Quote Originally Posted by Paradox Uncreated View Post
Sidicas is on the right track. Later poster is completely wrong.

Indeed it should be easy to understand, that if kernel blocks for 5ms, rather than 0.2ms, timing jitter is going to be present, and alter framejitter, and also 5ms lost of 20ms, to calculate frame.

Peace Be With You.
Assuming I am the "later poster" could you please tell me how I am completely wrong?
It could be a language issue but I don't understand how you fail to grasp that RT adds overhead. Yes, less jitter (i.e. smoother frame rates), but you also lose the ability to run for periods of uninterrupted time on the cpu, which is sometimes necessary in order for a frame to be completed. Now, I could see a setup where you prioritized frame completion and delivery (that is, it responds to the irqs from the gpu before almost anything else) over everything else but that would mean, necessarily, that you'd have a less responsive system since inputs would then be delayed. That kinda defeats the point.
Setting up a RT machine requires a lot of thought.