What do you consider an exploit?

I suppose that's what feels "exploitative" about it ... there are not currently enough costs associated with the manuever, IMO.

Yeah, if the consequences were greater, I wouldn't care. The point is, you can continually drain the AI's money with essentially no consequences in preparation for war. There isn't much I consider a genuine exploit, but this I do.

Likewise, I considered ROP rape in Civ3 to be an exploit. If you get open borders and moved a huge army in, you could win the game with the AI being completely unable to compensate. They changed this in Civ4 so you teleport outside their territory if you declare war. It wasn't as realistic, but it was to prevent a huge, unfair advantage. This was even though the AI could technically do the same thing. It wasn't the same in practice, though, just because humans do things on a grander scale that makes it essentially game-breaking.
 
Quick question: Are there any other "maneuvers" that are as lucrative on a per turn basis as resource/lump sum trading followed by a dow?
 
Utterly ridiculous. By approaching t = 0, EVERY TACTIC IN THE GAME approaches 0 opportunity cost, until it reaches 0 at t = 0. That is not a valid basis for refuting the claim that DoW consistently has opportunity cost. That a tactic is *sometimes* the dominant option is not a valid basis for ruling something bad for competition. It's also invalid to claim t = 0 is known until one is very close to it. You can't make a serious case that something is an "exploit" on this basis.

This isn't the only case. It is merely the most easily mathematically demonstrable one; the simplest existence proof. Many others exist. Eg: fake wars with civs on another continent when not tech limited (say, you need ten RAs with seven opponents to end the game), because the AI is incapable of executing amphibious assaults. In that case the only opportunity cost is the diplo hit...and that cost is nil if you've already re-upped the last round of deals with civs you aren't already in a DoF with.

Further, I went on to make the case (beyond the existence proof) that the endgame problem is not trivial and limited to the final turn. Distant civs many hexes away can safely be declared on many turns before the game ends, as long as you are not currently in a Research Agreement. The turn the game will end on can be known with near certainty thirty to forty turns in advance for all peaceful win conditons on normal speed. Adjust the victory estimate for game speed based on Research Agreement duration.

If you're going to insist on misrepresenting my argument by taking a portion of it out of context and ignoring the rebuttal of the quoted claim that I have already made (twice), there isn't much point in arguing with you. I understand why you get so upset about the behavior when you observe it, but perhaps you should realize that sometimes you do the same thing and exercise a little forgiveness? Perhaps when you observe such an issue, the problem is one of miscommunication (either reading comprehension, poorly worded argument or both) rather than an attempt at deliberate distortion of the truth.

Yes, your fake rules add some minor strategy, but it also removes strategic usage of the "exploit" based on situation.

The rule also reduces the influence of random factors. You have repeatedly insisted that game rules should attempt to reduce the influence of randomness. So which problem is more important: getting rid of random factors or preserving opportunities for strategic interaction?

Answer: preference. I'd argue that the "strategic interaction" you're chasing is chimerical; opportunities to declare fake wars for monetary gain are generally pretty obvious. Sure, they'll tend to happen late due to diplomatic opportunity cost, but knowing when you will win thirty turns from the end of a Quick speed game that lasts 120 turns offers a lot of chances for fake war, no?

In insisting on this t ---> 0 nonsense, you are still ignoring the point that war consistently has opportunity cost.

I proved mathematically that war does not consistently have opportunity cost. Other cases exist (see above). Perhaps it would have been helpful had I cited them, but the simple fact of the matter is that you are wrong and that this can be demonstrated by cold, hard, unassailable logic.



I'm not going to go any further because you have failed to address any of the major components of my argument, which can be summarized as follows:

- Exploit can and should be defined minimally and loosely (see first post). An actionable exploit (bannable for rules purposes) requires meeting a higher standard of enforceability.
- Unfortunately, the detrimental impact component of the definition of "exploit" is entirely preference based. Nobody is right or wrong in their preferences over how they would like to see the game behave.
- The result is that the answer to the question in the OP is functionally subjective. Attempts to impose some sort of order on others' responses or seek out a universal definition of "exploit" are therefore misguided. Deciding which "exploits" stay and which go from a game is a social choice problem (ie: deciding how to resolve conflicting preferences), and I'm willing to bet that Shafer left Firaxis primarily because he wasn't willing to solve that problem in a manner consistent with the preferences of a large share of the active community and/or other influential figures at Firaxis.

Dismissing the definition as "not useful" doesn't alter its accuracy.
 
Civ5 is a 4X game.

Kinda' hard to exterminate without first exploiting.

Can you even have an exploit in a game with rules designed to be exploited?

Exploit is about exploiting resources, no? I assume it doesn't mean exploit flaws in programming.
 
I think part of the problem with the term "exploit," as has been pointed out, is that it means different things to different people. And as has also been pointed out, this is subjective and a matter of personal preference.

1. "exploiting" a bug in the system. e.g., a game error that gives the player an advantage due to a programming error. I think most would agree that this is exploitative.

2. "exploiting" certain game mechanics to generate an "unfair" advantage. e.g., the "lump sum trade and then DoW" tactic. Some find this exploitative while others feel that, since the AI does it too, it's not our fault that there are insufficient diplomatic penalties for pulling this stunt.

