AU Mod: Resource Scarcity

[pvzh]

On resoursce scarcity

Why is there so much ado about nothing?

Jesse told long time ago, that "resource scarcity" is NOT a design decision. It is a indirect and UWANTED consequense of adding new bonus resourses, and promised that they will address this problem.

Now of course seeing like old regullars and "addicts" generally admire the "change": creates KAI, provides challenge, "gives meaning", etc. Jesse might be talked into keeping it as is because the fix requres more than just upping frequencies in the editor.

However, unlike the regulars and "addicts" like me average gamer is not so fond neither of KAI, nor other stuff is listed in this thread. Fighting for saltpepper once in the game could be fun, but fighting for all other resources onward just because... will end up with "sent-this-****ing-CD-in-the-garbage-never-buy-anything-again-from-them" and unlike "addict" they might even do it.

[jshelr]

The new AU game is a good example of the human player getting a big advantage from resource scarcity. It was possible to dominate iron and horses during the first two eras on the home continent. Here come the cavs, with no opposition.

[Tall Stranger]

Re: On resoursce scarcity

Originally posted by pvzh
Why is there so much ado about nothing?

Jesse told long time ago, that "resource scarcity" is NOT a design decision. It is a indirect and UWANTED consequense of adding new bonus resourses, and promised that they will address this problem.

If this is the case, please provide a link to this statement. I have yet to see anything from Firaxis directly stating that this was unintended. (Doesn't mean I don't believe you, I'd just like to see it confirmed.)

Now of course seeing like old regullars and "addicts" generally admire the "change": creates KAI, provides challenge, "gives meaning", etc. Jesse might be talked into keeping it as is because the fix requres more than just upping frequencies in the editor.

That may be true, but we have just as much right to urge Jesse to keep this as you do to urge him to change it. And I'm not clear why you think that "fixing" this "problem" requires more than just changing the frequencies in the editor. I was under the impression that doing so is all that is required. (I think I saw something at CFC which said that multiplying the frequency rate by 1.33 will make C3C generate resources in the same ratio as PTW.)

However, unlike the regulars and "addicts" like me average gamer is not so fond neither of KAI, nor other stuff is listed in this thread. Fighting for saltpepper once in the game could be fun, but fighting for all other resources onward just because... will end up with "sent-this-****ing-CD-in-the-garbage-never-buy-anything-again-from-them" and unlike "addict" they might even do it.

Keep in mind that this discussion is on the AU Mod, not the game as a whole. If you (or the other "average gamers" you claim dislike the current set-up) do not like the current resource allocation on your standard game, you can easily change it in the editor. The debate we are having is whether the current allocation should be changed in the AU Mod. There are many changes to stock rules in the AU Mod which are specifically designed to make the AI more challenging and which the casual gamer may not like. No one is forcing the "average gamer" to use this mod.

[pvzh]

I do not remmeber where Jesse said that and not quite sure how to find it, so you may say it is anekdoatal. Nontheless...

I'm not clear why you think that "fixing" this "problem" requires more than just changing the frequencies in the editor. I was under the impression that doing so is all that is required.

The reason for that that every resourse has a dead space around it (1 tile radius), if it is luxury or bonus resource it may have another copy within these 8-tiles. Obviously, these another copy provides some more dead space. The only exception I saw is fish on coast does not prevent resourses next to it on land, but that is in C3C. The result, that new sugar on plain or tobacco on grassland prevent iron or coal or <...> appear in the nearby mountain or hill. Upping appearance ratio by 1.33 still require enough land. Even before C3C (in PTW) I had many small maps where only 5 types of luxuries, or some tiny maps with no luxuries at all. I think, ratio fix may work only for standard maps with at least average land.

Rest in my post was about why Jesse might be motivated to fix it, but being back on topic: "... for AU-mod".

