I'd like to play around with some MMO code. Can anyone recommend any free or affordable pay engines I can use? What I'd like to be able to do is hook a single player engine I've written up to one. Also I don't know if it matters but my game engine is in C++ and C++ is all I know atm. [Edited by - Davaris1 on July 15, 2008 3:22:01 AM]

I had a look at the FAQ but I'm still not sure where to start. I saw Beej's Guide to Network Programming, but I'm not sure if I need to study that, because I don't want to reinvent the wheel. Should I start with Beej's Guide and then move on to a library or skip it go direct to a library?

A networking library is a world of difference from an 'MMO engine'. What exactly do you want to achieve? Bear in mind that making an MMO is rarely as simple as just taking a single player game and adding networking.

Quote:Original post by Davaris1
I had a look at the FAQ but I'm still not sure where to start. I saw Beej's Guide to Network Programming, but I'm not sure if I need to study that, because I don't want to reinvent the wheel. Should I start with Beej's Guide and then move on to a library or skip it go direct to a library?

A quote comes to mind: "The ability to speak does not make one intelligent".

While understanding the issues related to sockets and internet protocols never hurts, they are merely a single API related to dumb transfer of bytes between peers.

But all the complexity of non-trivial networking comes from many other directions. Knowing how to sendto() over socket will bring little benefit when using a library, since that will be the very first thing any library abstracts away.

And if focusing on sockets, thousand little problems pop-up just related to that.

Rather than choosing between raw socket API and socket abstraction, look into game engines. All of them have many limitations, yet they provide a high-level approach to actual problem at hand (replication, state integrity, persistence, logic, scripting, client/server interactions) with minimum fuss.

One common problem with such engines is characteristic limitations - there will always be a feature that is just hard or impossible to implement, or there's an error which you just can't solve. That may or may not be an issue, but getting full control over all aspects is simply a massive task in shear work required, even if considering you're familiar with all the underlying concepts.