3. "exploiting" the AI tactically. Since the AI plays a certain way (militarily, expanding, etc.), if we learn its tactics and adopt counters accordingly, we can be said to be exploiting the weaknesses of the AI. This is normal and expected in any game (and in life, for that matter), though it is exploitative in the dictionary sense of the word.

The debate here is primarily about whether certain player actions are in the second or third category.
 
Here's the question about the second one, does the AI do it intentionally or do they just not factor things in correctly?

I think the difference between two and three is whether or not it's a conscious strategy. Seeing how the AI has their units out and reacting to that is common sense and a necessity. Otherwise, you could lose the game. Choosing to put the AI in a specific position in order to then gain some advantage with risk or consequence is entirely different.
 
What I'm arguing is that the difference between 1), 2) and 3) is a matter of degree, since AI tactics and game inconsistencies are a subset of game mechanics. Where you draw the line is purely a matter of personal preference. There's probably even somebody out there that thinks, "The AI cheats, and since it cheats it's fine if I build Oxford multiple times even though I should not be able to." I (along with most of you) happen to disagree with that position, but no argument exists that will make that position objectively wrong.
 
Yeah, if the consequences were greater, I wouldn't care. The point is, you can continually drain the AI's money with essentially no consequences in preparation for war. There isn't much I consider a genuine exploit, but this I do.

Ignoring previous points is bad form, I see no reason to counter your points if you don't bother addressing mine.

It wasn't the same in practice, though, just because humans do things on a grander scale that makes it essentially game-breaking.

So basically everything is game breaking.

This isn't the only case. It is merely the most easily mathematically demonstrable one; the simplest existence proof. Many others exist. Eg: fake wars with civs on another continent when not tech limited (say, you need ten RAs with seven opponents to end the game), because the AI is incapable of executing amphibious assaults. In that case the only opportunity cost is the diplo hit...and that cost is nil if you've already re-upped the last round of deals with civs you aren't already in a DoF with.

The point is that this by itself isn't sufficient grounds to make something an exploit. DoW still has very real consequences, regardless of whether you care about them any longer or not. The decision on when/if to compensate for the consequences (or when they are acceptably minor) is part of the strategy of the game. To turn your argument around: is this a tactic you always want to execute at every opportunity in order to reach a fast victory? I'm betting you can't honestly answer yes to that, so your approach to "proving" exploit here is pretty weak.

Further, I went on to make the case (beyond the existence proof) that the endgame problem is not trivial and limited to the final turn. Distant civs many hexes away can safely be declared on many turns before the game ends, as long as you are not currently in a Research Agreement. The turn the game will end on can be known with near certainty thirty to forty turns in advance for all peaceful win conditons on normal speed. Adjust the victory estimate for game speed based on Research Agreement duration.

Even in this paragraph, you qualify the DoW! Think for a moment why you had to qualify it! You are also still ignoring that *all* opportunity cost drops as you near endgame since consequences for anything (good or bad) tend to run out of time to occur.

If you're going to insist on misrepresenting my argument by taking a portion of it out of context and ignoring the rebuttal of the quoted claim that I have already made (twice)

Aside from material the mods asked us not to discuss, you have no rebuttal I didn't address. All you've done is nitpick my "always" without providing a meaningful basis for calling DoW to break gold deals an exploit. I will concede that the opportunity cost is not *always* meaningful then, if that irrelevant point matters to you, but nevertheless point out that the evaluation of when doing this is helpful rather than harmful involves strategy and that the opportunity cost follows the precise pattern of the vast majority of in-game tactics - a point that ironically you've chosen to ignore in your "rebuttal".

The rule also reduces the influence of random factors. You have repeatedly insisted that game rules should attempt to reduce the influence of randomness. So which problem is more important: getting rid of random factors or preserving opportunities for strategic interaction?

Answer: preference.

Yeah, it's a tough question for me to answer too. The random factors I've hated the most are always the ones that involve no strategy or minimal strategy; when asked to choose between a random factor that adds strategy and its removal it becomes quite difficult.

opportunities to declare fake wars for monetary gain are generally pretty obvious

So is building libraries, only more so. Remember that we're both high level players. It's obvious to *you*, but watch a prince player play through the game with the mind to use this tactic every time the benefits outweigh the opportunity costs, and watch as he botches opportunities to use it, overuses it, or loses himself the game. On the other hand, he'll probably build libraries when it's cost effective, and with less coaching.

Sure, they'll tend to happen late due to diplomatic opportunity cost, but knowing when you will win thirty turns from the end of a Quick speed game that lasts 120 turns offers a lot of chances for fake war, no?

Depends how quickly you can get out of those fake wars. Also, you're now saying "this tactic looks really strong when I cook the settings to make it look strong".

What if I turned that on you, and dumped a huge marathon game on your lap? How many "fake wars" do you want to try?

You're still not differentiating this tactic from standard in-game tactics, making it increasingly hard for you to actually make an exploit case.

I proved mathematically that war does not consistently have opportunity cost.

