FVector_NetQuantize network performance?

Hi!

Should you use e.g. FVector_NetQuantize/FVector_NetQuantize10/FVector_NetQuantize100 when replicating? Couldn’t find any information about it? How much will it affect performance to switch to those types for replicated variables and RPCs or is it something that is handled internally in the engine?

Thanks!

The reason to use any of the NetQuantize vectors is to reduce the amount of data that needs to be replicated. The comments on the types are pretty self explanatory, but in general the numbers simply deal with the amount of floating point precision for each of the vector’s components. If you don’t need exact precision, say for a simulated cosmetic type, than there is no reason to not use one of the lower precision vectors to reduce your network data overhead.

The reason to use any of the NetQuantize vectors is to reduce the amount of data that needs to be replicated. The comments on the types are pretty self explanatory, but in general the numbers simply deal with the amount of floating point precision for each of the vector’s components. If you don’t need exact precision, say for a simulated cosmetic type, than there is no reason to not use one of the lower precision vectors to reduce your network data overhead.

Okay, I see. Thanks!