Physics Actor Replication Blueprint

Hello all,

I’m looking for a replication expert to weigh in on my blueprint I’ve designed for important item physics actors in my game. These items are critical to gameplay and to ensure fairness, I want to force all client actors to match the location and orientation of the server actors. For whatever reason, even though I set the actor to replicate, the item can show up in two different areas in the client and host games, despite all interactions being done by replicated actors, there seems to be some randomness to the physics in the engine.

To combat this, I came up with a rather simple script to give the parent blueprint for these critical actors. There shouldn’t be more than maybe 150-200 in a really busy game, and honestly I think most rounds would have less than 40 or so active at once. Does anyone know what the bandwidth/latency consequences would be for the script below?

Thanks for any feedback!

Looks like there will be a lot of jerkiness between the transform updates. You will probably need to interpolate/extrapolate between updates.

For the Bandwidth/latency consequences, you should the use the “net stat” command and/or use the network profiler to see the impact that this has on your game. Your bandwidth/latency targets will be specific to your project (only you can define them).

I removed the delay nodes and it doesn’t seem to have a significant performance impact, yet. Thanks for the tip about the “stat net” command, I used it and didn’t see a noticeable difference in network usage. I’ll leave it as-is for now, and see how it goes later in the development cycle.

Have you tried setting “Replicate movement” to true on the actors in question?

Wow, can’t believe that I missed such an obvious flag. Cheers!

I have a similar question. Did you ever found a solution to make a big bunch of actors replicating smooth and nicely for clients also with higher frequency?