player without Wampy. Also: a sweep while the codec is in standby prints `?` for every step; the probe now wakes it | (a) Wampy installed, nothing on the drive: install with `wm1a`. (b) A player without Wampy: put `ov_127x.tbl` and `ov_dsd_127x.tbl` from the cinder-sony-analysis release at the top of the drive, install with `wm1a`. (c) Any other file renamed `ov_127x.tbl` on the drive | Install log `volume table: installed ov_127x.tbl from … (SHA-256 matches Sony's)`; boot log `volume curve: wm1a applied`; `cinder-probe --volcurve` monotonic (vol 40 → PHV 84, vol 60 → 124). (c): `WARN: /contents/ov_127x.tbl is not Sony's ov_127x.tbl` and the stock curve | Boot log `needs Sony's ov_127x.tbl`: nothing reached `/system/vendor/unknown321/usr/share/cinder/audio_dac` — read the install log's `volume table:` and `WARN:` lines. A browser may have renamed the download (`ov_127x (1).tbl`) |
0 commit comments