Ooh that is interesting. Also noticed this passage:
"When people measure actual end-to-end latency for games on normal computer setups, they usually find latencies in the 100ms range."
That is really surprising to me. 100ms is HUGE. On NetHack, sshing into servers the difference between 50ms and 150ms is enormous. If do my experiment and I'm not too lazy I want to check on that 100ms figure, your link points to more sources which I didn't check on right now.
I don't know if I'm blind but I can't tell when that article was written exactly. But it mentioned MacBook 2014 and Lubuntu 16.04 so maybe it's around mid 2010, 2016 possibly? (I should check properly if I do my experiments)
The author in the Wayland vs X11 used mouse flicks on the video. Just now while writing I kind of had a thought if this needs some more deep thinking: how do I accurately measure "my finger touched this button and X milliseconds later character moved". Wondering what to consider as the "finger touched this button" event in a 240fps video. Maybe I can do what the author did and map a key to a mouse instead, because maybe there's a much-easier-to-see physical movement with a mouse. But then, am I really measuring faithfully my old NetHack-like environment? Or maybe this isn't really an issue and it'll be easy to tell.
Too much for my small brain to think in the moment :) that link gave me at least some validation that iTerm2 maybe really is a bit slow, or at least was at that time. There's also bunch of terminals on that page I've never heard of.
Today iTerm2 is my daily driver, but xterm is still really fast on my Fedora Linux system I have on the side.
adeon|1 year ago
"When people measure actual end-to-end latency for games on normal computer setups, they usually find latencies in the 100ms range."
That is really surprising to me. 100ms is HUGE. On NetHack, sshing into servers the difference between 50ms and 150ms is enormous. If do my experiment and I'm not too lazy I want to check on that 100ms figure, your link points to more sources which I didn't check on right now.
I don't know if I'm blind but I can't tell when that article was written exactly. But it mentioned MacBook 2014 and Lubuntu 16.04 so maybe it's around mid 2010, 2016 possibly? (I should check properly if I do my experiments)
The author in the Wayland vs X11 used mouse flicks on the video. Just now while writing I kind of had a thought if this needs some more deep thinking: how do I accurately measure "my finger touched this button and X milliseconds later character moved". Wondering what to consider as the "finger touched this button" event in a 240fps video. Maybe I can do what the author did and map a key to a mouse instead, because maybe there's a much-easier-to-see physical movement with a mouse. But then, am I really measuring faithfully my old NetHack-like environment? Or maybe this isn't really an issue and it'll be easy to tell.
Too much for my small brain to think in the moment :) that link gave me at least some validation that iTerm2 maybe really is a bit slow, or at least was at that time. There's also bunch of terminals on that page I've never heard of.
Today iTerm2 is my daily driver, but xterm is still really fast on my Fedora Linux system I have on the side.