Wrong. You proved mathematically that war does not consistently have *meaningful* opportunity cost. It always has some, unless you're actually ON t=0 where of course nothing matters in-game. The opportunity cost is always there; it's up to the player to determine when it's to high and when it is not.

Perhaps it would have been helpful had I cited them, but the simple fact of the matter is that you are wrong and that this can be demonstrated by cold, hard, unassailable logic.

Perhaps it would be helpful if you cited cases where this "exploit" differed from basic tactics like "sign research agreement" or "capture opposing cities, but the simple fact of the matter is that you are wrong and that this has been demonstrated by cold, hard (not warm and hard), unassailable logic...or at least your failure to provide any.

I'm not going to go any further because you have failed to address any of the major components of my argument, which can be summarized as follows:

- You've provided no reasonable basis for why the definition of exploit should be based on your preferences, and we've never disagreed on enforceability (possibly, this is why I didn't attempt to counter that part of your argument...)
- I'm not sure what meaningful response you want to "this is preference", other than that "preference" is not a valid means of banning something in a competitive setting, and is as valid as anything else in non-competitive settings.
- If discussing the basis of an exploit definition is misguided, why are you using misguided arguments against me? The purpose of this thread is to discuss a social choice problem; reasonable criteria for common grounds isn't a groundbreaking idea.
- While my guess on the purpose of Shafer's leaving the company is probably more educated than yours, I don't know his actual reason and neither do you.
- The definition of "exploit" in any dictionary you pick up suggests that it is a good thing to seek to do so actively in civ V. Not only the way people apply it, but even its base usage is inaccurate.

1. "exploiting" a bug in the system. e.g., a game error that gives the player an advantage due to a programming error. I think most would agree that this is exploitative.

"Most" would only agree if the bug allows one to actually overcome/break the stated rules (IE cheating).

2. "exploiting" certain game mechanics to generate an "unfair" advantage. e.g., the "lump sum trade and then DoW" tactic. Some find this exploitative while others feel that, since the AI does it too, it's not our fault that there are insufficient diplomatic penalties for pulling this stunt.

You've only demonstrated a fraction of the argument from the side claiming that this isn't exploitative, and the other points are pretty strong.

The debate here is primarily about whether certain player actions are in the second or third category.

Actually, a large part of the debate is whether #2 is a legit exploit. The only way #3 has been brought in is largely to compare the impact of tactics that fall under #2 or to debunk arguments by showing that arguments against a tactic in #2 also call tactics in #3 "exploit".

Exploit is about exploiting resources, no? I assume it doesn't mean exploit flaws in programming.

Separate programming flaws from intentional AI inadequacy, compensatory bonuses, and design decisions...and prove it. Nobody can do that, which is why this "flaws in programming" nonsense is a bit ridiculous to hinge an argument; better to use more valid approaches.

So far, the "this is an exploit" side of things has done no better than "personal preference of the majority dictates that things beyond cheating/bypassing game rules are exploits".
 
This is more an urban dictionary definition though. The actual use of exploit as a term for tactics in gaming is confusing; generally exploiting your opposition is a good thing by the actual definition of the word.

"Urban dictionary definition?" What does that mean - one with which you disagree?

The term is indeed confusing, because by the "actual definition of the word," exploiting your opposition is not necessarily a good thing. One of the two basic definitions of "exploit" refers to selfish acts, and does so in a negative light. Anyone could argue what that means - which in turn means that basically no definition of "exploit" in this thread is wrong. To paraphrase Martin, they're just inherently subjective.
 
The point is that this by itself isn't sufficient grounds to make something an exploit.

In your opinion, which is an opinion that is entirely based upon preference.

Aside from material the mods asked us not to discuss, you have no rebuttal I didn't address. All you've done is nitpick my "always" without providing a meaningful basis for calling DoW to break gold deals an exploit.

This is because you're still not listening. You make a rule against doing something because the implications of the behavior are undesirable and because the gains to the individual of engaging in the behavior exceed the opportunity cost in a sufficiently meaningful proportion of cases to make it worthwhile to expend the cost to enforce the rule. Most of the time, most people would prefer not to murder other people because doing so is costly. You make a rule against murder and enforce consequences to deal with the cases where the gains exceed the opportunity cost; in other words, where the opportunity cost does not bind and is therefore not "meaningful".

I've made a strong case that the opportunity costs of fake wars are not meaningful in the latter phases of most games, and that a rule against the behavior is therefore necessary to prevent the undesirable implications of the behavior - an increase in luck factors. I've gone on to point out that that's my preference, and that no one else is objectively right or wrong when they disagree with me.

Remember that we're both high level players. It's obvious to *you*, but watch a prince player play through the game with the mind to use this tactic every time the benefits outweigh the opportunity costs, and watch as he botches opportunities to use it, overuses it, or loses himself the game. On the other hand, he'll probably build libraries when it's cost effective, and with less coaching.

This is irrelevant. In a competitive environment, imposing additional rules to deal with "exploits" is only going to affect outcomes amongst the most competitive players. The Prince player loses the competition irrespective of the rule set.

