AU mod: The Colosseum

[punkbass2000]

Originally posted by Moonbars
I don't douvt your skill at all Punkbass, I am sorry if you took my comment the wrong way. maybe having 5 cities, at least one of which has been bulding settlers, is IMO not very many to then go on and build Cols in before 2000BC.

How BIG are your 3 cities? How many SPT do they produce, do any of them have granaries?

My capital is size 6 in 2190BC, producing 5fpt and 8spt and already has a Granary. My second city is size 5 producing 5fpt and spt is dependent upon MM, but it's 3spt and 5spt, alternating. It too has a Granary. If I had went that way with technology both these cities could get Cols fairly soon, I believe. The third city is, admittedly, not in a position to build one too soon.

EDIT: In fact, both cities have an "extra" forest to work, so at 5fpt they're producing a "bonus" spt. Call it 9 and 4, though I'd MM to 8 and 5, most likely.

[nbarclay]

I just checked back my in my Glory of Culture game to see how many culture-doubled colosseums the AIs that survived into the late game had. As of 1430 AD (with the last few turns of that played in a parallel universe), the Germans, Ottomans, and Iroquois were still around. Unless I missed one somewhere, the Iroquois capital was the only AI city with a colosseum producing double culture, and thus the only one that would get a tourist attraction bonus. I had about five, having prioritized colosseums more highly than normal for cultural reasons (but still not anywhere near as highly as I could have).

If the dates of AI colosseum-building in that game are at all typical, I think a tourist attraction bonus would almost certainly benefit humans significantly more than AIs. Indeed, compared with the current approach of reducing the maintenance cost for colosseums, a change to the tourist attraction concept wuld make AIs worse off.

[alexman]

That game is not typical, since the human was encouraged to focus on culture, but the AI did not know to do that.

Also, the higher the difficulty level, the farther falls the human behind the AI in building cultural buildings.

[nbarclay]

The game's special emphasis certainly affected human building patterns, but a tourist attraction bonus for colosseums would also affect human building patterns. As for AI building patterns, do you have any particular reason to think that game was atypical? In that game, only one city among the survivors of inter-AI warfare on an entire continent would have gotten a tourist attraction bonus by AD 1430. That seems to imply that AIs would get very llittle use of the tourist attraction bonus, that such a bonus would likely be mostly an extra toy humans can use to gain an additional advantage. (If, that is, we can figure out how to use it correctly - which brings us back to my earlier concerns about the complexity involved.)

As for the difficulty level issue, I was playing on Emperor. On Monarch, I'm guessing the AIs would be even slower in building colosseums since they wouldn't REX quite as quickly.

[Dominae]

Maybe this is a difference between stock and the AU mod, but in my (stock) games the AIs regularly have Colosseums everywhere by 1400AD. Not sure when exactly they get around to building them, but only one or two in total I consider atypical.

[Rommel2D]

The one colosseum he mentioned wasn't the only one built by 1430, it was the only one built before 430.

Nbarclay, I'm guessing GoC was the only game you've built early colosseums...?

I don't see any mystery to figuring out 'how to use' the tourism bounus- the sooner you build them, the sooner you start to get the bonus. It's a matter of cost/benefit calculation that still doesn't look to me like it's use would be any more gamebreaking than other existing strategies against the AI (although I'll point out my inexperience with the AU mod). Yes it is complex, but you make it sound like any modification should display an ability to be adapted to a simple, rote strategy.

[Dominae]

Originally posted by Rommel2D
The one colosseum he mentioned wasn't the only one built by 1430, it was the only one built before 430.

Doh!

[nbarclay]

Originally posted by Rommel2D

Nbarclay, I'm guessing GoC was the only game you've built early colosseums...?

You guess incorrectly. In Power of Fascism, the fact that conquest was not an option caused me to put a higher emphasis on culture - including colosseums. I checked a save from AD 320, and I had four at the time. I've also had games without special predefined goals where I built early colosseums because there weren't enough luxuries available to get a happiness boost from marketplaces, or at least to get as much boost from marketplaces as from colosseums. Granted, such games are the exception rather than the rule, but they do occur. (And I might add that the reduced maintenance cost in the AU Mod plays a role in such decisions.)

