[lozank]

I agree with the final verdict, mostly because today's systems are really good at optimizing those things. Unfortunately, testing power consumption on a laptop is inherently prone to errors. You have no way to detect if a laptop uses additional power from the battery to smoothen sudden power spikes etc. And even if those tests were done on a desktop (or a laptop without a battery), you can't protect against MacOS/Windows doing stuff in the background (for example system updates, windows defender stuff etc etc)

This is definitely a super interesting experiment, just take the one-time measurements without any kind of repetition with a grain of salt..