Hey Guys!

I'm currently a computer science student and my 3rd Project involves developing a online chess game. My goal is to learn a lot more than I currently know about distributed systems and produce the best design possible for the infrastructure of the game. For this I thought I could create my own client-server framework where I could address all the communication, synchronization , security and performance problems and then build the game on top of it.

I wanted to know opinions from people who have tried on whether this is being too ambitious or even the best way of approaching the problem. I really wanted to get a deeper knowledge about DS and maybe get to make a framework that other people can come to use.

If anyone can give me a hand with this it would be awesome as my supervisor doesn't seem to want to meet me until marking time !

Cheers !

I agree with hplus, I think is a doable project, also it seems interesting and fun to me, so go on! (can I suggest you to create a developer journal in this website so we can follow your progress, if you eventually start that project?)

I usually enjoy developing libraries and frameworks (sometimes more than games themselves, haha). When I develop a library or a framework first of all I think about how I would like to use it (i.e. when I'm developing the client, I would like to call the API that way, and have that methods and for the server I'd need that thing and so on), so I would recommend you to start the same way, also you should look at libraries that do similar things to get ideas. I don't know what languages/tools will you use, but KryoNet is a nice open source library for server-client communication written in Java. And try to keep things as simple as possible.

But be aware that the more time you spend developing the framework, the less time available to develop the game, so I only would develop a real framework if that's the real focus of your project.

A good goal, but you should probably write online board games (and start with simple ones) and take the reusable parts to make a framework. Otherwise you will likely miss features that are required and implement others that aren't.

Also, what kind of board games do you want this to be for? What about card games? Many board games have both cards and dice. Also, 2D or 3D? Or do you consider the graphical frontend separate from the game itself (probably a good architectural choice)