[FFmpeg-devel] [PATCH] aacenc_utils: unroll loops to allow compiler to use SIMD.

[Reimar Döffinger]

On Sun, Mar 06, 2016 at 07:35:58PM +0100, Reimar Döffinger wrote:
> Approximately 10% faster transcode from mp3 to aac
> with default settings.

Note to anyone wanting to optimize it further:
There is almost 25% on the table if you can replace
the pow() and cos() function uses by something more
efficient.
Unfortunately that won't be possible to do without changing
the output at least a bit, so I don't dare to do it,
plus one option, accessing the approximations that may be available
as CPU instructions, might not be possible without using assembler.
Beyond that, optimizations would probably have to attack
search_for_quantizers_twoloop in a way that makes it lower
quality but faster (it stands for another 25%, not including
subfunctions it calls).