No, no, not that kind of Alias. I'm actually referring to audio aliasing. If you've ever messed around with a bit crusher before, you've heard aliasing in action. Bit crushers simply recreate the lower bit depth and sample rates of old samplers and the artifacts the low specs of these machines would impart to audio. (Which we used to try to avoid like the plague, but now, predictably, has become trendy... listen to any current French house if you don't believe me...)
In digital audio, there is something called the Nyquist Theorem. It basically says that a digital audio recording can faithfully reproduce the frequency content of a signal equal to half of the sampling rate. So a sampling rate of 44.1k can reproduce frequencies up to 22k faithfully. Aliasing is what happens when you try to reproduce frequencies that are more than half the sample rate.
When you do this, there are not enough sample cycles to accurately recreate the sound's waveform and this lo-res "picture" ends up sounding like a lower frequency that was never in the original signal. Depending on the frequency content of the original sound, aliasing can sound crunchy, gritty, ringing, atonal, and any combination thereof.
Since aliasing mostly occurs from the inability to accurately reproduce higher frequencies, you can imagine that if you send a sound that is made up mostly of high frequencies, the aliasing will be more extreme - even to the point where all you get are the aliases. In fact, with the right sounds, you'll get all manner of digital shrieks and squawks. Try loading up a glockenspiel or high-pitched bell sound and running it through your favorite bit-crusher with the sample rate set extremely low (5-10k). The lower sample rate makes it near impossible to recreate these high pitched sounds and what comes out bears scant resemblance to the original signal. You may find you even get vastly different sounds from each different key.
If you have an old hardware sampler that allows you to sample at an extremely low rate (I keep my old Ensoniq EPS-16+ around for this exact purpose), you can get great, gritty soundscapes
by taking a long sample of a sound at a very low rate, and then playing sustaining notes much lower than the original root key of the sample so they stretch waaaaaaaay out. Instant industrial ambience! And with that, I will stop pretending you ever made it past the picture of Jennifer Garner and actually read this article.
Posts
8 comments:
I was looking fwd to an audio example. Maybe you can add one using your Ensoniq EPS-16+
Perhaps at a later date...I've got a month to finish my album, so I don't have the time at the moment.
There's even a synth which uses this as its sound generation method: Mr. Alias!
And it's FREE: http://www.thepiz.org/mralias2/
Thanks for the link Anu! I'll try this one :)
That is some great information, does this work the same with vocals?
reminds me of my first synth which was a kawai rompler. some of the weirdest sounds would come out of it when you transposed some of the sounds all the way up or down.
Really worthwhile data, much thanks for your article.
I canister answer that's a accomplished awareness as broad practicioners impose buy modalert afterwards buy provigil in lieu of long call advantage.
Post a Comment