I don't see any mystery to figuring out 'how to use' the tourism bounus- the sooner you build them, the sooner you start to get the bonus. It's a matter of cost/benefit calculation that still doesn't look to me like it's use would be any more gamebreaking than other existing strategies against the AI (although I'll point out my inexperience with the AU mod). Yes it is complex, but you make it sound like any modification should display an ability to be adapted to a simple, rote strategy.

The, "the sooner you build them, the sooner you start to get the bonus," view does not factor in the opportunity cost of what else could be built instead, or (in the case of more extreme strategies) the opportunity cost involved in researching Construction early instead of focusing research in other directions. Nor does it factor in the maintenance cost for colosseums until the tourist attraction bonus kicks in. With most improvements, the nature of the benefits provided is immediate enough and clear enough that cost/benefit analysis isn't all that difficult. But when an improvement has an immediate upkeep cost, but a key benefit doesn't start to kick in for over a thousand years and remains small for a few more centuries after that, analyzing costs and benefits is a lot more complex.

Also note that even in AU 504 and 505, where colosseums were a relatively high priority for me, they still didn't come until the AD's. In order to get much impact from the tourism bonus, players would have to build them even earlier than I did in those games, early enough that there would be an even bigger issue of opportunity cost.

I certainly don't want Civ to be a game that can be played with a simple, rote strategy game after game. Take a look at the fact that I sometimes build relatively early colosseums when other people are calling the choice not to build them a no-brainer if you doubt that.

But I do want to be able to have a clear picture of what the consequences of my choices will be and of how the consequences of different choices compare (subject to the vaguries of random factors and AI behavior where applicable). I can't find that picture with a tourist attraction bonus without a lot of complex mathematical calculations.

[Rommel2D]

Fair enough, but I want to try a brief summary of why I still think this change has worthwhile potential.

Back when I was well occupied playing the basic SP game, I didn't pay much attention to the AU. Discovering that the stroke- of-genius zero-range bombard was adopted from the AU impressed me a great deal. But looking at the Mod itself, the original Colosseum change was the first and largest item to send up flares warning me off.

This change, and more importantly the philosophy behind it, just seemed like too large of a departure from what I liked about the game. I don't know if it is so much a matter of historical accuracy, but the current AU Colosseum just looses all character to me. By making the improvement a likely choice to build in the stead of other hapiness improvements, it has been adjusted to be, well, just another hapiness improvement.

To me stock rules are better because it is distinctive. If you need to improve the mood in a particular city, but don't have Monotheism and/or enough luxuries, you can build this less efficient improvement. It may not fit into better players' strategies, but that's the Colosseum.

Now my interest has been peaked again by the tourism bonus. It holds the promise of making the improvement useful while maintaining an identity seperate from the strictly 'religious' ones.

It is the complexity of the situation that makes it seem so right. Actually 'complexity' isn't a good description as it implies that a certain answer lies within the calculation. It is the uncertainty of what events will unfold in the next thousand-odd years of the game that make it a difficult decision. There would never be a clear answer on what value any Colosseum would be to any particular game. The number and timing of colosseums built would always be a gamble, a characteristic very appropriate for an improvement that represents a place for people to gather and observe spectacle.

This is, of course, assuming that taking the 'Colosseum Gambit' wouldn't allow players to consistantly win at a level they could not otherwise...

[Rommel2D]

Originally posted by nbarclay
Also note that even in AU 504 and 505, where colosseums were a relatively high priority for me, they still didn't come until the AD's. In order to get much impact from the tourism bonus, players would have to build them even earlier than I did in those games, early enough that there would be an even bigger issue of opportunity cost.

That is exactly what makes this look like a good idea- in the course of a normal game, the Tourist Attraction will provide some benefit over stock, but not the game-breaking numbers being thrown about in the other thread.

[nbarclay]

Originally posted by Rommel2D

That is exactly what makes this look like a good idea- in the course of a normal game, the Tourist Attraction will provide some benefit over stock, but not the game-breaking numbers being thrown about in the other thread.

