PCVR/Link Development

Spec: Ubuntu 14.04, intel core i5 with integrated GPU (H67ITX-C-E/H67ITX-C-E), latest Intel graphics drivers.SDK 0.2.3 works fine, but with 0.4.3 the config utility says there is no device attached. When trying to toggle legacy app support, I see this in the console:Error: [SysDlg] Service and driver legacy mode are out of sync: Failed to fix it.Oculus world demo seems to be opening a window for a microsecond then exits with status 1. Log:X Error of failed request: GLXBadDrawable Major opcode of failed request: 155 (GLX) Minor opcode of failed request: 29 (X_GLXGetDrawableAttributes) Serial number of failed request: 80 Current serial number in output stream: 80(tried the fix at viewtopic.php?f=20&t=16268#p213055, didn't help).oculusd log:OVR::DeviceManagerThread - running (ThreadId=0x7f7f67988700).OVR::DeviceManager - initialized.OVR::Linux::HIDDevice - Opened '/dev/hidraw0' Manufacturer:'Oculus VR, Inc.' Product:'Tracker DK' Serial#:'M1EGG3PGPM39'Erro

Hello,I'm looking for a 3D Model of the Oculus Rift DK2 that I could include in my game.Where can I find one ?It would be really cool if that would be included into the SDK (.OBJ and .FBX) and "royalty free".It seems that the one made by "MannyLectro" is no long available for download (see topic Rift/Accessories, 3D Models -!DK2 AVAILABLE! at https://developer.oculusvr.com/forums/viewtopic.php?t=1514).Thanks !

So I just updated from 0.4.2 to 0.4.3 and there were a bunch of changes to the SDK which I needed to incorporate into my code. I listed the ones I was affected by here to aid others in their updating.ovrHmd_GetEyePose is depreciated so remove:eyeRenderPose[eye] = ovrHmd_GetEyePose(hmd, eye);add:ovrVector3f hmdToEyeViewOffset[2];hmdToEyeViewOffset[0] = eyeRenderDesc[0].HmdToEyeViewOffset;hmdToEyeViewOffset[1] = eyeRenderDesc[1].HmdToEyeViewOffset;ovrHmd_GetEyePoses(hmd, 0, hmdToEyeViewOffset, eyeRenderPose, NULL);eyeRenderDesc.ViewAdjust is depreciated so remove:eyeRenderDesc[eye].ViewAdjustadd:eyeRenderDesc[eye].HmdToEyeViewOffsetPosition returned by ovrHmd_GetEyePoses now includes the View Adjust translation already, so for head position replace:eyeRenderPose[eye].Positionwith:ovrTrackingState trackingState = ovrHmd_GetTrackingState(hmd, 0);trackingState.HeadPose.ThePose.PositionFeel free to add other tips in this thread!

I wanted to make a post because I hear a lot of people nay saying about games not specifically for vr and wanted to share my experiences with moding gta sa for the rift because I think it applies to the new gta5. I was able to get gta sa to work really well with the oculus in first person. I played threw the entire game / did lots of messing around and really enjoyed it. there were some hick ups but not many that really hurt the game play. it was still really fun. the best part was running from the cops, getting into a wreck then having to escape struggling to see out of the broken wind shield. there were some problems with perspective in some of the cars but they still got the job done. none of the problems were near as bad as the shadow problem in skyrim using the rift. I know that gta 5 will have its own set of glitches if its modded for vr but the fact that they are making it with first person already in it is going to make things a million times easier and up the quality of vr ga

So I'm making a program for the rift, using Game Maker studio. It does not natively support the rift but there is a community extension, which I have even contributed a tiny bit to myself: http://www.gmoculus.comWe have standard client rendering working, but not direct-to-rift SDK rendering, which would be a much better solution because of lower latency, automatic updating with the sdk, and because we are missing a lot of features in our own rendering, like chromatic aberration.The reason we have not gotten sdk rendering to work is because neither me or the main extension developer are very experienced in c++ and certainly not directx. But I think we've got the bulk of the work done, if someone knowledgeable would step in and help I'd think they could get it working pretty quickly. So I thought I'd email oculus and see if they could help, because who knows their SDK better than themselves, and I would have thought they would be very eager to help enable as many developers as they can.

