I'm not sure if something similair has been asked but I'm having a bit of "what's the point momment" with regard to multithreading a game I'm working on.
I did have a physics thread and an opengl thread which worked fine ish... To be honest I had to put the physics thread to sleep a lot as it was overrunning the opengl thread a lot. To try at get the thing running as fast as possible I thought, well I only ever wan't the update thread to run once per opengl thread so I decided just to update the physics thread from the opengl thread directly(calling physicsthread.update() from openglthread.draw) and it now runs as fast as possible without the physics thread overrunnning the gl thread...
It's ugly though, is it the best way to do it? Would I see any benefit of splitting the threads out again... I'm struggling to see the point when it's a single core processor to be honest.... Hmmmmmm.