[Lizzie] from LustLab sent in her Ball of Dub that turns a few accelerometer and a digital audio workstation and turns everything into an aural experience of wubs and dubs. The Ball of Dub can turn just about anything into dubstep, and does so with a fairly interesting user interface.
There isn’t a build log for the Ball of Dub, but the folks at LustLab did send in a basic overview of her project. Inside the ball, there’s a Razor IMU from Sparkfun that is attached to the ever-popular XBee wireless transceiver. A tiny program on an Arduino calibrates the gyroscope and accelerometer and sends that data to the DAW at 50Hz.
The host computer is running Renoise, a very popular tracker that can accept MIDI and OSC input. A Processing app parses the ball spin, free fall and impact, averages them over a period of time, and pipes that into the OSC input of Renoise. In [Lizzie]’s video, the ball spin is sent to a low-pass filter on the baseline track, and the average impact is applied to the vocal track.
This isn’t the first time we’ve seen some fairly strange ways to modulate wub; we saw real instruments covering Skrillex earlier this month. The Ball of Dub wins in the simplicity department, though.
That sounds just like a regular song playing, I can’t think of which one it is, but I’m going to be skeptical on this one till more is posted.
I love the sound, but the “EXPERIMENT 002 – Twitter Silly” video released on the same day, does make it look, a bit, like it could be a fake (maybe). I’m not saying that the technology behind it is crazy complicated or anything (by today’s standards), but two videos on different areas of sensor feedback and control does make it look a little bit shady. Still I think that it was just bad timing of releasing the videos.
Dunno about the other video, but this Ball of Dub is highly suspicious. Anyone taking that much time to build a complicated instrument is going to do WAY more video of it in action. The real-time averaged data doesn’t really seem to track the ball movements in my opinion. If you are reading Lizzy, please prove us wrong.
Occam’s razor suggests that it doesn’t take much talent to make something people will label as ‘dubstep’ …
Hi Brett, please see below video, hope this clears it up for you.
Hi Lizzie! Thanks for posting another video. There is still some sync issues, but I can see it’s real now. Can you juggle?? Make 2 more and make some sweet music!
The ball movement does not correspond to the WobWob sounds at all, but the volume of what seems to a per-recorded WobWob track does match the ball movement.
So I call this as being misleading in that the ball does not make Wobwob, just changes the volume of a certain track.
Seems legit to me, not that I am an expert but in the Experiment 1 video you can sort of see the input data above the user video. Not that I watched closely but it does look like its tracking feedback that coincides with the ball. Additionally it does say the feedback is averaged out so it would make sense that the track effects are not completely in sync with the motion. Just rising and falling with the acceleration.
Interesting project.
It seems to me that the same results could be achieved with a simple application of pattern recognition, applied to a webcam. However, that method might have too low of a sampling rate for use in music. That could be interpolated, though.
I don’t see any evidence of accelerometers or electronics in the sources given. Judging from her video “Twitter Silly”, she clearly know how to use pattern recognition. That could explain the color and texture of the “ball of dub”. Just maybe, this project is actually based on pattern recognition.
I don’t (well can’t) dance but I would love to see some accelerometer play done in this fashion but put in actual shoes (I guess dubstep shoes means something else) or some other article of clothing and see real dancers make something of it. I think I’m going to suggest it to some Koreans ans we’ll see it in a Kia Soul commercial real soon.
it seemed more like the ball of dub was controlling the volume, but not the make-up of the sounds.
And what kind of modulation it is? AM or FM?
The movement of the ball and the audio never synchronize effectively. It’s a cute concept, but it’s fake.
I stand corrected; in light of the new video, it makes quite a bit more sense.
I don’t know why there’s so much skepticism regarding this video. There are examples in Max/MSP that do simple object tracking via color that could easily be modified to send OSC to Renoise.
It looks like they are just taking the overall acceleration of the object and using that to control an LP filter/gainer combo on the bass track of an pattern looping in Renoise. It’s not rocket surgery.
Hi guys, I see there has been some skepticism about the ball, I made another video to better deonstrate the data you can get from the ball. Detecting freefall, spin magnitude, impact magnitude which I didn’t demonstrate in the dub example, I hope this clear it up!
http://vimeo.com/31366215
Lizzie @LUSTlab
hey lizzie! CRAZY SHIT! Keep pushing it girl! *HugZ*
Thanks, I’m now a believer. Was a bit of a sceptic before. I could even noticed that the ball isn’t balanced, which wasn’t obvious from the previous video.
Expert here.. well at least expert enough to know my scene, and thats Renoise. NON-FAKE. 100% MAD RESPECT to her for the vid!
Haha Dubstep made easy,I love the simplicity of the idea, thanks again hackaday
How about putting the same tech into something wearable.. I forsee a a sexy girl shaking her ass making sweet dubstep in the process.
Semi related to the video: what is it that causes sooo many youtube and other flash vids to get the audio out of sync with the video. It’s caused a ton of legit videos to get commenters calling them fake because the sound ends up looking badly dubbed thanks to flash’s screwup.
Has anyone made this? I’m thinking about making it and doing this type stuff is still new to me.