[FFmpeg-devel] dcadata.h: note on LFE FIRs

Kostya kostya.shishkov
Sun Aug 24 17:17:34 CEST 2008

On Sun, Aug 24, 2008 at 09:03:39PM +0600, Alexander E. Patrakov wrote:
> Hello,
> lfe_fir_128[] is a scaled up by a factor of 1.9999958 copy of lfe_fir_64[] 
> (but with less significant digits provided in the spec). Surely one wants to 
> round this factor to 2.0 (because LFE samples are just 8-bit signed PCM sound 
> samples) and use the common table (lfe_fir_64[]) for both cases of LFE 
> upsampling. The symmetry of this FIR table relative to its middle point is 
> already noted in the FIXME.
> No patch yet, because I should not be coding late at night.

Verified, seems so. Waiting for a patch but I can hack one myself too :).
> -- 
> Alexander E. Patrakov

More information about the ffmpeg-devel mailing list