@DTS Engineer thanks for replying. I indeed also opened a bug report, under: https://feedbackassistant.apple.com/feedback/21965760 Do you have any updates on this topic? This is having an impact on all SpriteKit apps, which is obviously major issue for apps already published to the AppStore.

Good to see this already reported, I am facing the same experience with my SpriteKit games. I am very eager to see this fixed/ investigated asap, as I rely on SpriteKit in all my games which are published in the AppStore. I am very curious to know what is causing these performance hits, and whether we can do something in our projects to mitigate the effects already. @DTS Engineer Would it help if I file the same feedback report in order to prioritize this issue?

For reference, the project is also available here: https://github.com/sanderfrenken/sktilemap-rect-in

Here are all the leaks reported by Xcode after a new instance of the scene is presented:

Hi @DTS Engineer , Thanks a lot for your reply. That is unfortunate, but maybe an enhancement request will provide a solution. I did also post the question on StackOverflow, and there is a way to make it work indeed. https://stackoverflow.com/questions/79128004/call-to-main-actor-isolated-instance-method-run-in-a-synchronous-non-isolated I think that is the same solution as you refer to when you talk about: very fragile assumptions that I won't recommend you make In practice, I do see this could work if you really have no other option, but sticking with Swift 5 seems more reliable indeed. I will create an enhancement request as you propose, thanks again for your help. Sander