Analyzing effects with pink noise

[ToneCarver]

Found something kind of fun and interesting last night that I thought I'd share with y'all. I set up a project with two tracks: one of pink noise, the other the same pink noise but phase reversed so the mix is pure silence. Added SPAN and other analyzers to the main out bus. Then I inserted selected effects (compressors, TapeBus, EQs) in one of the pink noise tracks and began to vary the settings. As the plugin changed the phase and/or amplitude of its track, the phase-cancellation effect was eliminated and non-silent output equivalent to the magnitude and shape of the change appeared on the main bus. I had a dificult time hearing the effect precisely (all pink noise), but SPAN and the other analyzers showed some interesting and sometimes suprising artifacts.

I could clearly see the difference between a "coloring" compressor (Marquis) and a more transparent one (Kjaerhus Classic). Sonitus compressor had a series of repeating humps in the span graph - some kind of phase warping perhaps ? I could see the different emphasis and curves of TapeBus and Cakewalk TapeSim plugs. Raising a peak band in EQ caused a raise in the output graph at the band frequency and a dip at some other lower frequency !? Is this pre-ringing ? spectral leakage ? operator error ?

Anyway, it was interesting to me and I figured it might be interesting to others too.

[JackDark]

This is very interesting actually. Could you post some screenshots of your sessions (that is if you do any future sessions)? That would be appreciated.

You might find this book to be of interest.

[smart]

cool experiment!

[bengeorge]

Yes, very very interesting.

[Atza]

interesting

[ToneCarver]

This is very interesting actually. Could you post some screenshots of your sessions (that is if you do any future sessions)? That would be appreciated.

I will try to post a few screenshots later. On reflection, using a plugin analyzer, like Christian Budde's, will give a more accurate understanding of the phase and amplitude changes because it provides measurements for them separately. The output from the nulled-pink-noise technique indicates phase and/or amplitude changes mixed together and it is difficult to tell whether it is the phase, the amplitude or both that are affected by the plugin.

You might find this book to be of interest.

Cool book. They have that at my local library - have checked it out and am just starting to read it. Thanks.

[ToneCarver]

One more thing ... running some eq-curves and filters over the nulled-pink-noise creates wind-type sounds that are a bit softer or less aggressive than just filtering the pink noise source directly. Just FYI if you are into generating wind and static type sounds.

[RTaylor]

One more thing ... running some eq-curves and filters over the nulled-pink-noise creates wind-type sounds that are a bit softer or less aggressive than just filtering the pink noise source directly. Just FYI if you are into generating wind and static type sounds.

Use something like a delay or reverb to break things up and get some "waves" going through it... then try noise reduction software. D/Noise and In-Tune can give you some pretty amazing results.