This is an old revision of the document!
Experimental and for linux pros only
Here is how to setup the Raspberry Pi to support normal OpenGL or GLES (both with acceleration) and be able to compile the Proton SDK examples. (tested with Raspbian “Jessie” on 6/18/2016)
From my tests I got it working two ways:
1. Set it up using the OpenGL experimental driver. Result: Proton examples only work in X11 in windows mode. Fullscreen (using the SDL2 harness) didn't work. Directions to setup
2. Set it up using GL ES, won't work in X11, works full screen by starting from console. This is more useful for me as it's more pedal to the metal and easier to get 60 fps, no x overhead. With this method we're setting up our own EGL context manually, but still using the SDL2 harness for other things
These steps are what I use to compile stuff by copying things from my Windows computer. A normal user would probably just grab the svn tree and run .sh files.
- Give the video card more memory on the pi. Run raspi-config and give it 128 or 256 (good for the pi3)
- Also enable the experimental OpenGL driver under the Advanced section in raspi-config
- Install cmake on the pi (sudo apt-get install cmake)
- Do “sudo rpi-update” to make sure your firmware is updated
- Run the following from a shell:
sudo apt-get -y install xcompmgr libgl1-mesa-dri && sudo apt-get -y install libalut0 libalut-dev && sudo apt-get -y install mesa-utils sudo apt-get install libglu1-mesa-dev freeglut3-dev mesa-common-dev
- Do the following from a shell window to install SDL2 from source:
cd ~ wget https://www.libsdl.org/release/SDL2-2.0.4.tar.gz tar zxvf SDL2-2.0.4.tar.gz cd SDL2-2.0.4 && mkdir build && cd build ../configure --disable-pulseaudio --disable-esd --disable-video-mir --disable-video-wayland make -j 5 sudo make install
- To test, type “glxgears” from a shell inside of the GUI desktop
Step 2: Proton stuff
- Install linux tools for Windows, rsync and ssh should work from the command line
- Setup a ssh key pair so you don't have to type passwords
- Edit SetupVars.bat and edit the IP to the pi
- Run CopyStuffToPi.bat. There should now be a ~/proton tree on the pi
- Edit the RTBareBones/linux/CMakeLists.txt file and make sure “OPTION(RASPBERRYPI_OPENGL “Compile for Raspi 3 OpenGL” ON)” is uncommented
- RTBareBones/CopyStuffToPiAndCompileLoop.bat and it should update that example and compile it in a loop
The last step of the build copies the rtbarebones binary to /RTBareBones/bin (as that's where the media is), so go to that dir and run “./rtbarebones” and it should work.