GOTM Difficulty Levels

[DynamicSpirit]

Making the hard games easier:
- builds false confidence in the lower level players
- does not satiate the hunger of the harder level players

Making the hard games easier via a good starting location has a similar effect to the adventurer bonus: It makes it a bit easier to get your civ established with a few cities at the start. It doesn't change the fact that in the long run, the AI can research faster, settle faster, build things faster, is more aggressive, etc. etc. In that regard the hard levels are still - ummm - hard. Besides, it is possible, if the player makes a few mistakes, for player-start-position-bonuses to turn instead into bonuses for a lucky AI civ, thus disadvantaging the player!

If it were up to me, I think the GotM should be at a high level all the time, and should present the toughest challenges with no handicaps for the lessors like myself. Training wheels usually create a situation in which it takes a person a lot longer to learn to ride the bike than those that learn without them. I think people would learn a lot more if the games were tough, and few people were able to win. Thus, the winning games would become more important, hold more info for people to learn from, and truly seperate out the elite players from the fray.

Would that be the same logic that says the best way to start teaching a 5-year old to read and write is by making him attend university English lectures?

I think usually the best way people learn is if they are given a challenge that's just slightly beyond what they can currently do.

 

[drkodos]

Would that be the same logic that says the best way to start teaching a 5-year old to read and write is by making him attend university English lectures?

No, it is not the same logic AND if you are waiting until a kid is five to begin that process, you have dropped the ball in major fashion. I might not have created the best analogy, but I do believe that holding the bar very high is the best way to get people to raise their own standards and improve. I believe in teaching to the top of the class, not the middle, and certainly not the bottom rung (the way the American school system does).

Also, for Adventurer Class, I have previously argued that it could indeed be a handicap for a player because settling a 2nd city right off the bat in the start of the game costs money and depletes the treasury faster, effecting research (as you mention).

No doubt I tend to be somewhat Machivellian in my learning theories. A lot of us misanthropic, Social Darwinist types are often like that.

 

[bio_hazard]

I'm not sure my goal with GOTM or civ in general is to get as good as I can as fast as I can. I usually can't play for long stretches at a time, and want to spend as little time as possible micromanaging. I'm perfectly happy being a poor to mediocre player, and glad GOTM is accomodating to people who don't regularly win at higher levels, whether its through periodic games at lower levels or starting bonuses at higher ones. I'm not sure 'social' is the right word for it, but I enjoy seeing how other people's games turn out. Don't get me wrong, I learn a lot too!

Just my perspective...

 

[Cactus Pete]

Returning to an ealier post by DaviddesJ . . .

I agree that lower level games are particularly time consuming if you wish to optimize your score. If we are going to continue to play them -- and I'm in favor of that, as participation seems a high value in the gotm context -- I'd suggest the moderators adjust the scoring so that winning in fewer turns gets a significantly greater weight.

 

[Déja]

I think that the best idea would be to alternate between high and low level games. For instance, GotM10 is Immortal, so GotM11 should be Noble and GotM12 could be Deity. By stepping up each time, players who got trounced at a lower level will be wary to continue playing.

I think that Noble would be a "happy minimum" that ensures that the games don't get TOO easy for the experienced players, but still gives players, such as myself, a chance to stay competitive.

 

[playshogi]

I might have posted this somewhere else, but I was thinking that each GOTM should have a save for each of the levels of difficulty Warlord-Deity. That way everyone can play at the level that is most enjoyable for them. This would eliminate the need for special classes.

 

[FuRRie]

I think a Window (Noble) Prince - Emperor (Immortal) seems good.
Let me explain why I think so, first of all, if you play a game or 2 a day, you can play Noble difficulty and even win sometimes.
I will always encourage people to play Succesion Games, with succesion games you learn fastenough to play on Prince/Monarch and even win, because your team can make up (and explain) some of the mistakes you make.
So reaching Prince level shoudn't be too hard, as an upper level for GOTM's, it doenst really matter, playing an Immortal game and losing badly isn't too bad, it learns you to improve your skills, maybe not after 1 game, but it does give you insights on harder AI's .

 

[DynamicSpirit]

Returning to an ealier post by DaviddesJ . . .

I agree that lower level games are particularly time consuming if you wish to optimize your score. If we are going to continue to play them -- and I'm in favor of that, as participation seems a high value in the gotm context -- I'd suggest the moderators adjust the scoring so that winning in fewer turns gets a significantly greater weight.

