A 3 Year Old's Dream

Before Bieber, there was Hamsterdance – what in 1998 counted for viral on the Internet. In today’s ever-geekier times, even obscure sound software can go viral. Photo (CC-BY) twodolla / Wendy.

In this age of the 24-hour news cycle and instant publication of stories, sometimes it’s good to slow down and wait. And thus, while for whatever reason I didn’t get around to mentioning the extreme audio stretching of a certain Justin Bieber track (see Synthtopia), I can’t let an ethereal, ambient reworking of 1998 Internet hit Hamsterdance go unnoticed, here shared on SoundCloud.

As it happens, while folks have taken notice of one of the tools, this strange Web meme opens a door on a lot of free and open source sound software goodness.

HamsterDance Stretched to 800% by Stefan Anion

Thanks to creator Stefan Anion aka Stefan Weise for sending his work our way.

The technique is catching on; now we get to play the game of spotting which producers slip this software into tracks coming out in the next 12-24 months. You can thank free and open source software: Paul Nasca aka Nasca Octavian PAUL has released two tools that use the magic of FFTs, a mathematical process by which it’s possible to transform time and frequency information quickly. HyperMammut (another cool tool that does huge, single-window FFTs on sound and image) and Paul’s Extreme Stretch (the tool used on Justin’s track) are GPLv2-licensed. Lifehacker even did their own how-to on the topic.

http://hypermammut.sourceforge.net/paulstretch/ Documentation and software page
http://sourceforge.net/projects/hypermammut/ SourceForge home of both Extreme Stretch and HyperMammut (Windows binary, Linux source)
Paulstretch Mac Port for PPC/Intel – your mileage may vary, as this is an unofficial port; let us know how it goes for you in comments, Mac users

Note that Paul is also the author of the terrific ZynAddSubFX open-source soft synth. (Look closely, closely at the title and guess at the feature set and synthesis technique.)

See also the software Mammut on which HyperMammut was based; the former is available for Mac, Windows, and Linux. Via Anders in comments:

I would like to mention that hypermammut is based on NOTAM ´s (Norwegian Center for technology in music and the arts) Mammut programmed by Øivind Hammer and Kjetil Matheussen

Download: http://www.notam02.no/index.php?/eng/Technology-and-text/Software/Mammut

I’d love to see some of the basic notions of this technique adapted to similar real-time patches. (Pd, ho!) Ideas, sonic wizards?

After I was first introduced to the program in the early 90s at a program at Oberlin’s TIMARA electronic tech center, I spent much of that decade addicted to convolution in Tom Erbe’s strange and wonderful SoundHack, another free sound-shaping tool for the Mac. I watched as the convolution process evolved from something I let run overnight (literally) to something that took an hour or so to something that became near-instantaneous, corresponding to the extraordinary forward march of processing speed.

It’s clear that some of these more obscure processes are going mainstream. It’s likewise evident that, as this audio has gone viral, SoundCloud really has become the “Flickr of audio” I predicted it might when it was launched back in 2008. So, I guess we’ll have to go and find some new sound design secret. We’d better just not tell the rest of you. I know I’m good at keeping lots of secr… doh!