I've noticed a number of times since getting that putting things into child events seems to slightly slow down performance compared to leaving things as regular, non-hierarchical events. Of course, one would assume that the opposite is true, since part of the purpose of child events is to save Fusion from having to repeatedly check the same conditions over and over.
I built a quick test today (attached) and, unless I made a mistake somewhere in the code, the results are concerning. On my PC at least, the old method of events blows child events out of the water in terms of performance. The test uses the Microtimer extension to measure performance in microseconds. Using the built-in profiler also shows child events to be slower (though the difference is less pronounced according to the profiler). Is this a known issue and, if so, is it being worked on?