Basically, what it comes down to is that either the tourist attraction bonus will be so small as to be almost completely irrelevant, or players will significantly alter their strategy for the specific purpose of pursuing the tourist attraction bonus. If the former, I don't see how the game is really improved. We can get a similar net benefit in a much simpler, more straightforward, and more consistent way with the current approach of reducing the maintenance cost. If the latter, we're no longer dealing with a "normal game" by current standards, but rather with a game where players deliberately alter their strategy in a significant way to build colosseums earlier.

Now let's consider the latter possibility in greater depth. If altering strategy to build more early colosseums is a good move, it gives humans an additional advantage at the expense of the AIs, because (at least if the game I checked is any indication) AIs don't tend to build colosseums particularly early. If altering strategy to build colosseums significantly earlier is not a good move, we end up with a situation where players don't really understand the costs and benefits all that well and, as a result, make a bad decision. Either way, I view the outcome as detrimental.

The tourist attraction bonus idea is an interesting gimmick, but I contend that it complicates the game with little prospect of actually making it better. That's not something I think a conservative mod should do.

[nbarclay]

Originally posted by Rommel2D

It is the uncertainty of what events will unfold in the next thousand-odd years of the game that make it a difficult decision.

Dealing with uncertainty about the future is an integral part of the game. I have nothing against the idea of having to deal with such uncertainty in and of itself.

What really bothers me is that even if players could know the future perfectly, it would still take fairly complex calculations to figure out how the benefits of building an early colosseum compare with the benefits of doing other things instead. The time delay makes it impossible to use knowledge of the benefits of each choice in the near future as a fairly reliable heuristic the way we normally can. And complexity that would be hard to deal with on more than a superficial, guesswork level in the face of a known future is vastly worse when it has to be weighed in the face of a variety of possible futures.

[Rommel2D]

Originally posted by nbarclay
And complexity that would be hard to deal with on more than a superficial, guesswork level in the face of a known future is vastly worse when it has to be weighed in the face of a variety of possible futures.

In a game based heavily upon the 'single die-roll', low integer combat system, contemplating a "known future", or even a quatifiable number of possible ones, strikes me as a bit absurd.

The uncertainty level being discussed here again seem like a good fit with many other elements of the game to me...

[nbarclay]

I think you're still missing the point that my only concerns about the uncertainty aspect are in relation to how they magnify an already serious complexity issue. I have yet to hear anyone suggest an approach to figuring out whether it's better to build a colosseum or something else that was more than essentially guesswork and that wouldn't require getting out a calculator (and probably a pencil and paper as well to keep track of the various steps involved in the calculations).

[alexman]

Originally posted by nbarclay
I have yet to hear anyone suggest an approach to figuring out whether it's better to build a colosseum or something else that was more than essentially guesswork and that wouldn't require getting out a calculator (and probably a pencil and paper as well to keep track of the various steps involved in the calculations).

And I have yet to hear you explain why this would be an issue with the Colosseum, when there are already dozens of decisions like this in every game of civ, and most have a much greater effect on the outcome of the game than the tourist bonus. Do I build Horsemen or Swordsmen? Do I attack my neighbor now, or after I get a stronger economy going? When and where do I build the FP? Do I build marketplaces before libraries? Granaries before Barracks? Temples or units for MP? Catapults or Swordsmen? Do I build the Pyramids or the Great Library with my SGL? Do I beeline to Military Tradition, or do I take the upper branch and get my Heavy Cavalry with Nationalism instead? I could go on and on...

The type of decisions involving what you call "guesswork" are what make strategy games interesting. It's not fun when you know the correct answer in every situation.

[nbarclay]

Originally posted by alexman

And I have yet to hear you explain why this would be an issue with the Colosseum, when there are already dozens of decisions like this in every game of civ, and most have a much greater effect on the outcome of the game than the tourist bonus. Do I build Horsemen or Swordsmen?

The complexity of such choices is a necessary consequence of offering players more than one type of offensive unit, which in turn adds considerably to the game's richness. If I felt like adding a tourist attraction bonus to colosseums would add anything halfway resembling a similar level of richness, I would feel differently.

Also note that horseman/swordsman type issues are complex enough that they have helped spark the creation of combat calculator programs. I definitely don't like adding an element to the AU Mod that would require development of an additional calculator program for players to get a clear picture of how the benefits of different build choices compare.

