The DataSystem::update function is invalidating some of the results

[clevijoki]

I did some profiling and std::fmt inside of DataSystem::update is 95% of the time spent executing the benchmark.

For some perspective with 2m mixed entities the benchmark takes 260 ms with pico_ecs on my machine. Disabling the last two lines of DataSystem::update will change it to take 13ms which is more in-line with what I would expect performance to be.

This makes the difference in performance across ECS libs much more dramatic.

[abeimler]

Any other ideas to do some smaller string operations in the system to fill the buffer ?

(I may try out std::to_string(), but idk maybe the string allocations is also a factor)

EDIT: Maybe I try out to make a System with something more "realistic" and useful, like a "RenderingSystem", but with a string buffer to "draw" a character into it or some other data and number crunching things.

[clevijoki]

I would avoid any string manipulation or allocation at all, most games try to avoid it. That way those things will be overhead in the libs, if they do it.

I don't think you need any real-world simulation but purely benchmarking actions under different conditions is fine.