Tomás Chor

> @tomchor since you're the main person using the tilted gravity feature, I'm wondering if you can help provide some insight into this "stratified fluid at rest" test. The main...

> Can't we just test directly that the output of `x_dot_b` is as expected (as well as the others)? Along with a test that the constructor works without error this...

Removed the "fluid at rest" test that was failing and included another (simpler) one for tilted buoyancy along with constructors. The new test checks that `x_dot_g_b`, etc., match for two...

@glwagner I'm trying to use this branch for my research so I tried resolving the conflicts as best as I could. I hope you don't mind. Otherwise I can always...

The errors of the `cpu-simulation-tests` are of this kind: ```julia JLD2 output writer [CPU]: Test Failed at /var/lib/buildkite-agent/builds/tartarus-2/clima/oceananigans/test/test_jld2_output_writer.jl:135 --   | Expression: wT == zero(FT)   | Evaluated: -1.2037062f-34 == 0.0 ```...

Just a note: it seems like this PR is currently changing the dynamics. For example comparing the convecting plankton example [in this PR](https://clima.github.io/OceananigansDocumentation/previews/PR1910/generated/convecting_plankton/) has significantly more noise than the same...

> Is this a GPU problem? Is it still there when running on CPUs? We had some very early issues with this. Ali may remember. > […](#) The equality failing...

We can apply 4 pretty immediately and start speeding up things if everyone agrees. @glwagner I'm curious as to why you think we should remove that example Also, do we...

I can definitely second that. Came here to post exactly that but found this issue. For me and my theme (solarized dark higher contrast) I can see the vertical separations...

Weird that it worked for you. Maybe I thought I was doing one thing and did another. If that's the case I apologize! I will definitely try to make the...