Do I attack my neighbor now, or after I get a stronger economy going? ... Do I build the Pyramids or the Great Library with my SGL? Do I beeline to Military Tradition, or do I take the upper branch and get my Heavy Cavalry with Nationalism instead? ... Catapults or Swordsmen?

These types of decisions are not simple, routine build choices for a single city, but rather are matters of basic strategy or are very closely intertwined with matters of basic strategy. (The Great Library has its greatest value when players adjust their strategy to maximize the advantage they get from it, so questions of whether or not to build the Great Library are intimitely tied to larger strategic issues.) Such complex, over-arching, long-term strategic issues need to be a lot more complex than simple build queue choices in individual cities. The game would lose its fun very quickly if complex metagame strategic choices were no more complex than typical build queue choices, and would be virtually unplayable if the kinds of build queue choices made several times every turn were as complex as choices regarding "big picture" issues.

When and where do I build the FP?

In my view, comparing the complexity of a proposed new feature to that of the FP issue is a very strong recommendation against it. Offhand, I can't think of any other regularly-used feature in the entire game where an effect is determined entirely by a player's choice, yet where it is so difficult to determine how the benefits of different choices compare. But at least the FP question is something that normally has to be answered only once per game - unlike the question of whether to build colosseums or something else.

Do I build marketplaces before libraries?

I have a simple heuristic for that that I view as highly reliable. Marketplaces first only make sense if either (1) they will provide a happiness benefit and the city will need that benefit before the markeplace would be finished if I build a library first, or (2) the sliders will be set in a way that produces more gold than science. Otherwise, a library first makes sense because it will produce culture and will produce as much or more increase in commerce. (And libraries get additional emphasis if I have a cty fighting a culture war with an AI city for tiles.)

Granaries before Barracks?

When I know what I'm planning to do with a city in the near future, the answer to that is generally obvious - and has been ever since I got a clear picture of the power of granaries.

Temples or units for MP?

In the short term, under governments that allow it, warriors for MP duty are vastly more efficient than temples. They cost only 1/3 as much even for religious civs, and don't require upkeep. The value of temples in the early game is more for culture, for a happiness boost beyond what MPs alone can provide (in which case the question is not either/or), or possibly for border expansions depending on the city build pattern (and especially when a city needs to fight a culture war against an AI city).

The type of decisions involving what you call "guesswork" are what make strategy games interesting. It's not fun when you know the correct answer in every situation.

I don't think you understand how I'm using the term "guesswork" in this context. My concern is not just that an element of guesswork would be involved in how to fit the new feature into an overall strategy, but rather that players would be left guessing at the impact of the feature itself. I view that kind of guesswork very differently from other types of things that might also be called "guesswork" depending on what definition one uses.

Offhand, I can think of four basic things that can require players to guess.

1) Lack of knowledge of the moves opponents are making. The need for that kind of guesswork contributes to the game by forcing players to plan for the different possible moves their opponents could be making and/or to learn how their opponents think.

2) Random elements that force players to plan not just in terms of a single known outcome, but in terms of a variety of possible outcomes. That need for contingency planning also makes the game richer, and is why I view Civ 3's penchant for occasional unrealistic combat outcomes as more good than bad.

In both of these issues, it is not the guesswork itself, but rather the contingency planning that the lack of certainty makes necessary, that truly adds to the strategic richness of the game.

3) Complexity. In tic-tac-toe, the game is so simple that it is possible to determine which moves are best simply by examining every possible outcome. Strategy becomes much more complex and interesting when the number of factors involved is high enough, and the ways in which the factors interact are complex enough, that a brute-force apporach to determining what strategy is best is no longer possible. An element of guesswork is involved, but it is a kind of informed, educated guesswork in which the individual elements are clearly understood and only the complexity of how they fit together makes it necessary to guess.

4) Lack of understanding of the game's mechanics. This is the type of guesswork that I view as detrimental to the game. It is this kind of guesswork that leads to the development of combat calculators, to your own in-depth analysis of the corruption system, and so forth. The more such guesswork can be either eliminated or reduced to a more manageable level, the more solid a foundation players have to build their strategies on.