Depends how quickly you can get out of those fake wars. Also, you're now saying "this tactic looks really strong when I cook the settings to make it look strong".

You don't have to cook the settings or get out of the wars. Last time I checked, every single game has a denouement during which it would be possible to make fake wars without resolving them. You prove that you don't understand what I'm telling you by citing the example of huge Marathon. Fake war during the terminal stages of a game gets better as you expand the number of players and the game speed. The AI will pay you more cash, there are more distant AIs to milk and those factors scale faster than building cost!

Wrong. You proved mathematically that war does not consistently have *meaningful* opportunity cost. It always has some, unless you're actually ON t=0 where of course nothing matters in-game.

1) Only the cases where opportunity cost does not bind matter to the necessity of a rule. Consider it this way: you can write up a maximization problem for any arbitrary behavior with opportunity cost as the s.t. portion of the equation. If the solution is non-empty, you will observe the behavior in question. If the behavior is undesirable, you need a rule against it so long as the enforcement problem is not large enough to outweigh the gains from enforcing the rule. (Notice how preference sneaks back into the terms for enforcing the rule and the cost of the enforcement problem.)
2) Since AFAIK the diplo hit doesn't land until the next turn, on t-1 the opportunity cost is still zero. This is also true for any and all cases where all of the terms in the opportunity cost function collapse to zero, and that's a much larger set than you seem to think. For instance, a diplo hit has zero cost if you're not negotiating deals and AIs don't have time to organize an attack.

I'm not going to go any further because you have failed to address any of the major components of my argument, which can be summarized as follows:

Clever rhetorical tricks don't constitute an answer to my criticisms. There's a preference component to the definition and any attempt to impose further order on the definition is nothing more than the attempt to impose your own preferences on others. That's been my core contention the entire time, and we've had several exchanges without you making any attempt to answer it. If you want to concede the point and confine further discussion to resolving the social choice problem, that's fine with me. If not, I'm going to persist in claiming that the definition I provided is intersubjective and that what we have is a social choice problem of resolving conflicting preferences rather than a problem of operational definition. This matters, because the proper solution depends on which type of problem is under discussion.

As far as solving the social choice problem goes, any headway there requires creating choice rules that the arbitrator can consistently apply. While that seems to be what you're after, there's a serious bargaining problem over those rules (resulting from players' preferences over those rules). The bargaining problem means that there will almost certainly be significant and contentious debate over those rules irrespective of their form (including no-holds-barred).

The rest of your points result from continuing to misunderstand my argument.

So far, the "this is an exploit" side of things has done no better than "personal preference of the majority dictates that things beyond cheating/bypassing game rules are exploits".

I don't recall suggesting majority rule as the resolution to the social choice problem. I don't recall anyone else suggesting majority rule either.
 
In your opinion, which is an opinion that is entirely based upon preference.

Ah, but to ban something in competitive settings of any kind, you need more than preference...and how is the game designed? So that you can do this, that's how. Prove it should change...or why would/should it?!

This is because you're still not listening. You make a rule against doing something because the implications of the behavior are undesirable and because the gains to the individual of engaging in the behavior exceed the opportunity cost in a sufficiently meaningful proportion of cases to make it worthwhile to expend the cost to enforce the rule. Most of the time, most people would prefer not to murder other people because doing so is costly. You make a rule against murder and enforce consequences to deal with the cases where the gains exceed the opportunity cost; in other words, where the opportunity cost does not bind and is therefore not "meaningful".

I've made a strong case that the opportunity costs of fake wars are not meaningful in the latter phases of most games, and that a rule against the behavior is therefore necessary to prevent the undesirable implications of the behavior - an increase in luck factors. I've gone on to point out that that's my preference, and that no one else is objectively right or wrong when they disagree with me.

Hah. Increase in luck factors as an argument, coming from someone who can supposedly predict his win within 30+ turns out! You're saying the end-game situations are luck based? I call BS. Prove it. What the AI has and even which ones exist on the final turns are DIRECTLY influenced by player actions. If you have so much control as to predict VC...I have substantial doubts that the ROI of the end-game based on this tactic is particularly variable or, at least, that it is particularly dependent on luck. We're not talking about ruins here.

Also, using a tactic built into the game is not murder. Your ONLY basis for this rule is preference, so far. How is one person's preference a viable definition of exploit outside of single-player only experiences? It isn't.

This is irrelevant. In a competitive environment, imposing additional rules to deal with "exploits" is only going to affect outcomes amongst the most competitive players. The Prince player loses the competition irrespective of the rule set.

Ah, but the underlying point stands; that this decision metric is decidedly more complex, with more factors to consider, than many other typical in-game tactics that are universally accepted (I was sure I pointed this out, but I guess not)...the point being that this trivializes your "minimal strategy required" argument, which is like saying civ V requires minimal strategy (which I wouldn't agree with even if you said it, despite flaws this game still has strategy).

1) Only the cases where opportunity cost does not bind matter to the necessity of a rule. Consider it this way: you can write up a maximization problem for any arbitrary behavior with opportunity cost as the s.t. portion of the equation. If the solution is non-empty, you will observe the behavior in question. If the behavior is undesirable, you need a rule against it so long as the enforcement problem is not large enough to outweigh the gains from enforcing the rule. (Notice how preference sneaks back into the terms for enforcing the rule and the cost of the enforcement problem.)

