i have already been advised to not use blueprints so much 'cause they have a big impact in performance, i wanted to know where using blueprints dont have such effect in-game performace and where programming is a better choice
Purely event-driven logic without heavy math is a great use case for Blueprints. Things that need to run every frame can become problematic but also don’t have to.
You’d be surprised how much abuse the Blueprint VM will take before it impacts performance a lot. If you’re not currently having performance problems, maybe just put that consideration off until you do.
I mostly agree with @Teddy_Police
The only part I disagree with is the last bit- don’t wait until you’re having performance problems to improve performance. Optimization is something you should do throughout development- it’s often really hard to find what’s causing your performance loss when everything is bodged together.
You don’t need to go hard into optimization with bit manipulation CPP, but try to keep to best practices.
Ie event-driven logic, as Teddy mentioned. That’s one of the biggest ones to adhere to.
Doing heavy calculations every frame should definitely be in CPP if you can manage, but Blueprint is good for most things.
Great blueprint is way better than bad CPP. Because of their visual component, it’s often easier for newcomers to create good code in Blueprint.
Though I still recommend looking into CPP- not because of performance, but because of all the cool things you can do in it that you can’t in Blueprint. Pointers, one of the most versatile and amazing tools in coding don’t even exist in Blueprints. You do in some way use them for objects and pass by reference, but they’re heavily limited. (though references aren’t actually pointers, they’re the closest blueprint has)
Not even mentioning that there are a lot of things in unreal that just aren’t exposed to blueprints. Ie the entire team system for the perception component.