It weakens AI's more than does the human.
1. Human can use bombard units and many of them do not require any resources: I could imaging a protracted Archer/Spear/Catapult fight against AI swords/spears to get an iron, but I will not expected it from an AI.
2. Humans are far better in trade than AI.
3. The fewer resources the easier to deny them (and not a Human will be denied).
4. AI's are very very bad at connecting and protecting the resources; they assume to much. It was a bane of Roman AI: plant a city with an iron in the 2 tile radius on mountain and "never" build a cultural building to get it in the border, nor colony. Romans have no or very little priority for cultural buildings, but AI is not programmed to handle this.
5. AI has no clue that it has to fight for resource. All it can do more or less effectively is to trade for them, but alas even so, I saw many large prosperous AI on Emperor bothering to build a harbour somewhere in the middle of the Middle Ages despite the fact that there are mamy deals awaiting them if they made so.
But again, you need extra-copies to have a trade, ant that is not the case with resource scarcity.

Obviously all we can do is: delete added bonus resources (they are rather weak anyway) or increase frequencies by 1/3.

[alexman]

Originally posted by pvzh
It weakens AI's more than does the human.

There are good arguments for and against this statement. If you by some miracle manage to follow this thread (thanks SR), you will find them.

Anyway, I agree with Catt on the AI's performance. My experience is limited, but from what I have seen in practice, the luxury scarcity in C3C does not make the AI weaker, on average.

On the other hand, I can more easily accept the argument that the scarcity limits human options, as I have seen many players experience it, and complain loudly about it.

But "more easily" does not mean without having doubts. The problem with the limited options argument, IMO, is that it applies more to players who like to play at an easy level compared to their ability (i.e. have fun), and thus are able to win the game by any victory condition they like, just because they have a better economy. If you are an age ahead of the AI, of course you will have unlimited options on how to win, unless you lack a resource or two.

So increasing the difficulty level also decreases options for the human, because any departure from an optimal strategy in a given situation becomes more costly. That doesn't mean that the AU mod should make harder levels easier though, I hope.

[pvzh]

It weakens AI's more than does the human.

There are good arguments for and against this statement. If you by some miracle manage to follow this thread (thanks SR), you will find them.

I did, and reread again on you request. There are only 2 arguments pro resource scarcity. It promotes warfare (rather dubious because Humans are better at that than AI), it creates KAI. I have never saw Sir Ralph's posts, but I doubt that he had anything to support resource scarsity.

That doesn't mean that the AU mod should make harder levels easier though, I hope.

And that it is exactly what happends with resource scarcity: on the levels like demigod and above you frequently need a beaten up civ to start your conquest from, and that scarcity and KAI's provides you them. It creates a nice ladder to scale you up. Kill the smallest -- grow a little, kill a bigger -- grow a bit more, etc.
My recent first demigod game was like that, and AU 501 had a good examples of that with Japanees, Sumerians and Hitties that all lacked resources pretty much. Would not I like a peaceful builder game, they would have all be dead by middle ages (before saltpepper).
The famous Aesons "So very cold" was again moving up by beaten up AI's.

The situation now: one AI killed another without Iron and became KAI. Sooner or later I am onto it and I pilladged/cut-off its Iron. Game ends for that AI (otherwise I would not started the attack). In Civ3 i would not really bother to cut it of because it would have also require to ally every AI with surplus Iron against it (might be very expensive on high difficulties), but with resource scarcity it is not necessary in most cases there are none or next to none tradable resources on the market.

To clarify my standing point on this: all stratigic resources including upto Industrial Age have to be on average 1 per player (it does not mean that they are guaranteed for every player: a land mass could be pretty much void of one or two resources and filled with several others). Thus, killed or diminished AI's will create trable surplus again on average, so AI's/KAI's can repurchase pillaged/lost resources from other AI's.

NB: AFAI, no modern resource is required for modern defensive units (MI, TOW), unlike industrial (Infantry). This is the reason why I stated upto industrial included. And all pre-modern resources except horses (MI does need oil, right?) are needed either for defender or crucail in economic development. Thus AI MUST have it relatively easy.

[nbarclay]

The resource is saltpeter, not saltpepper. I wasn't thinking about it before, but pvzh is right that C3C has largely destroyed the old concern that if you take away a civ's domestic source of a strategic resource or if the civ doesn't have one, they might trade for the resource in the middle of your war with them.

