So i know the developers have said that incorporating multiplayer into beamng drive would be almost impossible to do, if not already impossible. But has anyone considered making it a mod? Instead of trying to re-do and insert new code to make beamng drive multiplayer compatible, what if someone just makes a mod for it? They already did it with ATS (American Truck Simulator) and ETS2 (Euro Truck Simulator 2), multiplayer known as TruckersMP, so why not beamng? I don't know, it was just an idea I just had. I'd try and do it myself, but i have neither the time nor the technical know-how to pull it off. Heck, i can barely install mods.
I mean...in a way, it could work, but this is BeamNG Drive. Torque3D (the engine for the game) AFAIK would barely support this, and at the same time, the lag would be unbearable. But you do have a point, TruckersMP exists, so maybe it could be attempted, but who knows. All I'd say is to be patient on any dev updates with it.
There is only one reason why it is impossible Lag I'd like you to, right now, spawn 15 cars in beamng, ok? Can be any car you want Now, make them all t series... Put them on AI And try to drive one That is the reason why, as multiple vehicles in beamng drive really like to slow it down, if not, crash beamng drive, to give you an example of the nessicary computing power needed for a large multiplayer of, let's say, 60 cars at once, you would need the computer just as powerful as the one NASA uses to send astronauts to the ISS So, multiplayer, even 5 cars at once, can be a pain, and, the ability to slow down time would have to be removed, along with pausing and rewind
Alright, multiplayer would be neat, but you’re the eighteen billionth person to suggest this, and both @Car8john and @Daniel W. (MJ4Lyfe) have quite valid points.
Wow. Why did no one ever thought about that? "Lets not have the devs who know how the game works make a multiplayer component, let people who have no idea how to work on this specific engine do it for them!
A) People have done a botched multiplayer mod. BeamNG has been researching it AFAIK, they definitely were at one point. The community tends to be the one saying it's impossible more than the devs. B) I'm not sure which is worse, thinking that modders could be more capable of creating this than the company who created the game and have been researching it, or the apparent ignorance of the existence of a search bar.
This doesn't look like "we can never add this" to me (From Something is still over the Horizon) They were researching it at some point, as I said earlier. Not sure about the current status.
another very important thing that would make BeamNg multiplayer quite hard to make and play would be internet lag. Just look what happens when on ETS2 some brazilian decides to go on EU server with ping 800... Everything around goes flying. When you'd put this into very phisics oriented game it would be a huge mess with vehicles getting stuck in each other.
I think that there has been at least 5 people I have seen suggest this. Not only would Beam not be able to load enough cars to make multiplayer even fun or interesting, you wouldn't be able to have mods engaged because someone who doesn't have the mod installed wouldn't have it show up. I know in FSX SE add on aircraft show up even if you don't have them installed, but I don't think that that is the case with Beam.
ENDING THIS UNORIGINAL IDEA NOW! Ok, first, before i forget, BeamNG Multiplayer will mean BeamNG To set up a game server, which can cost thousands if not millions of dollars, and it will have to moderated, and how can mods moderate a site AND a game? Third, the lag would be crazy. On grid, small, pure, i get like 15 fps when i have 5 cars spawned. For us toasters, the lag would cook our toast. Fourth, Internet costs. You or somebody else pays for your internet (unless you steal your neighbors.) and BeamNG online= Internet costs to go up, not down. 5th, How would it even bee programmed? It would take years and years of hard work to complete it, mods would probably not work, mods would be harder to create, and etc. Multiplayer is a TERRIBLE Idea in my book of updates, Let's just stay happy for what we got, or look into career mode!!!
Ending suggestion purely to add multiplayer, yes, suggesting ideas on HOW to implement multiplayer, no. (I kinda like reading other people's visions on how X should be implemented). Sometimes if a very large group of people start to think about solving things, they might just find the answer. Your first and second point is already invalid, BeamNG doesn't HAVE to host the server, you could locally host it, for just your friends... ow... wait.... (just kidding) Your third point: valid. But I would already be happy if some kind of MP like the kind in Rigs of Rods was implemented, just driving with your friends... racing etc, without the physics... I mean... crashing is fun but there is more to MP. Your fourth point: maybe your one of the unlucky few with bandwidth limitations and having to pay for every bit you consume. But many of us have unlimited data to use for fixed price per month and just a fixed speed. Your fifth... if your not capable of figuring out how to "program" multiplayer, don't shove it of as some kids imaginary dream. Like I mentioned before, if you think hard, you might be able to figure something out. It's called "research". Some are happy to do it. If we were all to stay happy with what we have, we would never have something called progress... But that's my view on it, your entitled to your own.
The devs have researched how to implement multiplayer, and have run into a wall with the amount of data needed to be transferred each second. This has been discussed many times, and needs to stop. Please don't call people out for not contributing ideas when you don't do so yourself.
You want an idea? Ok, here we go (this has been on my mind for some time now, and might still not be feasible): You have a skeleton network with nodes and beams around a vehicle right? And thats the tricky part, because if you would send over every node's position you would indeed get into huge issues. But what if you take only a few nodes and send those over? Like on every corner of the car (you would talk about aprox 16 nodes). Then, when a collision happens, you take those nodes and let the other nodes react around those nodes (because they are connected). Its like crunching. Ofcourse you would have delays which would mean clipping. You could make the local player like a non physics object (aka normally you would drive through it). If the local player on the other end crashes into another player, the other end calculates where those 16 nodes end up, and transfers those back to the local player, making it look like the local player on the local computer crashed into the other players car. Data wise that really shouldn't be an issue. I don't know which values a nodes brings along, but I could imagine these needs to be send over, and I am going to assume the types here: 1. x = float = 32 bit 2. y = float = 32 bit 3. z = float = 32 bit 4. force on the node = some value possibly 32 bit possible direction of force: 5. x_f = float = 32 bit 6. y_f = float = 32 bit 7. z_f = float = 32 bit so one node would contain 224 bits. x16 = 3584 bits = 448 bytes. Now, the biggest issue I think beamng has what causes this bandwidth problem is the fact that (i read it somewhere) the simulation runs at 2000 times/s. But the idea i mentioned above could act as another skeleton above/around the existing simulated node/beam skeleton, running at it's own refresh rate. So you could probably take a refresh/tick rate of around 60 (which would be ideal): that would be: 60*3587 = 215040 bytes = 26880 bytes = 26.25kbytes / s. That should be achievable with even the slowest connections in my opinion (well, of course there are some that have even less, but most users would have more than 1 mbit connections = 1 000 000 bits pers second = 125000byte/s = 122.07kbytes/s ). If your worried about delay/ping issues, you could make this extra skeleton a bit larger than the car itself, making it look like the initial contact is happening after the delay has happend. This would obviously not be perfect, inducing clipping, still weird collision issues, but its a start, and it would at least give some kind of interaction between players. I could also see some syncing issues happening, but on a local network, this would work in my opinion... but hey... what do i know. That said, I already mentioned, for me interaction with physics between players is a nice addition for me, I would already be happy if we had a multiplayer system such as rigs of rods where there was no physics interaction between players, but you were still capable of playing together (you only saw a static model of the player), this would also make it performance wise possible, since you don't have to run physics calculation on other cars/players. But that's my 2 cents.
That wouldn't work like that because the cars would not crash well and errors would constantly freeze physics. You simply cannot reduce the 2000 fps that the game needs to run at to simulate metal. This has already been discussed before and every time the thread has gone down the drain, so I am abandoning this thread.