AU Mod: Difficulty Levels

[nbarclay]

The Problem:

The jump in difficulty from Emperor to Demigod is too big. I like how Demigod plays out in the later stages of the game, but I absolutely hate starting in a big hole compared with the AIs - and an extra settler on top of all the other AI advantages on Demigod definitely starts players in a big hole, not just militarily but in terms of REXing. As a result, I don't like playing on Demigod unless I make sure I have a really great starting position to offset the AI head start.

If I'm the only one who feels that way, it's not that big a deal. But if there are others who find Emperor too easy but who find Demigod too hard to be fun, the situation is worth doing something about.


Proposed Solution:

Eliminate the Chieftain difficulty level, move all the other difficulty levels below Demigod down a slot, and insert a new difficulty level ("Mythic Hero"?) in between Emperor and Demigod. The new difficulty level would have all of the properties of Demigod except for the extra AI settler. (This change also requires changing the AI difficulty level in the general settings to fit the new location of Regent.) An alternative approach would be to add a new difficulty level, but the new one would appear at the end of the list as if it were the hardest instead of fitting the normal progression from easiest to hardest.

Note that the situation with multiplayer games seems to be a lot more awkward. When I tried firing up hotseat games with modified versions of the AU Mod, the list of difficulty levels available reflected the standard rules, not the modified rules in the scenario. That meant that an added difficulty level was not available at all, while with the difficulty levels moved, the difficulty level choices were one off from where they were supposed to be in the Mod. So from a multiplayer perspective, our choice is either an arrangement where the new difficulty level is unavailable or one where game set-up will be confusing. (By the way, with the difficulty levels rearranged, the name of the difficulty level does reflect the modified rules when loading a saved hotseat game.) If we split off a separate MP version of the Mod, we probably ought to keep the standard difficulty level configuration in it.

One other note is that if we move the difficulty levels, scores will probably be as if players were playing one difficulty level lower than they really are. But do people really care enough about scores for that to matter?

Edit: I just came up with what would probably be a better implementation. Instead of eliminating Chieftain and moving difficulty levels down, we could add a new difficulty level and move the highest three levels up (with Sid moved into the new slot, Deity moved into Sid's slot, and so on). That way we could be essentially positive that Emperor and below would not be affected even if there are factors involved that are tied to which "slot" a difficulty level occupies without being reflected in the editor.

[pvzh]

Although I played only 2 demi-god games, I came to similar conclusion. Too hard to be fun in the begining.

Maybe, we can just modify existing difficulties without inserting anything? If anybody needs anything more he can add more difficult levels.

We can substitute the settler with 2 extra wokers: AI needs workers anyway (I hope it will not join them to city before atleast some improvements).

[punkbass2000]

I generally agree with your idea, Nathan. I think DG-Emp is clearly the biggest jump in difficulty, and never truly understood the seemingly perennial difficulties people would have going from Mon-Emp. Two settlers instead of one is huge. Give a Deity AI three settlers and it will be nearly as good as a Sid AI in many respects, IMO. I would go so far as to make this inbetween level Deity in all respects except only one settler. More Workers, cheaper everything, etc, but without the initial doubling.

BTW, instead of 'Mythic Hero', why just come out with 'Hercules' or 'Odysseus'? Hell, why not even 'Theseus'? It would be a good name as well as a nice homage to great player/contircutor. RaR has its 'DocT' level.

[Dominae]

What do the difficulty levels have to do with the AU mod?

Every jump in difficulty involves just one thing: how much/long are you prepared to be outclassed by the AI in the early-game? The longer any game lasts, the more Demigod feels like Emperor, and the more Deity feels like Demigod. The only problem with the Emperor-Demigod jump is that players want to feel powerful somewhere in the early-game, which is purely a matter of preference.

