The feature I long for the most is LAN support (local dedicated server). Tbh I don't know how big of a task it would be to implement localy hosted LAN party support. I feel that this is what keep me from actually playing this game regularly, because just driving around alone without a task or objective is something you grow tired of in an unfinished game. You can still do street races or destruction derbies without this being implemented in the game yet, so I think LAN support would breath new life into this project. I'd appreciate it if the devs could shed some light on this.
Yeah good point. It's certainly not going to be easy to synchronize the nodes that make up a vehicle. But I think you could push about 1000 nodes over a local area network. You would only be able to update about 40 times a second but it should be enough to get things started.
Even LAN may (at this time) be too slow to run BeamNG over it. However, I am looking forward to local machine multiplayer. (not just to play with friend(s), but to control multiple cars at once)
There are hundreds of threads on this same exact topic. Please use the search bar to get your answer before posting a thread.
Good point, I didn't think about the challenges related to the physics engine. Now I'm even more interested in what the devs have to say about this, because this makes me wonder if or how LAN multiplayer will ever be implemented. Somebody should make a PCI Express bridge device that'll connect several computers together, that would be cool. 32 GT/s versus 1 GT/s for 1000BASE network. I did in fact use the search bar and didn't find a single one about this. This thread has yet to be answered and I think there's a potential for further discussion here.
This guy is a perfect example of how shit the BeamNG community is. Instead of possibly giving input, he just complains that there is "hundreds of threads" and "just use the search bar". Back on topic: I'm guessing LAN will be implemented along with the regular MP. I agree, driving around alone without tasks does get somewhat boring from time to time, but tasks and missions have already been discussed by the devs to be added at a later time. I'm sure they will focus on finishing the game with a proper single player, then work on getting a multiplayer set up.
Your search is for 'multiplayer', which, in the common post-splitscreen era terminology, generally means long-distance, internet/server-based multiplayer. The OP is referring to more local or traditional multiplayer. Hardly apples and oranges, but at least it's not a "Can we haz multiplayer?" user. Edit: Nevermind, nixed by Six. Short version is, LAN has the same limitations server-based has. Also, half of the threads that show up in search are followed closely by users expressly stating that the OP should search around more. I just looked through a half-dozen threads, and didn't see even one direct answer about the technical limitations. I hate to tell you this, but you're highly unlikely to attract the attention of the devs. In most cases, informed community members stand in for the devs when answering direct questions, which is precisely what has happened here. And I'm not quite sure what discussion you're expecting to see occur from this point on.
However TCP and UDP do not differentiate between LAN and online play. The code for both is identical, or rather the code for both is actually 1 piece of code covering both circumstances. LAN implies Online and vice versa. While there are LAN only games and online only games, thats usually to do with whether dedicated servers are released or not and what the in game server browser supports, or if the game is server based at all. Fundamentally at least though, the LAN only games can be played online and the online only games can be played LAN (its just you can't exactly run your own private Titanfall server). As such, from a development point of view you can interchange multiplayer and LAN as you wish. A search for LAN covers multiplayer, a search for multiplayer covers LAN. I did some maths on the subject of sending a covet JBeam over LAN. Tdev did respond to that, said it was wrong, but that I was right with it being prohibitively high. My calcs did seem off but did exceed the 150mb/s of wireless n and thats before you account for wireless n bandwidth being shared between all devices. My calcs did assume that you tracked every node, every beam and at the full 2000Hz physics rate though. I think the reality is that they would use a much lower update rate and not track a full JBeam.
My apologies, then. I'm hardly very adept at technical knowledge. I just figured that shorter distance = less latency. I imagine there's a fixed delay factor for latency when a computer picks up information from a network, inasmuch as one would need to consider layover, ticketing, and security clearance into a plane trip alongside the actual flight time? Wouldn't a lower update rate mean a heavier vehicle? I mean, I know there'd need to be compromises to make the impossible happen, but I imagine that'd involve remaking the current vehicles to, effectively, ROR standards. This reminds me that I was once planning to create a thread about multiplayer, from the standpoint of when certain technological hurdles could be crossed to pave the way for its feasibility. Moore's Law and such. I'm not sure if I will, but at least I know that there'd be some tangible data to discuss.
Shorter distance (or more accurately less hops) would of course be less latency, however the number of hops between client and server is virtually ignored for both TCP and UDP, ultimately the game code would use one of the other (more likely UDP in this situation), server has an IP address, client tries to contact that IP address, client simply doesnt give a flying fuck how the data gets from the client to server. While internet is higher latency, the client can still theoretically contact the server. There is not usually a fixed delay factor. Some games use what is known as prediction to compensate for latency, but that is an actual thing programmed into the game itself, not a feature of networking (which tends to ignore latency, although networking interfaces may select the best route based upon known timing factors - this is beyond the scope of even windows itself though). Lower update rate would not effect physics, simply how often the game would update other clients of changes in its physics, say calculate 5 "frames" of physics but only broadcast every 5th over the network instead of every single frame.
Did I ever claim there weren't any? Or did I say I couldn't find any? Your arrogant attitude doesn't contribute to anything useful... I didn't expect much really, but was worth a shot. As previously said I actually did try to search for it, but I guess I made an error in that. Maybe the search somehow got limited to a specific part of the forum. What I really wished for was some thoughts from the devs and some realistic estimation on how and when we might see this (from a technical point of view).
There is a rather good sized chance we will not see it for a long time, if ever. It is hard enough to sync the cars being simulated on one CPU, let alone over networks, so interactive multiplayer is already kinda limited. Then the pure bandwidth required is rather large, even for a wired LAN network. Add that all in with the fact that the game isn't finished and there is a lot more to develop and I honestly don't see it within the next two years or more.
I don't know if you were being sarcastic or earnest with this, but assuming it's the latter, I just thought I should inform you that I wasn't quite agreeing with you. Yes, people should search before creating new threads, and the endless questions pertaining to multiplayer do get extremely tiring extremely quickly. My point above, though, was that answers are surprisingly hard to come by using the search functionality. Perhaps there is some kind of informational form that can be triggered by the word 'multiplayer' appearing in a new thread's title, or just a page in the wiki to prevent some of the above from happening. At least for the smart people. The only way I can see it being implemented at all within the next decade, is if all of the vehicles are simply simulated with a powerful on-site server, with inputs coming from, and rasterized visuals going to, the clients. Of course, input lag is the killer there.
Wasn't being sarcastic... Understandable In my honest opinion I think that the team needs to either do something with the current search engine on this website OR hire a small team of managers that can maintain this website and keep it in order. This would include revamping search engines, making changes to miscellaneous things, and/or changing the theme once in a while. I doubt the main developers have a whole lot of time to check the forums and make sure they're running smoothly. Let alone, reply to people's questions... Not saying they should leave, just a small team to take the load off of the developers who are trying to actually "make" the game... Anybody agree? Just my opinion for better feedback among the community and an overall improvement in the way things run around here.
Hire a team? More like turn a couple of level-headed community members into moderators. Simple enough. I see nothing wrong with the search bar as it is. I don't see why the theme would need to be changed. The company's colors look fine to me. The community already does a pretty good job of being helpful.
The PCI one is a good idea! Maybe the devs can pair up with a hardware manufacturer to get a card that will transfer the N/B over that so we can have kind-of-LAN multiplayer.