[alexman]

Originally posted by pvzh
on the levels like demigod and above you frequently need a beaten up civ to start your conquest from, and that scarcity and KAI's provides you them. It creates a nice ladder to scale you up. Kill the smallest -- grow a little, kill a bigger -- grow a bit more, etc.

Sometimes, that will happen, sure. But other times the weak AIs will get eliminated by other AIs before you can claim their land. On average, I have not seen evidence (other than speculations) that what you say happens more often than it used to happen in PTW.

The situation now: one AI killed another without Iron and became KAI. Sooner or later I am onto it and I pilladged/cut-off its Iron. Game ends for that AI (otherwise I would not started the attack).

Iron is already more than one per civ in C3C. Kill an AI and you get even more surplus. Clearly, the situation you describe can happen, but I don't think it happens often enough to justify changing the appearance ratio of Iron. If there were more Iron on the map, perhaps there would be no KIA in the first place.

To clarify my standing point on this: all stratigic resources including upto Industrial Age have to be on average 1 per player (it does not mean that they are guaranteed for every player: a land mass could be pretty much void of one or two resources and filled with several others). Thus, killed or diminished AI's will create trable surplus again on average, so AI's/KAI's can repurchase pillaged/lost resources from other AI's.

This assumes a certain elimination rate of civs in the game. I'm not sure we have enough experience to assume anything about that. What if the lack of resources makes weak AI civilizations become eliminated faster in C3C? Then you might get back to PTW levels of resource surplus, but with more killer AIs.

[nbarclay]

Originally posted by alexman

Iron is already more than one per civ in C3C. Kill an AI and you get even more surplus. Clearly, the situation you describe can happen, but I don't think it happens often enough to justify changing the appearance ratio of Iron. If there were more Iron on the map, perhaps there would be no KIA in the first place.

The formal stats on iron and horses are a bit misleading. When some of the iron is on islands (and the following logic is the same for horses), civs to research Map Making, build galleys, send out settlers, found cities, and build harbors in those cities before the iron can do them any good in their core. Further, without either the Great Lighthouse or Astronomy, the iron is useless to the civ's core cities even then if the island is separated from them by sea rather than just coast tiles. Also, in the absence of an overseas trading capability, AIs in one part of the world can be stuck without iron while another part of the world has more than it really needs. And some of the iron may just be far enough from civs' cores that it doesn't get a road hooked up to it and get connected to the core road network until long after Iron Working is researched. Further, what happens when one civ has two irons but doesn't have trade links with another civ that needs it in the early game?

In contrast, by the time saltpeter comes into play, the home continents are essentially entirely within someone's cultural borders or other and islands near a home continent are settled. Astronomy is around the corner if it hasn't been researched already, and Navigation is likely not much farther away. Islands are even more developed by around the time industrial resources come into play, and more developed still by the time of modern ones. Thus, it doesn't take nearly as high an appearance rate of aluminum or uranium as it does of horses or iron to more or less ensure that there will be enough to go around.

This assumes a certain elimination rate of civs in the game. I'm not sure we have enough experience to assume anything about that. What if the lack of resources makes weak AI civilizations become eliminated faster in C3C? Then you might get back to PTW levels of resource surplus, but with more killer AIs.

What I think we really have is greater inconsistency among games played on the same difficulty level and with similar starting conditions. Sometimes resource scarcity helps a killer AI emerge (and I do see a pretty clear trend that AIs kill each other a bit more in C3C than they did in previous versions). In other cases, the player gets extra help finding weak targets to take out. And sometimes both can happen in the same game.

That, in turn, creates an interesting issue regarding people's perceptions. Those who mostly notice and remember situations where resource scarcity worked in their favor look at the situation one way, while those who mostly notice and remember situations where it seemed to help a killer AI emerge look at the situation a very different way. Unfortunately, the number of games that would be required to get a statistically valid sampling would be so high as to make a truly objective evaluation impractical, so we're stuck with subjective biases coloring the issue.

Personally, I think the game is better with less variation between easy and hard games played on the same settings and from similar starting positions. That makes it easier for players to get the kind of challenge they want without getting more than they bargained for. And I think the traditional resource distribution was better in terms of being able to predict how challenging a game is likely to be.

