@Mith - IRC is the de facto choice for chatting when you don't have your own dedicated chat server. We run on shared hosting, which means we don't get a dedicated server, and any localized chat system we build is going to suffer from the same lack of PC power that our current system suffers from. Saying that we are limiting ourselves to IRC is really an invalid statement, since IRC is quite versatile. Here's some more super duper general information.
IRC Benefits:
1) Familiarity. Not only am I familiar with IRC, but so are a lot of our members. They would be able to use an IRC client without training, and I would be able to rig a client so the rest of our users would need very little training.
2) Versatility. If one of our users doesn't happen to like the client I set up, they can always download and install mIRC, or whatever other IRC client they prefer and still participate.
3) Scalability. IRC can handle the load we are demanding.
The downside to IRC:
1) Familiarity. Yes, this is a benefit and a downside. There are some users who don't know IRC at all, and will have trouble adapting. Even if I give them the simplest client, they will still whine, bitch, complain because its something new and different and its not what they are used to. This is just how people are.
2) Lack of Control. Our current system integrates nicely with our forum software, and allows us a large measure of control. An administrator doesn't have to be there the whole time to babysit because we know that only our members are going to be in that room. They have to be a registered forum member to be able to use it. IRC doesn't have this. We can't use our forum IDs to register for the chat, or see who's chatting on a forum page. It will be open for anyone to join. And if an admin or moderator isn't there, then there is no protection for the average joe user. In the past, this hasn't been too much of a concern, but it should always be mentioned.
3) Effort involved to impliment. Choosing IRC will put the chat room down for a week at least while I figure out how to put it in place. Probably longer. In the meantime, people will have to use yahoo, or what we currently have anyways.
Reverse all of the benefits and downsides for our current system. Its familiar with everyone who's used it, and quite easy to use. Its already implimented, so no effort is needed. We have a large measure of control, Its not very scalable. Its not very versatile, so users are stuck with our way or the high way.