NatNet, VRPN, TrackD, and Plugins
Post Reply
Posts: 6
Joined: Mon May 05, 2008 4:35 am


Post by ajaleo » Mon May 05, 2008 4:37 am

Hi all,

I'm programming a Java application that retrieve real time data from rigit body tracking tool throught natnet. The point is that I'm trying to interpret the byte array I get and I've found a lot of problems because it is not compatible with the syntax I have (NatNet Bitstream Syntax 1.4.0). Probably I'm missing something so please, is possible to access the source code of the class NatNetServer so I can contrast with my implementation.

Thanks a lot

Posts: 1139
Joined: Thu Jan 30, 2003 5:00 am
Location: Corvallis, Oregon

Re: NatNetBitStream

Post by Birch » Mon May 05, 2008 1:01 pm


Please email us directly ( with a reference to this forum post, we will work with you to try and resolve the issue you are seeing.

Posts: 4
Joined: Sat Nov 08, 2008 6:43 pm

Re: NatNetBitStream

Post by bgamp » Thu Nov 13, 2008 9:24 am

Hi, I am also having a similar problem. We finally got the java program to read the multicast udp stream, but the data received doesn't match the interface control document provided. Is there an update to this doc? I am using Doc 1.4(draft).

Specifically, the Mocap Data Packet doc says if there is no models, to skip over the entire packet.... ok, so that's not the case. next is the number of markers - correct.

I get 4 markers, exactly what I have in the system, but I am getting 52 bytes instead of the 48 xyz bytes expected.

Next I get the number of rigid bodies, 1 - correct. ID for rigid body - correct. The RB x,y,z & quaternions "seem" correct.

Markers in RB - 3 - correct.

But then it seems to give the positions of the markers again, but they are different from the original positions of the markers. Are these relative to the centroid of the RB?

Above, I wrote the quaternions "seem" correct because while the rigid body isn't moving, shouldn't they all go do zero?



Post Reply