Unfortunately, the way I see it, a huge part of the complexity added by a tourist attraction bonus would fit into this fourth category - at least unless players perform some pretty sophisticated calculations. Players would not just be dealing with the vaguries of how to fit a well-understood mechanic into a complex grand strategy, but would be left guessing at the benefits that the mechanic itself will provide.

Yes, it's not that hard to know on an intellectual level that building a colosseum in a particular year will result in such-and-such additional gold per turn later in the game. But the next step, from that superficial knowledge of effects per turn into how the costs (including opportunity cost) and benefits compare, is one I can't find any way to make without sophisticated calculations. And without that next step, the true meaning of the game mechanics is at least as much a matter of guesswork as of genuine understanding.

[Rommel2D]

I think the majority of players enjoy this game because they are able to tackle such complex problems and test their intuitive intellectual abilities against them. What you seem to be describing sounds more like an accounting exercise.

So where does this 'conservative mod' label come from? I don't see it mentioned in the philosophy section of the main thread anywhere. If there's a movement to give Civ3 a dignified burial before 4 is released, I'll desist in my commentary...

[nbarclay]

Quoting from the first post in the AU Mod C3C Version thread,

Philosophy

The main purpose of the AU mod is to challenge the player with a need for deeper strategy, while changing as little as possible. Deeper strategy is accomplished by presenting the player with more non-trivial decisions, and by improving the AI.

Presenting the player with more strategic decisions is otherwise known as ‘balancing’: strong elements of the game are made slightly weaker, and weak elements are made slightly stronger. The game is more challenging when the best approach is not always obvious, or at least when the best approach depends on the given situation.

Stronger opponents encourage the player to rely more on sound strategy, and less on one-dimensional approaches that work against weaker opponents. Therefore, improving the performance of the AI is one of the most important goals of this mod.

Game-play modifications that do not improve the AI nor balance the game are not part of the philosophy of this mod. However, cosmetic changes that add flavor but do not affect game-play might be considered.

The fact that this is intended to be a conservative mod is reflected in things like, "while changing as little as possible," and "Game-play modifications that do not improve the AI nor balance the game are not part of the philosophy of this mod." In addition, the spirit of conservativism is reflected in long-standing traditions that date back to the original versions of the AU Mod before even PtW, much less C3C, was released (if I remember the flow of events correctly). There has always been a certain amount of disagreement regarding how to balance the goal of conservatism against other goals, but conservatism has always been one of our goals.

The big advantage to a conservative philosophy is that it helps avoid giving players reasons not to use the Mod. If everything in the Mod clearly makes strategic choices more interesting and/or helps the AIs compete more effectively, even if players don't always like the changes from a perspective of individual taste, they can consider the Mod worth using because it gives them a better game. But if we start incorporating changes that people value essentially just as matters of personal taste, we risk turning people away from the Mod unnecessarily. Consider your own [addressing Rommel2D] reaction to the original changes to the colosseum: you didn't view the changes as having a particularly good objective reason behind them, and the changes offended your sense of what you felt like the colosseum was supposed to be like. A few posts ago, you characterized that change as, "the first and largest item to send up flares warning me off." Any time the AU Mod makes changes without there being a sufficient net benefit to clearly justify the departure from stock, it risks annoying players in such a fashion and possibly driving them away.

[nbarclay]

Originally posted by Rommel2D
I think the majority of players enjoy this game because they are able to tackle such complex problems and test their intuitive intellectual abilities against them. What you seem to be describing sounds more like an accounting exercise.

The problem is, the mathematics of a tourist attraction bonus are too complex for intuition to be trustworthy. With most choices in Civ 3, I don't have to go to the point of an "accounting exercise" because I can get a sufficiently clear picture to feel comfortable that I'm making good (if not always 100% perfect) choices without the need to go to that extreme. But when I ask myself the question, "How would I know whether I'm making good choices regarding the tourist attraction bonus without engaging in an accounting exercise?" I can't find an answer. Even playing out different games using different strategies and comparing the results would have only limited value because the cause and effect for colosseums would be so far apart.

