You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
t27f: pixel sweep -- no overlay on any mode 5 crossfade frame, every lesson, both languages, three voices #10651
A measurement on the Railway lab of specs/t27f/lesson_reel.t27 in LR_MODE=5, as on #10647. If it finds an overlay on a crossfade, the fix goes in lesson_reel.t27 (and its seal) in a follow-up PR. No gate, cap or ledger change.
User Scenarios
A viewer watches any lesson reel. At each of the two cuts the scenes crossfade for 16 frames, and no caption, tag, card or address is drawn over the blend.
Render every lesson (27) x both languages x three voices (1300/4200/6800, 900/2600/4100, 700/4200/8000) in the tall orientation, lossless, from solid-colour clips. Render lesson 0 in both orientations.
On every crossfade frame (start = ceil(offset x 30), 16 frames), count the distinct colours of the whole frame. A blend of two solid clips is one colour; more than one means something is drawn over it.
Activity
t27f skill: sweep every mode 5 crossfade frame, 324 renders, 0 flagged (