Nathan

[Catt]

I'm saying the same thing alexman, I think, but . . .

Originally posted by pvzh
I did, and reread again on you request. There are only 2 arguments pro resource scarcity. It promotes warfare (rather dubious because Humans are better at that than AI), it creates KAI. I have never saw Sir Ralph's posts, but I doubt that he had anything to support resource scarsity.

I think the principal argument put forward by Dominae among others, repeatedly, was that relative scarcity makes resources truly strategic -- saying that it unduly promotes warfare has been an argument advanced in opposition to C3C's resource levels, and one that hasn't been acknowledged as fact (at least for me). There are others there, too!

And that it is exactly what happends with resource scarcity: on the levels like demigod and above you frequently need a beaten up civ to start your conquest from, and that scarcity and KAI's provides you them. It creates a nice ladder to scale you up. Kill the smallest -- grow a little, kill a bigger -- grow a bit more, etc.

This may be true in certain circumstances; the opposite may also be true -- that you may have much more trouble killing the smallest and moving up. But is your view that C3C's resource scarcity makes warfare both easier and more attractive (versus other arguments that scarcity makes conquest necessary)? Map features and start locations just as easily make the ladder you refer to more accessible to the AI as to the human. Indeed, if you assumed that, on any given map, there would be 2 AIs who have no critical resources and would make ripe targets, those 2 are more likely to be convenient targets for another AI than for the human.

My recent first demigod game was like that, and AU 501 had a good examples of that with Japanees, Sumerians and Hitties that all lacked resources pretty much. Would not I like a peaceful builder game, they would have all be dead by middle ages (before saltpepper).

Remember that AU 501 was not a random map. Resources were moved away from the human's starting position deliberately (on the other hand, I didn't find the map an extreme anomaly). If you have the fortune (good or bad) of starting to a bunch of resource deprived AIs, your conquest will indeed be easier if that is the route you go. But it is just as likely that you'll end up deprived with a bunch of resource-enabled AIs, and likely that one or more resource-deprived AIs will be located across the seas and out of your reach for a long time. I don't think you've made the case that lower resource levels means easier conquest for the human, on average.

The situation now: one AI killed another without Iron and became KAI. Sooner or later I am onto it and I pilladged/cut-off its Iron. Game ends for that AI (otherwise I would not started the attack). In Civ3 i would not really bother to cut it of because it would have also require to ally every AI with surplus Iron against it (might be very expensive on high difficulties), but with resource scarcity it is not necessary in most cases there are none or next to none tradable resources on the market.

This happens among AIs, too. The issue is not whether individual AIs might suffer, IMHO, it is whether the overall game (facing numerous coimpetitors) is more engaging and strategic. You can't really be everywhere at once and if you could, the game is over in any event. Especially where a KAI emerges, there is an increased chance of surplus resources available on the market. It is axiomatic that with less total resources overall, resource denial is a more powerful tactic, and I agree that the human is better at it than an AI (especially when the C3C AI won't bombard resources, dammit ).

To clarify my standing point on this: all stratigic resources including upto Industrial Age have to be on average 1 per player (it does not mean that they are guaranteed for every player: a land mass could be pretty much void of one or two resources and filled with several others). Thus, killed or diminished AI's will create trable surplus again on average, so AI's/KAI's can repurchase pillaged/lost resources from other AI's.

NB: AFAI, no modern resource is required for modern defensive units (MI, TOW), unlike industrial (Infantry). This is the reason why I stated upto industrial included. And all pre-modern resources except horses (MI does need oil, right?) are needed either for defender or crucail in economic development. Thus AI MUST have it relatively easy.

The standard number of resources available worldwide has always scaled downward as one progresses through the tech tree; this was true in Civ and PTW. And this makes sense to me in that one might expect that the number of civs alive in the modern age is almost always less than started the game. But I guess I just disagree with the need for resources equal to an average number of civs. If one plays the game to win, its hard to imagine the other 7 AIs as anything other than opponents (assume 8 civ game). My point is that resource scarcity may indeed hurt one or more particular AIs -- but often enough the party that benefits from a resource-deprived AI is another AI. Even on a pangaea map (let alone more isolated maps), a resource-deprived AI is more likely to be a convenient conquest target for another AI than for the human (just based on the number of AIs in the game). Development of a large AI empire early in the game makes for a much more engaging later game, IMHO. I'd much rather play a game that quickly gets to 4 or 5 relatively strong AIs than one that remains 7 moderate AIs well into the Middle Ages. The AI is strengthened by a larger empire of course, and I think it makes for a better game if those larger empires get established early so that conquered cities become early,productive powers for the AI.

