Question Unity Multiplayer 120+ms ping two builds on the same computer.

Title says it all.
computation for ping:

Code (CSharp):

1.  
2. NetworkManager.Singleton.LocalTime.TimeAsFloat - NetworkManager.Singleton.ServerTime.TimeAsFloat
3.  

Server/Host: 0ms
Client: 120+ ms

 

Not sure if this is a true measure of ping.

And the background app might be purposefully slowed down since it's in the background. It does happen with the editor playmode for example.

 

Thanks for the reply. Both instances of the game ran on two builds. Editor was closed while running (this is what i suspected at first). the game feels like it had 120ms ping.

 

I don't use NGO so answering the best I can with that in mind.
I reviewed docs
https://docs-multiplayer.unity3d.com/netcode/current/advanced-topics/networktime-ticks/
And from the wording I suspect you're making the right calculations.

Is it possible you are using a low tick rate, perhaps around 10?

 

Hi there! thank you for the reply.

Our current tick rate is at 30. I will be experimenting on checking my ping in a new scene with bare network objects. I suspect the amount of network objects in the scene might have something to do with the delay. Otherwise, i cannot think of any other reason as to why this is happening.

However, if you don't mind me asking, I am quite curious. how much of an impact does network prediction provide for a game in comparison to setting up a dedicated server? My friend and I argued about this. I told him that, he should not worry too much about doing network prediction to enhance player experience from the massive ping as we have not yet setup a dedicated server. I find it unnecessary at the moment. I would love to hear your profession thoughts on the matter.

 

I suppose it really matters what you are trying to predict. If you are creating a combat game and important elements of your gameplay are behind based on user ping, I'd say that is probably a deal breaker as a player.

If you mean prediction for little things like picking up a piece of fruit on the ground and waiting for the server to remove the object, who really cares if it's immediate or takes 100ms.

The scenario matters a whole lot to be honest; ideally fake as much as you can only using prediction where you really need it. This will create a much easier to work with code base and less burden on the client/server.

Not sure what you mean by this though sorry, "how much of an impact does network prediction provide for a game in comparison to setting up a dedicated server"