Organized NS
homicide
Join Date: 2003-11-10 Member: 22451Members
I am hoping to develop a website and plugin that facilitates organized NS2 games through pick-up-games, scrims, and matches. The primary goal is to reduce the overhead associated organizing private games and make it accessible for casual players by removing the need to download or monitor additional software.
<!--sizeo:1--><span style="font-size:8pt;line-height:100%"><!--/sizeo-->
- Servers running the plugin are automatically registered with the central ONS database
- Registered players get reserved slots in public ONS servers
- Allows servers to verify player identity for pugs, scrims, and matches
- Central server keeps track of player pug, scrim, and match record
- Registered players can communicate across all servers and read real-time announcements
- Registered players can create, view, and join pick-up-games without ever leaving the game
- All players in public games are notified when a pick-up-game has started
- Once enough players have joined the pick-up-game they are automatically transferred to an automated private server
- Private servers running the plugin can be reserved for pick-up-games, scrims, or matches by registered teams
<!--sizec--></span><!--/sizec-->
I am looking for people interested in working on this project (programmers) or people planning on running servers when NS2 is released.
<!--sizeo:1--><span style="font-size:8pt;line-height:100%"><!--/sizeo-->
- Servers running the plugin are automatically registered with the central ONS database
- Registered players get reserved slots in public ONS servers
- Allows servers to verify player identity for pugs, scrims, and matches
- Central server keeps track of player pug, scrim, and match record
- Registered players can communicate across all servers and read real-time announcements
- Registered players can create, view, and join pick-up-games without ever leaving the game
- All players in public games are notified when a pick-up-game has started
- Once enough players have joined the pick-up-game they are automatically transferred to an automated private server
- Private servers running the plugin can be reserved for pick-up-games, scrims, or matches by registered teams
<!--sizec--></span><!--/sizec-->
I am looking for people interested in working on this project (programmers) or people planning on running servers when NS2 is released.
Comments
Players will have two options:
1) The website
2) In-game GUI
Depending on the power we have in LUA (which is promised to be vast) I am hoping I can make an in-game GUI client-side using a server plugin. The game server would then communicate with a central server (the website and mysql database).
I am considering using HTTP requests from the server side LUA plugin to the central web server to fetch xml data.
However, this might have too much overhead if we want to have "chat". I considered using a continuous low level socket connection between the game servers and the central server. This would require me to write my own custom server, I was thinking Java based.
The method of using HTTP requests to send and receive xml data seems like the most flexible option and by far the easiest. This works well for most of the plugin features. It would also parallel the website backend and cut down on development time. So thats still my plan thus far.
Still looking for suggestions if anyone has a better method.
There are different paths to take in this matter.
Dedicated program - not prefered. You'll need to constantly update it, make sure distributions get to people, version mismatches, high maintenance, etc. Another con is the possibility of dodgy programs and peoples paranoia in that regard (warden anyone?).
Plugin - Easiest in usability for the players. However, every feature will have to be defined and perfected. This again is high maintenance. Plus you can't really get started untill you get your hands on an alpha or beta-build. Pros will be that the integration could be seamless, providing an in-game experience. No external load on any server if you can use steam/steamcommunity/steamworks API's.
Server-side (website) - prefered. Instead of providing a fully integrated experience, you can provide an environment in which users can meet and then set up the game themselves. By using game and voip clients launch syntaxes, you could simply provide a checklist for people in the gather. ESL has a nice system in place that closest matches my description. Another pro would be the guiding of traffic to the site itself. Cons would be the rent and load on the server.
Its just a matter of how deep you are willing to go and how much stock you put in binding players to a community or integration into the game.
Agreed, except it is not noob friendly. They have to download special software, know how and what server to join, know how and what channel to join.
I'm sure we would be interested.
I doubt CEVO will pick up NS2; they seem to support only a few extremely popular games.
Given that NS2 will be running its own engine and its no longer free I doubt it will have the same community overlap that NS1 had with TFC and CS. I suspect much of the NS2 scene will look to NS specific community sites. Europe has ensl.org and Australia/Asia has nsplayer.net. I imagine they will start NS2 leagues and have a good foundation of dedicated players moving over from NS1. A few people have mentioned the possibility of nsplayer.net branching out to the US community for NS2. There is certainly no longer an obvious home for a competitive US community like CAL. Basically, I have no idea, but I am hoping to support whatever site picks up a competitive US scene.
We already have ENSL for European players. If you are interested in developing an american side for our website, go ahead. I'm ready to give any decent (dedicated, fair and reputable) person full rights to organize a competitive community for NS for the people over the pond. When I made the website, I really wanted to automize as much as I could, which means that I hope I can make it as easy as possible with NS2 (with NS1 I have <a href="http://www.ensl.org/rounds/5749" target="_blank">per-round statistics</a>).
