My point being: it's very easy to "switch" to Godot when you've created nothing of note in other engines to begin with. You're not going to actually ship a game, so concerns about platform compatibility, stability, collaboration and so on are significantly less important than in an actual commercial project.

I think that this is a bit of an unkind take, but also a truthful one. I've shipped a few (mediocre) games in my freelance developer days with Unity and actually finishing a game and getting it out the door is more time consuming than most might believe, same as with software and the 80/20 principle. Especially when the majority of game developers are quite starry eyed and have fantasies of grandeur that lead to attempting projects with unchecked scope creep.

On the other hand, there's nothing wrong with engines that let you iterate very quickly and come up with game jam worthy games, ones with smaller scopes, oftentimes not even 3D graphics. That's why even arguably technically limited platforms like GameMaker have also had good games developed in them that have seen success.

It's just that there's definitely a bit of a disconnect between what the loudest voices will say/praise and what the more corporate projects will require.