Re: Fwd: Re: Re[2]: [linux-audio-dev] peakfiles and EDL's

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: Re: Fwd: Re: Re[2]: [linux-audio-dev] peakfiles and EDL's
From: Steve Harris (S.W.Harris_AT_ecs.soton.ac.uk)
Date: Thu Mar 01 2001 - 02:33:15 EET


On Wed, Feb 28, 2001 at 01:57:56PM -0500, Paul Barton-Davis wrote:
> >At least I never felt any neccesarity for two seperate peak meters
> >showing the positive and negative peak values of the same signal
> >seperatly. Maybe you can you convince me, why I would need this an
> >for what purpose.
>
> i think thats because your peak meters are almost certainly computing
> RMS levels, even if they say "peak meter".
>
> in theory, this would actually be a good thing for a peakfile, but its
> a lot, lot more computationally expensive than min/max determination.

True, but is it actually much slower? Its only

while (foo) {
        sumsq += val * val;
}
rms = sqrt(sumsq / length);

Isn't it? There's no conditionals, so its pipeline friendly.

A useful feature might be to sum the amplitude of the freq bands using
those perceptual loudness curves, but that would be very CPU heavy.

- Steve


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Thu Mar 01 2001 - 09:14:32 EET