[Rommel2D]

Nbarclay, you're recent summary of your argument was twofold:

Basically, what it comes down to is that either the tourist attraction bonus will be so small as to be almost completely irrelevant, or players will significantly alter their strategy for the specific purpose of pursuing the tourist attraction bonus.

What is the exact dividing point between these, the only two possibilities?

I generally prefer to think along the conservative path. I like Apolyton because there are many like-minded people here. But the 'conservative' label carries alot of baggage with it. When people start to use the term in arguments as a self-referential justification it is a good indication of a lazy intellect. Everthing must be distilled into a simple, boolean container. Only the extremes are worth considering. Mega-dittos, dude!

The tourist attraction actually looks likely to be a 'more conservative' change from stock. There seems to be a few accomplished players considering that the TA might have useful impact without opening an exploitable loophole. If it can be demonstrated that this is indeed a game-breaker, end of discussion. Otherwise, it gives the game more strategic options without something as drastic as altering maintenance costs.

I didn't raise the issue because it merely offended my tastes- this change has been in the mod since the beginning, I believe. Firaxis has passed over the colosseum change repeatedly as they have incorporated others into the expansions. Yes, most of the rest of the mod has been too, but here is an opportunity to tweak this aspect of it. I agree that the mod should be 'conservative', but the methodology for exploring possible changes should not be. Using that as an argument before significant playtesting has been done induces stagnation. If this is too radical to consider, perhaps its time to declare the AU mod Gold?

The problem is, the mathematics of a tourist attraction bonus are too complex for intuition to be trustworthy.

So now you're judging everyone else's level of intuition? Or is the mod not supposed to provide any challenge beyond your current playing ability? Perhaps we could just figure out some way to exclude the TA from the Mythic Hero level...

[nbarclay]

Originally posted by Rommel2D

What is the exact dividing point between these, the only two possibilities?

The dividing point is the point at which players view the impact as big enough to be worth significantly altering their playing styles. Granted, what I said is a bit simplistic in that it left out the possibility that players might overestimate the impact of the bonus and significantly alter their playing style even though they would be better off not altering it, or altering it only a tiny bit.

I generally prefer to think along the conservative path. I like Apolyton because there are many like-minded people here. But the 'conservative' label carries alot of baggage with it. When people start to use the term in arguments as a self-referential justification it is a good indication of a lazy intellect. Everthing must be distilled into a simple, boolean container. Only the extremes are worth considering. Mega-dittos, dude!

I think you may be confusing the traditional definition of "conservative" and the modern political definition. To illustrate how the two differ, consider the issue of vouchers for K-12 education (using American nomenclature). In American politics, the vast majority of support for vouchers comes from people we label as political conservatives. Yet the reality is that a voucher system would be a radical departure from our traditional way of doing things, and is thus not conservative at all according to the classical definition of the term. On that issue, and on a few others, the political use of the terms "conservative" and "liberal" can get turned upside down from the traditional meanings of the terms.

When I refer to the AU Mod as a conservative mod, my doing so has nothing whatsoever to do with conservative or liberal political ideologies. Rather, it has to do with the classical definition of "conservative," based on the desire to conserve the feeling of the stock game in the absence of strong reasons for change.

The tourist attraction actually looks likely to be a 'more conservative' change from stock. There seems to be a few accomplished players considering that the TA might have useful impact without opening an exploitable loophole. If it can be demonstrated that this is indeed a game-breaker, end of discussion. Otherwise, it gives the game more strategic options without something as drastic as altering maintenance costs.

You're the only person I remember seeing express the view that reducing the maintenance cost for colosseums is a "drastic" change. From my perspective, the only times when the change in maintenance cost ever influences my decisions are when colosseums would be very close to being worth building even at the higher maintenance cost. I don't view tipping the balance in what were already close calls as "drastic."

In contrast, I see potential for vastly more drastic effects on players’ choices in the tourist attraction bonus. Depending on when players discover Construction, there is a very real possibility of their building colosseums they would not have even considered building otherwise at anywhere close to the same time under the stock rules..

Further, under the stock rules, the effects of a colosseum are essentially the same regardless of when it is built, aside from the culture doubling issue. (And if what is built ahead of a colosseum is also something else that produces culture, even the culture doubling issue can become irrelevant.) Having the timing of colosseums become as important as a tourist attraction bonus would make it is itself an enormous departure from stock.

