Tweet
Tentative Project - Part Three
It seems to me that a good game story should be divided in to 3 acts; The Start, The Struggle & The Winning (or Close Losing)
It seems as though every game can only 'dish out' a finite challenge, so while a beginning or intermediate player can get a good game, with generous helpings of each of the above 'acts', an experienced player who knows and exploits the limitations of the game engine cuts the middle act short. Then its only a question of mopping up. That becomes tedious awefully quickly.
I was thinking about this problem towards the end of the CtP1 product cycle, and looking forward to being able to maybe accomplish something in CtP2 towards this end.
It seems as though, by all accounts, the AI is defending well, but generally attacking badly. This could be due to the fact that these players are getting the AI onto its back foot, which as you can see in the AI files, minimizes its offensive capability.
Something needs to be changed in the game dynamic, and its unlikely that any game design company will do this, because they are in the business of catering to the lowest common denominator. Difficulty levels should sort this out, but never do, since they are just 'resource modifiers or multipliers' for the AI and player. If a player can achieve a greater critical mass than is necessary to overpower one single region then thats all that is necessary.
In short, a human has a critical military advantage; Divide and Conquer. A human doesn't wage wars on all fronts, but power projects in a narrow sense. The AI of the computer also 'power projects' since single targeted city attacks 'work'. What the computer does BADLY is defend against power projection. It doesn't predict troop movements or see paterns. It doesn't understand interdiction or any grand strategy at all.
What to do? Keep the status quo? In my opinion, no, at least in a mod for those that choose to have a challenge.
There are some specifics that may be helpful in resolving these game play issues;
1) Eliminate the game engine exploits. The key exploit in Civ games has been and is still ICS. If you're not sure what ICS, please search for ICS... I won't repeat whats been said here a hundred times before.
CtP2 is interesting in that it uses a different dynamic of resource gathering for cities. Essentially you gain a percentage of resources within a ring until you get to a new critical size, then you start getting increasing resources within that ring.
The reasoning behind this new model was a move away from micromanagement towards abstraction, that I am all in favor of; it allows the player to deal with the empire level stuff, yet still allows for a degree of micromanagement by city and tile improvement placement.
The ICS problem still exists because 10 size 1 cities are more beneficial than 1 size 10 city. This is easy to solve by changing the properties of improvements upwards in general, making it ultimately detrimental to keep cities small. Bigger cities should be just plain better.
There should be some amount of struggle to get them going and growing, so building that settler at size 2 will drop you back another 20 or 30 turns.
Unmodified cities in deserts or tundra should either starve or grow so weakly without serious tile improvement placement.
The PW/Tile Improvement scheme of CtP is a great component of an ICS solution.
2) The rich get richer.
This topic was started by myself in the following topic Wonders o' the world... the root of all evil? and continued to some degree in a Civ3 General discusion.
I won't repeat this topic over and over, except to say that Feats of Wonder are more of the same with regard to the rich getting richer.
I'd advocate keeping FOW's, but limiting Wonders to one per Civ, per age. That would seem to be a happy medium.
3) AI Fudging=GOOD & AI Cheating=BAD
There were two main ways Civ2 made up for the weaknesses of a finite state AI. First were the player limitations and decreased starting resources. Secondly and to me most annoying, was the way the AI players cheat; ignoring game mechanics blatantly. These are well known by now, and have been documented.
I am not opposed to playing withly handicapped rules versus the AI, but there is a critical point; the point at which reality is stretched, that the game starts to be flawed.
CtP1 was a positive step in this regard; the AI didn't cheat but had increased resources and multipliers for food and science and so on. Since they are 'hidden' they do not offend the senses in the same way.
The level at which the AI fudges will need to be adjusted based on what other changes are made.
I think that perhaps the most important will be the production bonuses; EVEN IF the AI can stay in the game, keep offensive it's USELESS UNLESS it be given the units to put in the stack, by appropriate bonuses. I believe that this can only make for a more satisfying tactical and strategic experience.
4) Guns vs. Butter
It seems too cheap for the player to build and maintain large armies in CtP1&2. Since ultimately military strength was for the majority of history the key to power and sucess, its key that the human should not be able to easily get into a lead this way.
For the reasons I mentioned above, having twelve good units, and 'enough mobile defenders' is enough to conquer the world in CtP2. Building (and perhaps more so keeping) large armies should ultimately be more difficult for the human player.
There should be a clear choice between serious defense spending and a productive society.
5) Defenders should have a serious advantage.
In the Medmod, it was plain difficult to take a town defended in strength by the AI . You were going to take losses, inevevitably. This felt *VERY* right, particularly the counter bombarding and so on. I have a memory of some of the tactical situations that 'turned the tide', in a way that I never did in Civ2. Thats the mood that you want to capture in a civ game.
I think its clear, that the advantage should be with the defenders and to this end, the defensive bonuses should be 'beefed up', along with introducing more rounded unit choices, that the AI can use too.
