Rift compatible Tracker based on STM32F3DISCOVERY

[yetifrisstlama]

Hi,

I thought you might be interested in this:

http://yetifrisstlama.blogspot.fr/2014/ ... verse.html

It is firmware for the STM32 development board, which emulates the original tracker of the Oculus Rift and is hence compatible with all of the existing games.

If you want to try it, I have posted the source code and the binary file on my website.

It is still pretty much a work in progress and not all of the features have been implemented yet. So far I was able to test it with the Oculus World demo and the Red Frame demo on Linux and Windows and it worked well for me.

Let me know if you got it working also with other applications. Also comparing its performance to the original tracker would be interesting.

Cheers!
YFL

[Fredz]

Nice work. I don't know if you're aware of the Adjacent Reality tracker which was made by the same guy who designed the tracker in the Oculus Rift. He's made the hardware design and the firmware available under an open source license here : http://adjacentreality.org/

Maybe it could be helpful to have a look at this code for your project.

Also did I understand right that you tried the Redframe demo on Linux ? Where did you get it ?

[geekmaster]

Interestingly, the default VID/PID for the STM32F3DISCOVERY board was still in the OculusVR SDK source code last time I checked. Perhaps it was used in Oculus prototypes before they developed the Rift controller with its own unique VID/PID that is also in the OculusVR SDK source code. Beware that the source code has a comment saying the non-Oculus VID/PID may be removed in some future version of the SDK.

[yetifrisstlama]

Nice work. I don't know if you're aware of the Adjacent Reality tracker which was made by the same guy who designed the tracker in the Oculus Rift. He's made the hardware design and the firmware available under an open source license here : http://adjacentreality.org/

Maybe it could be helpful to have a look at this code for your project.

Also did I understand right that you tried the Redframe demo on Linux ? Where did you get it ?

Cool, I didn't know about the Adjacent Reality tracker -- but that seems to be a really nice project. The biggest disadvantage of the STM board is still its size and weight. In the long run, I will probably go with a custom board. Luckily, the firmware and the source code can be reused.

The RedFrame demo I only tried under Windows

Interestingly, the default VID/PID for the STM32F3DISCOVERY board was still in the OculusVR SDK source code last time I checked. Perhaps it was used in Oculus prototypes before they developed the Rift controller with its own unique VID/PID that is also in the OculusVR SDK source code. Beware that the source code has a comment saying the non-Oculus VID/PID may be removed in some future version of the SDK.

Yes, I stumbeled over this comment as well. Luckily the VID/PID is defined in the tracker firmware and can be changed very easily.



A short status update on my side:

I managed to hack the EDID of my LCD panel.
It is now recognized as a HMD by LibOVR, which allows me to use the configuration utility and I can also carry out the magnetic calibration successfully.
I know that the magnetic data is sent, as these indicators (X,Y,Z) all wiggle around during calibration.

However, there is still Yaw drift in the Oculus World Demo. It looks like the sensor data from the compass is not used at all.

By pressing F6, I can display "Magnetometer Yaw Angle Marks". They seem to stick to the floor. Are they supposed to move when I play around with a magnet next to the tracker? I couldn't find any information on this at all.

Maybe someone who owns an original Rift can shed some light on how these Marks should behave.

Cheers YFL.