Multiplayer?

I was wondering if Kongregate had any plans to offer any sort of Multiplayer API / socket-server for developers to make multiplayer games for the site. It’s not really something I’ve seen too often—the only times I have, it’s when the site developers are creating their own multiplayer games for the site rather than it being a community / user-generated setup. It might be more trouble than it’s worth for Kong, but was thinking it could be interesting to have some sort of multiplayer integrated with the chat / user-accounts, etc. that developers could use for some Kongregate-exclusive multiplayer games.

Any thoughts on the subject?

We really want to do this. We definitely plan on having a Multiplayer API in the future. Our current plan is to finish our own multiplayer game first (the CCG) and then leverage work we’ve done for that for an API. The current estimate for that game is May, so hopefully this could be something we’re working on this summer.

Awesome. :) Glad to hear. Though, I understand that it would be a extremely huge hit on the servers, and might not be feasible. Hope everything works out!

Boxhead : More room online, it’s my dream… :p

I definitely want to do this at some point, as I think multiplayer flash games have a lot of untapped potential. We definitely have a lot of experience on this team with online multiplayer games, so I think we can come up with a decent solution for this.

Yay.. I’m excited to hear that! I will definitely make use of it when it comes out.

That would be Great! * dreams about Choas Faction Mulitplayer *

I wouldn’t put alot of err… faith in real-time multiplayer flash. Its simply almost impossible.

Turn based would certainly be possible though.

I wouldn’t put alot of err… faith in real-time multiplayer flash. Its simply almost impossible.

I’m sorry but thats just not true.

There are tons of huge multiplayer flash games out there like http://www.dofus.com/ and so many others.

dofus is a great game because french people created it (3 guys from my school actually :D). this game is huge and they have even bigger games coming up soon

I wouldn’t put alot of err… faith in real-time multiplayer flash. Its simply almost impossible.

Maybe you just don’t know how to do it then, because even with my limited experience in socket server-type stuff, I managed to put together a (very simplified) two-player version of Q-Zoid that actually runs kinda decent. I said kinda, but I think a real, smooth networked multiplayer experience is not too far off for Flash, especially considering the phenominal demand for it and the ever-improving technology. IMO, it’s only a matter of time…

dofus is a great game because french people created it (3 guys from my school actually :D). this game is huge and they have even bigger games coming up soon

That’s awesome Pix! I saw the trailer for their new game and the art is amazing. The gameplay is not my style, but the game is popular and the best looking 2d MMO out there.

Well maybe things have improved. I certainly haven’t bothered with it for awhile.

Too much synchronising issues to deal with. I’d rather stick to turn based.

anyway. Don’t need to bash me. I said simply ALMOST impossible. I didn’t say it was out of the picture. Stop looking at my post count. There may be plenty of so called multiplayer games out there, but how many of them actually work? Dofus works because it isn’t just flash. It is probably compiled with zinc. Now why do you think they need an installation file? Most multiplayer flash (swf, not exe) I’ve seen all have had crappy connection issues, and really screwed up synchronisation. Now there may be one or 2 of them that’s actually worked really well. But don’t forget that there’s more to the picture. Alot will depend on the kong servers (already stressed out), and also on the implementation of the API. The API will be the absolute factor in this. If it is crap, everyone’s multiplayer will be crap. End of story.

That said, I hope I am proven wrong, and Kong might come out with something great.

Multiplyaer Kongregate-exclusive games would be great. It would be awesome to have your user account play a role in a game, and make a character and duke it out. I really see where you’re coming from on this one, I support you. Great idea.

Dazzer, I dont think anyone is “bashing” you. I’m also not sure who is looking at your post count or what that has to do with anything as well (?). Whether you said almost or not, its still an incorrect statement. It is difficult, but its not “almost impossible”.

Additionally, just because there are not many multiplayer Flash games out there doesnt mean its nearly impossible, it just means many developers dont have the skill or have not had the budgets to explore optimal solutions. I dont think Zinc is the determining factor for Dofus working as it is really still just Flash. Whether or not they used Zinc and Zinc’s broadcast protocols, I dont know, but its not needed to create a multiplayer Flash game; you can do it just the same on the web.

