11-14-2009, 11:39 AM
Yeah, making all teams identical is a bit of a cop-out in that while it's guaranteed to be balanced, it's less interesting all around.
As long as you're willing to keep updating and tweaking the teams, you should be able to stay on top of the "flavor of the week" issue and at least squash any obvious imbalances quickly.
My personal favorite RTS is still Battle for Middle Earth 2, which launched with 6 teams, no two alike, and did a very good job of balancing them. I eventually just played as "Random" and still got to the top 10 in the official rankings.
Incidentally, I think the secret to balance in an RTS and the reason why BFME2 and to some extent Starcraft did so well is that you need unit production levels to be fairly high. Company of Heroes (and that followup Warhammer 40k game that used the same engine) had problems largely, I think, because your units were so expensive and thus so few. When the enemy came out with a Rock, if you didn't already have a Paper, you were probably screwed. Whereas in BFME2, units were cheap enough that you'd probably have some mix out anyway and if the enemy started spamming Rocks, you'd have enough Paper on hand to hold him off while you built more Paper.
I also think that having a bigger rotation than "rock->paper->scissors" helps, as well as making it a "weak" system. In BFME2, for example, cavalry mulched archers and archers mulched infantry and infantry mulched pikes and piked mulched cavalry. But with this bigger rotation of counters, it meant that swords and cavalry were not really counters for each other.
Consequently, when cavalry met swordsman or pikes met archers, it could turn into a lengthy fight as neither was terribly good at killing the other. Furthermore, while cavalry beat archers, dollar for dollar, 3 bucks worth of archers would still beat 1 buck worth of cavalry. It wasn't like Company of Heroes where 1 tank could destroy the entire enemy army unless they had at least 1 of something specifically meant to counter tanks. Starcraft did it fairly "weak", too, in that while X could beat Y, enough Y could still beat small numbers of X.
If you make it too weak it can lead to single unit spamming, though.
In our "living world" game it would be interesting to randomly generate factions, with stats meeting certain parameters, and see how "evolution" plays out.
Maybe part of the process of creating the game world would be to actually run it in accelerated mode for "2 years" and let the NPC factions battle each other until the obviously weak ones were wiped out and it became a natural world of several strong factions.
As long as you're willing to keep updating and tweaking the teams, you should be able to stay on top of the "flavor of the week" issue and at least squash any obvious imbalances quickly.
My personal favorite RTS is still Battle for Middle Earth 2, which launched with 6 teams, no two alike, and did a very good job of balancing them. I eventually just played as "Random" and still got to the top 10 in the official rankings.
Incidentally, I think the secret to balance in an RTS and the reason why BFME2 and to some extent Starcraft did so well is that you need unit production levels to be fairly high. Company of Heroes (and that followup Warhammer 40k game that used the same engine) had problems largely, I think, because your units were so expensive and thus so few. When the enemy came out with a Rock, if you didn't already have a Paper, you were probably screwed. Whereas in BFME2, units were cheap enough that you'd probably have some mix out anyway and if the enemy started spamming Rocks, you'd have enough Paper on hand to hold him off while you built more Paper.
I also think that having a bigger rotation than "rock->paper->scissors" helps, as well as making it a "weak" system. In BFME2, for example, cavalry mulched archers and archers mulched infantry and infantry mulched pikes and piked mulched cavalry. But with this bigger rotation of counters, it meant that swords and cavalry were not really counters for each other.
Consequently, when cavalry met swordsman or pikes met archers, it could turn into a lengthy fight as neither was terribly good at killing the other. Furthermore, while cavalry beat archers, dollar for dollar, 3 bucks worth of archers would still beat 1 buck worth of cavalry. It wasn't like Company of Heroes where 1 tank could destroy the entire enemy army unless they had at least 1 of something specifically meant to counter tanks. Starcraft did it fairly "weak", too, in that while X could beat Y, enough Y could still beat small numbers of X.
If you make it too weak it can lead to single unit spamming, though.
In our "living world" game it would be interesting to randomly generate factions, with stats meeting certain parameters, and see how "evolution" plays out.
Maybe part of the process of creating the game world would be to actually run it in accelerated mode for "2 years" and let the NPC factions battle each other until the obviously weak ones were wiped out and it became a natural world of several strong factions.