Someone recently asked my why we decided to go with the CNC3 engine instead of one of the more "popular" game engines out there like Company of Heros or Dawn of War 2. My first knee-jerk reaction was to say "Because the original mod was on CNC Generals", but then I thought about it.
There are a lot of different game engines out there that players can choose from if they want to build a mod. CNC is definatly not one of the nicer ones out there, requiring a player to recompile the mod after every little change just to see if it went into effect properly. Talk about annoying. Other game engines like the original Dawn of War are much more user-friendly in that they use Lua Scripting instead of XML. So really, why CNC3?
The truth is once I started thinking about it I think CNC was actually the proper choice, though we may not have realized how good of a choice it was at the time. Games like Company of Heros and Dawn of War 2 are amazing games, full of great replay value and strategic choices to make to win the game. But if you step back and tried to put a Gundam into these games, it wouldn't work. These games are designed around squad level tactics and you would end up with either games being about 4 or 5 mobile suits vrs 4 or 5 other mobile suits or they would simply be too big to field on the maps the games come with.
The other engine that was very popular at the time when we started was Supreme Commander. This engine would have actually been a good choice in a lot of ways, huge maps (I mean HUGE maps), and 2 very straight forward resources (vrs CNC's Limited Resource and As much as you have space for and can afford resource). However, have you ever actually PLAYED a game of supreme commander? A 45 minute game feels like it took 2 hours, and a 3 hour game feels like it took all day. On top of that a lot of the game is played from such a high view most of the time that you just see dots moving around on the screen. In the end, it's not the effect we were wanting to go with, on top of the bigger question of "would the models we already have be compatible with it?"
CNC3 is showing that it has quite a bit of potential left in it. The only unit cap is what your computer can handle, and the AI is being found to be easy to manipulate as long as your careful (read - make back up copies after every verified working change).
So yes, the CNC engine was a good choice, and we're definatly not moving on from it. Honestly the engine I would want to make this type of game on doesn't exsist, but if you want an idea to what it would be like try to go back and play a few games of "Empire Earth", and then imagine that with a very healthy dose of "Command And Conqure Generals" and the stability of CNC3. That would be my ideal choice.