Razer Hydra Backpack

There is a beta of the Sixense SDK that allows for base rotation. Check out the link below:
http://sixense.com/forum/vbulletin/showthread.php?4481-Beta-Sixense-SDK-allows-for-base-rotation-offset

I haven't tried to actually access this function, though. Which engine are you using?

Awesome thanks!
I'm using UDK

I haven't messed with UDK so I'm unfamiliar with their Hydra integration. Might be as easy as swapping out a .dll and some header files. Let us know how the experiment goes :)

Could you post some info on your setup? I'm currently looking to get a setup to go around showing off the oculus. (also this will work great setup with webcam on the Rift, adding crazy effects to the webcam stream will make for awesome augmented reality)

I have a question. I've been looking at a full motion simulator chair, and how to make it work with the Oculus Rift.
I've been looking at the Atomic models of motion simulator chairs www.joyridesimulators.com
I called the electric engineer who made it, and made him aware of the following problem.. and he agreed it was a challenge that needs a solution.

Here's the problem:
In case you have a rotating "ground surface", such as in a motion simulator, would it be possible to "offset" Oculus Rift motion/rotation with rotation data from the motion platform? If you are playing a flight or space sim, you don't want the motion platform to affect your Rift head rotation. You want the head rotation to be "rooted" to the motion platform somehow. And since all the Rift sensors are primarily focused toward the Earth, I think the rift might be easily confused by the motion platform. One solution is: Offset all Rift motion by the opposite of the motion from the motion platform - but you need the correct "arm length" so that it's based on where your "head" position is at in neutral.

I made this flowchart to quickly sketch my idea out. The question is: Is it possible to make such a real time or dynamic offset override to the Oculus Rift head tracker sensors? What would be the total latency cost to such a proposal? Any changes you think should be made to the flowchart? Thanks

http://24.media.tumblr.com/efba5f8d0f87d232b22dfca2d3a3e9c4/tumblr_mqrv303gIb1sdtc3vo1_1280.png

An optional paired sensor might be a good alternative. Use it in the same sense that the noise cancellation is done? This is just a theory, but it may be possible to just have a duplicate of the rift sensors as something you can put to act as a "control" of sorts for environmental inputs, cancel out what the rift is detecting with what that environmental one is.

My guess would be issues with syncing the data as a major drawback, not really sure.

One thing is certain, external motion from simulators or rides or whatnot will mess up the rift.

Other possibilities would be secondary sensors mounted on the device to also show position/rotation of the rift to help offset what the frame is doing.

"Felix12g" wrote:
An optional paired sensor might be a good alternative. Use it in the same sense that the noise cancellation is done? This is just a theory, but it may be possible to just have a duplicate of the rift sensors as something you can put to act as a "control" of sorts for environmental inputs, cancel out what the rift is detecting with what that environmental one is.

My guess would be issues with syncing the data as a major drawback, not really sure.

One thing is certain, external motion from simulators or rides or whatnot will mess up the rift.

Other possibilities would be secondary sensors mounted on the device to also show position/rotation of the rift to help offset what the frame is doing.

I saw some people use a Razer Hydra as a positional control of the Oculus head (by attaching it to the head).
Wouldn't it be better (less latency and sync issues) if you could "override" the Oculus Rift's sensors with a Razer Hydra type setup? Razer Hydra's positional tracking is very good 1:1. The "base station" needs to be 3 feet away from the sensor. So what if the "base station" was inside the neck of the chair, and the sensor was on the rift (like the Sixense Stem) - now all you need to do is take the sensor data from the Sixense Stem and override the Oculus Rift sensor data with that. I think that would be the best solution. No need to offset anything, as the "base station" would be a embedded into the motion simulator.

"Felix12g" wrote:
... One thing is certain, external motion from simulators or rides or whatnot will mess up the rift. ...

External motion from simulators or rides will certainly NOT mess up the Rift. What gave you the idea that it would? What made you so certain? Where is your proof?

Turning while standing on an ODT or seating in a rotating chair would be an EXCELLENT way to prevent motion sickness. It would be even better to have positional tracking to allow free movement over a large area by walking OR RIDING. Your "certainty" is completely contrary to all the evidence regarding haptic feedback. Here is one (of many) places you can begin learning about this stuff:
http://www.mtbs3d.com/phpbb/viewtopic.php?f=140&t=16484

Even the Rift motion sensors are fast enough and sensitive enough to sense (and compensate for) vibration or other "external motion". In fact, while my Rift was sitting on my desk, and I was plotting the accelerometer data, I could see footstep "ripples" in the graph while somebody walked across the hardwood floor in the next room. I can even see (and detect) very light tap and double-tap gestures to various spots on my Rift (as I mentioned in various posts).

External motion simulators or rides would be a most excellent way to experience VR in the Rift, despite your claim to the contrary.

Here is how the professionals do it:



Popularity of Virtual Reality Immersion in Theme Park Attractions of North America:
http://aquila.usm.edu/cgi/viewcontent.cgi?article=1105&context=honors_theses

Even the Hydra controller position sensing should be relative to the Hydra base, so moving in a simulator or ride should not be a problem, if kept well enough away from nearby moving magnetic or ferrous objects.

If one of these won't mess with the tracking on a dev kit I'd be very happy.




My assumption was based on gross movements of a simulator being sensed by the kit and passing it on as supposed movement of the player.

Hmm, I guess a better way to put it was an assumption based on the gravity direction changing in relation to the player's orientation as well as acceleration felt by the gross rotations?

geekmaster, the sensors in the Rift sense movement/rotation in relation to Earth, right? 3 Accelerometers, 3 Magnetometers, 3 Gyroscopes. In that korean rally driving video, I mostly just see vibration, very limited axis rotation. I'm still thinking the sensors in the rift will be confused by a motion platform making large movements.

But if the sensors in the rift won't be confused by a motion platform, that would be SWEET.

Felix, that Force Dynamics setup looks UBER. Too bad I don't have the space for it. :D