Ultimate Steamworks Networking - Assets Project (Fish-Net, Mirror and Netcode)

Flem100 said:
I am not sure if I understand correctly, but are we discussing leaving out the "self"-Host option? View attachment 11186
I really love that when I test, so I hope it's not that :)
Click to expand...
No, I had deactivated the smooth move of the host in server authoritative model without deactivating the host option, just for testing on the development side, but I fixed it, all work ?
 
Last edited:
denisfrs said:
That's great news! Keep it up.
I'm not sure i understand the "Dynamic Authoritative Model" though.


I agree that would be good to make a poll about what people want / are willing to pay for. In my case, i need the Server Authoritative with Fish Net.
Click to expand...

In Dynamic Authoritative Model, you can choose the model or put to Auto* Authoritative Model perfect for the tests development side.
*(launch in host = client model, launch in server only = server model)

The others DISABLE options is define compiler symbol for disable/delete on compilation the unless code, perfect for optimise and protect your game.

Screen - Authoritative Models.png
 
For those wondering why so much time I had to recode the base of the integration to be able to create the client side prediction of abilities and item abilities, now when the client fires he sees the muzzle flash and hears the detonation but the impact is on the return of the server to all Clients.
It takes time to get a quality integration, I have to see the changes internally with Opsive but only once my work satisfies me.
 
If you compare the features of your asset with the PUN Multplayer add on, how big of a difference are we talking?
 
Flem100 said:
If you compare the features of your asset with the PUN Multplayer add on, how big of a difference are we talking?
Click to expand...
The Client Authoritative Model is the same as PUN (where I had to create the Host Migration which is included in PUN) which allows cooperative play.
But the real change is the Server Authoritative Model (where I have had to create the Client Side Prediction and Server Reconciliation which kept creating delays for me) which allows to create a Dedicated Server to play competitive games.
The advantage with my integration is that the 2 Authoritative Models are included, secure and removable (with the Defines Symbols).

If anyone has any other questions to ask, even for more details on what I just said, don't hesitate
 
Last edited:
Pesadelo do Espantalho said:
The Client Authoritative Model is the same as PUN (where I had to create the Host Migration which is included in PUN) which allows cooperative play.
But the real change is the Server Authoritative Model (where I have had to create the Client Side Prediction and Server Reconciliation which kept creating delays for me) which allows to create a Dedicated Server to play competitive games.
The advantage with my integration is that the 2 Authoritative Models are included, secure and removable (with the Defines Symbols).

If anyone has any other questions to ask, even for more details on what I just said.
Click to expand...
That is an important advantage I agree. Is the asset going to work in the same way as most if not all existing network-frameworks, with a NetworkBehaviorComponent attached to all game objects that need to be synchronzied?
How many of the game features will be supported in the first version: Walking around, shooting, melee, interaction, walking on platforms, walking upside down (gravity), switch from 1st person to 3 person and vice versa? How about the different addon packs, like swimming? I think you mentioned earlier, that riding and driving will come later. I don't know if all that is actually supported in the PUN multiplayer asset btw.
 
Flem100 said:
That is an important advantage I agree. Is the asset going to work in the same way as most if not all existing network-frameworks, with a NetworkBehaviorComponent attached to all game objects that need to be synchronzied?
Click to expand...
Yes it depends on the network library, Mirror must have a NetworkIdentity but Fishnet and Netcode can have synchronized child game object components.

Flem100 said:
How many of the game features will be supported in the first version: Walking around, shooting, melee, interaction, walking on platforms, walking upside down (gravity), switch from 1st person to 3 person and vice versa? How about the different addon packs, like swimming? I think you mentioned earlier, that riding and driving will come later. I don't know if all that is actually supported in the PUN multiplayer asset btw.
Click to expand...
First I will try to make ALL the basic Abilities (and Item Abilities) work (the 2 perspective include).
Knowing that I will have to review my Client-Side Prediction and Server Reconciliation for Abilities Ride and Drive...
Once the integrations (Fishnet, Mirror and Netcode) released I will take care of the integrations of the Inputs (which will surely be released not long after because there is not much to do).
Then I would take care of the Add-Ons (swimming etc) if needed.
Then I will add content in my integrations, customizable menus, customizable game modes, etc..
And I will see if other integrations with the Controller require special integration with my networking integrations (probably vehicle integrations).
When everything is in place I will take care of finishing the integration with the Ultimate Inventory System.
 
Pesadelo do Espantalho said:
Yes it depends on the network library, Mirror must have a NetworkIdentity but Fishnet and Netcode can have synchronized child game object components.


First I will try to make ALL the basic Abilities (and Item Abilities) work (the 2 perspective include).
Knowing that I will have to review my Client-Side Prediction and Server Reconciliation for Abilities Ride and Drive...
Once the integrations (Fishnet, Mirror and Netcode) released I will take care of the integrations of the Inputs (which will surely be released not long after because there is not much to do).
Then I would take care of the Add-Ons (swimming etc) if needed.
Then I will add content in my integrations, customizable menus, customizable game modes, etc..
And I will see if other integrations with the Controller require special integration with my networking integrations (probably vehicle integrations).
When everything is in place I will take care of finishing the integration with the Ultimate Inventory System.
Click to expand...

Any ETA. Im still excited to play with networked ucc, but really just core ucc and movement/combat. Is there an updated ETA on that release?
 
Hi @dpr010, how are you?
I confess to having taken a step back on the project during July,
I introduced the MoveToward lately, I could release it in this state,
but I would like to make Ride and Drive compatible with the Client-Side Prediction and Server Reconciliation,
before releasing the project.
 
Pesadelo do Espantalho said:
Hi @dpr010, how are you?
I confess to having taken a step back on the project during July,
I introduced the MoveToward lately, I could release it in this state,
but I would like to make Ride and Drive compatible with the Client-Side Prediction and Server Reconciliation,
before releasing the project.
Click to expand...
Hey, maybe you can release this as a beta package so we all can buy it and start to test and give you some feedback.
 
Last edited:
Pesadelo do Espantalho said:
The problem is that I have a lot of changes to submit to Opsive, I need a final version before I can send them... :unsure:
Click to expand...

Speaking for myself here, but, as a beta package i wouldn't mind make changes on opsive's code to test / research how i'll implement your asset on my game later on.

Anyway, just suggested that because i'm really interested on using a self hosted solution for my game (fishnet) :)
 
denisfrs said:
Speaking for myself here, but, as a beta package i wouldn't mind make changes on opsive's code to test / research how i'll implement your asset on my game later on.
Click to expand...
Sorry, I don't prefer to take money from you, before Opsive has made the changes and approved my integrations, it could harm me... :eek:

denisfrs said:
Anyway, just suggested that because i'm really interested on using a self hosted solution for my game (fishnet) :)
Click to expand...
Thank you for your interest in my project, I'm going to try to work on it, but I also have to enjoy a bit of my holidays, because afterwards I have to get back to the Uis and to finally be able to really work on my game. :)
 
A "final version" must include ride and drive? Well then you have your answer. But if you can release the package without those to begin with, but let them arrive in a future update, I would prefer that. As long as the core engine is done and basic features are implemented and you feel like the architecture is so good that you can build upon that, then you should be able to release the package.
 
denisfrs said:
Speaking for myself here, but, as a beta package i wouldn't mind make changes on opsive's code to test / research how i'll implement your asset on my game later on.

Anyway, just suggested that because i'm really interested on using a self hosted solution for my game (fishnet) :)
Click to expand...
Same for me.

Personally I dont even need the ride or drive.
 
You must log in or register to reply here.
Top