I myself play Demigod regularly, and do not feel my strategic options are all that restricted compared to Emperor; I can still do Archer-rushes when the situation calls for it, I can still push for a tech/branch lead in the Medieval era, and I can incorporate an early Wonder (sometimes two) into my plans. Yes, the are some games where I am behind until the Industrial era, but that just means that, on Demigod, starting location actually means something (on Emperor the game is too easy with self-imposed restrictions). I am not saying these things to brag, just to point out that Demigod is a perfectly fair difficulty level for players of a certain level of experience. I am still working on making Deity feel "fair" to me!

(By the way, the extra Settler on Demigod is the main reason Demigod-level AIs are nice and challenging later on in the game, compared to Emperor. Taking that away and leaving just the production bonuses will not really do all that much, compared to Emperor. Replacing the Settler by two Workers will not do much either, other than give more of them for the human player to buy off!)

The point of AU, other than fun, is to encourage players to improve their game. This means that one of the indicators of success in AU is when players say something like "I've decided to up to Emperor for AUXXX, because Monarch has become too easy for me". Many players have done it (just read the DARs!), and many more players will. So, IMO, creating a new difficulty level is only helping players stand still, not progress. Please, try Demigod: it is not so bad.

[Tassadar500]

I generally agree with Dominae. I'll add some reasons why later, because it's really late at night here and I need to wake up very early tomorrow.

[nbarclay]

I'll be taking things a bit out of order:

Originally posted by Dominae

Please, try Demigod: it is not so bad.

I've played enough on Demigod to determine that while it can make for interesting games if I have a really good starting position, I don't like the kinds of games it produces otherwise.

I myself play Demigod regularly, and do not feel my strategic options are all that restricted compared to Emperor;

I disagree. Starting in a big hole seriously undercuts a player's ability to play a heavily economic-oriented game early - especially without leaving oneself wide open to military attack. And to make matters worse, the AI advantages make it a lot harder to use economic strength as a deterrent against AI aggression.

Every jump in difficulty involves just one thing: how much/long are you prepared to be outclassed by the AI in the early-game?

I disagree. Up through Emperor, the question is one of how AI advantages (or, on Chieftain and Warlord, disadvantages) match up against human skill (and willingness to micromanage). If you play on a difficulty level where your skill is not great enough to offset the AI advantages, you start in a hole. But if you're good enough, it is possible to play a competitive game from practically the beginning all the way up through Emperor.

But in an economic contest, there is a limit to how much in the way of AI advantages it is theoretically possible to overcome with superior skill. Past that point, in the absence of some other advantage (such as a dramatically superior starting position), you end up stuck starting in a serious hole no matter how well you play. The only ways out of that hole are through military conquest or by waiting until the game reaches a stage where trading judo can overcome the stronger economies of the AIs. That, in turn, restricts what kinds of games it is possible to play on the highest difficulty levels.

[quote]The longer any game lasts, the more Demigod feels like Emperor, and the more Deity feels like Demigod. The only problem with the Emperor-Demigod jump is that players want to feel powerful somewhere in the early-game, which is purely a matter of preference.[quote]

If you look at things a certain way, just about anything can be dismissed as nothing more than a matter of preference. But the question of whether or not players have something resembling economic parity with their AI rivals has an enormous impact on what strategic options are available. Thus, there is more involved than just a matter of taste.

Further, all else being anywhere near equal, a game that caters to a relatively wide range of preferences is better than one that caters to a narrower range. I don't like pvzh's idea of changing the Demigod difficulty level because that would undercut the fun for players who enjoy playing on that level the way it is. But inserting a new difficulty level in between Empeoror and Demigod would not present that problem.

I am not saying these things to brag, just to point out that Demigod is a perfectly fair difficulty level for players of a certain level of experience. I am still working on making Deity feel "fair" to me!

To me, the quesiton is not one of "fairness" but rather one of balance. Genuine fairness is impossible in a game between a human being and an AI when the game is as complex as Civ is, if only because human players can learn from and adapt to how the AIs think but the AIs can't do the same in regard to us. So AIs have to be given advantages to offset the human advantage. The game works best when the balance of human and AI advantages is at a point the player considers fun. For some players, what is fun is having to struggle to win at all, while for others (including myself), what is fun is expecting to win but still having the AI put up a good fight.