Too much synchronising issues to deal with

All multiplayer games have latency and synchronization issues to deal with, whether they are Flash, C++, or anything really. Such is the nature of the web. Those who get it working well have simply learned to deal with it using tricks and clever programming.

Alot will depend on the kong servers (already stressed out)

You would not use the same servers for a multiplayer system as your http or database servers, if you want anything at or above acceptable performance. Both Jim and BenV had previous jobs at pogo.com, and Jim has a lot of prior experience (read his linkedIn page) so they know a lot about connection optimization. Jim especially has experience in packet optimization. I doubt they would dump all the multiplayer server software on the same servers we are using now.

The API will be the absolute factor in this. If it is crap, everyone’s multiplayer will be crap. End of story.

The API is not the sole determining factor, certainly not the absolute one. The servers, the server software, server locations, how the developer chooses to call the API, as well as the implementation of data protocol on the other side of the API all have determining factors on the baseline performance. Additionally, good front end programming is used to mask unavoidable internet latencies. But yes, if the actual server communication protocols implemented are bad then the multiplayer will be bad. However, just because they are bad at times doesnt absolutely mean an API is bad or that Flash is not capable. There are many many factors.

Many people have a conservative attitude towards expanding technology, and thats fine. If you think the idea of multiplayer Flash is bad because it is a bit difficult then ok, but certainly dont discourage others! The only way technology improves is because some will refuse to accept the “we cant do that” mentality and choose to push the envelope until a way is found. I remember back 5-6 years ago Flash developers stated it was impossible to build a pinball or physics engine in Flash; they said you just couldnt do it – so no one really tried. Our company didnt accept that, so I built one. I doubted as well at first, but I’m sure glad I didnt accept the popular vote.

The same goes for any of this. The envelope needs to be pushed so that it will become more feasible. AS 3.0 will hopefully allow it to be easier as well. We should all be encouraging it as it opens up a lot of possibilities for a greater gaming experience.

Well, I don’t want to sound like I’m bashing on anyone. Dazzer, it seems like you have a lot of good dev experience and I’m not doubting that. Actually, the multiplayer experiment I was referring to was made possible by other people involved with IndieFlashArcade (not myself), through their PHP skills. The thing is, even though it was laggy, it still ran, which makes me think it is possible to do better.

Additionally, good front end programming is used to mask unavoidable internet latencies.

I think this is the major realization that we came to in regards to multiplayer Flash… don’t rely on the technology alone.

Like i said

I hope Kong will come up with something great. And yes , AS 3.0 will allow things to be easier as well. But, as far as I understand, Kong is still firmly in its Flash 8 roots. And while I know that there is some effort being made to push into the AS3 realm, I think that would not only require quite abit of an overhaul of the current system, and also you got to make sure your target demographic is ready to download new players if they haven’t already. So, with my current experience with Flash 8 real time multiplayer games, I just won’t put too much faith in it. That is why I am devoting my efforts into AS3.

I Guess i should rephrase my original statement.

It is possible. But like Indie said, it took a whole team to get it working, and even then it was still laggy. I just don’t see it being something that can be done by one person. Not to forget that we’re looking for quality content here. Noone wants to play a laggy, poorly implemented game feature. Sure, it is possible… but is it something you want to play?

I’d rather have something that works well. Sure, I’m really not looking to revolutionise the industry. That’s just my view. Note I ain’t saying that anyone is wrong. You have your opinions, I have mine. It would be great to see someone do something awesome, that has not been done before. Patrick Mineault and Aral Balkan, for example, with their own Remoting solutions.

Show me a multiplayer browser game that works in real time. Because I certainly haven’t seen one. I remember there was a tank wars thing… but that was turn based. There was Artillery wars. That was also turn based. There was a french site full of blobs blobbing around a field. That was turn based. That RPG game I always see on the ads… that’s not even multiplayer. There was Stick Arena. But there wasn’t any lobby system, the match making system didn’t work very well, and you could spend your day slashing in thin air, and getting blown up by people that were invisible. Even MSN Messenger Games, which runs on Flash, had some problems, though I must say they probably did a good job, even if its only 1 game that is real time. Solitaire Showdown.

