Leap Motion

[knucracker]

Anyone have one?  Experiences?
I'm seeing too many reviews like this one...
http://arstechnica.com/gadgets/2013/07/hands-on-with-the-leap-motion-controller-cool-but-frustrating-as-hell/

[Grauniad]

I've decided not to complete my pre-order from last year.

[thepenguin]

It seems the problems are less with the device itself, and more that the app developers are no good.

(though you can't blame them, transitioning to a 3D model of software can't be easy from a 2D mindset...)

[knucracker]

Well, they never bothered to even respond to my application for beta... so no CW3 for them.   I'm sure I would have put all the others to shame... uh, hrmmm (tongue being removed from cheek).

The whole 'click in the air' thing worried me a long time ago when I first heard about this.  I wondered what they would do and they did the obvious thing, and it works about as well as you might expect... which is not very well.

My plan of use (that ship sailed, btw) was only to care about 1 point and to not care about depth.  I don't care how close you finger is to the screen.  I just want to replace the mouse pointer.  Since your hand is in the air, a key would replace clicking.  So a strategy game would work like it does now, except the mouse pointer would have been controlled by the position of a finger and 'contact'/'mouse down' would have been replaced by the space bar or another key.  Sounds primitive and like an under-use, and it might be.  After I never heard back, I never put any more thought into it.

But at this point the product looks like it might founder, so I'm not sure I will ever be bothered to try on the retail version. 

[pawel345]

But I needs to register depth to notice a "click" in the air. I think in a few years such input method might become more popular as the introduction of products like Google glas will require input without clicking on physical objects.

[zakari]

I've seen some internal demonstrations where I work.

The device is very reactive, feedback has almost no latency to it, but indeed the challenge comes from the interaction and how it's done in the software.

I think it's an interesting product with a lot of potential, we have yet to see some good commercial software that make use of it.

[knucracker]

"Air Clicking" is likely where a lot of the problems reside.  A human arm/hand without bracing is a blunt instrument with terrible precision.  Mice work so well because you stabilize with your wrist.  Touch screens (like iPads) work so well for similar reasons (pay close attention to how you hold and use a touch display).

'Minority Report' or 'Iron Man' interfaces work because the area of operation is massive and the things being interacted with are huge.  So give me a 60"+ display and a track area 75 sq ft (5'x5'x3') with 'buttons' that appear to be inches high... and you got a deal.  Sounds more like Kinect doesn't it?

The precision and latency of Leap has always been the impressive technical claim.  That part looks about as real and cool as you can get (unless the LOS is broken, then fingers seem to disappear).   I guess you could make a really awesome theremin with a Leap, but piano or guitar...  not so sure, and I think that further demonstrates the issues.

But, I'm always hopeful... so maybe somebody will lead the way, figure out a way to really make it go, and make me look like a Pakled...