[This message has been edited by TheLimey (edited November 26, 2000).]
It seems as though every game can only 'dish out' a finite challenge, so while a beginning or intermediate player can get a good game, with generous helpings of each of the above 'acts', an experienced player who knows and exploits the limitations of the game engine cuts the middle act short. Then its only a question of mopping up. That becomes tedious awefully quickly.
I was thinking about this problem towards the end of the CtP1 product cycle, and looking forward to being able to maybe accomplish something in CtP2 towards this end.
It seems as though, by all accounts, the AI is defending well, but generally attacking badly. This could be due to the fact that these players are getting the AI onto its back foot, which as you can see in the AI files, minimizes its offensive capability.
Something needs to be changed in the game dynamic, and its unlikely that any game design company will do this, because they are in the business of catering to the lowest common denominator. Difficulty levels should sort this out, but never do, since they are just 'resource modifiers or multipliers' for the AI and player. If a player can achieve a greater critical mass than is necessary to overpower one single region then thats all that is necessary.
In short, a human has a critical military advantage; Divide and Conquer. A human doesn't wage wars on all fronts, but power projects in a narrow sense. The AI of the computer also 'power projects' since single targeted city attacks 'work'. What the computer does BADLY is defend against power projection. It doesn't predict troop movements or see paterns. It doesn't understand interdiction or any grand strategy at all.
What to do? Keep the status quo? In my opinion, no, at least in a mod for those that choose to have a challenge.
There are some specifics that may be helpful in resolving these game play issues;
1) Eliminate the game engine exploits. The key exploit in Civ games has been and is still ICS. If you're not sure what ICS, please search for ICS... I won't repeat whats been said here a hundred times before.
CtP2 is interesting in that it uses a different dynamic of resource gathering for cities. Essentially you gain a percentage of resources within a ring until you get to a new critical size, then you start getting increasing resources within that ring.
The reasoning behind this new model was a move away from micromanagement towards abstraction, that I am all in favor of; it allows the player to deal with the empire level stuff, yet still allows for a degree of micromanagement by city and tile improvement placement.
The ICS problem still exists because 10 size 1 cities are more beneficial than 1 size 10 city. This is easy to solve by changing the properties of improvements upwards in general, making it ultimately detrimental to keep cities small. Bigger cities should be just plain better.
There should be some amount of struggle to get them going and growing, so building that settler at size 2 will drop you back another 20 or 30 turns.
Unmodified cities in deserts or tundra should either starve or grow so weakly without serious tile improvement placement.
The PW/Tile Improvement scheme of CtP is a great component of an ICS solution.
2) The rich get richer.
This topic was started by myself in the following topic Wonders o' the world... the root of all evil? and continued to some degree in a Civ3 General discusion.
I won't repeat this topic over and over, except to say that Feats of Wonder are more of the same with regard to the rich getting richer.
I'd advocate keeping FOW's, but limiting Wonders to one per Civ, per age. That would seem to be a happy medium.
3) AI Fudging=GOOD & AI Cheating=BAD
There were two main ways Civ2 made up for the weaknesses of a finite state AI. First were the player limitations and decreased starting resources. Secondly and to me most annoying, was the way the AI players cheat; ignoring game mechanics blatantly. These are well known by now, and have been documented.
I am not opposed to playing withly handicapped rules versus the AI, but there is a critical point; the point at which reality is stretched, that the game starts to be flawed.
CtP1 was a positive step in this regard; the AI didn't cheat but had increased resources and multipliers for food and science and so on. Since they are 'hidden' they do not offend the senses in the same way.
The level at which the AI fudges will need to be adjusted based on what other changes are made.
I think that perhaps the most important will be the production bonuses; EVEN IF the AI can stay in the game, keep offensive it's USELESS UNLESS it be given the units to put in the stack, by appropriate bonuses. I believe that this can only make for a more satisfying tactical and strategic experience.
4) Guns vs. Butter
It seems too cheap for the player to build and maintain large armies in CtP1&2. Since ultimately military strength was for the majority of history the key to power and sucess, its key that the human should not be able to easily get into a lead this way.
For the reasons I mentioned above, having twelve good units, and 'enough mobile defenders' is enough to conquer the world in CtP2. Building (and perhaps more so keeping) large armies should ultimately be more difficult for the human player.
There should be a clear choice between serious defense spending and a productive society.
5) Defenders should have a serious advantage.
In the Medmod, it was plain difficult to take a town defended in strength by the AI . You were going to take losses, inevevitably. This felt *VERY* right, particularly the counter bombarding and so on. I have a memory of some of the tactical situations that 'turned the tide', in a way that I never did in Civ2. Thats the mood that you want to capture in a civ game.
I think its clear, that the advantage should be with the defenders and to this end, the defensive bonuses should be 'beefed up', along with introducing more rounded unit choices, that the AI can use too.
[This message has been edited by TheLimey (edited November 26, 2000).]
Comment