If you guys want to set up a sort of team to work on this stuff, I say great. Go ahead. I certainly didn’t mean to discourage anyone here. Just know what sort of absolutely insane stuff you’re heading into. I guess I was inaccurate in thinking most people here would be independent developers. And trust me, you probably have more experience than I do.

Seeing as I don’t even have a game to upload, I’ll keep my mouth shut now :). Hope I haven’t offended too many people hehe.

Ive made a multiplayer flash game, which was mainly an experiment to see if I could do it: http://69.56.136.242/quadrantwars. I have a dedicated server to run it.

I think having a shared multiplayer server that lots of people could use would be tough unless you are not expecting much traffic. Things slow down quite quickly, especially if people havent done their flash code that efficiently

The reason that I like Kongregate is because we can have discussions like this, and I think everybody has a right to their opinion so I certainly don’t take offense ;)

... like Indie said, it took a whole team to get it working, and even then it was still laggy.

But, to clarify, it was only two people. I do work with an awesome team, but I am generally the Flash coder. Additionally, it was completed in a matter of hours, not days or weeks.

The point that I was trying to make was that it takes more than just a server technology to make a good multiplayer in Flash. We used PHP to handle the data transfer over a socket connection, but there are many ways to do it. The hard part is compensating for the lag which is inherent in this type of game. Coder mentioned this, also. Ultimately, the game itself needs to be able to compensate the lag and make everything appear smooth to the player, even if it’s actually ‘lagging.’ That is where the real work is, and I/we did not take the time to do that… (and by the way we are independent)

Show me a multiplayer browser game that works in real time.

Click Upon Dots is an excellent real-time multiplayer game right here on Kongregate, which was developed by the very talented Jiggmin.

Quadrant Wars is a good effort, but visibly lags just as much as our own experiment, if not more so.

Unfortunately I can’t show my example to the public because I don’t have a dedicated server.

Well implemented, I give it that. But hardly very complex is it? I suppose, for a game of this level of complexity it is okay. Its not using kong servers is it?

So… what I can say is , the current technology allows a game of this complexity to be realised without a hit in playability. But as the level of complexity increases, so does the playability of the game.

Thus, the direction to take would be to try adding more complexity, to judge the limit of the current technology.

Out of curiousity, what method do you use to get “real time” data? Unless the server is “pushing” data to you, you would have had to call data every few frames, no?

Its not using kong servers is it?

Nope, Jiggmin runs his own server I believe…

We used PHP to handle the data transfer over a socket connection

It’s all about sockets my friend :)

So you had to run a socket server on your own server? :) I did have a look around for open source servers… red5? and smartfox.

But I do hope Kong comes up with something extra ordinary.

Or at least something that’s sufficient for turn based games – sticks out a tongue -

Upon dots is , i think, just an experiment… Seriously this game is not great….

Show me a multiplayer browser game that works in real time. Because I certainly haven’t seen one.

You’ve got to look more! But thats also kind of the point, that there needs to be a pioneering in the area since not a lot of people have been making multiplayer Flash titles, let alone doing them well. Its time to step into that area.

Stick Arena is another. It uses MMOCHA, which is one of the fastest real time Java sockets servers made specifically for the purpose of real time multiplayer gaming in Flash. The XGen guys who built the technology are my partners and we are currently working on several titles. 2DArray and Luksy are also working on Stranded 2 with real time multiplayer features, as well as a few other developers on other projects.

Also, something else noteworthy is Multiplayer Snake. Its a real time Flash to cell phone game, which is even more difficult (no one ever seems to play though because the Entelepon guy got lazy and didnt do well in finishing the promotion and the rest of the project. Regardless, it works).

I agree with you though that the general quality of the games may be poor if inexperienced independent developers attempt to do it without really doing it well, and thats definitely a problem. I think that the whole multiplayer feature should be limited to only some game experiences made by experienced devs, instead of just letting every mcreds bog down the servers with their “games” ;). Trust me when I say that Kongregate is well aware of it, since we were discussing much of this almost a year ago before the site went up when I was spec’ing and building the original Flash technology and guidelines for the site. Now its just time to wait and see.