Hi,
On Fri, Mar 25, 2016 at 3:55 PM, Ganesh Ajjanagadde
wrote:
> On Fri, Mar 25, 2016 at 12:11 PM, Paul B Mahol wrote:
> > On 3/25/16, Ganesh Ajjanagadde wrote:
> >> On Fri, Mar 25, 2016 at 9:36 AM, Nicolas George
> wrote:
> >>> Le sextidi 6 germinal, an CCXXIV, Ganesh Ajjanagadde a ecrit :
>
On 3/25/2016 4:55 PM, Ganesh Ajjanagadde wrote:
> If anyone cares here, I do not know why we can't use inline asm or
> intrinsics
With Inline asm and intrinsics you're 100% dependent on the compiler
doing the right thing. The latter more so than the former.
So a function compiled with gcc 4.9 migh
On Fri, Mar 25, 2016 at 12:11 PM, Paul B Mahol wrote:
> On 3/25/16, Ganesh Ajjanagadde wrote:
>> On Fri, Mar 25, 2016 at 9:36 AM, Nicolas George wrote:
>>> Le sextidi 6 germinal, an CCXXIV, Ganesh Ajjanagadde a ecrit :
Depends on if it is small or not. Yes, in many codecs, FFT's are short
>
On 3/25/16, Ganesh Ajjanagadde wrote:
> On Fri, Mar 25, 2016 at 9:36 AM, Nicolas George wrote:
>> Le sextidi 6 germinal, an CCXXIV, Ganesh Ajjanagadde a ecrit :
>>> Depends on if it is small or not. Yes, in many codecs, FFT's are short
>>> length ones, e.g 512. However, on long lengths, e.g 8192+
On Fri, Mar 25, 2016 at 09:29:56AM -0700, Ganesh Ajjanagadde wrote:
> On Fri, Mar 25, 2016 at 8:23 AM, Hendrik Leppkes wrote:
> > If performance is the only reason one might want an external library,
> > then its a much better idea to try to optimize the code we have, that
> > way more people auto
On 3/25/2016 12:25 PM, Hendrik Leppkes wrote:
> Also for the record, most of the external libraries that duplicate
> internal functionality existed before the internal functionality was
> improved/implemented, or the current internal variant is still not
> quite on the level.
> Sometimes its just c
On Fri, Mar 25, 2016 at 9:36 AM, Nicolas George wrote:
> Le sextidi 6 germinal, an CCXXIV, Ganesh Ajjanagadde a écrit :
>> Depends on if it is small or not. Yes, in many codecs, FFT's are short
>> length ones, e.g 512. However, on long lengths, e.g 8192+, as seen
>> from the benches, there are som
Le sextidi 6 germinal, an CCXXIV, Ganesh Ajjanagadde a écrit :
> Depends on if it is small or not. Yes, in many codecs, FFT's are short
> length ones, e.g 512. However, on long lengths, e.g 8192+, as seen
> from the benches, there are sometimes 2x variations at the moment.
And how much of the actu
On Fri, Mar 25, 2016 at 8:23 AM, Hendrik Leppkes wrote:
> On Fri, Mar 25, 2016 at 3:34 PM, Ganesh Ajjanagadde
> wrote:
>> On Fri, Mar 25, 2016 at 12:35 AM, Clément Bœsch wrote:
>>> On Thu, Mar 24, 2016 at 05:50:48PM -0700, Ganesh Ajjanagadde wrote:
Ganesh Ajjanagadde (4):
configure:
On Fri, Mar 25, 2016 at 4:23 PM, Hendrik Leppkes wrote:
> On Fri, Mar 25, 2016 at 3:34 PM, Ganesh Ajjanagadde
> wrote:
>> On Fri, Mar 25, 2016 at 12:35 AM, Clément Bœsch wrote:
>>> On Thu, Mar 24, 2016 at 05:50:48PM -0700, Ganesh Ajjanagadde wrote:
Ganesh Ajjanagadde (4):
configure:
On Fri, Mar 25, 2016 at 3:34 PM, Ganesh Ajjanagadde wrote:
> On Fri, Mar 25, 2016 at 12:35 AM, Clément Bœsch wrote:
>> On Thu, Mar 24, 2016 at 05:50:48PM -0700, Ganesh Ajjanagadde wrote:
>>> Ganesh Ajjanagadde (4):
>>> configure: add fftw3 detection
>>> lavc/fftw: add initial fftw wrapper
>>>
On 25 March 2016 at 14:34, Ganesh Ajjanagadde wrote:
>
> like libopus based decoding instead of native FFmpeg decoding,
>
The decoders differ in output on broken/invalid files made using libopusenc
(at extremely low bitrates, way below the 5k minimum the spec says).
> the multiple AAC encoders/
On Fri, Mar 25, 2016 at 12:35 AM, Clément Bœsch wrote:
> On Thu, Mar 24, 2016 at 05:50:48PM -0700, Ganesh Ajjanagadde wrote:
>> Ganesh Ajjanagadde (4):
>> configure: add fftw3 detection
>> lavc/fftw: add initial fftw wrapper
>> lavc/fft-test: add FFTW3 tests
>> lavc/fft-test: update benchm
On Thu, Mar 24, 2016 at 05:50:48PM -0700, Ganesh Ajjanagadde wrote:
> Ganesh Ajjanagadde (4):
> configure: add fftw3 detection
> lavc/fftw: add initial fftw wrapper
> lavc/fft-test: add FFTW3 tests
> lavc/fft-test: update benchmark code
>
Why?
Using an external library for such an essent
Ganesh Ajjanagadde (4):
configure: add fftw3 detection
lavc/fftw: add initial fftw wrapper
lavc/fft-test: add FFTW3 tests
lavc/fft-test: update benchmark code
configure | 4 +++
libavcodec/Makefile | 1 +
libavcodec/fft-test.c | 82 +
15 matches
Mail list logo
