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
#1369 laid the groundwork for Linux framebuffer graphics via the Notcurses abstraction. It needs further work, though. At a minimum, we need wipes and rebuilds. I believe that we can do the former by writing a black block where the cell was, and damaging the corresponding glyph. The latter ought be as simple as replaying the relevant chunk of the copied data. This ought give us performance/complexity comparable to Kitty+animations, which is pretty damn good.
The text was updated successfully, but these errors were encountered:
This isn't going to be too hard. I just implemented fbcon_wipe(), though it still needs to save the auxvec. This will be an old-kitty-style auxvec, holding the original alpha values. in fbcon_blit(), we need check the TAM and zorch out anything that's annihilated...just got that done. we need pack the auxvec from there as well. we need rebuild, and then some glyph invalidation, and we're done! it looks fantastic =].
Both wipe and restore now appear to be working. The frame count on ncplayer now works as expected. The greatscott image in intro is now properly cleared. Good stuff!
#1369 laid the groundwork for Linux framebuffer graphics via the Notcurses abstraction. It needs further work, though. At a minimum, we need wipes and rebuilds. I believe that we can do the former by writing a black block where the cell was, and damaging the corresponding glyph. The latter ought be as simple as replaying the relevant chunk of the copied data. This ought give us performance/complexity comparable to Kitty+animations, which is pretty damn good.
The text was updated successfully, but these errors were encountered: