vr

[rouncer]

ive never used a vr product before, never even went to a virtual reality
arcade cause it didnt interest me.
ive seen movies though, and i sorta know how they work by keeping relative
or absolute 3d positions.

so if i could get a viewer and a positioner (i just made those two words up),
id like to hook it up to a 3d modeller.

can you get this stuff for windows?, where do you buy it?

[geon]

I believe the big 3D software packages support stereoscopic view. (LCD shutter glasses seems to be the cheapest/most practical solution.)

I am not sure about 3D "positioners", though. There are some things like a pen connented to a robot arm, that can sense the 3D position, or 3D motion capture units, but those are really expensive. (like $ 10000+)

Personally I would just go for a good mouse with a 3D "trackball".

[.oisyn]

I am not sure about 3D "positioners", though. There are some things like a pen connented to a robot arm, that can sense the 3D position, or 3D motion capture units, but those are really expensive. (like $ 10000+)
Surely the Wii controller isn't that expensive ;)

[monjardin]

MEMS gyros and acceleromters are making IMU (http://en.wikipedia.org/wiki/Inertial_Measurement_Unit)'s alot cheaper.

Xsens sells some neat little boxes for under 2000 euros. (http://www.xsens.com/index.php?mainmenu=products&submenu=human_motion&subsubmenu=MTx)

If you are more electronically inclined then you can build your own for a lot less. You can buy a 3-axis accelerometer (like this one (http://www.st.com/stonline/products/literature/ds/12094.pdf)) for under $15. When it's sitting still you can get your pitch and roll from the acceleration due to gravity, but you'll want gyros to get full attitude data.

EDIT: Analog Devices has a line of MEMS gyros (http://www.analog.com/en/subCat/0,2879,764%255F801%255F0%255F%255F0%255F,00.html) around $30.

[geon]

Surely the Wii controller isn't that expensive ;)

Would accelerometers give enough precission to work with a modelling software? What I've seen of the Wii videos shows mostly waving in a general direction, not precise selection and manipulation.

[.oisyn]

The wii controller has absolute positioning, not (just?) accelerometers (the PS3 controller has accelerometers). You need to place a strip of (invisible) LEDs on top of or underneath your tv set. In the controller is a small camera that detects the strip and the controller's position and orientation is calculated based on the projection of the strip on the CCD.

This obviously requires a line of sight between the controller and the strip, but with multiple lightsources around the room you should be able to get the full 6 degrees of freedom.

You can use the controller to exactly point at objects on your screen, so it's accurate enough for VR I would say.

[rouncer]

i guess youd need to calibrate any relative IMU (thanks for the terminology)
before you start... that wii controller sounds interesting.
but those MTx's from Xsens look good - especially since it comes with a
c++ dev kit, which sounds learnable. my idea being you strap the mouse to
one for buttons and you strap the other one to your head with some kind of
head monitor. cause you couldnt really mount your lcd screen to your
shoulders with any success. (even though they are pretty light) :)
does anyone know anything about these?

EDIT: maybe nintendo1 came with game glasses ages ago, with the power
glove, with the power glasses.

[monjardin]

It looks like the Wiimote uses the USB HID class to communicate over Bluetooth. In other words, you can make it work with your PC. Here is a video (http://youtube.com/watch?v=9iBaKsh5z_o) of it in action. It uses Analog Devices (http://www.analog.com/en/content/0,2886,764%255F%255F99573,00.html) ADXL330 (http://www.analog.com/en/prod/0%2C2877%2CADXL330%2C00.html) accelerometers. A Wiimote is $10 cheaper than the ADXL330 evalution board!