4 ms·
There's a difference between noticing latency and considering it a problem. My car doesn't go at the speed of sound. That doesn't mean I think "man, this car i
by p-e-w 3y ago
There's a difference between noticing latency and considering it a problem.
My car doesn't go at the speed of sound. That doesn't mean I think "man, this car is slow" each time I drive it.
But if the same car suddenly stops, and I have to open all the doors and close them again in order to keep going, that sure is a problem.
- wiseowise 3y ago> There's a difference between noticing latency and considering it a problem. Just because you don't consider it a problem, doesn't mean it's not a problem. Weren't you the one lamenting about ML having potential to being 100-1000x faster if only there were real™ software engineers to implement them? > My car doesn't go at the speed of sound. That doesn't mean I think "man, this car is slow" each time I drive it. This isn't even remotely valid example. Put your car on track against faster cars and you'll quickly arrive to `man, this car is slow`.
- thefaux 3y agoI'm sure then you'd be perfectly content if your otherwise perfect car were replaced with one identical in all ways except that the acceleration was 1/4 of the previous rate.