Re: flac -- exhaustive model search vs. -A <*>?
Linda Walsh <[email protected]> Fri, 22 Jun 2012 02:47:59 -0700
| Newsgroups | gmane.comp.audio.compression.flac.user |
|---|---|
| Message-ID | <[email protected]> |
Martin Leese wrote: > Linda Walsh <[email protected]> wrote: > ... > >> That brings up another question then... What is 'hi' (v. low?)... >> > > Zero, or close to zero, is low. 12 is high. > ---- They mention 32 as a top end, so maybe 12 isn't that slow? I'm got encode times around 30-45 seconds on a 2 CPU machine, which has fallen to about 10 secs/album on my current server @2.7Ghz using ~1080% cpu... > > > > Despite the title of -M, -m the one that is > exhaustive. This is stated at: > http://flac.sourceforge.net/documentation_tools_flac.html#flac_options_adaptive_mid_side > The documentation is so clear on music optimization progs...(Lame is about the same). Most people still think 320CBR is the best, though the authors have stated it is no better than -V0 and just wastes bits. _______________________________________________ Flac mailing list [email protected] http://lists.xiph.org/mailman/listinfo/flac