So, if you’re asking this question, I guess the answer is: “you don’t even realize it’s happening”…
Rust marketing; since Rust has became popular, GC became a problem, out of no where, and segfaults became the most terrible bug a software can be afflicted with.
If C# and Unity are working well for them, good for them.
C# uses ref counting, which has much less impact on that. I do however use D for my game engine, which some say is infamous for its GC pauses, but they only happen if you allocate with the GC itself, and otherwise if you work work the GC, you can get some performance gains compared to dumb manual memory management, but not so much if you actually can manage your own memory, or write your own automatic memory management system, like I did on top of numem. Otherwise the pauses are negligible for smaller projects, and some of the larger projects like The Art of Reflection uses GC only for loading assets, with my engine also going in that direction.
Games might be an area where it isn’t too bad since you can disable the GC while doing all your physics and graphics, and then just after you’ve dispatched a frame trigger a GC with a hard time limit.
I don’t know if Unity works like that but it should.
Welcome GC pauses :) I wonder how do those manifests in real life.
There are plenty of games made with C# and Unity.
You surely played a few of them, let’s be honest.
So, if you’re asking this question, I guess the answer is: “you don’t even realize it’s happening”…
Rust marketing; since Rust has became popular, GC became a problem, out of no where, and segfaults became the most terrible bug a software can be afflicted with.
If C# and Unity are working well for them, good for them.
TBH I almost don’t play games and I’m genuinely curious whether GC pauses are noticeable. And not only Rust, there is Swift as well.
C# uses ref counting, which has much less impact on that. I do however use D for my game engine, which some say is infamous for its GC pauses, but they only happen if you allocate with the GC itself, and otherwise if you work work the GC, you can get some performance gains compared to dumb manual memory management, but not so much if you actually can manage your own memory, or write your own automatic memory management system, like I did on top of numem. Otherwise the pauses are negligible for smaller projects, and some of the larger projects like The Art of Reflection uses GC only for loading assets, with my engine also going in that direction.
Um, no, C# most definitely uses GC, not ref counting, and you can’t not use it.
Ref counting is still GC…
Games might be an area where it isn’t too bad since you can disable the GC while doing all your physics and graphics, and then just after you’ve dispatched a frame trigger a GC with a hard time limit.
I don’t know if Unity works like that but it should.
I don’t think that’s even possible - to have that much control over GC engine that is.
It definitely is.
https://stackoverflow.com/a/32007523/265521