The problem is, the gap between Emperor and Demigod is too big. Players can be caught in an in-between position where they consider Emperor too easy but where Demigod puts them at too much of a disadvantage for them to enjoy it.

(By the way, the extra Settler on Demigod is the main reason Demigod-level AIs are nice and challenging later on in the game, compared to Emperor. Taking that away and leaving just the production bonuses will not really do all that much, compared to Emperor. Replacing the Settler by two Workers will not do much either, other than give more of them for the human player to buy off!)

I think you underestimate how much difference the other factors make. A Demigod AI has just over a fourteen percent advantage over an Emperor AI in production capacity, which means more military units and more city improvements. Free support for an extra unit per city (plus four extra units for the empire as a whole) means that unit support costs don't cut into AIs' income as soon, leaving more income available for other uses and allowing backward AIs to support larger militaries. At the same time, the human player has to contend with higher tech costs and a little bit worse corruption. So there would definitely still be a significant impact on how competitive the AIs are as the game goes on.

I might also point out that in terms of empire size, overseas AIs' extra settlers pretty much cancel each other out. The extra settlers provide a little bit of a head start transitioning from REXing to building, but I very seriously doubt that that advantage is anywhere near as great as the advantages from higher production, faster research (compared with the human), and so forth.

Overall, the difference between Emperor and "Mythic Hero" would be very similar to the difference between Monarch and Emperor. That's especially true in the later stages of the game when the loss of a free happy face on Emperor tends to make less of a difference.

The point of AU, other than fun, is to encourage players to improve their game. This means that one of the indicators of success in AU is when players say something like "I've decided to up to Emperor for AUXXX, because Monarch has become too easy for me". Many players have done it (just read the DARs!), and many more players will. So, IMO, creating a new difficulty level is only helping players stand still, not progress.

You don't consider it progress for players to go from Emperor to a level in between Emperor and Demigod? To me, that definitely constitutes progress. And players who don't feel like they are ready to make the jump all the way from Emperor to Demigod might very well be interested in moving up from Emperor to an intermediate level.

Also note that some players reach a point where they could probably move up to a higher difficulty level but are uninterested in doing so because they don't consider the higher level fun. I've never liked the "feel" of the game on Deity, and Demigod without a big advantage in starting position isn't much better.

[nbarclay]

Originally posted by punkbass2000
BTW, instead of 'Mythic Hero', why just come out with 'Hercules' or 'Odysseus'? Hell, why not even 'Theseus'? It would be a good name as well as a nice homage to great player/contircutor. RaR has its 'DocT' level.

That's an interesting thought, but it doesn't fit the normal convention for naming difficulty levels. At present, the only difficulty level with a specific person's or entity's name attached is Sid, named after the person who created Civ. All the other difficulty levels are named after types of persons or entities, not specifically named persons or entities. So "Mythic Hero" would be more consistent with the normal convention.

[Arrian]

I like the idea.

Hell, I pondered the possibility of a difficulty level between Monarch & Emperor, but I doubt I'd get much support for that (To me, there is quite a jump there, since you drop from 2 born content to 1, and also have additional AI bonii do deal with. It is less of a jump than the Emp - Demi one, since a free settler is HUGE).

-Arrian

[nbarclay]

I just posted an edit to the top of the thread with another idea for how the additional difficulty level could be implemented.

[Modo44]

I'd say banana, but the longer this goes, the more I agree with Dominae.

Some personal experience, to show why:

Originally posted by nbarclay
Starting in a big hole seriously undercuts a player's ability to play a heavily economic-oriented game early - especially without leaving oneself wide open to military attack. And to make matters worse, the AI advantages make it a lot harder to use economic strength as a deterrent against AI aggression.

Do you mean that you find yourself in a situation, which forces you to attack the AI early on? Otherwise they just seem to be overwhelming you? You can't build early, because they come and kick you? They always seem to out-Rex you with ease? Well, this is exactly how I felt on Emperor for a while. I was overwhelmed, I got clobbered, I could never get out of the initial hole, except with great starting locations. Now I can do better, I can work with what I get, and I'm not so worried by the AI anymore. Basically, I'm better used to this difficulty level, and all changes that it brings.

