Hacker News new | ask | show | jobs
by kragen 1147 days ago
with respect to the memory bandwidth thing, typically with ddr4 ram you get 30 or 40 gigabytes a second, jason cook tells me. a single main memory copy at 530 megabytes a second eats up one of those 30 or 40. if it's 4k (3840x2160), four times that, 10% of the computer, or 20% at 120 hertz. if you go from one copy per frame to two copies per frame, instead of using 20% of the computer's memory bandwidth just to update the screen, you're using 40%. or 50% if you are also writing that memory before it goes through the two copies

if you're trying to do something else on the computer, even on a different core, that's bottlenecked on memory bandwidth (as opposed to cpu or i/o or something) that's like using 50% of the computer, which means the whole computer is effectively half as fast

i think it's worth a significant amount of complexity to get your display system to suck up 30% of your computer instead of 50%, even when you're doing something that updates the full screen every frame, such as smooth scrolling, a 3d fps, or watching a movie, and that's why I think it's important to try to minimize copies in the pixel paper path

also keep in mind that many people dual-wield monitors these days, and 3840x2160, though common, is not as big as they get