Its difficult to advice with the limited context provided, but if you have additional data which you would like to inject into the simulation you could directly retrieve it on the Unity server and inject it into the simulation and let it sync to all clients via netcode.
but that highly depends on what type of game you are planning to create, what data you need to retrieve and how often that data changes etc.
A little more information on the experience you like to build would be helpful.
Server (database for storing data about objects, for now mysql)
Server (transmitter for data between users)
Client (during authorization, requests data from the server about objects)
(what type of object, where should it be, etc.)
Client “another player” (entering the room to another client receives data from the server about objects as well)
Clients (exchange activity data via server)
at the moment I have a working version of this scheme through sockets, and on the server PHP acts as a receiver and transmitter.
I still don’t know how the server should be arranged in C # and plus with entities, so I would like to understand how it should be arranged and where to start …
oh yes, and also, I don’t need to completely exchange all data such as physics animation, etc., the main thing for me is to transfer simple data such as movement and actions, a kind of minimal package …
I don’t think you would need the netcode package for that, this sounds like you are trying to roll your own networking solution.
Netcode is really intended for a full real-time simulation of a fast-paced game with a lot of powerful features (think of games like overwatch, cs:go etc.).
It will take care of syncing state, lag compensation, prediction, rollback, networked physics! etc. etc.
So if you want to create your own multiplayer solution, you probably won’t need this package, but to get you started:
transport package (to create your own networking solution on top of unity transfport which will support web sockets), this is very low level but gives you a lot of power to create anything you want with it.
If you want to learn more about dots netcode and what it can provide, its best to start with the manual, the tutorial might still not have been fully updated since the latest release, so if you get stuck, just ask here again.
Hope that helps, if that’s even more confusing don’t hesitate to ask
that is, with the help of unity and the transport package, an application is created that is placed on the server and this application is a transmitter?
can it be a regular dedicated server, for example on Linux? or somehow specially configured?
It technically can, the dedicated server can be built for Linux, and you can also feed it command line arguments etc.
But it is not intended to be used like a normal application server (with accessing a database, serving thousands of requests of different users etc.).
If you only need to communicate with a backend service, you probably won’t actually need a unity server, but you rather use directly http from unity to communicate with it.
You can also create a rest service and use one of the rest apis like the following asset:
Most games will have a backend persistent server (or multiple) to store long-term data, and a realtime server (which can be a unity server) to handle the real-time gameplay.
For real-time gameplay, the unity server retrieves information from the persistent storage and injects it into the simulation.
But again all of that depends heavily on your game architecture.