In my view, a reduction in the maintenance cost keeps the overall feel of the AU Mod much closer to stock than a tourist attraction bonus would.

I didn't raise the issue because it merely offended my tastes- this change has been in the mod since the beginning, I believe.

I'm almost positive that the change to the colosseum is new to the C3C version of the AU Mod, and hence not something Firaxis had available to consider when looking at what elements from the Mod to incorporate into C3C.

Using that as an argument before significant playtesting has been done induces stagnation. If this is too radical to consider, perhaps its time to declare the AU mod Gold?

"Too radical" is a relative concept, not an absolute one. Our change to cavalry was pretty radical, but the imbalance between cavalry and musketmen in the stock rules was so serious that nothing short of a radical solution could address it. In contrast, the stock rules for the colosseum don't actually cause any harm at all. That makes it a lot harder to justify significant changes to the feel of the game just to make colosseums more worth building.

If you want to do additional playtesting and post your results, you’re very welcome to. But I don’t view an argument that additional playtesting would be useful as a good reason to go ahead and incorporate a change into the Mod, where people who use the Mod would be conscripted to participate in the playtesting whether they want to or not.

So now you're judging everyone else's level of intuition?

For all I know, most players might make reliably good choices based on intuition alone. But also, for all I know, it is entirely possible that a large percentage of players or even a majority would end up over-valuing or under-valuing early colosseums in comparison with other choices. Since I have no good way of knowing even how accurate my own intuition would be, much less the intuition of others, I have to conclude that intuition is not trustworthy. That’s not to say that intuition will necessarily be inaccurate, but just that there is no solid basis for trusting it to be accurate.

In other words, my characterization of intuition as untrustworthy is not so much a matter of passing judgment regarding everyone else’s level of intuition as it is recognizing that I have no reliable way to judge how accurate it might or might not be. It might be accurate, but there are enough pitfalls that it might not be. And I see that problem first and foremost in regard to the question of how good my own intuition would be. When I have a profound distrust for my own intuition, why should I expect the intuition of others to be dramatically more trustworthy?

Also note that the mathematics involved are complex enough that the only way players could even know how accurate their intuition is would be to work the math and see how their intuition compares. Unless players do that (or have someone do it for them), they could easily go on thinking their intuition is serving them well even if it isn’t.

Now let’s turn your challenge around. Are you not trying to judge everyone else’s level of intuition when you presume that intuition alone can lead to reliably good choices? If you object to something you construe as my judging your level of intuition, on what basis do you justify judging my level of intuition?

Or is the mod not supposed to provide any challenge beyond your current playing ability?

If I viewed it as offering a serious strategic challenge, I would view the issue differently. But to me, in the absence of pulling out a calculator, the choices presented by the tourist attraction bonus look more like a guessing game than like a serious strategic challenge. And I’ve never been especially fond of guessing games.

-----

To show how complex the mathematics are, consider the following scenario. It's 350 BC, and you have a city that's producing 10 shields and 20 commerce after waste and corruption. You are considering two possible build orders: Library-Marketplace-Colosseum and Colosseum-Library-Marketplace. For simplicity, we'll assume the city is already maxed out in size. We'll also assume that at the time we are doing our building, the sliders are at 4.4.2 (science 40%, luxury 20%).

With the library first, you get the benefits of the library 12 turns sooner and the benefits of the marketplace 12 turns sooner. For each improvement, the commerce lost building the colosseum first is 12 turns times 0.5 library/marketplace bonus times 0.4 slider setting times 20 after-corruption commerce. That comes out to 48 lost commerce per improvement, or 96 total. In addition, maintenance costs in the build cycle are 12 gold higher if the colosseum is built first, so the total commerce lost building the colosseum fist is 108.

