search

happycube / ld-decode

Software defined LaserDisc decoder
GNU General Public License v3.0
307 stars 80 forks source link

Handle more variation in VITC #827

Closed atsampson closed 1 year ago

atsampson commented 1 year ago

@harrypm's Eden Project sample failed to decode for a couple of reasons: it was using VITC lines we weren't looking at, and the time between the sync pulse and VITC start meant the end of VITC was (just) outside the active region.

Fix this by looking at all the possible lines, and giving the decoders the complete composite line to look at rather than just the active region.

atsampson commented 1 year ago

Reworked after testing with some commercial tapes - The Matrix VHS (S016985) has VITC with timing that's out of spec in the other direction (in fact, it starts in spec, then jumps a bit's width to the left as the film starts). So allow any start position after the colourburst.

harrypm commented 1 year ago

So I realised timecode cant be done properly with FFMPEG normally, however, if you copy the DV codec method of chaptered sub-clips embedded into a single file with each having their own date/timecode in theory that can be applied to tapes with VITC timecode with a new ''sub-clips made per each x amount of timecode difference or RF drop off between segments thought of this after reading this doc and going over some HDV Xmas clips.

Just thought I would copy this over from the discord.