"undesirable" behavior. That's like defining "exploit" as "anything that's an exploit". It's a joke. The whole purpose of this exercise is to demonstrate WHY the behavior is undesirable. You assert luck; there's reason do doubt this and little evidence that luck affects end-game opportunities to a material extent, or that allowing this tactic introduces more luck than exists without it. Can you prove the 1 reasonable basis for 'undesirable" behavior you've given to this point, to at least help the side of the argument that considers this and similar activities an exploit?

2) Since AFAIK the diplo hit doesn't land until the next turn, on t-1 the opportunity cost is still zero. This is also true for any and all cases where all of the terms in the opportunity cost function collapse to zero, and that's a much larger set than you seem to think. For instance, a diplo hit has zero cost if you're not negotiating deals and AIs don't have time to organize an attack.

And why are you "not negotiating deals"? Not negotiating deals IS an opportunity cost. Also, most times not only does the opportunity cost at t-1 become very small, but so does the benefit; diplo is a possible exception, but in practice what is the benefit of this to say, t-1 domination? If you KNOW domination is t-1 for example (IE you win by marching into the last capitol next turn) or that t-1 for space has come (pretty likely to know this), what BENEFITS are you getting, for the minimal cost? A building? A unit? A never-to-be realized RA? A luxury resource? Tell me, how even THIS breaks competition.

Also, if you only have an issue at t-1, then why are you arguing the tactic is exploitative, as opposed to arguing it's exploitative at t-1 only, for example?

Clever rhetorical tricks don't constitute an answer to my criticisms.

Neither does ignoring points (you still are) and coming up with wild scenarios where something might possibly be construed as a little abusive but which has a lower incidence of a material advantage against alternatives than common tactics.

There's a preference component to the definition and any attempt to impose further order on the definition is nothing more than the attempt to impose your own preferences on others. That's been my core contention the entire time, and we've had several exchanges without you making any attempt to answer it.

My answer to this was implied but not stated concisely:

It's not really valid to impose preferences onto others in a competitive setting without a basis on which competitors can reasonably agree. If you can't use only the objective component of the exploit definition or come up with an agreeable application of the subjective portion to the competitors, it has no place in competitive settings at all, and people are going to have to compete around systemic inadequacies (this has had both bad AND good consequences historically). If my stance on this was unclear before, I apologize; I thought you were aware of it already.

As far as solving the social choice problem goes, any headway there requires creating choice rules that the arbitrator can consistently apply. While that seems to be what you're after, there's a serious bargaining problem over those rules (resulting from players' preferences over those rules). The bargaining problem means that there will almost certainly be significant and contentious debate over those rules irrespective of their form (including no-holds-barred).

Yes, there will be social choice issues, but that's always going to come down to 2 ultimate solutions 1) the competitors find grounds on which they're willing to compete with their made up rules or 2) they play the game using the rules as designed. In the latter case, there are no exploits (only actions that are easily and objectively identified as cheating, such as using glitches to escape game rules, hacking units to have more strength, maphacking, etc). In the former, once such an agreement is in place nobody should have any complaints, although it really leaves question about the room for the term "exploit" vs simply "banned tactics" (which is less confusing anyway).

How well any given competitive setting handles the rules is of course debate-able, but outside the scope of this thread.

I don't recall suggesting majority rule as the resolution to the social choice problem. I don't recall anyone else suggesting majority rule either.

Indeed, it seems I was mistaken about this aspect of your argument.
 
Ah, but to ban something in competitive settings of any kind, you need more than preference...and how is the game designed? So that you can do this, that's how. Prove it should change...or why would/should it?!

Now you're starting to get it. I have a preference. You don't happen to agree with it. I can't prove to you that there's a reason why you should agree with me. No such proof exists, because we're dealing with preference. As long as a set of preferences meets basic criteria of completeness and transitivity (see Mas-Colell, Whinston and Green, Ch.1 for the canonical explanation) the set of preferences is rational. You can't say that preference set A is valid but preference set B is not given that they both meet the basic criteria of rationality, even though colloquially we might say that set A is "sane" (Martin Luther King) and set B is "crazy" (Adolf Hitler).

In other words, your set of preferences is in no way superior to mine. The result is that we have to come up with some way of sorting through your preferences, mine and those of others. That's a social choice problem. Ideally, we'll come up with some decision rule that is socially efficient - that gives as many of us as much of what we want as possible, given conflicting preferences over things such as impact on the game and enforcement costs.

Hah. Increase in luck factors as an argument, coming from someone who can supposedly predict his win within 30+ turns out! You're saying the end-game situations are luck based? I call BS. Prove it.

AI cash on hand varies sharply with the dirt they have, which in turn is random. The distribution of the AIs with good dirt, versus the ones with bad dirt, is also random. In some games, the distribution is such that you can safely fleece cash from AIs at the end of the game. In others it is not.

