Max Thiele

Ah I probably did not implement them because I never needed them, should be pretty straight forward though

I just wanted to check / fix this and realized that the uniform[*]iv mappings already exist since a while. Am I missing something? Only the unsigned variants are not implemented...

The problem is the Vec2 array uniform. When the Uniform2fv puts the values to the buffer it expects the buffer to be in Std140 layout, meaning that the individual Vec2f...

Not thoroughly tested, but I pushed a [commit](https://github.com/fabmax/kool/commit/d822bf512063eede43431c66c00c68e019fa9c35), which should fix that

As far as I know infix functions only work between two arguments, so, witrh an infix return, you would need to write something before it to work: ```kotlin kslFunctionBody.apply {...

Yes you can declare it as an infix function, but it doesn't make a difference because you can't invoke it 'infix-style' as long as there is no argument left of...

I myself have absoluetly no experience in iOS development, therefore I haven't looked into it so far. Obviously iOS doesn't run Java / kotlin directly, but it should be possible...

In general this sounds interesting. However I probably won't be able to contribute much to an additional rather sophisticated project (due to time-constraints...). First, lets see how your iOS implementation...

ANGLE definitely looks interesting. Using it as a compatibility layer on iOS would solve a lot of problems. Lets hope they implement it soon.

The track is not loaded from a model but procedurally generated. It is defined by a few control points in `VehicleDemo.makeTrack()` (line 181 in [VehicleDemo.kt](https://github.com/fabmax/kool/blob/edef672ca0f491b5b612203be7a9e102d411993b/kool-demo/src/commonMain/kotlin/de/fabmax/kool/demo/physics/vehicle/VehicleDemo.kt#L181) ) The first Vec3f defines...