I have been analyzing various wav files, and I am curious why some songs, when I analyze them using the plot spectrum, go all the way down to -90db, but other songs only go to -80db or maybe -82, etc. What does it mean?
No, what I mean…is why do some songs show -90db at the bottom and others don’t. I guess, I’m asking more along the lines of how is the dynamic range affecting the read out, but I am asking without an understanding of what dynamic range actually is, and I want to know what is it that I am looking out when I read the plotted graph.
The values shown in the graph are “normalized” such that a 0 dB sine tone will show as a peak of (approx.) 0 dB. Thus, a data point on the graph at -90 dB indicates that the average peak amplitude at that frequency is -90 dB.
Audacity automatically adjusts the y scale so that the maximum range is equal to the maximum data point.
Exactly how Audacity sets the minimum y scale is somewhat mysterious, but takes into account low data points and the Preference settings in “Edit > Preferences > Interface > Meter/Waveform dB range”.
It doesn’t say much.
-84 dB is very very very quiet. One could reasonably say “silent”.
-90 dB is a bit quieter than -84 dB.
I’ve never checked, but as a conjecture:
It may indicate the lowest data point in the -84 dB example is around -84 dB, whereas the lowest data point in the -90 dB example is around -90 dB.
To see if this is the case, try exporting from Plot Spectrum, then open the exported text file and look to see what the lowest dB value is.