PCVR/Link Development

This thread is all for noob related questions, as I'm sure I will have some and didnt want to create a new thread for each one, and I hope I'm not the only one, although everyone around here seems pretty knowledgeable.My first question involves what engine to start out with, I dont have a problem with c++ or UE4, I am going to build a new PC for DK2, and my current PC here at work is as followsIntel Core i7-3770 Processor (8MB, 3.4GHz)8GB, NON-ECC, 1600MHZ DDR31GB AMD RADEON HD 7470Will this suffice for simple starter noob development? Just want to create some simple shapes and polygons (someone suggested blender and UE4) and tinker around, get used to the environment, and just want to make sure I am starting off on the right foot with the right tools.

Hi All,After doing the ray tracing demo, I wondered if any of the optimisation I used there could be applied to rasterization. In the ray tracer, I warped the rays I shot out, avoiding the need to do an oversized image and a post process warp. This worked very well. In fact I'd be inclined to do a whole engine on RT at some point in the future (possibly based on a hybrid Rasterisation / RT thing, much like the recently announced Wizard GFX design from imagination tech). This lead me to thinking on how I could avoid losing the 'warp the rays' advantage. So my idea is to warp the G-Buffer instead. The theory goes like this, the oversized buffer is to accommodate the greater resolution in the middle of the warp, but ends up throwing away a lot of the extra detail around the edges. Since most of the processing power goes into shading, you warp, then shade, i.e:1. Render to oversized G-Buffer2. Warp to target sized G-Buffer3. Shade the smaller target (ignoring the black borders outside the

So as I'm developing with the Rift and Unity, a major pillar of this effort is maintaining a solid 60fps for the smooth window-like view through the Rift to the world inside the Unity engine. Whenever framerate starts to drop, the VR suffers and when it gets below 30 fps, low framerate starts to even make me get motion sickness.After lightmapping, now it seems the biggest culprit to low framerate is Unity's Water (Pro), likely since it is reflecting virtually everything, being used like an ocean. I noticed in the Tuscany Demo, they used a simple static texture for water, which at first I found sad, but now understand why.Has anyone had any luck with alternative solutions for Unity + Rift framerate-efficient water? We're talking ocean-scale water within just a 1K to 2K map. I'm about to reluctantly experiment with static non-reflective textures, just as in the Tuscany demo.-XPS: I also raised the question at the Unity forums if anyone is interested: http://forum.unity3d.com/threads/

When Oculus gets a consumer release, will Oculus itself be a development kit, or devs will have to buy some other kit to continue producing content?

Is there a head model I can use?That is, how the eyes depend on the neck and rotations.I am using the old Oculus SDK, and I am doing most of the math myself.

We developed a prototype that demonstrates crouching functionality for the Omni. Note: this prototype is extremely clunky, and not close to any product we would ever release to market. We are still investigating how to make this functionality available to our backers and customers. While crouching is not difficult to achieve mechanically, adding these moving parts would increase the cost structure of the product significantly (far beyond the reach of our current price point). Not to mention the reliability and maintenance headaches that typically accompany moving parts. That being said, perhaps a good-looking variation of this movable upper structure could one day be available as a high-end upgrade?

Hello,Thanks to various members of the forum here I got 90% of what I need. Thank you.I am just trying to resolve one last bit and I hope someone can point me to any easy errors I am making but not seeing.My code works except when a person looks left the camera actually rotates right. Vice versa with looking right, camera rotates left.Any idea what I am doing wrong here? Thank you:Here is how I get head orientation.OVR::Quatf OculusRiftOrientation = PredictedPose.Orientation;glm::vec3 CurrentEulerAngles;glm::quat CurrentOrientation;OculusRiftOrientation.GetEulerAngles<OVR::Axis_X, OVR::Axis_Y, OVR::Axis_Z, OVR::Rotate_CW, OVR::Handed_R> (&CurrentEulerAngles.x, &CurrentEulerAngles.y, &CurrentEulerAngles.z);CurrentOrientation = glm::quat(CurrentEulerAngles);And here is how I calculate the LookAt:/*DirectionOfWhereCameraIsFacing is calculated from mouse and standing position so that you are not constantly rotating after you move your head.*/glm::vec3 DirectionOfWhereCam

I wrote an easy to follow how to guide for Minecrift http://riftmod.com/how-to-setup-minecraft-for-oculus-rift/

