[FFmpeg-devel] [PATCH 3/3] Allow either float or int16_t input and output in ff_iir_filter().

[Kostya]

On Thu, Jan 20, 2011 at 10:30:41AM -0500, Justin Ruggles wrote:
[...]
> 
> Currently the code in iirfilter.c is not completely generic.  It assumes
> that the output coefficients can be normalized to integers.  This is
> fine for Butterworth and Biquad (which are what I need for ac3enc), so I
> don't need to modify it, but it probably should be changed if we want to
> use different filter types in the future.

I'm rather pleased you say so - it was actually intended to be simple
lowpass Butterworth filter but got more generic interface to be
future-proof and now it finally pays off.
 
> -Justin