waterphysics are off

Not a bug, it's how the game's engine works.

First of all, real life water would act normally in this case. The force of gravity is stronger than the force keeping the left-side gases from expanding. It would only start making sense in real life if all of that gas on the left side was removed, but still, ONI's engine still wouldn't allow it.

It's still possible to do what you're trying to achieve, though. Mass is disregarded when calculating gas displacement, so you can take this to your advantage and use a combination of two gases to block off one side from rising.

Every tile of oxygen, carbon dioxide and chlorine here are at 1 kg. The water naturally gets pulled down by gravity but eventually stabilises to the same amount of both sides.

Give it a while. It works mostly, but is nowhere near as fast as in reality.

Uh, I'm sorry to tell you this, but your "In Real Life" picture isn't what would happen in real life (particularly when the containers have open tops). ONI is doing what reality would do. If one column is taller, then the bottom of that column is higher-pressure than the bottom of the other column. So water flows from the taller column to the shorter column, until they equalize in height.

49 minutes ago, creidieki said:

Uh, I'm sorry to tell you this, but your "In Real Life" picture isn't what would happen in real life (particularly when the containers have open tops). ONI is doing what reality would do. If one column is taller, then the bottom of that column is higher-pressure than the bottom of the other column. So water flows from the taller column to the shorter column, until they equalize in height.

That statement is only true if both columns have the same gas pressure. If the gas pressure is different in both columns, then the height of the water level will also be different.

I agree that it would be nice if ONI had better physics simulation in that respect. But it is not easy.