News for hackers. Stuff that matters.
Re:Skeletal Tracking Beta by bravura on Wednesday May 28, 2014 @01:00PM (#7811402)
I hacked the Leap controller at the SF Music Hack day. My team and I built an app called leapJ: multiplayer DJ'ing with your body. (https://www.hackerleague.org/hackathons/music-hack-day-san-f...)

Leap is a great little device. My main complaint is that the API is too low-level. We would just fire events roughly every 20ms, in a markovian (stateless) way. For this reason, the instrument control was somewhat janky, because we didn't have the time to de-jitter hand paths / infer a tracking spline.

My wishlist for leap is:

* Higher level APIs, that apply the sort of sophisticated filtering and tracking that many people are liable to roll themselves.

* Multiple leap devices per computer. Needing a separate computer per leap is a big barrier to working with several leaps at once.

* ARM support. I'd love to have a RaspberryPI for each leap. Then they would be truly mobile. For right now, I think I'm going to have to get a mini ITX motherboard,

Re:Skeletal Tracking Beta by sitkack on Wednesday May 28, 2014 @01:38PM (#7811664)
You need a smoothing library. I welcome our low level access, you can always pile what you need on top, the converse is not possible.

What would be nice is some intelligence pushed down into the Leap, like the ability to run Lua or JVM code directly so that realtime multidevice is possible.

Re:Skeletal Tracking Beta by scott_karana on Wednesday May 28, 2014 @02:19PM (#7811900)
I think he meant augmenting your API with higher-level stuff, but not removing any of the low level access. :-)