[FFmpeg-devel] [PATCH] Move lpc utility code from flacenc.c to?lpc.{c|h}

[Michael Niedermayer]

On Sat, Aug 16, 2008 at 11:06:56AM +0530, Jai Menon wrote:
> Hi,
> 
> On Saturday 16 Aug 2008 10:44:58 am Jai Menon wrote:
> > Hi,
> >
> > On Friday 15 Aug 2008 4:44:35 pm Michael Niedermayer wrote:
> > > On Fri, Aug 15, 2008 at 03:07:12PM +0530, Jai Menon wrote:
> > > > Hi,
> > > >
> > > > On Friday 15 Aug 2008 2:58:15 pm Ramiro Polla wrote:
> > > > > Hi,
> > > > >
> > > > > On Fri, Aug 15, 2008 at 3:14 AM, Jai Menon <realityman at gmx.net> wrote:
> > > > > > Attached patch moves a few functions from the flac encoder so that
> > > > > > it can be shared by the alac encoder.
> > > > >
> > > > > For MLP I used the entire lpc_calc_coefs(). Is that not ok for alac?
> > > >
> > > > No, currently the order estimation does not require such extensive
> > > > code. It doesn't really help with compression or speed. A simpler
> > > > approach is used instead.
> > >
> > > please also use lpc_calc_coefs()
> > >
> > > > > And the functions should be prefixed with ff_, and lpc.c be made as
> > > > > an svn cp ; svn diff from flacenc.c
> > > >
> > > > ...and is there any particular reason/advantage for using 'svn cp'?
> > > > Just curious.
> > >
> > > "svn blame" cannot relate lines of code to authors&commits if svn cp
> > > hasnt been used
> >
> > I made the changes and created a new patch. I used svn cp this time and the
> > resulting patch was quite large (to put it mildly). I wanted to check if
> > this is normal before posting it. It comes in at around 94kb ;-)
> >
> 
> I'll just post it in case this is okay.

thats fine though a little better would be to post them both ways
that is the files and them diffed against (the rather large) ancestors.

That way i can see immedeatly what has changed AND how the final file
looks.

[...]

>   * Levinson-Durbin recursion.
>   * Produces LPC coefficients from autocorrelation data.
>   */
> -static void compute_lpc_coefs(const double *autoc, int max_order,
> -                              double lpc[][MAX_LPC_ORDER], double *ref)
> -{
> -   int i, j, i2;
> -   double r, err, tmp;
> -   double lpc_tmp[MAX_LPC_ORDER];
> +void ff_compute_lpc_coefs(const double *autoc, int max_order,
> +                          double lpc[][MAX_LPC_ORDER], double *ref);
>  
[..]
>  /**
>   * Quantize LPC coefficients
>   */
> -static void quantize_lpc_coefs(double *lpc_in, int order, int precision,
> -                               int32_t *lpc_out, int *shift)
> -{
> -    int i;
> -    double cmax, error;
> -    int32_t qmax;
> -    int sh;
> +void ff_quantize_lpc_coefs(double *lpc_in, int order, int precision,
> +                           int32_t *lpc_out, int *shift, int max_shift, int zero_shift);
>  

These functions should stay static, they contain specific implementations
that can be access through lpc_calc_coefs()


[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080816/85b25d14/attachment.pgp>