Reply To: Syphon QTKit and IP Cam Inputs

Home Forums Syphon Syphon Implementations – User Syphon QTKit and IP Cam Inputs Reply To: Syphon QTKit and IP Cam Inputs


Looking over the changes now and just so I understand all changes made here some statements for correction if needed:

a) I don’t need a framebuffer to be able to draw at all (which I thought) having a context is enough to create and modify a texture in OpenGL.
b) I don’t need to update the context as I am not technically changing anything that needs to be updated (e.g. painting meshes etc.) but only set this one texture buffer.
c) I don’t need to make the context current as I am the only one managing it..? This is the thing I was wondering most about.. Its clear that for one capture session I’ll get served from one thread but couldn’t it be that another capture device arrives on another thread? Is having a reference to the context and being sure you are the only one using it enough to be able to use it? Is makeCurrentContext just for “thread locking” the context?
d) The use of GLMacro is just more efficient when not using all the apple-related GL stuff I suppose? The “switching GL context” is caused due to the class I used being apple-specific GL use?

Again thanks very much for the help, its much appreciated, this is what I love OSS for 😉