I see your point, as well as everybody else’s points, however, I never said optimization wasn’t important. 1 remote event isn’t going to have an impact on your game was my point. Pointing out a typo wasn’t a good thing to do, of course, and I recognize that. I still stand by my point though, so we can all agree to disagree I guess.
1 Like
This isnt going out to just you, its going to every here.
Correct me if I’m wrong, but wouldn’t a GUI change from the server still have to replicate to the affected client(s)?
If so, wouldn’t firing a remote to change a value vs changing the value on the server not really matter at all? Both are going to use bandwidth. It might be worth further testing.
1 Like
Replication is way less expensive than packets.
1 Like
I don’t understand what you mean by this. Packets are sent in the process of replication. Please elaborate.
Edit: I read up on this further and it’s pretty interesting. Remote event firing is handled via the same system through with the engine handles replication. According to the Task Scheduler they’re both handled under “Replication Receive Jobs”. So the two methods are probably about the same in terms of speed, at least for smaller applications that change less properties/transmit less data.
2 Likes
I have got some interesting results:
This benchmark was ran 5000 times each. Though still, the difference may seem large but replication is more efficient and only a little bit more costly.
(NetworkController is an optimized RemoteEvent)
1 Like
What change were you having replicate and what data, if any, were you sending through the remote?
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.