Much of the above is really well-educated guesses on the impact of C3C resource scarcity. My real point is that, despite the incredible Civ experience and skill represented by the posters on this board, my own experience with C3C have not followed expectations (mine or others'). My experience is limited, as I am sure others' is, but thus far I've seen little evidence that resource scacrity helps the human more than the AI. On the contrary, to me it seems to promote a more even distribution of power -- weak AIs are swallowed up by strong AIs or the human, and they are swallowed up earlier than in PTW. The Industrial Age and even the Modern Age are more interesting now to me personally (YMMV), and I think the "pinch" of expected resource scarcity is at least partially offset by the increased likelihood of early AI exits.

If I knew that C3C resource scarcity was truly a significant contributing factor to larger AI empires earlier in the game, I'd be agitating to keep it unchanged in future patches. Thing is, I don't know -- I can only speculate. What I'm asking is whether my speculation is anywhere close to reality, and whether others' speculation about harm to the AI competitor with scarcity is anywhere close to reality. I hear frustration at numerous expriences with not having resources close to home, but I don't hear numerous conclusions, based on experiences, that scarcity hurts the AI (I hear only well-educated guesses that it will).

Catt

[Catt]

Originally posted by nbarclay

That, in turn, creates an interesting issue regarding people's perceptions. Those who mostly notice and remember situations where resource scarcity worked in their favor look at the situation one way, while those who mostly notice and remember situations where it seemed to help a killer AI emerge look at the situation a very different way.

I think one might keep this in mind, but at the same time, we've got to give credence to the view that those with long-standing experience with the game (as most posters here have) are able to evaluate the situation without undue influence of one or more specific experiences. I'd venture that most of the community that both follows and posts in the AU forum, and customarily address gameplay issues in the context of the AU Mod discussions, are capable of bringing a view that is relatviely unbiased to how the game works under C3C's scarcity.

On the other hand . . .

Unfortunately, the number of games that would be required to get a statistically valid sampling would be so high as to make a truly objective evaluation impractical, so we're stuck with subjective biases coloring the issue.

I agree that we're a long way from any statistically valid sampling and conclusion. But again on the subject of subjective biases, I think posters on the whole have done a pretty good job of pointing out that "I don't like W," "X is no fun," "Y is more engaging," or "Z is more fun." The challenge that I see is, as objectively as possible, reporting a broad base of actual experiences, rather than a narrow focus on a few instances (good or bad) or the alternative of more educated guesses on how the resource level will play out on average. Civ is a complex system, and (where's the "beat a dead horse" smiley?) I am just not convinced that our collective educated guesses are hitting the mark.

And I think the traditional resource distribution was better in terms of being able to predict how challenging a game is likely to be.

Not sure if I agree, but I think I lean that way.

Personally, I think the game is better with less variation between easy and hard games played on the same settings and from similar starting positions.

Not sure if I agree, but I think I lean the other way. The desired granularity of each game's difficulty is highly subjective, as you note. I have little doubt that my tastes won't (and shouldn't) be taken to have more weight than others' tastes so long as the vast majority of players feel that the game offers sufficiently granular difficulty challenges through whatever means.

(and I do see a pretty clear trend that AIs kill each other a bit more in C3C than they did in previous versions)

Do others see this as a clear trend, too?

Catt

[nbarclay]

Originally posted by Catt
Even on a pangaea map (let alone more isolated maps), a resource-deprived AI is more likely to be a convenient conquest target for another AI than for the human (just based on the number of AIs in the game).

