It's pretty hard to measure end-to-end delay, Nvidia is only getting to it now with https://www.nvidia.com/en-us/geforce/news/reflex-low-latency...