So all this talk about Utopia made me thought about the one genre that people thought about making (perhaps a MMORTS) but never materialized or at least never got big in any meaningful numbers. We'll start assuming if you make such a game it's actually balanced because obviously the game would never get anywhere if you find some way to always win due to an imbalance issue. But even if the game is balanced, there are problems with this genre that requires more than a balanced game to suceeed.
The first thing I can think of is that the game has to be somewhat tactical. Utopia, for example, appears to have no tactical elements whatsoever. That is to say after you select your order to attack you never move your unit X from hex Y to hex Z and use attack Q. Now you'd say obviously this cannot be the case because then you'd just be able to attack someone while they're not physically there, and you'd presumably always win against the computer assuming the AI is not better than an average person. You also could have a problem where if your side has the best player in the world it's entirely possible he's so good that you just have him attack everyone and he'll just beat anybody even if they're present to defend and never lose and that's not very strategic at all.
But then a fight that is entirely strategic is just as bad, because it implies if you have X versus their Y your X will either always win or always lose, within some factor of randomness. It appears Utopia solves this by making the information hard to get so you don't always know what is the enemy's Y versus your X, but I don't like a game focused on total blindness. From a simulation point of view, it just doesn't make sense that you'd invade somebody without even the slightest clue what they might have. Yet in a strategic-only game, if you even have a clue of what they might have then the outcome is essentially already decided!
From playing ROTK 11, I think the concept of Action Points would be a good way to solve this. Let's say your army is of 1000 unit of power. You can make it such that a player can only control 100 unit of power of guys, versus an enemy that also only controls 100 unit of power. Further, you can only control say 3 such battles on the offense or 6 such battles on the defense per day, which means at best your 100 unit can wipe out 300 units of enemy defenders, and then after that you'll have to leave it to the AI. Here is the earlier assumption is very important: the AI absolutely has to be able to at least make you suffer for a win. If AI has 100 units against your 100 units, you'd want the AI to say at least make you lose 50 unit of your power before losing.
Also, this means your army size has to be relatively small, and the number of guys a player can command has to be also relatively small relative to the already small size army. There is no realistic way you can design a strategy game where 20000 units of power doesn't completely smash 10000 units of power assuming comparable skill on both side. You'd have to make a game where it is easy to get a decent army but prohibitively expensive to make a huge army.
I believe Civ 4 has a system where cities cost X^2 to maintain which indirectly prevents you from just having a zillion cities pumping out a large number of armies. ROTK troops are dirt cheap to hire, but there is a hard cap of 100K troops per city and troops require a ton of maintenence (much more than it takes to hire them) which means a lot of time you'd purposely not hire the max possible just because it'd make more economical sense to have say 60K people around, and then hire more people as they get killed, unless you know a mega attack is imminent. You'd want the people who are actually good at playing the game (whatever that might be) to have some tactical advantage for actually playing the game, but not so much that it negates the point of the game being a strategy game.
Note that I set it up such that a player can participate in twice the number of defensive battles compared to offense, so this would give a considerable edge to the defender. This means to hit a strong player you might really have to think about how to attack in waves to make him run out of his ability to participate on the defense, so to speak. Here it might be a good idea to purposely lose a small battle by letting the AI handle it while saving your best players when you know one of the enemy's best player is coming (we'll assume the defender can always see whether the a human player is participating in the attack).
I'm not sure how easy any of this would be to implement, and certainly coming up with a model where the AI can reliably always trade at least some resources with the best player is going to take a lot of effort (otherwise this game will always be won by the side with the most people). But if such a game exists, it would be pretty cool, no?
The first thing I can think of is that the game has to be somewhat tactical. Utopia, for example, appears to have no tactical elements whatsoever. That is to say after you select your order to attack you never move your unit X from hex Y to hex Z and use attack Q. Now you'd say obviously this cannot be the case because then you'd just be able to attack someone while they're not physically there, and you'd presumably always win against the computer assuming the AI is not better than an average person. You also could have a problem where if your side has the best player in the world it's entirely possible he's so good that you just have him attack everyone and he'll just beat anybody even if they're present to defend and never lose and that's not very strategic at all.
But then a fight that is entirely strategic is just as bad, because it implies if you have X versus their Y your X will either always win or always lose, within some factor of randomness. It appears Utopia solves this by making the information hard to get so you don't always know what is the enemy's Y versus your X, but I don't like a game focused on total blindness. From a simulation point of view, it just doesn't make sense that you'd invade somebody without even the slightest clue what they might have. Yet in a strategic-only game, if you even have a clue of what they might have then the outcome is essentially already decided!
From playing ROTK 11, I think the concept of Action Points would be a good way to solve this. Let's say your army is of 1000 unit of power. You can make it such that a player can only control 100 unit of power of guys, versus an enemy that also only controls 100 unit of power. Further, you can only control say 3 such battles on the offense or 6 such battles on the defense per day, which means at best your 100 unit can wipe out 300 units of enemy defenders, and then after that you'll have to leave it to the AI. Here is the earlier assumption is very important: the AI absolutely has to be able to at least make you suffer for a win. If AI has 100 units against your 100 units, you'd want the AI to say at least make you lose 50 unit of your power before losing.
Also, this means your army size has to be relatively small, and the number of guys a player can command has to be also relatively small relative to the already small size army. There is no realistic way you can design a strategy game where 20000 units of power doesn't completely smash 10000 units of power assuming comparable skill on both side. You'd have to make a game where it is easy to get a decent army but prohibitively expensive to make a huge army.
I believe Civ 4 has a system where cities cost X^2 to maintain which indirectly prevents you from just having a zillion cities pumping out a large number of armies. ROTK troops are dirt cheap to hire, but there is a hard cap of 100K troops per city and troops require a ton of maintenence (much more than it takes to hire them) which means a lot of time you'd purposely not hire the max possible just because it'd make more economical sense to have say 60K people around, and then hire more people as they get killed, unless you know a mega attack is imminent. You'd want the people who are actually good at playing the game (whatever that might be) to have some tactical advantage for actually playing the game, but not so much that it negates the point of the game being a strategy game.
Note that I set it up such that a player can participate in twice the number of defensive battles compared to offense, so this would give a considerable edge to the defender. This means to hit a strong player you might really have to think about how to attack in waves to make him run out of his ability to participate on the defense, so to speak. Here it might be a good idea to purposely lose a small battle by letting the AI handle it while saving your best players when you know one of the enemy's best player is coming (we'll assume the defender can always see whether the a human player is participating in the attack).
I'm not sure how easy any of this would be to implement, and certainly coming up with a model where the AI can reliably always trade at least some resources with the best player is going to take a lot of effort (otherwise this game will always be won by the side with the most people). But if such a game exists, it would be pretty cool, no?
