Facebook Multiplayer Game

[MexicanBandit]

Hi there, I am looking to make a turn-based multiplayer board game on Facebook. I am new to making Facebook games as this would be my first. I'm just looking for any advice or any routes/theories you would recommend.

 

The game would be,

1. Turn Based

2. Play against friends or find someone in the world to play against (Could use request dialogs between friends)

3. Keep records of wins/loses

4. Chatting between the two users

 

Any suggestions or links to would be appreciated. I have some knowledge with the use of JavaScript, PHP and MySQL 

 

Thanks

 

[Gio]

PHP and MySQL aren't really suitable for this sort of thing, you should probably look into node.js, mongodb (via mongojs) and socket.io. If you are familiar with JavaScript this will all be straightforward, as it's all JS-based and very easy to use.

 

Having done something like this before (see here), I would suggest one thing above everything else: don't make concurrent multiplayer a requirement for your game! Concurrent multiplayer generally doesn't happen until you have a decent number of daily users (depends on the game, but say about 5,000 or so). If that is a requirement, you are going to have a hard time building up an initial audience for your game.

 

You know how you say "find someone in the world to play against"... if that "someone" needs to be connected and playing your game, the chances that you will actually find anyone to play against are very, very slim until you have a farily large user-base. During the first weeks or months, this is a very real problem, and it keeps being a significant problem even after that (eventually you'll have small group of friends that play together regularly, and it's going to be less of an issue... but still).

 

If you can make a "play-by-email" sort of thing, it's going to be much easier. The infrastructure to support your game is also going to be much easier: socket.io is very nice, but scaling a socket.io app isn't exactly the easiest thing in the world. In a non-concurrent scenario, socket.io isn't even strictly needed.

 

Client-side, of course I would recommend our WADE engine (which is free) plus its isometric plug-in (almost free ), which is what we used and simplifies a lot of things. But I may be biased, having written most of that myself.

 

The main challenge on the client side, I would say is designing a UI that works on a large screen as well as a small one (landscape and portrait). Most users will be playing on desktop if it's a facebook game (I have posted some stats here), but you will have mobile users too, especially if you allow your game to be played without facebook (so it will end up on html5 game protals, etc).

 

 

There are some technical challenges in making a nice facebook-based board game, but I would say that, by far, the biggest challenges are non-technical and related to the design of your game and UI, integration with social features, building a cost-effective marketing campaign on facebook, and so on.

[MexicanBandit]

I'm quite new to these technologies so I'll have a look into them

 

I forgot to mention that I already have the game created which I have to develop to use on Facebook. One problems is that it was originally intended to be played on one PC alone so I have to find a way to allow the user to make one move and then send a request to another user player and that they can only them make one move. Allowing the user one move is my most current problem.

 

Can I ask when you found the statistics from?

 

Thanks for replying!

[Gio]

Those stats were collected with Google Analytics and refer mainly to the month of June 2013, which is the game's first month on facebook.

 

Following that, we made the game available on HTML5 portals (making all the facebook stuff optional), and the stats changed quite dramatically as a result: mobile platforms and Android in particular became much more prominent. As of today, it's 40% Android, 28% Windows, 20% iOS.

 

If you don't expect users to be playing this in real-time (but you think they'll be making one move, then closing the game and waiting for the other player to move), then it's quite different and certainly socket.io is no longer a requirement - in fact, you may get away with PHP and MySQL for something like that (although node and mongo would still be more efficient).