My point is, this feeling is always present at some point, when moving up a level. Perhaps not between all levels, because that might depend on the player. Personally, I felt only a small difference between Regent, and Monarch, but I know that others think of it as a big one. Same goes for other difficulty jumps, IMO - some people find them extreme, some don't. That's how this game works, and I don't think this can be changed.

[Tassadar500]

Originally posted by Modo44
I'd say banana, but the longer this goes, the more I agree with Dominae.

Some personal experience, to show why:


Do you mean that you find yourself in a situation, which forces you to attack the AI early on? Otherwise they just seem to be overwhelming you? You can't build early, because they come and kick you? They always seem to out-Rex you with ease? Well, this is exactly how I felt on Emperor for a while. I was overwhelmed, I got clobbered, I could never get out of the initial hole, except with great starting locations. Now I can do better, I can work with what I get, and I'm not so worried by the AI anymore. Basically, I'm better used to this difficulty level, and all changes that it brings.

My point is, this feeling is always present at some point, when moving up a level. Perhaps not between all levels, because that might depend on the player. Personally, I felt only a small difference between Regent, and Monarch, but I know that others think of it as a big one. Same goes for other difficulty jumps, IMO - some people find them extreme, some don't. That's how this game works, and I don't think this can be changed.

Exactly, it's all about adjusting.

[nbarclay]

Originally posted by Modo44
I'd say banana, but the longer this goes, the more I agree with Dominae.

Some personal experience, to show why:


Do you mean that you find yourself in a situation, which forces you to attack the AI early on? Otherwise they just seem to be overwhelming you? You can't build early, because they come and kick you? They always seem to out-Rex you with ease? Well, this is exactly how I felt on Emperor for a while. I was overwhelmed, I got clobbered, I could never get out of the initial hole, except with great starting locations. Now I can do better, I can work with what I get, and I'm not so worried by the AI anymore. Basically, I'm better used to this difficulty level, and all changes that it brings.

My point is, this feeling is always present at some point, when moving up a level. Perhaps not between all levels, because that might depend on the player. Personally, I felt only a small difference between Regent, and Monarch, but I know that others think of it as a big one. Same goes for other difficulty jumps, IMO - some people find them extreme, some don't. That's how this game works, and I don't think this can be changed.

You're doing a good job of reasoning by analogy here, but I'm not convinced that the analogy is valid. As long as there is room for players to play better, players can compensate for higher difficulty levels by improving their own games. But once players reach a point where they play the economic game about as well as is theoretically possible, there is no longer room for them to improve their economic game to compensate for AI advantages on a harder difficulty level. At that point, the "have to dig out of a hole" situation is permanent, not just temporary until the player learns how to play better. To play on a higher level, players have to either use their military to dig out of the economic hole or resign themselves to lagging behind until relatively late in the game.

When I first made the jump from Monarch to Emperor more than two years ago, I went through a period similar to yours where I felt like I was starting in a hole. But since then, I've had more than two years to refine my strategies and learn strategies from others. If there is still significant room left to move my game up a notch to where I could play on Demigod without an advantage in starting position and without starting in a hole economically, I don't have a clue where it might be. And neither, as best I can tell, does anyone else. Look at Dominae's attitude of expecting to be behind for a considerable length of time on Demigod.

So what I have is a choice of playing on Emperor where I routinely massacre AIs (for example, my cavalry, infantry, and artillery against musketmen in the current AU game) or playing on Demigod where I'm likely to start in a big enough hole that it seriously undermines my fun. What I want is a difficulty level in between that gives me a bit more of a challenge without giving the kind of "start in a big hole" feeling that Demigod does.

[Dominae]

Originally posted by nbarclay
I've played enough on Demigod to determine that while it can make for interesting games if I have a really good starting position, I don't like the kinds of games it produces otherwise.

