Signed-off-by: Marton Balint
---
tests/Makefile |1 +
tests/extended.ffconcat| 114 +
tests/fate-run.sh | 20 +
tests/fate/concatdec.mak | 21 +
tests/ref/fate/conca
On Tue, Nov 3, 2015 at 4:52 PM, Ganesh Ajjanagadde
wrote:
> On Tue, Nov 3, 2015 at 4:49 PM, Ganesh Ajjanagadde
> wrote:
>> [...]
>
> Same as v1, except with appropriate Boost license included inline and
> commit message rework to include the perf numbers.
Carl pointed out that the copyright mess
On Tue, Nov 3, 2015 at 7:56 PM, Rostislav Pehlivanov
wrote:
> The binary increase is 16 kbytes per file * 2 (since it's a template for
> the fixed and float decoders) = 32 kbytes. This is not very significant at
> all, even for the most memory and storage-space constrained (with an MMU)
> device.
Signed-off-by: Paul B Mahol
---
libavcodec/Makefile | 2 ++
libavcodec/allcodecs.c | 2 ++
libavcodec/avcodec.h| 2 ++
libavcodec/codec_desc.c | 14
libavcodec/wmaprodec.c | 94 ++---
libavformat/riff.c | 2 ++
libavformat/ri
This improves accuracy for the bessel function at large arguments, and this in
turn
should improve the quality of the Kaiser window. It also improves the
performance of the bessel function and hence build_filter by ~ 20%.
Details are given below.
Algorithm: taken from the Boost project, who have
On Wed, Nov 04, 2015 at 01:11:45AM -0500, Ganesh Ajjanagadde wrote:
> This speeds up build_filter by ~ 50%. This gain should be pretty
> consistent across all architectures and platforms.
>
> Essentially, this relies on a observation that the filters have some
> even/odd symmetry that may be explo
Hi Paul!
On 2015-11-04 14:18 +0100, Paul B Mahol wrote:
[...]
> diff --git a/libavcodec/wmaprodec.c b/libavcodec/wmaprodec.c
> index 19dc335..7471961 100644
> --- a/libavcodec/wmaprodec.c
> +++ b/libavcodec/wmaprodec.c
> @@ -86,6 +86,8 @@
> * subframe in order to reconstruct the output samples.
On Fri, 23 Oct 2015 18:08:00 +0200
Michael Niedermayer wrote:
> From: Michael Niedermayer
>
> Signed-off-by: Michael Niedermayer
> ---
> libswresample/aarch64/audio_convert_init.c |8 -
> libswresample/arm/audio_convert_init.c |8 -
> libswresample/audioconvert.c
Le quartidi 14 brumaire, an CCXXIV, Marton Balint a écrit :
> + run ffprobe${PROGSUF} -show_streams -show_packets -v 0 -fflags
> keepside -safe 0 $extra_args $concatfile > $packetfile
Maybe -of compact, that makes the reference files quite smaller. I had not
realized they would be that big,
On Wed, Nov 4, 2015 at 9:44 AM, Michael Niedermayer
wrote:
> On Wed, Nov 04, 2015 at 01:11:45AM -0500, Ganesh Ajjanagadde wrote:
>> This speeds up build_filter by ~ 50%. This gain should be pretty
>> consistent across all architectures and platforms.
>>
>> Essentially, this relies on a observation
On Fri, Oct 23, 2015 at 07:18:06PM +0200, wm4 wrote:
> On Fri, 23 Oct 2015 18:08:00 +0200
> Michael Niedermayer wrote:
>
> > From: Michael Niedermayer
> >
> > Signed-off-by: Michael Niedermayer
> > ---
> > libswresample/aarch64/audio_convert_init.c |8 -
> > libswresample/arm/audio_co
On Wed, 4 Nov 2015 16:15:36 +0100
Michael Niedermayer wrote:
> On Fri, Oct 23, 2015 at 07:18:06PM +0200, wm4 wrote:
> > On Fri, 23 Oct 2015 18:08:00 +0200
> > Michael Niedermayer wrote:
> >
> > > From: Michael Niedermayer
> > >
> > > Signed-off-by: Michael Niedermayer
> > > ---
> > > libs
Signed-off-by: Kyle Swanson
---
Changelog| 1 +
doc/filters.texi | 36 +++
libavfilter/Makefile | 1 +
libavfilter/allfilters.c | 1 +
libavfilter/asrc_anoisesrc.c | 222 +++
libavfilter/version.h
Thanks for the updated patch, see comments below.
Le quartidi 14 brumaire, an CCXXIV, Kyle Swanson a écrit :
> Signed-off-by: Kyle Swanson
> ---
> Changelog| 1 +
> doc/filters.texi | 36 +++
> libavfilter/Makefile | 1 +
> libavfilter/allfilters.