On the other hand, human players tend to make a strong, deliberate effort to take advantage of a neighbor that is at a disadvantage due to being missing a vital resource, while whether an AI will take advantage of such a situation is a lot more hit-or-miss. Further, AIs that are winning wars not infrequently make peace rather than finishing the rival they've half conquered off. Thus, while there are more situations where an AI could profit from a neighbor's lack of a resource, the issue of how often and how effectively AIs actually do exploit such lacks is another matter entirely.

[Tall Stranger]

pvzh: I'm sorry you don't think that the numerous reasons a number of us have laid out for supporting resource scarcity are even worth acknowledging in your discussion. For the record, I have never stated that it "promotes warfare." That is, as Catt points out, a claim made by scarcity's opponents, not its supporters. In fact, I laid out a number of ways that one might be able to obtain rare strategic resources without doing much in the way of fighting. I'd be interested in hearing why you think they won't work or aren't applicable. (For example: I may be completely wrong in my analysis there.)

Catt: It should come as no surprise, based on my arguments, that I do see AIs killing AIs as more common now as compared to PTW. And I agree with you that 4-5 decent AI civs make for far more interesting a game than 7 mediocre ones.

IF this is a trend, logic tells us that a number of factors could be involved (there are probably more; these are just the ones I came up with off the top of my head):

1. Changes to aggression levels between PTW and C3C may have made AI-AI wars more likely, hence making an AI civ's destruction more likely;
2. Generic improvements to the AI have made it more capable in combat, meaning that a given AI-AI war is less likely to grind into a stalemate;
3. Resource scarcity has rendered some AIs less able to defend against another AI, leading to their destruction.

(#2 is probably the least likely, but I will say that the AI is much more likely to aggressively counterattack now than it ever did in PTW.)

While I would certainly support some experimentation to determine if the game needs balancing, I fear that we will immediately run into the problem of defining "game balance." I think the hardest question we'll need to answer is "Since any change is going to benefit one play style and harm another, which one do we choose?" I'm not seeing much consensus on this question.

[Catt]

Originally posted by nbarclay

On the other hand, human players tend to make a strong, deliberate effort to take advantage of a neighbor that is at a disadvantage due to being missing a vital resource, while whether an AI will take advantage of such a situation is a lot more hit-or-miss. Further, AIs that are winning wars not infrequently make peace rather than finishing the rival they've half conquered off. Thus, while there are more situations where an AI could profit from a neighbor's lack of a resource, the issue of how often and how effectively AIs actually do exploit such lacks is another matter entirely.

Same proviso regarding my limited sample size . . .

I seem to be seeing a lot more "fights to the death" between AIs. In my games I am seeing either a stalemate situation followed by peace (similar to PTW) or a relatively thorough eviscerating -- which seems to contribute to the possible trend of early AI exits. Haven't a clue whether this is an anomaly, is the product of a multitude of factors other than scarcity, or may be influenced by scarcity (and the relative military balance of power that I suspect may play a role in AI peace willingness).

Originally posted by Tall Stranger

Catt: It should come as no surprise, based on my arguments, that I do see AIs killing AIs as more common now as compared to PTW.

Thanks for the info.

While I would certainly support some experimentation to determine if the game needs balancing, I fear that we will immediately run into the problem of defining "game balance." I think the hardest question we'll need to answer is "Since any change is going to benefit one play style and harm another, which one do we choose?" I'm not seeing much consensus on this question.

I suspect that Firaxis will change it back to PTW regardless -- I don't have the link, but there was a post by Jesse that the introduction of new terrain types and resources affecting resource levels which was not intended. And I doubt that the AU Mod community would propose restoring C3C resource levels if it is pushed back to PTW levels.

Catt

[nbarclay]

Originally posted by Tall Stranger

Catt: It should come as no surprise, based on my arguments, that I do see AIs killing AIs as more common now as compared to PTW. And I agree with you that 4-5 decent AI civs make for far more interesting a game than 7 mediocre ones.

How often have you actually seen two AIs out of seven killed at the hands of other AIs without human involvement? I don't recall any games offhand where I saw that happen, but I'm not positive. Of course between my military aggressiveness and my research aggressiveness (which tends to lead to shorter games), I probably don't give the AIs quite as much chance to have two killed by other AIs as a lot of players do.

Nathan