By "interesting" you mean the types of games you have grown accustomed to playing: heavily economy-oriented early-game (multiple Granaries, many Workers, low military), followed by a quick military buildup and a victory somewhere around Military Tradition or the early Industrial era. It seems to me that the fact that you are not able to do this as easily on Demigod has nothing to do with the Emperor-Demigod jump (which, for many players, is like the Monarch-Emperor jump - see Arrian's post), but with how you like the game to play out. I am sure Arrian used to love to grab a whole wack of early Wonders on Monarch and this made his transition to Emperor annoying, but he decided to make the switch because he wanted more of a challenge and in Civ3, more of a challenge means coming out of the hole later on.

What do you say to the Regent players that get out-REXed by the AI? Learn to REX better! Build fewer Wonders and improvements early on! One can give the same advice for the Emperor-Demigod jump. Consider that, on Demigod, sometimes the AI REXes so fast that you may need to pump out a few Settlers without a Granary! Is this something you are fundamentally against doing, even though it may be the best move strategically?

I disagree. Starting in a big hole seriously undercuts a player's ability to play a heavily economic-oriented game early - especially without leaving oneself wide open to military attack. And to make matters worse, the AI advantages make it a lot harder to use economic strength as a deterrent against AI aggression.

Yes, Demigod makes it harder to exploit the AI's reticence to attack early on until you are good and ready. The AI does bolder things when it has "more stuff" - this is what the difficulty levels are all about. If you want the AI to be bolder without the associated stuff, I suggest you increase the AI Aggressiveness values in the Setup screen.

If you play on a difficulty level where your skill is not great enough to offset the AI advantages, you start in a hole. But if you're good enough, it is possible to play a competitive game from practically the beginning all the way up through Emperor.

You start out in a hole on Emperor, too, by the way. It's just that it's a hole you have grown accustomed to. Many players find Demigod barely playable (too easy) anymore, and look to be challenged by Deity. They, too, complain about the Demigod-Deity jump. True, Emperor-Demigod may be a bit more difficult, but many players have taken up the challenge, and overcome it.

On Emperor, I can reasonably expect to Archer-rush the nearest civ in almost every game, essentially providing me with a victory right then and there (I wrote a thread about this a while back). Is this the kind of strategic option I enjoy in a difficulty level? Not really. On Demigod, Archer-rushes are not so easy. It's not really important to me that this is because the AI gets an extra Settler, or whatever, as long as I am sufficiently challenged. Like I said above, if you just get used to Demigod, you find that it is not so restrictive.

But in an economic contest, there is a limit to how much in the way of AI advantages it is theoretically possible to overcome with superior skill. Past that point, in the absence of some other advantage (such as a dramatically superior starting position), you end up stuck starting in a serious hole no matter how well you play.

The point is to play well to get out of the hole. Just because you are in a hole does not mean your decisions are irrelevant; even on Sid, the first 40 turns are critical, despite the fact that the AI has 5 cities and you have yet to found your second!

The only ways out of that hole are through military conquest or by waiting until the game reaches a stage where trading judo can overcome the stronger economies of the AIs.

Ah, that happens to be the way Civ3 works. The alternative to these two is, um, no warfare and no trading, which is a challenging variant indeed. Show me a recent game of yours where you won via purely economic means. Even if you could, like in SMAC, is this a good thing? It just means that you are playing SimCity-like variant where it's you against yourself. There need to be outward forces that challenge you while you do your "economic thing". Else you can just look at your starting location and say "I win" (which, sadly, often happens in Civ3 anyway).

The problem is, the gap between Emperor and Demigod is too big. Players can be caught in an in-between position where they consider Emperor too easy but where Demigod puts them at too much of a disadvantage for them to enjoy it.

Like I said above, AU is about getting players to learn about Civ3 strategy while having fun. You, I presume, know more than you need to know about Civ3 strategy to beat Deity level, you just choose not to use to beat the higher difficulty levels. The Emperor-Demigod gap is only "too big" because you only find a certain type of game fun.

I think you underestimate how much difference the other factors make. A Demigod AI has just over a fourteen percent advantage over an Emperor AI in production capacity, which means more military units and more city improvements.