Damn, with just 2 days to go until June, Yei have announced that the PrioVR won't make the June shipping estimate.https://www.kickstarter.com/projects/yeitechnology/priovr-suit-up-game-on/posts/859068From the update post:As with any new technology, if we waited until the suit was “perfect” we’d never ship a product at all. There will always be improvements to be made, and we’re only just brushing the surface with where we can take inertial motion tracking. That being said, we’ve set three benchmarks that we need to meet before we can say it’s ready to ship. (1) Developer tools. This is first and foremost – solid developer tools, matured APIs, and seamless integration into Unity and Unreal engines. We know you don't have a lot of time to spend experimenting with integration into your game or VR experience, so we want to make sure you can have confidence in our APIs and that there’s a near-to-drop-in experience for integration into main engines. This means clear documentation, code examp

Project Gauntl33tWe are working on a haptic feedback glove. The glove will be wireless and provide hard and soft feedback to the user to give the users a sensation of holding objects in the game. Our device should make a good pair with STEM, MYO, PrioVR, and other new tracking solutions. If you are interested please join our facebook group. https://www.facebook.com/groups/536642796371093/Newest Prototype! Gauntl33t Lance. Early ImagesThis prototype will use raspberry pi and use a bluetuth dongle to be wireless.First PrototypeWorking Glovehttps://drive.google.com/file/d/0B1vsMIDwrysBYXdhN3lQU1hrUm8/edit?usp=sharingRev VR Podcast Special Thanks to Reverend Kyle for letting us show off first prototype on his show.Start of Game DemoFirst Google HangoutSecond Google Hangout

Link: https://github.com/wwwtyro/node-ovrsdkThis version of the bindings supports both Windows and Linux. Hope to have OSX .dylib compiled and tested soon.

Hi All,I've just updated my Ray Tracing demo to try out Fovial rendering, that is, rendering more information in the middle of the view, then rendering at a lower resolution outside that radius, then compositing the two together.You can download and have a go here.Note: You will need a fairly recent Nvidia Based GTX gfx card (I'm running on a GTX475 and getting about 50 FPS).You also need a DK1, as I've not set any fallback mode. You would need to stare straight ahead to get the idea, looking around with your eyes will be strange, but something like this will be possible with eye tracking, to move the sweet spot around with your eyes. :) Further work: Would probably need to blur the lower resolution version a bit more to make it less 'artifacty' (yes I used that as a word) 8-)

Would it be possible to get/make some form of middle ware to be able to use a different camera a la PS3's Eye Move Motion camera and a bunch of ir LED's attached to the DK1's outer shell so it can be tracked in the (more or less ) same way as Crystal Cove will be doing ?I've seen a few methods of doing this with somne middleware for new apps in unity and the like, but it'd be awesome if it was possible to do this with the old DK1 too.Refferance(s):http://www.youtube.com/watch?v=XIbaxnpYUsMhttp://www.youtube.com/watch?v=ynM3YHaU8To

Hi! :)I know this has probably already been asked a lot of times, but how do I get the Oculus SDK onto my Unreal Engine 4 editor and start using it? I can only find the Unreal Engine 3 version of Oculus VR.Thanks!

Hello,recently I've received my raspberry Pi and a IR camera for it.I've searched in differents topics and I haven't found a answer.I'm a beginner and I just want to display what my camera see on the oculus screen, I don't know how to process the image for displaying, can you help me please? :? thanks

I have been using the latency tester while doing some engine integration work. When I tried setting the head tracking prediction to the exact value that the tester was reporting, the experience was kinda magical. The world felt more stationary and permanent as I moved around.I think there might be great value in incorporating the latency tester into the final Rift. You could put the photo sensor right in the lower corner of the display where it won't interfere with the image, so that every second or so it can take a latency test sample and dial in the right prediction as the framerate changes over time, automatically.I found that as long as I didn't go over about 50 ms the benefit of rotational accuracy far outweighed the downside of jerky accelerations, and that seems to be easy to achieve even on mediocre hardware. I definitely won't give another demo without first dialing it in using the tester.

