On Fri, 3 Dec 1999, Mathew Hendry wrote:
> > From: "Acy Stapp" <[EMAIL PROTECTED]>
> >
> > I unfortunately can't get VTune to work reliably on LAME.
> > If someone wants to point me toward any other math-intensive
> > inner loops I will be glad to optimize those as well.
>
> The other big one is the FFT code; always shows up near the top of
> profiles. But there are quite a few high performance FFT
> implementations around (e.g. FFTW) that we haven't tried yet.
>
Indeed (not that I know any of them);
but, as I remarked some time ago;
why not look for a fast FFT which considers an additional parameter: cut
off frequency; eg when lame-encoding input only up to 16kHz, or when the
input file is of low F_sample etc.
(And while on this subject; I haven't got a clue on what LAME psycho model
actually computes; is this a stupid remark ? If lame only encodes up to
16kHz or input is of low F_sample; the psycho model can drop any masking
and other calcs following the fft, if above the appropriate freq; or is
this already in there ?)
regards,
Patrick.
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
