> And with reference counting the cost of RC bookkeeping is smeared predictable across the run of the program. The biggest downside of naive reference counting

> And with reference counting the cost of RC bookkeeping is smeared predictable across the run of the program. The biggest downside of naive reference counting usually isn't the cost of reference counting itself, it's being left stuck holding the bag when you're unlucky enough to be the last deref on a giant object graph. You can move the destruction onto a separate finalizer thread like what most runtimes that use tracing GCs do, but then you end up running into CPU scheduling issues if you're not overprovisioned.