Most of these production bonuses are wasted, as you may know. In my experience watching the AI in Debug mode, the production bonuses are most relevant in the Demigod-Deity and Deity-Sid jumps.

Free support for an extra unit per city (plus four extra units for the empire as a whole) means that unit support costs don't cut into AIs' income as soon, leaving more income available for other uses and allowing backward AIs to support larger militaries. At the same time, the human player has to contend with higher tech costs and a little bit worse corruption. So there would definitely still be a significant impact on how competitive the AIs are as the game goes on.

The only thing that is really relevant here is the tech costs. But in practice this just means that it takes you longer to get a tech/branch lead. Once you have it, it's smooth sailing, just like Emperor.

I might also point out that in terms of empire size, overseas AIs' extra settlers pretty much cancel each other out. The extra settlers provide a little bit of a head start transitioning from REXing to building, but I very seriously doubt that that advantage is anywhere near as great as the advantages from higher production, faster research (compared with the human), and so forth.

The extra Settlers do not "cancel each other out" with respect to the human player, which is what difficulty levels are all about (the AIs are always equal to each other at any given difficulty level). Watch a few games in Debug mode at Emperor and Demigod, I think you will find that the extra Settler is really what makes all the difference. Put another way, which would you rather have: slightly reduced tech costs, or two Settlers at the beginning of the game?

You don't consider it progress for players to go from Emperor to a level in between Emperor and Demigod? To me, that definitely constitutes progress. And players who don't feel like they are ready to make the jump all the way from Emperor to Demigod might very well be interested in moving up from Emperor to an intermediate level.

I consider the Emperor-Mythic and Mythic-Demigod jumps way to small to show any significant improvement. Mythic difficulty would be the difficulty level for Demigod-level players who do not want a real challenge. I consider this against the aims of AU. IMO, too many players play the AU games at one difficulty level lower than they normally do/should (for various reasons), and adding a intermediate difficultly would only reinforce this.

Also note that some players reach a point where they could probably move up to a higher difficulty level but are uninterested in doing so because they don't consider the higher level fun. I've never liked the "feel" of the game on Deity, and Demigod without a big advantage in starting position isn't much better.

Yes, well, this is the real issue, is it not? Like I said above, once you get the hang of it (three games, tops) you (especially you!) will find that Demigod is not so bad.

[joncnunn]

I've run my own more extreme experiment and found that Demigod level production advantage with all other AI advantages elimated plays easier the standard Monarch.

[nbarclay]

Originally posted by Dominae


By "interesting" you mean the types of games you have grown accustomed to playing: heavily economy-oriented early-game (multiple Granaries, many Workers, low military), followed by a quick military buildup and a victory somewhere around Military Tradition or the early Industrial era.

A heavily economy-oriented early game is definitely my preferred style of play. That doesn't mean that I feel like I ought to be able to play every game that way. But I want situations where a heavily economy-oriented early game doesn't make sense to be the exception, not a rule enforced by such massive advantages for the AIs that I can''t keep up in an economic race no matter how well I play.

Given the opportunity, I usually do try for a "quick military buildup and a victory somewhere around Military Tradition or the early Industrial era." But the ability to do that as often as I can these days is something I dislike more than I like.

What do you say to the Regent players that get out-REXed by the AI? Learn to REX better! Build fewer Wonders and improvements early on! One can give the same advice for the Emperor-Demigod jump. Consider that, on Demigod, sometimes the AI REXes so fast that you may need to pump out a few Settlers without a Granary! Is this something you are fundamentally against doing, even though it may be the best move strategically?

How often do you have a game where you can't get a competitive REX with building granaries but can if you don't build granaries? It's hard for me even to imagine such a thing. Most if not all of the time, situations where you have to go without granaries on Demigod are situations where the AI is going to out-REX you pretty seriously no matter what you do. "Learn to REX better" works beautifully in a Regent-to-Monarch jump where learning to REX better can enable the player to pretty consistently out-REX AIs in spite of the AI bonuses, but the situation with Demigod is completely different.