On 11/4/2015 3:19 PM, Kyle Swanson wrote:
> Signed-off-by: Kyle Swanson
> diff --git a/libavfilter/asrc_anoisesrc.c b/libavfilter/asrc_anoisesrc.c
> new file mode 100644
> index 000..d008d67
> --- /dev/null
> +++ b/libavfilter/asrc_anoisesrc.c
> @@ -0,0 +1,222 @@
> +/*
> + * Copyright (c) 2015
Hi,
On Wed, Nov 4, 2015 at 12:30 PM, Nicolas George wrote:
> Thanks for the updated patch, see comments below.
>
> Le quartidi 14 brumaire, an CCXXIV, Kyle Swanson a écrit :
>> Signed-off-by: Kyle Swanson
>> ---
>> Changelog| 1 +
>> doc/filters.texi | 36
On Wed, Nov 4, 2015 at 1:10 PM, James Almer wrote:
> On 11/4/2015 3:19 PM, Kyle Swanson wrote:
>> Signed-off-by: Kyle Swanson
>> diff --git a/libavfilter/asrc_anoisesrc.c b/libavfilter/asrc_anoisesrc.c
>> new file mode 100644
>> index 000..d008d67
>> --- /dev/null
>> +++ b/libavfilter/asrc_an
Le quartidi 14 brumaire, an CCXXIV, Kyle Swanson a écrit :
> I can change this, but most filters I've seen have used floating point
> sample formats. Anyone else have any opinions on this?
Floating point is easier, that explains why a lot of code use it. But it is
less efficient, and most importan
On Wed, Nov 4, 2015 at 2:18 PM, Kyle Swanson wrote:
> Hi,
>
> On Wed, Nov 4, 2015 at 12:30 PM, Nicolas George wrote:
>> Thanks for the updated patch, see comments below.
>>
>> Le quartidi 14 brumaire, an CCXXIV, Kyle Swanson a écrit :
>>> Signed-off-by: Kyle Swanson
>>> ---
>>> Changelog
On 11/4/15, Kyle Swanson wrote:
> Hi,
>
> On Wed, Nov 4, 2015 at 12:30 PM, Nicolas George wrote:
>> Thanks for the updated patch, see comments below.
>>
>> Le quartidi 14 brumaire, an CCXXIV, Kyle Swanson a ecrit :
>>> Signed-off-by: Kyle Swanson
>>> ---
>>> Changelog| 1 +
Signed-off-by: Marton Balint
---
tests/Makefile | 1 +
tests/extended.ffconcat| 114
tests/fate-run.sh | 20 +++
tests/fate/concatdec.mak | 21 +++
tests
> +static double pink_filter(double white, double *buf) {
> +double pink;
> +
> +/* http://www.musicdsp.org/files/pink.txt */
> +buf[0] = 0.99886 * buf[0] + white * 0.0555179;
> +buf[1] = 0.99332 * buf[1] + white * 0.0750759;
> +buf[2] = 0.96900 * buf[2] + white * 0.1538520;
> +
On Wed, Nov 4, 2015 at 4:51 PM, Kyle Swanson wrote:
>> +static double pink_filter(double white, double *buf) {
>> +double pink;
>> +
>> +/* http://www.musicdsp.org/files/pink.txt */
>> +buf[0] = 0.99886 * buf[0] + white * 0.0555179;
>> +buf[1] = 0.99332 * buf[1] + white * 0.0750759
On Wed, Nov 4, 2015 at 10:07 AM, Ganesh Ajjanagadde wrote:
> On Wed, Nov 4, 2015 at 9:44 AM, Michael Niedermayer
> wrote:
>> On Wed, Nov 04, 2015 at 01:11:45AM -0500, Ganesh Ajjanagadde wrote:
>>> This speeds up build_filter by ~ 50%. This gain should be pretty
>>> consistent across all architect
> Alternatively, it should not be hard to find on the web. I personally
> am ok with your patch with floats, leaving fixed point for future
> work, but Paul and Nicolas have greater expertise and knowledge of
> this stuff and their opinion is definitely more important.
FWIW, SoX and Audacity both
On 11/4/15, Kyle Swanson wrote:
>> Alternatively, it should not be hard to find on the web. I personally
>> am ok with your patch with floats, leaving fixed point for future
>> work, but Paul and Nicolas have greater expertise and knowledge of
>> this stuff and their opinion is definitely more imp
"Fast seek" uses linear interpolation to find the position of the
requested seek time. For CBR this is more direct than using the
mp3 TOC and bypassing the TOC avoids problems when the TOC is
corrupted (e.g. https://crbug.com/545914).
For VBR, fast seek is not precise, so continue to prefer the TO
This uses the trigonometric double and triple angle formulae to avoid
repeated (expensive) evaluation of libc's cos().
Sample benchmark (x86-64, Haswell, GNU/Linux)
old:
1104466600 decicycles in build_filter(loop 1000), 256 runs, 0 skips
1096765286 decicycles in build_filter(loop 1000),
This changes the default to not clip, and thus matches libavresample as
well.
Documentation has been accordingly updated (with some typo fixes).
API users have complained about this, see e.g
https://github.com/mpv-player/mpv/commit/bf69edb1c2bac2ac11343d8e086f0540a0e4b5df.
Signed-off-by: Ganesh
29 matches
Mail list logo