ALT.CTRL.GDC is an exhibit at GDC that shows off games with weird, experimental control methods and input devices. It was one of the coolest things at GDC last year, and they just opened the call for submissions: http://www.gdconf.com/news/altctrlgdc_needs_your_one-of-a.htmlVR is pretty great for playing around with input, and at last year's exhibit, I got to try out Please Don't Spacedog, a VR game that was played with a panel full of buttons and dials. John Polson (the organizer) asked me to spread the word here, since he thinks there are probably a ton of cool Rift games that could be part of it. I think so too! :D

I currently do not own a rift, but as a developer, I've been looking for examples of semi-flawless player head-tracking/body movements to study for a small game I'm making. I want to be able to create a realistic fantasy character that'll give the player the actual feeling of being inside of the body, maybe aside from arm/hand movements. Has any game pulled this off? If not, can any rift users refer my to the best body experience they've had so far (possibly open-source)?Thanks in advance!

Following the "eye-Patch" discussion on Reddit, I made a suggestion that could either be genius or simply nausea inducing.Rendering two view points each time to update the screen is very process intensive.So why just don't? Only render one eye and then the other. Think of it as "Asynchronous Low Persistence"Every frame you send, you only render one eye and the other one is left black.In the next frame, you render the view point of the other eye.Obviously, this would only work at 75fps.Please don't reply to this post with opinions on why you think it wont work.I'd be interested to see if someone could implement and test this theory.

Hi, we just added DK2 support into CopperCube, by upgrading from the 0.2 SDK to 0.4. And it was quite a PITA do do that unfortunately, because quite a lot seems to have changed since the 0.2 days, and we also had to make some modifications to the SDK, also because we wanted it to run in D3D9. CopperCube is an editor for creating 3D apps, here is a short video of it in action:After having done this upgrade, I have a question/suggestion for the SDK: In order to let the user model their 3D world in any scale they want to, for the DK1 / SDK 0.2 version, we had an option to specify a world scale in the editor. We would simply scale the IPD value in the Oculus SDK for this, and it worked great. We had to make some modifications in the SDK for this, but it worked. The 0.4 changed everything quite a bit, so before trying to do the same there again, wouldn't it be a good idea to introduce such a feature to the SDK? Like a simple float for changing the world scale? But maybe there is something l

I'm not too experienced with programming or unity, but I've been developing a game, but I am unsure on how to get speech recognition integrated to it. I've watched a few tutorials to make one on c# but i can't seem to get it to work inside of unity. Any help would be appreciated.

As I have not implemented a rift project yet I know I'm asking a bit of a noob question.... but regardlessI poked around to see the guidance on stereo rendering, but am wondering if there is any standard method for switching between normal and stereo rendering yet.So that when swapping from an extended mode dual screen, back to a single screen we can render normally... Or is there a way yet to toggle from direct to rift mode back to regular single rendered monitor screen. ?

I know a few of you fortunate souls have a Hydra to play with, but those are long gone from all by the priciest Ebay sellers, and they don't really track you're full hand, with all of the fingers, just the position of your palm. Unless your plan is to develop for the EXTREMELY tiny future STEM market ($300 a set, not developer price, retail), why not shift your development to the Leap Motion? It's cheap, at around $80 ($100 with the $20 Rift mount). It tracks the whole hand, including fingers. And most, if not all, game/hand interactions will require you to look at your hands regardless. So the Leap's tracking will always be where you need it, when you need it. The tracking, in it's current incarnation that I've been playing with, is a little spotty at times, but good enough to push buttons, turn levers, knock things over, and most in-game actions. The tech should get better over time. Adventure gaming, location exploration, seems to be the most interesting thing the Rift does, so why

Attach the phone to the arm (wristband) and use the accelerometers to calculate position. Use also cameras and additional sensors of the phone to refine.Good Luck, looking forward to your app. :D

With the update to 0.4.3 there was at least one change made to the C API which affects most/all custom Oculus SDK integrations (see this thread). There was no communication of this change, or of other potentially existing changes to the C API in the change-log. The integration documentation still shows the old, incorrect, syntax. I understand this is extra work for internal Oculus developers who could be spending time elsewhere in the SDK, but the lack of it makes updating SDK versions painful as the resulting errors which could be trivial to solve ahead of time, are instead unforeseen.I would like to request that changes/additions/depreciations to the C API be announced with each release.Please add a comment if you are interested as well.

So I was wondering if it was possible to somehow configure a Unity build to run on the rift on startup?I am able to run on it on the rift manually, but it would be great automatically with no interaction from the user. I could do this in an internal Unity script, or maybe in an external script that runs the Unity build and automatically moves the build's window to the rift. Thanks!Edit: It appears that this thread should be in the Unity oculus integration section (sorry). Also, I did find a thread that was similar to this one:viewtopic.php?f=37&t=14372But I'm not sure if there will be any strange side effects if I modify OVRShimLoader as the thread indicated...

