On Wed, May 9, 2012 at 3:23 PM, Anton Khirnov <[email protected]> wrote:
>
> On Wed, 09 May 2012 10:09:29 -0400, Justin Ruggles <[email protected]> 
> wrote:
>> On 05/09/2012 03:13 AM, Anton Khirnov wrote:
>> > ---
>> >  libavutil/samplefmt.c |   17 +++++++++++++++++
>> >  libavutil/samplefmt.h |   13 +++++++++++++
>> >  2 files changed, 30 insertions(+)
>> [...]
>> > +/**
>> > + * Fill an audio buffer with a constant value.
>> > + *
>> > + * @param audio_data array of pointers to data planes
>> > + * @param byte to fill the data with
>> > + * @param offset offset in samples at which to start filling
>> > + * @param nb_samples number of samples to fill
>> > + * @param nb_channels number of audio channels
>> > + * @param sample_fmt audio sample format
>> > + */
>> > +int av_samples_set(uint8_t **audio_data, int b, int offset, int 
>> > nb_samples,
>> > +                   int nb_channels, enum AVSampleFormat sample_fmt);
>>
>> Is there really a need to set multiple samples to a single value other
>> than (logical) 0?  Since it works with different sample formats, I don't
>> see how giving it a byte value is even useful for anything else other
>> than 0 or 128.  Why not just have the function set samples to zero and
>> handle the 0x00 vs 0x80 internally based on sample format?
>>
>
> Yeah, I suppose you're right.

What about float audio?
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to