Skip to content
Snippets Groups Projects
Select Git revision
  • rav1e_16b
  • rav1e_15b
  • rav1e_14
  • rav1e_13b
  • rav1e_12
  • rav1e_11b
  • rav1e_11
  • rav1e_10
  • last_overlay
  • fix_rect_tx_var_tx
  • rav1e_9
  • new_frame_context_signaling_4
  • always_sef
  • rav1e_8
  • rav1e_6
  • rav1e_7b
  • rav1e_5
  • rav1e_4
  • rav1e_3
  • ravie_5
20 results
Compare
user avatar
Use fdct8x4 instead of fdct4x4 where the block size allows it.
Ronald S. Bultje authored 
This allows for faster SIMD implementations in the future (currently
there is no speed impact).

Change-Id: I732647e9148b5dcb44e6bc8728138f0141218329
5cfd82bc
History
user avatar 5cfd82bc
History