You start to hear the difference at extremes... The attached WAV has 2-level (digital) noise but now with very sparse positive pulses. (Low 97% of the time. High 3% of the time.) The long-term power spectral density is still that of "white noise", the same as the previous examples. As you reduce the probability of the pulses it starts to sound more like frying bacon, "popcorn noise", or eventually like "surface noise" on old vinyl records! -Richie, On 2026-03-03 10:31, Tom Wiltshire wrote: > +1 agree with Richie and Brian. > > The "noise colour" (frequency spectrum) and amplitude distribution are > independent. It kind-of blew my mind when I first discovered this, > since (in my head at least) you'd have thought that such a significant > difference in the signals would be audible, but it's really not. You > can make "white noise" with all of these different flavours, as Richie > has done, and it'll affect the output of an S+H sampling them, but not > the sound as audio. > >> On 3 Mar 2026, at 10:01, rburnett@richieburnett.co.uk wrote: >> >> If the ZIP attachment makes it through it contains five examples of >> "white noise" created with different amplitude distributions: >> >> 1. Gaussian distribution >> 2. Uniform (all amplitudes equally likely) >> 3. Triangular distribution >> 4. Two discrete levels (Digital PRBS) >> 5. Three discrete levels (Two digital sources added together) >> >> See if you can tell the difference by listening! >> >> -Richie, >> >> >> On 2026-03-03 07:21, brianw wrote: >>> The probability distribution of the values in a random sequence are >>> completely independent of the frequency spectrum of those values as a >>> signal... >> <noise.zip>________________________________________________________ >> This is the Synth-diy mailing list >> Submit email to: Synth-diy@synth-diy.org >> View archive at: https://synth-diy.org/pipermail/synth-diy/ >> Check your settings at: >> https://synth-diy.org/mailman/listinfo/synth-diy >> Selling or trading? Use marketplace@synth-diy.org
Message
Re: [sdiy] Various colors of noise.
2026-03-03 by rburnett@richieburnett.co.uk