You start out in a hole on Emperor, too, by the way. It's just that it's a hole you have grown accustomed to.

Militarily, you start out in a pretty significant hole on Emperor, especially against AIs that start with Bronze Working or Warrior Code (and that therefore get better starting units). But economically, the hole is only a very small one, consisting of AI production advantages and an extra worker.

If players can't overcome the AI production advantages with superior REXing skills, the hole gets bigger over time until it becomes the kind of significant hole you're trying to paint a picture of. But for players with good enough REXing skills (and sufficient willingness to micromanage), the economic hole can generally be overcome so quickly that it hardly feels like a hole at all.

Many players find Demigod barely playable (too easy) anymore, and look to be challenged by Deity. They, too, complain about the Demigod-Deity jump. True, Emperor-Demigod may be a bit more difficult, but many players have taken up the challenge, and overcome it.

Looking in the editor, Demigod is a lot closer to Deity than it is to Emperor. Demigod adds an extra settler and an extra worker over Emperor in addition to military units, while the only extra units added with Deity are military ones. The difference in the AI cost factor is slightly bigger in percentage terms going from Demigod to Deity, but the difference in free unit support is smaller in percentage terms going from Demigod to Deity.

The reality is that the differences between Demigod and Deity are almost identical to those between Emperor and my proposed "Mythic Hero" level: a production advantage, extra unit support, and such. Indeed, Mythic Hero adds a complication that Deity does not: an extra AI worker (or two in pvzh's variant). So if the difference between Demigod and Deity is great enough to draw grumbling about its size, it is hard to see how the difference between Emperor and Mythic Hero could be too small to be significant.

On Emperor, I can reasonably expect to Archer-rush the nearest civ in almost every game, essentially providing me with a victory right then and there (I wrote a thread about this a while back). Is this the kind of strategic option I enjoy in a difficulty level? Not really. On Demigod, Archer-rushes are not so easy. It's not really important to me that this is because the AI gets an extra Settler, or whatever, as long as I am sufficiently challenged. Like I said above, if you just get used to Demigod, you find that it is not so restrictive.

The point is to play well to get out of the hole. Just because you are in a hole does not mean your decisions are irrelevant; even on Sid, the first 40 turns are critical, despite the fact that the AI has 5 cities and you have yet to found your second!

To me, the whole "start in a hole and have to dig out" paradigm is not fun, at least when the hole is big enough to be obvious. It's really essentially a kludge to make up for the fact that the AI isn't competitive enough to give the best players a serious challenge otherwise. But while the kludge helps the game from the perspective of players who want the strongest competitive challenge possible, it can seriously undercut flexibility for players who want to build up the kind of civilization that's fun for them.

Ah, that happens to be the way Civ3 works. The alternative to these two is, um, no warfare and no trading, which is a challenging variant indeed. Show me a recent game of yours where you won via purely economic means. Even if you could, like in SMAC, is this a good thing? It just means that you are playing SimCity-like variant where it's you against yourself. There need to be outward forces that challenge you while you do your "economic thing". Else you can just look at your starting location and say "I win" (which, sadly, often happens in Civ3 anyway).

I don't play purely economic games because my military policy is an extension of my economic policy. Even when I'm pursuing "fastest space race victory" honors in a CivFanatics GOTM, military conquest plays a significant role because it provides an economic advantage. But the kind of game I generally enjoy most is the kind where the economic game plays the central role, with other aspects of the game supporting it.

Keep in mind that SimCity has sold a lot of copies in spite of its not having any real competition in it at all. People get their fun not out of trying to build a better city than an opponent, but rather from trying to build the best (or most interesting) city they can for themselves.

