Re: [Audacity-nyquist] RE: Nyquist trouble saving sound file

[Eivind L. R. <el...@br...>]

> Furthermore on the prompt:
>
>> (peak (aref (my-snd) 1) 1000000000000000000)
> [ gc: total 17640, 1320 free; samples 102KB, 100KB free ]
> 0.39938
>> (peak (aref (my-slope-snd) 1) 1000000000000000000)
> [ gc: total 17640, 1220 free; samples 99KB, 96KB free ]
> 14768.1
>
> So there must be some scaling going on here, or different representatio=
n
> of the sampledata.
>
> Thanks again!
>
> Eivind

Ah, got it:

> (setq array1 (snd-fetch-array (aref (my-slope-snd) 0) 10 0))
#(3.89099 -10.8032 4.18854 -1.09863 4.55475 0.19455 12.2108 -6.63757
-11.5356 -5.0354)
> (setq array2 (my-slope (aref (my-snd) 0) 10))
#(4.05312e-05 -0.000112534 4.36306e-05 -1.14441e-05 4.74453e-05
2.02656e-06 0.000127196 -6.91414e-05 -0.000120163 -5.24521e-\
05)
> (dotimes (n 10) (print (/ (aref array1 n) (aref array2 n))))
96000
96000
96000
96000
96000
96000
96000
96000
96000
96000
NIL
>

Where my-slope is a function which just subtracts two neighbour-samples.

Is it guaranteed that the values are not out of bounds (wrap-around or th=
e
like)?  I am trying to smoke out some glitches in damaged wav-files, so
the deltas can be pretty steep.

Eivind