About pick-up games, you should check out our website-based <a href="http://www.ensl.org/gather" target="_blank">gather</a> system. There you only have to click and steam will autoconnect to the server. You have to join the voice com manually though. We have ENSL plugin which uses HTTP to connect to Apache HTTP server to fetch player data but this is not necessary for pick-up games.
Web is the future, IRC is too oldschool for pick-up games.
Good ol' days.
:D
IMO, if they are short in time, they would be wise to implement the client side API and allow the community to develop something good that they can then incorporate into the game. Just as the game can incorporate the best community maps, it could also incorporate the best community client side interfaces, whether they are HUDs, configuration interfaces or server browsers.
For example, NS never had a config-per-class system even though it was often asked for. With a simple system for constructing interface components etc anyone with half a clue could build such a system.
If you look at WOW today and how it has, over time, incorporated several of the most popular plugins into its core feature set ( most prominent being questhelper-like features ), I think this approach makes sense.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->eXtine: Speaking of Left 4 Dead, L4D2 utilizes a lobby system for joining games and this feature was mentioned as possibly making its way into TF2. What are your thoughts on a 6v6 lobby system in TF2 and the status of any sort of implementation?
Robin: The guys behind the L4D2 matchmaking system are working on broadening it out into generalized steam matchmaking. Once they have it figured out, we'll look at how it'd fit into TF2.<!--QuoteEnd--></div><!--QuoteEEnd-->
<a href="http://communityfortress.com/tf2/news/robin-walker-on-competitive-tf2-and-future-updates.php" target="_blank">Interview here.</a>
Hopefully it will work with all steamworks games?
I don't have L4D2 but I've played L4D1 and they've basically made it so you can click a join game button for your chosen map and it will automatically put you into a Warcraft3 custom map style lobby where you can pick teams etc. You can also make a game with any number of friends and then let it place random people to make up numbers. I think there might be an option for your team of 4 to play against another random team of 4 as well so you don't just get randoms all the time.
You might be able to make things even smoother with closer dev support, but short of having something built into the game this seems like an idea solution.
Also, an IRC gui plugin might be handy (autojoins a specific channel) - especially for getting newer players into organized games and clans.
Beware that plenty of noobs don't understand this whole IRC thingy.....
It took me a while to figure out that if I wanted to play at a more competitive level, that's where I had to go. And then there was figuring out IRC for the first time. and figuring out the IRC interface people had built....
Suffice to say while nice and easy (yay command lines in general) it's got quite the initial learning curve for the uninitiated.
Why do one system when you can have two at twice the price (kudos to anyone who knows where that's from).
Player stats would probably be better for a game like NS2 but that still leaves the system open for exploitation and could cause the games to be nothing more than people watching their K/D.
The other option I can think of is a system where your teammates vote on your performance. Clearly this could be exploited with friends constantly up voting themselves or a random cool guy down voting everyone.
An option that combined all three of the above ideas would be the most ideal but also extremely complex.
It's also not particularly nice being a scapegoat. For example drag a terrible team to near victory as a commander and you might still get blamed for the loss. Obviously comms make mistakes too and sometimes deserve the blame, but way too often it's also the team creating scapegoats.
Wrong. Maybe if you do something like KPD but there's something a lot more sophisticated. Its called Microsoft's <a href="http://research.microsoft.com/en-us/projects/trueskill/" target="_blank">Trueskill</a>. It uses Bayesian probability mathematics to calculate how much a player increases the victory likelihood of the team. It has a lot better convergence properties than ELO meaning it learns player's skill a lot faster (after a few games) because of the distribution area σ stored for every player. MS uses this on xBox to make sure most of the games will be balanced.
L4D's system is great for casual gaming, all it needs is some kind of a filter so I don't end up on same random server with extra bull######.
Its hard to get servers full with random enforcement, and also the problem with dedicated public servers is it requires people to join up and play for an hour to get it running. I've personally witnessed many servers dying because those people couldn't be arsed anymore. This is a game theoretical problem easily fixable with proper match-making system.
I'm well versed in Bayesian networks and if you think for a minute this won't be abused within an inch of farcicality you're deceiving yourself. Abuse of Bayesian networks is as obvious as people who abuse relevancy measures when they mark an article as irrelevant simply because it did not solve their problem. Thus it's measure of relevancy is severely skewed by an incompetent userbase.
Statistics are not the answer, lobbies are not the answer, anything that can be easily abused is not the answer. Random assignment is the only unabusable answer.,