We are using a three-tiered, configurable architecture to minimize lag. We are also fully integrating
our system into the Lithtech development architecture to take advantage of Lithtech's capabilities.
The three major components of our architecture are the RealTime Server, the MasterMotion Server,
and the MasterMotion client.
view system diagrams
We hope to increase the framerate by optimizing the physics and visibility calculations in Lithtech.
Currently the Lithtech world runs at the following frames per second:
1 student ..............................30fps
1 student, 1 master .............22fps
1 student, 4 masters ...........15fps
3 students, 4 masters .........13fps
Lag is estimated to be 1/3 of a second,an unacceptably high number. Reducing lag will be critical
in the fourth cycle of our project.
A major obstacle in creating a usable training system is noisy marker data which causes jittery
movement of the virtual characters.
We are spending a large portion of our time experimenting with different filtering techniques that can be
applied to the data in real time.
The team is interested in how virtual reality can be used to learn physical movements.
Our investigation will focus on how various visual feedback techniques affect the accuracy
of participantsí motion over multiple trials and if any of the techniques only
possible in virtual reality outperform those used in a traditional learning situations.
| YIN | YANG | THE
ENTERTAINMENT TECHNOLOGY CENTER