I have noticed this weird things with current Dev Kit. When looking at some textured object that is right in front of you, your eyes tend to both look at it and simultaneously focus at it. Because the Rift is focused into infinity, the texture just gets blurry. However if you keep your other eye closed and then relax the open eye to focus into infinity, then the texture gets sharp, but it looks like it's far away.I don't know how big a problem this even is, but I suppose it isn't even fixable. What about our eyes, do they get used to see objects very near, but focus into infinity at the same time? I'm not an eye expert, but it would be nice to hear from someone who has studied this more.

C++ devs, this one is for you!If you're not using our Unity VR assets (or some of the amazing assets that you can find for Leap + Oculus + JS), this post from Leap engineer Jimmy He will help you get started with correcting for distortion, orienting objects and cameras within your scene, and using the Image API for raw image passthrough:http://blog.leapmotion.com/vr-essentials-need-build-scratch/Also, be sure to tune in to our Twitch TV session on Wednesday, where our engineers and design leads will be on tap to talk all things VR:http://www.eventbrite.com/e/all-about-vr-a-twitchtv-spotlight-for-leap-motion-3d-jammers-tickets-14056811301?aff=eorg

Has been charged? I just got a message the payment has cleared.. and was just curious/excited etc.. :)

Hi,I can get everything to run smoothly in OpenGL in both extended and direct modes using the 0.4.3 SDK. The only problem I seem to be having is that my code isn't rate-limited as it is very simple so my framerate calculation is showing it running at a silly 400-1000FPS which is much higher than the 75Hz required. Either my calculation is wrong, or I'm using more CPU than necessary and begs the question if there is a way to VSync directly and limit my framerate. I'm using GLFW to create the windows. It is possible that the Oculus SDK gives a special OpenGL context and doesn't glFinish frames unless it decides to render them and so I end up clearing the screen and re-rendering a few times before it decides it wants to glFinish and only then do I pay the high price of stalling and warping. Just a guess here.Thanks!

Hi All,I'm having trouble building my very small application. I have a single c file (helloworld.c) as follows:#include "OculusSDK 3/LibOVR/Include/OVR.h"int main(int argc, char** argv) { ovr_Initialize(); return 0;}And I'm using g++ as follows:> g++ helloworld.c OculusSDK\ 3/Samples/CommonSrc/Platform/*.cpp OculusSDK\ 3/Samples/CommonSrc/Render/*.cpp OculusSDK\ 3/Samples/CommonSrc/Util/*.cpp -L OculusSDK\ 3/LibOVR/Lib/Mac/Release/ -lovr -o VROUTPUTSo, I'm including my c file, all the CommonSrc code included in the Oculus SDK, LibOVR, but i get the following error:OculusSDK 3/Samples/CommonSrc/Util/RenderProfiler.h:27:10: fatal error: 'OVR_Kernel.h' file not found#include "OVR_Kernel.h" ^1 error generated. Does anyone have a very simple application they have built on mac with all the correct libraries etc and a compile command that seems to work? Am I using the wrong compiler? Is there a "hello world" tutorial for actually building and compiling an application?

Hi,I am thinking of building a simple scene to experiment with binocular rivalry. The basic idea is that for the same object with different textures, the brain will only pick one of them. Is there an easy way to have one material use different texture for each eye in UE 4?

Hi there,Background:I'm very new to 3D Programming and I'm currently implementing Rift support for the Car Simulator we have at our institute at university. (This would be my Bachelor Thesis). The main function of this simulator is to test control engineering systems and it provides the graphical display as a added feature. A user can freely move the camera around the whole landscape where the car is driving in addition to features like fixing the camera to a inside view of the car.The simulator Software already exists and was not programmed by us. However I managed to get some info about how rendering is done and they are using a combination of this TCL - Skript language (probably because of the control sys part) and an OpenGL - TCL extension.What I have:The (beauty) is that I can *fairly* simple add features to the simulator by loading an additional TCL script on startup. What I have done is write a .DLL that wraps the Rift functionality (for sensor data) and gets loaded when the dis

Having trouble locating anything like this from all the search noise around 'video'.Are there any sample DK2 projects that render video onto a 3D surface?

http://jmonkeyengine.com/ - for people that work with Java - it would be nice to integrate it.

Hello everyone!This is my first post here on the oculus forums and here goes.While looking for a wireless display solution for the new oculus development kit, I noticed that most transmitters that offer -2ms latency cap out around 1080p at 60hz. Does the DK2 require a 75hz input or is that taken care of at the "box" before it reaches the headset?example: 1080p 60hz output from desktop - to signal box (upclocked? 75hz) - headset.If i am totally wrong thats fine. I just cant seem to find any info on it.