possible bugs and glitches?
I think the invade earth client may have a ram leak somewhere in it. Or maybe it's because of the way the gameplay saves every single turn?
The client process starts at around 30-40 megs, much like the server. However, after two games and 26 turns, it was nearly 90 megs... and continued to rise until every player was disconnected from the server.
The army usage/placement slider is fantastic, but it was a little glitchy on startup. I dunno if it was a one-time thing or not.
Assume current player only seemed to last for one turn. When a player gets disconnected, it's next to impossible to get back in their spot. And the feature that pauses the agents is kinda buggy too. I think it has something to do with the gui updating the actual variable in real time. When you try using both the hot keys and the gui it gets confused.
Ideas and suggestions:
It would be nice to be able to right-click and interact with the players in the game so that you can change AI at will... pause specific players, and replace particular AI's (especially ones that used to be players) with players in the room or yourself.
Adding a column in the master server list for # of connected players, so you can tell if a server is empty before you join.
And last but not least... new game>from url might be helpful in allowing players to open a custom game from within an online library. I'm currently hosting two custom maps from my computer available for download @ 24.163.77.145/invadeearth
Thanks again.
Re: a few ideas and some possible bugs or glitches
Thanks, I'll definitely take those under consideration. I don't know anything about a memory leak, I can check that out later....
As for replacing players/AI, unfortunately that would require a lot more rewriting of the program. Those options were really kind of experimental... the players controls just weren't designed that well originally. Invade Earth was the first project of its kind for me, and so it's been a learning experience. My main goal now is to make it as useful as possible (for the people that do use it) with the little time that I have to spend on it. Ideally I'd start over... but it's a good 20k lines of code, and that would take a long time even with the knowledge I have now.
So thanks for testing and suggestions and stuff. I'll add what I can when it comes to nifty tweaks and bug fixes (if I can figure them out), but there probably won't be a lot of huge updates etc.
I've never had trouble with pausing agents, though. I thought the synchronization was worked out mostly on that one.