I agree that the scoring system is poor, and I do really wish Ainwood and AlanH would look into the issues, as I think the poor scoring system is one of the biggest weaknesses of the GOTMs (I know it's inherited from Firaxis, so it's not the fault of anyone associated with GOTMs).

The solution I'd personally like to see implemented is simply to remove the population element from the score (or at least, heavily reduce its weighting).

In terms of gameplay, doing that will mean that people are rewarded for managing their economy well, developing science, and - obviously - winning early. In other words, they'll be rewarded for playing Civ skillfully, as opposed to the current system where people are to a large extent rewarded for what looks to me to be fairly mindless milking.

In terms of real-world realism, I think there's another good argument for removing the population element from the score: On the whole, countries in the modern world do not earn respect for having a high population. They earn respect for cultural and technological achievements etc. Does anyone, for example, look at the US and think, what a great country that is to have 300+ million people? I'm sure lots of people admire the US [1] for lots of other reasons but I've never heard of that one as a reason! So removing the population from the score would I think make the Civ scoring system a bit closer to reflecting the things that matter when you're rating how 'good' a civilization is.

[1] I'm taking the US as an example here because it's big. I'm not making any comment either way on whether I think the US is in fact admirable

 

[Déja]

[1] I'm taking the US as an example here because it's big. I'm not making any comment either way on whether I think the US is in fact admirable

I think China would make a better example. Do you admire China for their massive population?

 

[EEO]

I agree that the scoring system is poor, and I do really wish Ainwood and AlanH would look into the issues, as I think the poor scoring system is one of the biggest weaknesses of the GOTMs (I know it's inherited from Firaxis, so it's not the fault of anyone associated with GOTMs).

This consideration is probably one of the most important in order to alter current gameplay. For quite some time I thought of creating a new thread on it but was not sure whether people are ready to discuss it. Thanks for initiating the debate.

The solution I'd personally like to see implemented is simply to remove the population element from the score (or at least, heavily reduce its weighting).

In terms of gameplay, doing that will mean that people are rewarded for managing their economy well, developing science, and - obviously - winning early. In other words, they'll be rewarded for playing Civ skillfully, as opposed to the current system where people are to a large extent rewarded for what looks to me to be fairly mindless milking.

Well, we are immediately in the big issue ... But maybe it is not that simple.

I certainly agree that people should be rewarded for playing Civ "skillfully" as opposed to extreme techniques like milking. By simply removing the population element, you will make obsolete the current milking but favor other extreme techniques like total military rush.

Have a look on GOTM8 and Obormot final spoiler. He did not milk, but he used with an extraordinary organization slavery, slavery bug and total military rush to obtain both best score and fastest domination.

Let me say here that such a play requires much skill as well as Hendrikszoon milking did. In fact, I think that Hendrikszoon (GOTM2) and Obormot (GOTM8) have created perfect techniques according to the current score formula and specific settings (pangea-like maps for Hendrikszoon and archipelago-like maps with medium-high difficulty level for Obormot).

But in a game called Civilization, the score formula should not reward "barbarian" behavior ( crazy conquest, slavery ...). Nevertheless, we cannot prohibit such behavior because it represents an historical reality. What we can do, is to give penalties in civ score that would make conquest, treachery (e.g. war on a friendly civ) and slavery less attractive however efficient they are. Some examples :
- Let us say that each population point of a conquered city at the moment it is conquered counts as "-1" (instead of "+1" currently) in the total population of the conqueror at the end of the game. It implies that a 8 pop city conquered begin to give score bonus to its conqueror only when it reaches 17 or more pop. This simple modification should seriously reduce "crazy warmongers" benefits and give sense to the civilized aspect of the game (as long as players are going for score) ...
- For a razed city or for slavery pop rushing, let us say "-3" for each pop.

I'm taking the US as an example here because it's big. I'm not making any comment either way on whether I think the US is in fact admirable

I have read many of your posts and have seen that you have a great sense of humour. So I can't resist to give you a (very known in Europe) sample of how US is described outside its borders:
"America is the only nation in history which miraculously has gone directly from barbarism to decadence without the usual interval of civilization."
from Georges Clemenceau, French statesman (1841 -1929)

 

[bio_hazard]

EEO-

Interesting ideas. I think you might have hit on something for reducing somewhat the score due to population. What if population for score purposes was the sum of (your city's actual pop * percent of the city that is cultrually yours)? This means you could still get credit for having a big metropolis full of happy citizens, but not so much for your neighbor's big cities that you took 3 turns before.

