15% faster refine_bidir_satd, 10% faster refine_bidir_rd (or less with trellis=2)
re-roll a loop (saves 44KB code size, which is the cause of most of this speed gain) don't re-mc mvs that haven't changed
Please register or sign in to comment
re-roll a loop (saves 44KB code size, which is the cause of most of this speed gain) don't re-mc mvs that haven't changed
VideoLAN code repository instance