I'm using the Optitrack SDK with a TrackIR 4 Pro. (Why don't you release the TrackIR SDK?)
I've been studying the system in order to obtain precise callibration, and I've come across a variety of issues.
The docs (for the Vector class) state that the center of rotation for the clip is supposed to be near a person's head's center of rotation when the clip is worn on a cap. However, my analysis suggests that this is not so.
I've measured the center of rotation for the roll axis to be about 135 mm ABOVE the main bend in the clip, while the center for the pitch axis to be 80 mm above the same bend. (For yaw, it appears to be at the bend.)
In order to make these measurements, I attached a bracket to the clip and lengthened it as necessary in order to find the point about which the clip could be rotated without changing the resulting position.
In addition, I also noticed that the Z axis scale appears to be 1.5 times larger than the X/Y axes for position.
This is using the Optitrack_1.0.028.beta.03.exe SDK.
Does anyone else observe these discrepancies?