I'm not sure how much we should project our own moral ideas on what constitutes a good game of Civ. I'd be less inclined to support penalties for legal tactics. "Classy" or "Moral" games would be better set up as variant games where, for example, Slavery was not allowed. Otherwise, I don't know if we'd ever agree what constitutes shrewd play and what is over the line in a general competition like GOTM.

 

[Obormot]

The problem as I see it is that civ can be played in many different styles, but having a scoring system means that we are limited to one single "right" way of playing it, if we want a high score. And that is true for any kind of scoring system. Penalising slavery and removing points from population will just mean that everybody should go lets say for a space race win every time to get a good score. I can't see why thi would be better then using slavery to get a fast domination. Some people may like it because the most advantageous approach will now be less "barbaric" and more "civilized", but there will be as many people who will say going for a space race win is boring.

That is why I like fast finishes and the epthatlon: they allow you to play for different victory conditions all the time and still compete with others for an award. Of course, that is still only seven different ways to play the game, compared to countless variants one can think off, but unfortunately we cannot make more awards and still keep the games competitive.

I don't think the current system can be significantly improved. There was an interesting idea though that was discussed some time ago about creating a "variant of the month" for each game setting one unique victory condition for each game and giving one extra award for that. I think that would add some new variety to the GOTM if the GOTM stuff are willing to have some extra work. The first time we try it, it can be a game without slavery, chopping and early war (did I forget something? Oh, yeah, the CS slingshot).

 

[DynamicSpirit]

I think China would make a better example. Do you admire China for their massive population?

Agreed. The reason I used the US was because lots of people do admire the US, but I'm not sure that any significant numbers of people outside China admire China?

Thinking about it that means that China would be the perfect example to illustrate my point. On the current Civ scoring system applied to reality, China would win the game by a mile, but I'm fairly sure there's very little in how China has been managed that would suggest any victory is deserved

 

[Cactus Pete]

This is a thoughtful and productive discussion that, hopefully, the moderators will heed. I don't want Obormot's comments to stifle it. He is certainly correct that the scoring rules will alter game play no matter what they are. It does not follow, however, that this means the particular scoring rules we choose to use are all equally undesirable.

Any alteration of the scoring system that reduces the need to spend time in tedious play in order to maximize score is worthy of consideration (and certainly reducing the value of population meets that criterion). I would again suggest that increasing the value given to fast finishes will almost certainly advance this goal as well.

Of course, there are other competing concerns related to game's basic premise of building a civilization. The trick will be to reduce milking without undermining that aspect of play, and I hope that my fellow fanatics will continue to discuss the optimal approach.

 

[DynamicSpirit]

The problem as I see it is that civ can be played in many different styles, but having a scoring system means that we are limited to one single "right" way of playing it, if we want a high score. And that is true for any kind of scoring system.

I disagree I'm afraid. I think it is perfectly possible to get a scoring system that equally rewards different play styles.

Thinking in abstract terms, one way to do that is to ensure that the things the scoring system rewards are to some extent in conflict. Imagine say, that in a game you might want to get lots of A or lots of B. If the game is designed in such a way that getting A tends to harm your ability to get B, and vice versa, then a scoring system that adds points for both A and B will mean you get at least two choices of playing style that are encouraged: Maximize A or maximize B.

Back to Civ scoring, the points for wonders are a great example of that (or at least, it would be a great example if it ONLY scored wonders that you've built yourself, not ones that you've captured. I'm not sure if that's the case?): Building a wonder means committing a lot of resources, and in the short term that damages your civ (although in the long term you may benefit a lot from the wonder). Rewarding early finishes is another example: Finishing early tends to mean you have less science, less pop, less of everything else that is scored).

And this argument I think shows up another of the big problems with rewarding population: Your population doesn't hinder other things, it tends to help everything else: If you have a high pop, that almost inevitably leads to high land area, high science (eventually, obviously not initially if you're on a conquering spree), high production so more wonders, etc. In a sense, you get rewarded twice by the scoring system: Once for the population, then again for all the other things that having a high population almost inevitably leads to.

