pixels_per_meter and gravity

[Novakasa]

Hi all, I am a but confused trying to understand the pixels_per_meter setting. Reading https://rapier.rs/docs/user_guides/bevy_plugin/common_mistakes#why-is-everything-moving-in-slow-motion it all makes sense to me. pixels_per_meter controls the scale of the world and thus should make gravity apply differently for different scales. However, when running the "get started" example (2D) from https://rapier.rs/docs/user_guides/bevy_plugin/getting_started_bevy#basic-simulation-example, changing the setting doesn't seem to change the effective acceleration of the ball on-screen.

The way I understand it, gravitational acceleration should accelerate with g=9810 pixels/(second^2) when pixels_per_meter is 1000.0 compared to g=981.0 when the setting is at 100.0. I was thinking maybe the mass also changes when I use a different pixels_per_meter setting, but that wouldn't really make sense since the gravitational acceleration should be independent of mass.

What am I missing?

[Novakasa]

This was answered in the discord server - Gravity strength is also given in pixels/s^2, so you have to set the gravity consistently as well.