If you haven't noticed this, you aren't paying enough attention to the diplomacy screens on lower difficulty levels as the game winds down. It isn't regularly an issue on Deity; the major random factor there is whether or not an AI or two get killed off and become unavailable for RA partnerships.

Ah, but the underlying point stands; that this decision metric is decidedly more complex, with more factors to consider, than many other typical in-game tactics that are universally accepted (I was sure I pointed this out, but I guess not)...the point being that this trivializes your "minimal strategy required" argument, which is like saying civ V requires minimal strategy (which I wouldn't agree with even if you said it, despite flaws this game still has strategy).

It's about as complicated as choosing between RAs and rushing buildings and units in a non-Domination game. Suppose a 120 turn Quick Standard Pangaea. For the first 90 turns or so, you're almost always going to choose "no" on fake war. For the last 30, you're going to start regularly choosing "yes" as the opportunity cost term vanishes. The situation is similar with the RA/building/unit dynamic. For the first 90 turns, you're going to choose "RA" with one or two exceptions. After that, you're going to choose "building" with an obvious correct choice every time. There is strategy in both cases, but not a whole lot.

I'll agree that CiV requires strategy. There aren't a lot of meaningful choices available along any given dimension, but there are a lot of dimensions and the result is a game with significant strategy. Note that we can infer both the presence of strategy and the comparatively minimal role of luck from the HoF gauntlet results; the same names tend to land at the top of the gauntlets each time, and the results look more like those of a series of chess tournaments than those of a series of lottery drawings (no strategy) or a series of poker tournaments (strategy, lots of luck).

The whole purpose of this exercise is to demonstrate WHY the behavior is undesirable.

Preference dictates where that line is. You're still effectively arguing that everyone else's preferences are wrong and yours are right. There's no possibility of making headway on this issue until you admit that other people have valid viewpoints on this issue that differ from yours, rather than insisting that some universal operational definition that everyone can or should agree upon exists.

Neither does ignoring points (you still are) and coming up with wild scenarios where something might possibly be construed as a little abusive but which has a lower incidence of a material advantage against alternatives than common tactics.

You're continuing to miss the forest for the trees here. I'm ignoring points that are tangential and have no bearing on the validity of my core argument. There's no point in getting bogged down in rhetorical tricks, magician's distractions and attacks on portions of an argument which have been distorted by taking them out of context.

Further, neither the magnitude or relative magnitude of material advantage alone matter IMO. It's the distortion of the strategy space that matters, and therefore comparing the relative power of mechanics that don't directly compete for the player's resources is unhelpful.

For example, RA blocking is exploitative, but the effects of the behavior itself aren't that harmful IMO. Some undesirable micromanagement is introduced, but that's about it. The problem with RA blocking is that RAs then become both dramatically undercosted and a means to bulb things that you shouldn't be able to bulb. The result is that alternative uses of :c5gold: can't compete in peaceful win condition games, and we have the HoF gauntlet results to prove it. The obvious solution is to increase the :c5gold: cost of RAs, such that rushing buildings and units and burning SP choices on :c5science: policies become more competitive.

If my stance on this was unclear before, I apologize; I thought you were aware of it already.

It's not that your stance is unclear. It's that your preference for no-holds-barred is precisely that: a preference. Any set of agreed upon additional rules which is enforceable and can be understood by players is just as valid as no-holds-barred. The only question left to settle is which possible set of rules meeting those conditions best resolves the social choice problem.

In the former, once such an agreement is in place nobody should have any complaints, although it really leaves question about the room for the term "exploit" vs simply "banned tactics" (which is less confusing anyway).

Jim Fearon has a 1998 article in International Organization that is well worth your time here. The problem is that the clearer, more comprehensive and more intersubjective the proposed rule set is, the harder it becomes to get everyone to agree upon it. The clearer the rule set becomes, the less wiggle room is available for actors that don't like the rule set, so at the margin those actors continue to hold out and press for concessions rather than agree to the rules.
 
Quick question: Are there any other "maneuvers" that are as lucrative on a per turn basis as resource/lump sum trading followed by a dow?

I can describe one that I consider an exploit. Let me give you an example.

I have an emprire with 1 incence, 1 marble and 2 wine. One of my wine resources are not improved.
When my last wine needs one worker turn to improve, I stop my worker. This way I can push the button and instantly improve my wine.
  • Now I set all my cities to maximise gold income.
  • Then I make a deal for all of my resources and all of my gold per turn, in exchange for a lump sum. All in one deal.
  • Now I have a worker start a fort on my improved wine. This removes the improvement and I loose the wine, this way canceling the deal.
  • Finally I finish the wine that needed one turn, so I do not loose any happiness. And I set my cities back from gold focus.

This way I did not loose anything, I got a big pile of gold, and the AI does not recent me at all.
 
"Urban dictionary definition?" What does that mean - one with which you disagree?

Probably not the kind of reference that should be explained too heavily on CFC...but an urban dictionary is one that will give you the definition of how current/momentary slang is used (you also get better usage examples of expletives etc :p). Exploit in the gaming usage of the term applies in this context; the way people use it in gaming is quite different from how it's used anywhere else, and yet that usage is not a widely accepted part of the English language yet...imo because it's somewhat confusing the typical meaning of the word. We should probably find a term that fits better.

