More like
“You can have both worlds if details do not matter and you do not need a 1:1 accurate result”.
The reason why it works is that instead of having 65535 components on screen, when you scale up by 5 its very likely that a huge amount of them get occluded entierly.
(Not 65535/5 likely much more than that. On a 2km landscape you barely ever see more than 300 components save for flat under developed maps).
This would indeed improve performance.
But you’d have to put in meshes for just about everything, and use decals to apply forest paths and the like.
Whatever you save performance wise ends up spent in both development time and rendering meshes…
Not sure if it’s worth the effort when you can easily go the extra hour and just use meshes alltogether…