Subject: RE: [linux-audio-dev] disk-based realtime sampler prototype relea sed ! please try it.
From: MOULET Xavier FTRD/DMR/ISS (xavier.moulet_AT_rd.francetelecom.fr)
Date: Thu Jul 13 2000 - 18:57:00 EEST
> > The resampling should be done always to downward so that
> disk reading
> > speed never increase. I will check simple resampling filter from
> > Dattorro's Effect Design paper but you could test different playing
> > speeds without interpolation or with linear interpolation.
> If we would
> > resample upward, a simple linear interpolation between
> nearby samples
> > would not work, I see.
I don't see why ?
>
Basically the DSP of resampling is the following : if you want to downsample
with no distorsion, you can do that (and it is the ONLY WAY to do that) with
two lowpass filters, and just removing samples/inserting 0's. Linear, Second
Order, third order (I don't remember which one is called cubic
interpolation), whatever , adds distorsion. But I admit it is WAY faster to
do that. Just, if you want to upsample to disk, please use the lengthy
method, it may be better (read perfect, NO distorsion at all if sampling
frequency if above . It may be interesting to see if linear + 1 lpf (XX
order) could be better than second order interpolation. I'll check.
This archive was generated by hypermail 2b28 : Thu Jul 13 2000 - 19:29:24 EEST