There is a lot of stuff going on in the code, so the only important bit (I think) is: I have been bashing my head against this RPC call for a few hours now, and despite extensive googling I cannot figure out what is going wrong.Īt the moment, I am doing all testing within unity editor, (player set to run in background if that matters).įor some reason, my RPC call isn't going through. Is there something I need to do to the vertex data before I compare it to the vertices along the ray? Is my ray not aligned properly even though it visually passes directly through the cube? No other vertices fall within the min and max of the cube. LAST x:-9.43373 y:-683.9638 z:343.73978įinally, if you've stuck with me this far, here is a snippet of some of the vertices along the line. The first and last values of my ray trace (just to prove that it is accurate): So, you would expect that for example vertex (0, 0, 13) would return true and, if a ray passes through a cube defined by these values some points on that ray would have x,y,z values that fall between the max/min. These are max/min values of all of the vertices for the cube taken from the transform feedback buffer: Here are some numbers - I'll try to keep it simple but I really want to illustrate my problem. The camera, projection matrices for building the ray are used before any rotation or transform are applied. The vertex data I am passing in to the picking routine are taken directly from the transform feedback buffer. However, if I trace the ray NONE of the points along the ray fall within the boundaries of the cube. I have a cube and a ray that passes directly through the cube if I click on it. The one routine I have, which I have tried to port from an old project, did work but now the numbers are way off.
I have tried multiple ray trace / collision detection implementations and none of them work or are accurate to the point where I'm forced to assume that there is something drastically wrong. So how do I get the exact OpenGL rendering surface/buffer size ? LOGI("JNIGlue.cpp:Init:RB-Height: %d", H) LOGI("JNIGlue.cpp:Init:RB-Width : %d", W) GlGetRenderbufferParameteriv(GL_RENDERBUFFER, GL_RENDERBUFFER_HEIGHT, &H) GlGetRenderbufferParameteriv(GL_RENDERBUFFER, GL_RENDERBUFFER_WIDTH, &W) LOGI("JNIGlue.cpp:Init:EGL-Width : %d", W)
UNITY NETWORKVIEW RPC VS RPC CODE
How can I do that ?ĭisplayMetrics metrics = getResources().getDisplayMetrics() įrom the JNI C++ code I am going following with NO luck (again output: Size = 320 x 526):ĮGLDisplay display = eglGetCurrentDisplay() ĮGLSurface surface = eglGetCurrentSurface(EGL_READ) ĮglQuerySurface(display, surface, EGL_WIDTH, &W) ĮglQuerySurface(display, surface, EGL_HEIGHT, &H)
UNITY NETWORKVIEW RPC VS RPC ANDROID
Hope that makes sense, and that I'm right.On every single android device. However, if you're moving the sun across the sky, it's better to AllBuffer it so that all players get the same amount of light no matter how good their ping is. So, if you're trying to sync your player character with the rest of the world, you'd want to use OthersBuffered so that when you move forward, you will get immediate response ins$$anonymous$$d of waiting for everyone else to buffer it. I'm not the best person to explain this to you, but I do believe AllBuffered will buffer your parameter to make it more realistic to everyone, where as OthersBuffered will give you feedback based on your actual location, and buffer everyone else so that no one viewer of you has any better idea of where you are than the other. AllBuffered sends the RPC and buffers everyone including you. OthersBuffered in this case, is sending the RPC, and buffering those whom are not you. As far as I know, different buffering allows for different syncing to be applied.
0 kommentar(er)
