Currently, we generate lots of cloud blocks during world gen and each of those blocks has an entity attached which is used for conditional rendering based on whether or not the player is wearing goggles. This definitely takes a toll during world gen and I suspect having all these entities is causing undue calculations to happen on the server tick.
Instead, could we build a system similar to the trapar storm (except without the "build up" and "peter out" work).
When a chunk is loaded:
Generate a set of random block positions and store them as data
Probably clumps of block positions in predefined "shapes"
This would significantly cut down on world entities but would still allow us to conditionally render shapes in front of goggle-wearing players and apply lift to boards.
Currently, we generate lots of cloud blocks during world gen and each of those blocks has an entity attached which is used for conditional rendering based on whether or not the player is wearing goggles. This definitely takes a toll during world gen and I suspect having all these entities is causing undue calculations to happen on the server tick.
Instead, could we build a system similar to the trapar storm (except without the "build up" and "peter out" work).
When a chunk is loaded:
This would significantly cut down on world entities but would still allow us to conditionally render shapes in front of goggle-wearing players and apply lift to boards.