Originally posted by TheOne
View Post
Furthermore, the API design is to be 'few calls/transitions' between VM and engine. To clarify: say an action like you want to animate moving an image across the screen. Instead of querying the scripts every frame for updates, you specify transformation chains, like "i want to move this image from here to here in this amount of time, tweened with this formula" and it is up to the engine to calculate the current position and update whenever it is reasonable. This means the engine can optimize for different scenarios, like emitting fewer frames when on low power or prioritize input latency and animations aren't fixed to some 60Hz nonsense, if the screen has a better refreshrate - so will the animations.
Leave a comment: