Hey Guys, I'm still worried and confused as to how PA will handle mods in multiplayer in order to not fracture the player base. It's a problem all of PA's predecessors had and I'm wondering how PA will overcome it. If, by enabling any mod, vanilla players are restricted or hindered from joining or seeing my games then mod implementation of PA has failed IMHO and whatever I wanted the mod for will have to be in vanilla for it to get played. Does anyone, uber or otherwise, have an answer or idea for this?
Actually, PA does improve on other systems with the client-server model. PA's modding system is an improvement over other games because client-side UI mods can exist all on their own. You can go ahead and download any of the mods in this forum without needing to worry about what other players are using. The other type of mod will need to run on the server. These mods change the game simulation itself, so the server will need to be modded. It is entirely possible that some server mods will be small enough (balance tweaks, for example) that the client will not need to download anything new in order to play. The third type of mod, and possibly the most common once the game is released, will be large enough that both the client and server will need to be modded. This can't really be avoided, because there isn't really another way to introduce new units into the game. The server will need unit stat information and the client will need the model. The best we can hope for is that Uber will try to reduce the situations where both the client and server need to be modded. Hopefully the game lobby will be able to help us make sense of it all. I know that balance changes on the server still require the client to have the new values so that it can do a better job of predicting the action between updates, but I remember reading somewhere that they might be able to have the server send out that information before the game. This is a complicated answer for what is looking like it might become a complicated system, but a good lobby interface can go a long way to helping clear up the confusion.
Wouldn't we be able to use the model that for instance Enemy Territory uses. When you connect to a server that has mods installed, it will automatically download the necessary client mods and then join the server.
I hope that's an option for mods that are small enough, but I expect some mods to end up being huge. Automatically downloading the mods is incredibly useful, but it gets less practical when you have to download a few hundred mb of data before you can connect.
I'm really hoping to hear from Uber on this. They said mod support will be deep in PA, and I keep getting people tell me "oh just make/wait for a mod" for whatever feature I'm asking for. If that's not viable I need to know before the game is released so I can push for that feature in vanillia since thats the only way anyone will end up using it. Honestly I would rather have the game be completely un-modable if having mods meant only other people with that mod can see / join my game effortlessly because any other solution splits our playerbase which is unthinkable!
True. Let's say one has a total conversion, replacing everything. Currently the media stuff takes 1.1 GiB of the 1.4 GiB, which after compression is about 415 MiB. I can't find a reference for this anymore, but I believe the minimum requirement on the Internet connection is 2 Mb/s, so that's roughly 0.24 MiB/s. The mod would then take just under than half an hour to download. That does seem a bit long, but not unreasonable since 2 Mb/s is the bare minimum. It would be best to alert the user to this fact before automatically downloading the mod of course and stress that it needs to happen only once. Additionally, a link to the website of the mod could be shown so the user can download it later and just play a round on a different unmodded server for now. Actually, I've also seen a number of 4 Mb/s floating around. In that case, it would take only 15 minutes.
Actually, we can do whatever we want in the lobby as a client-side mod. I'm sure we'll get a better lobby from either Uber or modders. There are already a few lobby mods out there.
I would want the downloading to occur at the server browser level though. The other players in the lobby might not want to wait 15 minutes (or longer) for you to finish downloading. That does mean there would be a good chance the game would have started before you are done downloading, but you'll be ready for next time. This is also why I suggested additionally showing a link to the website of the mod, so the player can download the mod later and just play on a different unmodded server for now. I agree the lobby is clunky right now, but like cptconundrum I also think that will be improved during the current beta phase we are in.
All of that sounds like a big "Nope" when i put myself in the shoes of someone browsing the servers looking for a game. It needs to be more seemless than that. Maybe have mods register with Uber and have them download as part of the next update to the game? Then when in the lobby hosts can turn them on and off at will like the built in "2x resource" mod in Sup com FA.
That's pretty much how FAF does it, but centralizing the mods doesn't really lead to more openness. I think there is a lot less of a problem than you're saying there is. The 15 minute wait to download a mod does a lot less to fragment the community than simply the choice to play a different mod. There will always be this pro/con discussion when it comes to mods for games with smaller communities, but modding is one of the core principles behind this game and is not likely to change. Luckily, this game actually has a lot more players than anyone ever expected. Garat has said that there were over 30,000 users logged in at some point in the first 8 days of January.