That is why I like fast finishes and the epthatlon: they allow you to play for different victory conditions all the time and still compete with others for an award. Of course, that is still only seven different ways to play the game, compared to countless variants one can think off, but unfortunately we cannot make more awards and still keep the games competitive.

Yep I'd agree there. The eptathlon is a very good thing about the GOTM award system. Unfortunately, the trouble is, it's the highest-scoring medals that tend to be the headline-grabbing thing. (Though personally, I increasingly pay more attention to reading down to the fastest finish awards, and I now take very little notice of the medals)

 

[EEO]

Interesting ideas. I think you might have hit on something for reducing somewhat the score due to population. What if population for score purposes was the sum of (your city's actual pop * percent of the city that is cultrually yours)? This means you could still get credit for having a big metropolis full of happy citizens, but not so much for your neighbor's big cities that you took 3 turns before.

Surely interesting. The difference with my proposal is that you have no negative effects on score by conquering a city, just a lesser bonus.

I'm not sure how much we should project our own moral ideas on what constitutes a good game of Civ. I'd be less inclined to support penalties for legal tactics.

I see your point. But you must be aware that, by elaborating a score formula for civilization, you choose implicitly to reward some kind of ethics. The constraint is to remain as universal as possible. By pointing out city razing or slavery as not civilized acts, I think I join a consensus in the world today.

 

[Obormot]

Penalising conquest and slavery is not a good idea, because these tactics are usually good for your empire. Implementing a scoring system that will penalise people for making good decisions is wrong, because then we won't be Civilization IV any more, but some other game with self imposed restrictions.

Even if you look at it from RL point of view (I am not a big fan of this though - a game is just a game) there is no point in penalising conquest and slavery. We admire the deeds of Alexander, Napoleon and other greatest conquerors as well as we admire the Piramids that were built by egyptian slaves.

OK, now back to the main problem: we have 6 different victory conditions and things that are good for one VC are not good for the other VCs. Conquering cities and increasing population is almost always good for conquest and domination, very often it is good for space-race and dimplomacy, but it is usually not so good for a culture victory. Teching fast is very important for space race and diplo, but not so important for other VCs. Culture is not very usefull for anything but culture. And so on.

The only way I see how we can get around this, is to have different scoring formulas for each VCs. This will be very difficult to balance, so that's why I am so sceptical about this. But if you want to try it, go ahead, I will be happy if you manage to create a formula that balances everything reasonably well.

 

[EEO]

The problem as I see it is that civ can be played in many different styles, but having a scoring system means that we are limited to one single "right" way of playing it, if we want a high score.

Ooh ! I think you go to a wrong conclusion because you assume a wrong hypothesis. What you say could be true, if the scoring formula doesn't contain any variable linked to victory conditions. It's the case with the current formula and, yes, the discussion has not yet proposed to modify these things (but we are just at the beginning).

Let us do it the simplest way: let's multiply the current formula by a function "f(v)" depending of the victory condition obtained by the player.
We can compute f(v) empirically by the ratio "best score/best score(v)" on a significant number of GOTMs.
This is, of course, simplistic because a pertinent correction of the current formula should taking into account the victory conditions in the core formula according to other variables and not in a linear way ... But I don't need to go deep into the maths for now.
I just want to highlight the fact that a light modification of the formula can make much more difficult the choice of the ideal victory condition.

Let's consider the numbers for GOTM8 (no time to go further):

f(domination)= 1.00
f(diplo)= 1.57
f(conquest)= 1.60
f(space)= 3.29
f(culture)= 6.57

The numbers have surely to be adjusted by including more GOTMs (diplo multiplier seems too small, and culture multiplier seems too big), but imagine it's adjusted. You get the idea: now, by construction, it should be difficult to choose the right way of playing and I 'll bet the players would go rather for a victory condition matching their own style of play ...
Thus, this simplistic example illustrates that the system can easily be improved.

 

[Obormot]

That approach would lead to some weird culture games winning gold!

 

[Déja]

You get the idea: now, by construction, it should be difficult to choose the right way of playing and I 'll bet the players would go rather for a victory condition matching their own style of play ...
Thus, this simplistic example illustrates that the system can easily be improved.

I love this idea. I've wanted to go for a cultural victory in a GotM, but you just can't compete score-wise with a domination victory