You can't say that preference set A is valid but preference set B is not given that they both meet the basic criteria of rationality, even though colloquially we might say that set A is "sane" (Martin Luther King) and set B is "crazy" (Adolf Hitler).

The only basis for saying a preference set I have is that some rules are not rational IE they stop some activity but not others when the activities are functionally identical or very similar.

AI cash on hand varies sharply with the dirt they have, which in turn is random. The distribution of the AIs with good dirt, versus the ones with bad dirt, is also random. In some games, the distribution is such that you can safely fleece cash from AIs at the end of the game. In others it is not.

If you haven't noticed this, you aren't paying enough attention to the diplomacy screens on lower difficulty levels as the game winds down. It isn't regularly an issue on Deity; the major random factor there is whether or not an AI or two get killed off and become unavailable for RA partnerships.

On lower levels, however, one can easily dictate who has dirt at all, while on higher ones it's not so easy to bleed the AI dry...which makes luck a pretty small factor here. Have you quantified this over a bunch of games? I certainly haven't.

Further, neither the magnitude or relative magnitude of material advantage alone matter IMO. It's the distortion of the strategy space that matters, and therefore comparing the relative power of mechanics that don't directly compete for the player's resources is unhelpful.

Not necessarily. If a tactic provides a tremendous amount of power over non-mutually exclusive mechanics one can still have some basis it's too strong; indeed this is often the case people cite with RA. Taken to the extreme, imagine a mod where warrior rushing was so strong it worked on every level with minimal effort. The warriors do not directly compete for resources beyond their initial investment (which would have been made regardless), but they trivialize a large portion of the rest of the game. Granted, not many things are off kilter enough to this extent, but it's certainly a valid consideration when it happens! Back to Ra blocking:

For example, RA blocking is exploitative, but the effects of the behavior itself aren't that harmful IMO.

I would argue they're pretty harmful...they change the expected ROI of RA dramatically, as you point out. You also point out that this materially affects strategic options; literally people must do this if they have any hope to compete with the fastest games! In practice, this removes a good amount of options from the game that, if ROI were as intended, would be viable. I guess our only break in opinion here is whether this is actually harmful, but I find its impact consistently greater than any gold-then-dow trick in terms of impact on a game to game basis. I agree with the solution btw; I've long advocated doing away with the micromanagement of this feature entirely and just raising its cost...the alternative being the old model of it just giving fixed extra beakers also being acceptable. In either case, the approach could be balanced against other options and moves the AI one step closer to the human in that there's no longer a big gap created by obscure micro in this case.

It's not that your stance is unclear. It's that your preference for no-holds-barred is precisely that: a preference. Any set of agreed upon additional rules which is enforceable and can be understood by players is just as valid as no-holds-barred. The only question left to settle is which possible set of rules meeting those conditions best resolves the social choice problem.

So, what's the best way to resolve a social choice problem? Obviously they've been addressed many times over the years...is there an approach to it that tends to yield the best *practical* results quickly?
 
The only basis for saying a preference set I have is that some rules are not rational IE they stop some activity but not others when the activities are functionally identical or very similar.

Ah ha! This is the root of the misunderstanding. You're conflating preferences and rules, but they are very different economic concepts. Preference sets over outcomes are rational or not, but rules have no rationality. Rules perform a function. Sometimes we can say that a rule is suboptimal, in that an alternative rule exists which would, say, produce identical results at lower enforcement costs, or produce better results for at least one person while producing no harm for anyone else. We can say that a rule is incentive compatible, meaning that it causes everyone to act honestly and not try to game the rule by lying. But we have to judge rules based upon the outcomes they generate, and we do so in light of how well the generated outcomes maximize preference.

Tax avoidance and tax evasion are similar behaviors. One is legal and one is not. The reason is the implications of those behaviors. The government declares some things "tax evasion" because several constituencies all agree that the implications of the behavior are undesirable. Offshore tax havens take assets out of the U.S. and no legislator wants that, so the government tends to be nasty about them. By contrast, the government wants to promote personal saving to reduce stress on Social Security and Medicare, so it creates programs like IRAs and 401(k)s that permit individuals to legally avoid taxation on present income.

On lower levels, however, one can easily dictate who has dirt at all, while on higher ones it's not so easy to bleed the AI dry...which makes luck a pretty small factor here. Have you quantified this over a bunch of games? I certainly haven't.

You also incur diplomatic and :c5happy: costs by taking the dirt away, build orders are crowded in peaceful games, and games are short. I find that there's no way that you're going to run over everyone on the lower levels and still optimally progress towards a peaceful win condition. YMMV, but I find that for peaceful games expansion and getting production and :c5science: or :c5culture: buildings up consumes the first half of the game, with barely enough room in the build order to sneak in a passable anti-barb power projection force to take advantage of camp quests. As you up the difficulty level you are compelled to build enough units to survive, but that force doesn't permit running rampant all over the continent.