I, too, was part of the kickstarter lol. It is known, but actions speak louder than words and now Uber needs to put their $ where their mouth is. The game isn't even out yet! Have for foresight man... 10 years after this game comes out what will the state be? When Uber stops supporting us and it's all up to the modders, thats where I'm afraid of fracturing because that is where it hurts us the most. Right now I play FA on steam. That automatically means i don't get to play with anyone on FAF because: DUH DUH DUH! its a mod... Which I have no problem with and I wish i could play with them but it's too late for FA... I honestly don't think centralizing the mods is great either, but it's a heck of a bunch better than anything else I've heard suggested yet.
I agree and you'll be downloading a lot of stuff you will likely never use. Especially for larger mods, which is the problem at hand, this can easily double or triple the patch size, right? I kind of envisioned the mod downloading from the server browser like this: So you have an icon for each mod the server uses. The icon includes a progress bar indicating if you have that mod or how far along it is with the download. Mousing over the mod icon could display the name of the mod, a description and a link to the website for manual downloading. This seems pretty seamless to me.
I'm not sure what you mean. They are doing an excellent job supporting modding. What you seem to be asking for is either less modding or a centralized system that relies on uber to distribute the data. Once uber stops supporting this game, what will happen to their centralized mod server? The updates to the mods will stop coming in and people will need to rely on the old fashioned way of getting mods anyway. I'm sure someone would come along and do what FaF did at that point, but in that case can't we leave it to the next people to deal with this problem? (If it really is a problem anyway) Again, client-only and server-only mods won't even be an issue at all. No other RTS that I can think of has allowed for this, but PA is built from the start with it in mind.
If everything goes well there will be a myriad of mods out there. No way Uber can validate (check for viruses, etc) them all and include them into patches. There just is no way around you having to download that 500mb total conversion before you can play it. If there is a game hosted right now with it and your internet takes 10 minutes to download 500mb you will have to wait that time. No way around it. Uber however can make it easier on you by providing: - A simple, automated way to download any missing mod before you join a game - A simple to use mod repository where you can install mods that look interesting. If you want to be ready you could just install all mods as they pop up (assuming a system to enable/disable them) and thus have them ready later on. To my knowledge they have very advanced plans for this direction. Time will tell what comes out of it.
no... I'm asking for an explanation on how we can avoid ever fracturing this community because of mods and obviously I gave a shitty suggestion myself. If I had a great suggestion I would have shared it to begin with... That is why I'm worried. Just because all past games failed to deal with this isn't an excuse for PA, since I got the impression from the Kickstarter mods would be a 1st class citizen from the beginning. Denim, thats a nice thought but it really requires mods to be recognizable and have a reputation. If i make a map, anybody can join and play it and realize its an awesome map. If I make a new mod, If I don't have a website and a team and some sort of community built around it already I doubt anyone would bother to try it.
But your awesome map suffers from the same problem, doesn't it? The map is downloaded automatically from the server you are hosting (or one of Uber's servers). A mod would also download automatically from the server you are hosting (but not one of Uber's). You don't need to have a website per se, but you do need some method of getting the mod to the users as cola_colin pointed out. Either the mod is downloaded directly from your game server or it has to be hosted on a website somewhere. You have the same problems regarding recognizability/reputation. How do you advertise the awesomeness of your map/mod? Showing the mods in the server browser avoids fracturing, because everything is visible to everybody. It's up to them to invest the time to actually download the mod, which would take only one click and some waiting.
+1 To avoid vendor lock-in, the mod distribution should work without Uber's intervention, so that we can play the game even after Uber is no more. A game server could tell the game client that what mods it has and an url where to download it (plus a sha1 of the mod pack to make sure that you'll have exactly the same version of the mod). That way no central server will be needed (except for the list of game servers, but that should be configurable after the server software is released).
Judging from the js parts of the lobby I would wager that, once we have access to the server, this could be modded in. The whole lobby after all is a webpage. You could easily load that from anywhere and put in any content. Including a "download mod here" thing. Only issue would be how to trigger a mod installation from within PA. It would be cool if PA had an API for that maybe.
But having access to the server, doesn't give you the ability to change the client, right? I mean the lobby is indeed just a website, but it is located at the client. Unless you mean that the server should have the ability to load a custom lobby web page in the client. Also, this requires the user to actually join the lobby first. What happens when the game starts while a modless user is still in the lobby? Wouldn't it be better to do this from the server browser somehow and prevent user from joining games they don't have the required mods for yet? I fully expect PA would eventually come with its own built-in mod manager making my and raevn's mod managers obsolete. Of course, one could mod the mod manager in that case.