well I hope you get better results from it out of the box than I did. I actually decided to ignore my real work today and play with the SDK a bit. I am definitely making progress, but the more I progress the more I despair for the potential of this product.
I'm thinking about this concept of an array microphone now and wondering. If the microphone is gathing input and doing complex mathematics to isolate sound and form direction "beams" etc., what would happen if one of the four microphones in the array was a little bit "off". I'm wondering if this could be what is happening to me. For one thing, the angles that I am getting back are off. When I am sitting to the left of center it tells me that I am directly in front of the Kinect. This reminds me why I stopped using raid disk arrays many years ago... Maybe all this Kinect needs is the ability to recalibrate itself somehow.
In my opinion this beta SDK is extremely limited, missing important stuff, poorly documented etc. This is understandable for something very new. BUT, my fear is that Microsoft won't really follow up with the fixes, features, documentation, expanded language models etc. (i.e. anything other than english) because ultimately there won't be any money to be made from doing all that work. Hopefully I am wrong.
There are other problems, acoustic echo cancellation, which should prevent your music etc. from going back into the microphone does not seem to work at all. It is possible that in my case it is not working because I am using digital coax out to my 5.1 amp, but anyone using an htpc is probably going to be using that, toslink, or hdmi audio and having the same problem.