Not necessarily. If a tactic provides a tremendous amount of power over non-mutually exclusive mechanics one can still have some basis it's too strong; indeed this is often the case people cite with RA. Taken to the extreme, imagine a mod where warrior rushing was so strong it worked on every level with minimal effort.

I remember that mod; it was called the initial release. The strategic problem there is that the ROI on early :c5production: is maximized via spamming Warriors under all conditions. Knock over enemy city, annex, build Settler in it, burn it and replace it was more effective than building Settlers because you could rush to IW and Steel and repeat the process to expand further.

I would argue they're pretty harmful...they change the expected ROI of RA dramatically, as you point out. You also point out that this materially affects strategic options; literally people must do this if they have any hope to compete with the fastest games!

The problem is that they're blatantly undercosted, not the lack of randomness. Certain techs are so important to each win condition that randomness would create major game spam in a competitive environment. I think that the cost is moddable, but I could be wrong.

Blocking RAs doesn't collapse the strategy space much. Scholasticism is at least as effective on Deity if you use Meritocracy to get into Medieval right away, and on other settings blocked RAs tend to enhance what you would be doing anyway rather than drive out strategies. The main change is that things you wanted anyway get moved up in the build order because they are available earlier. A lesser implication is that you end up playing the city-state game less; that's somewhat more harmful from a strategic standpoint.

In other words, blocking RAs is unquestionably an exploit under the definition I gave; you shouldn't be able to do it given the mechanic's description. But the harm is minimal and we're debatably better off with the ability to block. I think that better solutions exist, but they all involve a full Firaxis rework of the mechanic.

So, what's the best way to resolve a social choice problem? Obviously they've been addressed many times over the years...is there an approach to it that tends to yield the best *practical* results quickly?

Now that depends on who you ask. Everyone in the group would like to see their preferences obtain, and the rules that are instituted dictate who gets what they want and who does not. If you're a neutral third party trying to create rules to fix the situation, the approach depends on what you want to maximize. That's why, way back when, I made the argument that it's essential that we figure out what it is we want to maximize as a group before we start mucking around with instituting rules.

My advice for the HoF would be to start by opening a thread which asks everybody what they think is important. Then try to derive a consensus that almost everyone can agree with. Appoint a few experienced players that can say what the likely consequences of a given rule change would be to hash out a set of rules that appears to produce the results the community can agree on. Then go back to the community and talk about it before instituting it, so that you can address any concerns and fix any errors that might have been made. In hindsight, that's what should have been done in the first place during the six months after release before HoF started up, but hindsight is 20/20 and there's no sense kicking ourselves.

In this case, the HoF is functionally a dictatorship (again, let me stress that there is nothing ethically wrong with this). Denniz has the mandate of heaven because he supervises the work. So you should discuss anything you want to do with him before you start the process.
 
The problem is that they're blatantly undercosted, not the lack of randomness. Certain techs are so important to each win condition that randomness would create major game spam in a competitive environment. I think that the cost is moddable, but I could be wrong.

Yep, it's moddable.

Thalassicus' balance mod does a really nice job with this - in addition to balancing out things like maritime city states and a slew of other things. I'm not sure what the current iteration does, but I believe one recent version gave a fixed # of beakers depending on your era. The beakers go to your current tech, and overflow is not lost. Seems like a really elegant solution. I've also played a mod (foget which) which changed RAs to provide a per-turn boost to beakers - another sensible solution.

Much of the March patch was lifted directly from Thalassicus' earlier mod versions - and it was the first time I found the base game playable. Prior to that I only played with his balance mod. It does a really nice job with most aspects of the game. It also prevents getting far advanced techs from GS pops - it really makes these seem quite balanced compared with other options.

Despite the strength of his mod, fixing RAs and other things, it still feels easier than the base game - mostly for me due to the changes to Citadels of all things. You can get a GG before your first social policy or worker on marathon. A GG turned into a significant tile improvement (something like +2 production, +1 gold) that also connects resources is just an incredibly strong opening. Other than a warmonger/marathon opening though, I don't feel it's far off the base game in difficulty, and it basically removes/modifies the 'cheese/exploit' mechanics that clearly cause a lot of controversy.

Although off-topic for this thread, ideally the HoF will go develop a mod such as was the case for Civ 4. Both RAs and GPT trades are both easily fixed (and have been fixed) by mods. It prevents the need to police player actions, and removes player doubt about whether their gameplay inadvertently gets labeled exploitative. All we need is a motivated modder to coordinate with the HoF staff and come out with an official HoF mod - most of the work needed to address exploits has already been done in existing mods.
 
Marathon is EZ mode, it's simply not comparable. :lol:


Easy-mode for a warmonger on the aggressive(i.e. ME). It's actually harder than regular if you're on the defensive - especially when behind in tech. Marathon magnifies the effect of any military disparity - whether in your favour or the AI's.

Play a peaceful deity marathon game. It's a completely different beast from a deity warmonger marathon game. Rush-buying is expensive. Units are SLOW to train. You typically have a small standing army in a peaceful game. If following a builder strategy, an unexpected DoW can easily get you into a spot you can't recover from - whereas on standard you are more able to get the units you need to survive an initial assault.
 
Top Bottom