Perhaps the first thing anyone wants to do with the SDK is see if they can get the Euler angles from the tracker. Unfortunately the documentation is even more wrong than it was in v0.31, because Posef that the documentation tells you to use doesn't exist.Here is how I read the sensor instead (in Doom 3's angle system):#include <windows.h>#include "OVR_CAPI.h"#include "Kernel/OVR_Math.h"void ReadHmdOrientation(float *roll, float *pitch, float *yaw){ if (hmd) { ovrSensorState ss = ovrHmd_GetSensorState(hmd, 0.0); if (ss.StatusFlags & (ovrStatus_OrientationTracked)) { OVR::Transformf pose = ss.Recorded.Pose; // can't use prediction without frame timing float y = 0.0f, p = 0.0f, r = 0.0f; pose.Rotation.GetEulerAngles<OVR::Axis_Y, OVR::Axis_X, OVR::Axis_Z>(&y, &p, &r); *roll = -RADIANS_TO_DEGREES(r); // ??? *pitch = -RADIANS_TO_DEGREES(p); // should be degrees down *yaw = RADIANS_TO_DEGREES(y); // should be degrees left } }}I hope that helps someo

OVR_Linux_SensorDevice.o appears twice in the OBJECTS definition, lines 81 and 113. Not a problem for static compilation, but throws a wrench into building a shared lib.If there's a more appropriate place for reporting these, let me know. :)Edit: OculusSDK/LibOVR/Makefile

Link: https://github.com/wwwtyro/nw-ovrsdk-helloworldLet me know if you run into any problems. Enjoy!

How important is yaw and what are the best ways to input it on different devices (keyboard and controllers).

Hello,As turning on ODT bring a lof of problem, for positional tracking, compatibility with oculus, body tracking,Cables....I was thinking of this idea :_Have the user no more able to turn his physical body by rotation._Instead, when he want to turn, Make the floor rotate under his feet, instead of his body.By using the feet lateral adherence that he already use to turn his body(hardware improvment)Or_improve the feet movment detection to even detect when the user try to change body orientation and not only body walking direction..(software improvment)The result would be exactly the same.It would be an ODT compatible with the oculus rift positional tracking technology.It would be an ODT even compatible with a TV and not only with Oculus.

So I've been thinking about implementing one of the newer example based skinning algorithms out there, I've got one of the early ones from 2003 but as I am migrating 3d packages I figure now is the time for new ideas. I've done some research and come across real-time enveloping with rotational regression Which looks great and significantly improves over traditional smooth bind skinning. However it was back in 2007, and I haven't kept as current on research as I would like. Does anyone know if there are any more recent algorithms along the same lines? There is a lot of example based skinning work out there but all the research I see ties the computation of the skinning weights to the creation of a skeletal hierarchy as well, which will not work for me.

Hi,NICTA have just open sourced a plugin (and live demo) which shows basic integration between the Oculus and the WebGL Virtual Globe Cesium. Note that both Cesium itself and the Oculus plugin are released under the flexible Apache 2.0 License.https://github.com/NICTA/cesium-oculus-pluginhttp://cesiumjs.org/

Hello,So I have my application ported to the rift using the SDK and I can read sensor data fine. It's calculating the sensor data and converting it over to GLM that is giving me problems.I looked at some examples but I am not clear how I can incorporated the position and orientation from the device with the world coordinates of where the person is at?Thank you for your time.Here is my current wacky code:ovrEyeType l_Eye = (*MainRiftDeviceDesc).EyeRenderOrder[l_EyeIndex];ovrPosef l_EyePose = ovrHmd_BeginEyeRender((*MainRiftDevice), l_Eye);glViewport(MainEyeRenderDesc[l_Eye].Desc.RenderViewport.Pos.x, MainEyeRenderDesc[l_Eye].Desc.RenderViewport.Pos.y, MainEyeRenderDesc[l_Eye].Desc.RenderViewport.Size.w, MainEyeRenderDesc[l_Eye].Desc.RenderViewport.Size.h);/*How can I convert this to a GLM::mat4???*/ OVR::Matrix4f l_ProjectionMatrix = ovrMatrix4f_Projection( l_EyeRenderDesc[l_Eye].Desc.Fov, 0.3f, 10000.0f, true); /*End Question.*/glm::quat CurrentOrientation = glm::quat((l_EyePose.Ori

I'm experiencing an issue with my DK1 that I did not expect. Occasionally (maybe 3 or 4 times in 20 minutes) I can tell that roll (tilt head from shoulder to shoulder) is drifting significantly (more than 3 or 4 degrees). If I sit really still and watch it, it will slowly get worse and then "snap" back to the gravity vector. I know the accelerometer isn't faulty because otherwise it wouldn't know what to eventually snap back to, right?It seems like the drift correction is not converging on the gravity vector fast enough, especially when no movement is taking place (so no interference from other acceleration sources). Is there some setting I can tweak for this? Or some way to turn off correction so I can roll my own?I'm using WIndows 7 and have this issue on all demos I've tried, plus multiple versions of the SDK, including the latest example project and Unity prefab.