The beauty of Civ is that it combines simulation and competitive elements, thereby appealing to people who like a wide range of combinaitons of the two. For you, the competitive element is extremely important for its own sake, and so you play on difficulty levels where you feel a serious competitive challenge. In contrast, I get most of my fun out of the simulation aspect, with the "competitive" aspect playing a supporting role in the simulation. Ideally, I want the AIs to give me some serious competition as the "supporting cast" of the simulation, but if they have no realistic chance at all of actually beating me, that's fine. And I'm certainly not interested in shifting the simulation aspect of the game to a paradigm I enjoy significantly less just so the AIs will be more competitive. My real competition is not the AIs, but rather trying to build the "best" empire I can (in whatever way I define "best" in a particular game).

Both ways of playing are perfectly legitimate, and learning at Apolyton University can help people be more successful no matter what mixture of simulation and competition they enjoy in their games. I see no need for Apolyton University to try to push players into a difficulty level and style of play that undercuts their fun.

Most of these production bonuses are wasted, as you may know. In my experience watching the AI in Debug mode, the production bonuses are most relevant in the Demigod-Deity and Deity-Sid jumps.

Statistically, one would expect that instances in which the production bonus results in increased waste would be offset by other instances where the production bonus results in reduced waste. (Cases where an AI has plenty of shields for a settler or worker but is waiting for a city to grow would, of course, be an exception, but those shouldn't be an issue except in the REXing phase.) Have you actually conducted a careful analysis to determine how cases of increased and reduced waste balance out? Or is it possible that you're the victim of an illusion where you noticed cases of increased waste but weren't looking for offsetting cases of reduced waste?

The only thing that is really relevant here is the tech costs. But in practice this just means that it takes you longer to get a tech/branch lead. Once you have it, it's smooth sailing, just like Emperor.

Even after players reach that point, higher tech costs make it a bit harder for players to stretch a lead out. That's not likely to actually tip the balance in who wins, but it at least makes the really ridiculous mismatches (such as infantry and artillery against musketmen, or tanks against riflemen) less likely.

The extra Settlers do not "cancel each other out" with respect to the human player, which is what difficulty levels are all about (the AIs are always equal to each other at any given difficulty level). Watch a few games in Debug mode at Emperor and Demigod, I think you will find that the extra Settler is really what makes all the difference. Put another way, which would you rather have: slightly reduced tech costs, or two Settlers at the beginning of the game?

It's not just tech costs. It's also lower production costs and less food required for cities to grow. And if i were on a land mass by myself and I didn't plan to use granaries any more than AIs do, I think I'd choose the other advantages over the extra settler. (With lots of granaries, the growth difference between Emperor and Demigod is normally irrelevant, so it would be a closer call.) The main thing that makes the second AI settler so nasty on Demigod is how it can cramp the human player's ability to get a competitive REX.

By the way, in a debug game with the human civ not really doing anything, you don't even see the impact of a difference in tech costs. Remember that the costs for AIs is the same on any difficulty level; it's the human's costs that change.

I consider the Emperor-Mythic and Mythic-Demigod jumps way to small to show any significant improvement. Mythic difficulty would be the difficulty level for Demigod-level players who do not want a real challenge. I consider this against the aims of AU. IMO, too many players play the AU games at one difficulty level lower than they normally do/should (for various reasons), and adding a intermediate difficultly would only reinforce this.

The jump from Emperor to Mythic Hero would be comparable in magnitude to the jump from Monarch to Emperor or the jump from Demigod to Deity. And I don't see how you can argue on one hand that most of the AI advantages on Demigod come from the extra settler, but on the other hand that the jump from Mythic Hero to Demigod would be too small to show any sginificant improvement.

As for the impact on what difficulty level people play AU games at, how many people are playing on Emperor right now, and how many on Demigod? It seems to me that the potential for a "Mythic Hero" difficulty level to encourage players to move up in difficulty is a lot greater than the risk of its holding them back.

Yes, well, this is the real issue, is it not? Like I said above, once you get the hang of it (three games, tops) you (especially you!) will find that Demigod is not so bad.

I've played a few games on Demigod, some deliberately seeking out really fantastic starting positions and at least a couple (including one of the AU games) without. Can I win on Demigod? Yes. Do I enjoy the feel of games on Demigod? Only if I have a good enough starting position that the AIs can't use their advantage from their extra settler to crowd me too much.