Hey! Sorry, I only just saw this tag!
Generally, the best way to get quantitative data like what you’re looking for is to do an A/B test - build a test scene, and measure it with and without your effect. If you’re only after GPU thread costs, then you can simply use the console command “show particles” to turn particle effects on and off, but afaik that will still run the simulation and other logic on the CPU, it will just not render them.
I would probably build a small test blueprint for this, to be able to toggle stuff on and off at runtime with a keypress, to reduce iteration times.
In general, make sure you always test this stuff in a close to real-world conditions as possible - so never test in editor! Ideally test directly on your target platform, but if that’s not feasible, at least use “Play as Standalone Game” if you’re testing from the editor so that you don’t get editor overhead polluting your performance measurements.
As for those specific stats, I’m not entirely sure! RT generally stands for “Render Thread” (aka draw thread, as also seen in stat unit), and GT is for “Game Thread”. Particles typically have some costs on both of those.
Hope that helps a bit!