Leon Friedrich

I made an attempt at creating benchmarks and adding some basic simd to compare Numerics vs robust vs robust /w simd in #2939. But I don't feel like I know...

> You know when this PR will get reviewed? Or is it still work in progress? I honestly thought I'd merged this before I went on holiday two months ago....

I'm still unsure about whether I'm doing the benchmarking wrong somehow, but even when just taking the benchmark functions directly from [dotnet's githhub](https://github.com/dotnet/performance/blob/main/src/benchmarks/micro/libraries/System.Numerics.Vectors/Perf_Matrix3x2.cs), the matrix multiplication seems to be twice...

I forgot Queues don't have serializers, so I guess for now it'll just not be serialized

Instead of doing this refactor, which is quite likely to introduce bugs along the way, we should just finally switch to using `System.Numerics.Matrix3x2`. We've already made that switch for vectors...

Is there a specific reason its still marked as a draft? Because other than making sure it all works I'd be happy to fix conflicts & transition to matrix3x2 this...

#5100 partially fixes it, but does not fix the assert getting tripped if you switch focus from one control to another.

A similar error seems to have happened again recently: ``` Caught exception in entsys System.InvalidOperationException: The LinkedList node does not belong to current LinkedList. at Robust.Shared.Physics.Systems.SharedPhysicsSystem.DestroyContact(Contact contact) in /home/runner/work/space-station-14/space-station-14/RobustToolbox/Robust.Shared/Physics/Systems/SharedPhysicsSystem.Contacts.cs:line 344...

I think this might just be some known issue on some linux distros where you have to install/configure openal. E.g., see space-wizards/space-station-14/issues/19671. Though IMO the launcher should probably check for...

> Is it possible to do obsoletions? I really don't want to tank breaking changes here. AFAIK I removed all the breaking changes, though content tests are still failing because...