Recent Posts

Well, if you want to knock up some diagnostics I can ask the customer to give them a whirl and report back.

Cas

I agree with Mazon. There is no need to integrate Rift into LWJGL. A standalone project would be sufficient. I haven't tried JRift yet but I will do so now that it has OS X support

Btw: basic Razer Hydra access from Java worked very well for me using JNA. There was no other native code necessary to get that working. It's basically only calling the functions in the Hydra .so/.dll.

If and how all of that consolidates into some "LWJVR" project one day remains to be seen but IMHO that's a separate project.

EDIT: As a proud Rift owner I can tell you that all you've heard/read about it is really true! The good and the bad ... but really there is more good than bad. Especially things where you sit in some cockpit work really really well 

I don't think LWJGL is going to do any oculus rift integration library. This sound like a self-contained project?

I'm not aware of any win8 surface pro users, that can debug and/or verify this :/

@ajr_1 I've got the same issue. The mouse cursor looks exactly the same as in your picture.

JRift is just a wrapper for sensor input. I would argue that the 'wheel' has not yet been defined and therefore it's rather hard to reinvent it.

JRift (jni lb) just got Mac support I believe so maybe you don't have to reinvent the wheel

Mike

Throwing my hat in the ring as well. I'm waiting for my Rift to arrive, and seeing as it won't be long now, figured it was a good time to start playing with things. I'm using a Mac as my primary dev machine and not to keen on having to deal with windows, so I'll be braving Rift dev with the Mac SDK and Java. It's been years since I've played with OpenGL (read old SGI Indy boxes) but it does not look like I'll have an issue ramping back up. LWJGL looks like as good a place as any to start with integration.

I think the real question as far as LWJGL is concerned is what part of the core of the lib could be enhanced to support the Rift and what bits are really Rift specific. I think if you pull back a bit, there are a number of VR specific things that are going to need to exist whether you use the Rift or not. Things like modeling the users head relative to the 3D setup is really just a camera, making sure the ears are placed right for sound, etc. This starts to go a bit farther when you start to talk interaction things like Kinect or even the Razer Hydra would have a cleaner integration point if they mapped to a 3D avatar wireframe (Just a basic stick model is fine)

I guess the big question is, what are folks feelings on this. Is it preferable to start a VR package inside of LWJGL to handle these things, or a parallel child project? Personally, if done right, adding support for these things has benefit to the project as a whole as the more 'turn key' interfaces there are to input/output devices, the better, but I know that there are also likely a number of folks that could care less about VR and don't want any of it cufting up the performance of the stuff they are playing with. What i don't see making any sense though is forking anything as that tends to be a loose/loose situation for everyone.

I'm planning to play with getting this thing up and running for myself. This is a total side project and so I can't commit to any timelines, but I am planning on getting my hands dirty on some of the things like the JNI interfaces to sensors, and the like. I also tend to be a stickler for good design. i.e. I'm not above hacking a kludge together for a quick demo and some instant gratification, but anyone that has had to maintain shared code past one iteration/project knows that thinking things out a bit saves a ton of pain down the road. I'm a total noob to the Rift and this project, and would like to get involved in a meaningful mannor. I would nominate that we start a thread/section/etc on what more abstracted portions of VR have value in other areas and perhaps come up with a clean, abstracted way to start pulling in support for the Rift, but with the idea of making it easy to support any other VR type effort that comes down the pipe.

Thoughts?

From user:

"The game starts in a window. When I enable fullscreen under options, it only shows me the upper left quarter of the game"

meaning, DisplayMode.getWidth()/getHeight() are returning values twice what they should be. Or is it half. Brain too lazy.

Cas

A quick question. I'm looking to use LWJGL's quaternion class as a way to represent the rotation of objects in my world in a better way than the gimbal system (yaw,pitch,roll) but I'm not sure where to start using the class.

For a quaternion (x,y,z,w) x,y,z represent the axis of rotation and w the angle of rotation about that axis. To use the quaternion for math, these values must be normalized; i.e. the whole vector (x,y,z,w) must have a length of 1. To rotate a point p by this quaternion q I do (q*p*q^-1) and series of rotations can be combined by multiplying them together in reverse order. (i.e. if q1 then q2 then q3, the combination quaternion would be q3*q2*q1.)

When I create a lwjgl Quaternion, does it normalize automatically (i.e. can I do (1,0,0,Math.PI/4) for a 90 degree rotation about the x-axis?) does anyone have a link to some sample code of taking an intended rotation or set of rotations and using LWJGL's Quaternion class to translate the points?

Many thanks.