Turns during the building timeframe are 20 years each. So if the colosseum is built first, it will be completed in 110 BC. If the marketplace and library are built first, the colosseum won't be completed until 250 AD. So with the colosseum first, the tourist attraction bonus would initially kick in in 890 AD, while with the library and marketplace first, it won't kick in until 1250 AD. That's a difference of 36 turns. Now let's assume our slider settings in that timeframe are 2.8.0 (80% science, no luxury), with a university in the city but not a bank, and that none of the tourist attraction bonus will be lost to corruption. To reflect average effects, I'll retain fractions instead of rounding them. From the tourist attraction bonus, we could expect something like 0.2 slider setting for taxes times 1.5 to reflect the marketplace effect times two commerce from the tourist attraction bonus, plus 0.8 science setting times 2 to reflect library and university times 2 commerce. That's a total of 3.8 commerce per turn after improvements are factored in. Multiply that by 36 turns between 890 and 1250 and we get 136.8 commerce. Then from 1250 AD until 1390 AD, the two approaches generate the same amount of commerce. So the effect on the game through 1390 AD is that the library-and-marketplace-first strategy has an advantage of 108 commerce until 890 AD, at which time the colosseum-first approach starts to catch up until by 1250, it’s pulls ahead by 28.8 commerce. But in the meantime, we've lost whatever benefit having the commerce available sooner would have provided.

Then, starting in 1390 AD, the colosseum-first strategy starts another burst of pulling ahead again, and will continue to pull ahead for the rest of the game. With a bank added to the city and still none of the commerce bonus lost to corruption, it would gain at a rate of four commerce per turn until 1640, and a stock exchange or research lab would add a little more. I won’t try to project past 1640, but suffice it to say that the colosseum-first strategy would keep pulling ahead.

If the game is going to end in domination before 1250, the colosseum-first strategy would almost certainly be a mistake. If it is going to end somewhere in between 1250 and 1390 (or not long after), the question becomes whether a little less commerce earlier or a little more later is more valuable. And past 1390, the “more later” starts playing a bigger role.

In the face of that kind of mathematical complexity, I have very little faith in intuition as a source for a reliably good understanding of the costs and benefits of each choice. And without a good understanding of the costs and benefits of each choice, whatever strategic thinking is involved rests on a weak foundation.

[Dominae]

Originally posted by nbarclay
To show how complex the mathematics are, consider the following scenario. It's 350 BC, and you have a city that's producing 10 shields and 20 commerce after waste and corruption. You are considering two possible build orders: Library-Marketplace-Colosseum and Colosseum-Library-Marketplace. For simplicity, we'll assume the city is already maxed out in size. We'll also assume that at the time we are doing our building, the sliders are at 4.4.2 (science 40%, luxury 20%).

With the library first, you get the benefits of the library 12 turns sooner and the benefits of the marketplace 12 turns sooner. For each improvement, the commerce lost building the colosseum first is 12 turns times 0.5 library/marketplace bonus times 0.4 slider setting times 20 after-corruption commerce. That comes out to 48 lost commerce per improvement, or 96 total. In addition, maintenance costs in the build cycle are 12 gold higher if the colosseum is built first, so the total commerce lost building the colosseum fist is 108.

Turns during the building timeframe are 20 years each. So if the colosseum is built first, it will be completed in 110 BC. If the marketplace and library are built first, the colosseum won't be completed until 250 AD. So with the colosseum first, the tourist attraction bonus would initially kick in in 890 AD, while with the library and marketplace first, it won't kick in until 1250 AD. That's a difference of 36 turns. Now let's assume our slider settings in that timeframe are 2.8.0 (80% science, no luxury), with a university in the city but not a bank, and that none of the tourist attraction bonus will be lost to corruption. To reflect average effects, I'll retain fractions instead of rounding them. From the tourist attraction bonus, we could expect something like 0.2 slider setting for taxes times 1.5 to reflect the marketplace effect times two commerce from the tourist attraction bonus, plus 0.8 science setting times 2 to reflect library and university times 2 commerce. That's a total of 3.8 commerce per turn after improvements are factored in. Multiply that by 36 turns between 890 and 1250 and we get 136.8 commerce. Then from 1250 AD until 1390 AD, the two approaches generate the same amount of commerce. So the effect on the game through 1390 AD is that the library-and-marketplace-first strategy has an advantage of 108 commerce until 890 AD, at which time the colosseum-first approach starts to catch up until by 1250, it’s pulls ahead by 28.8 commerce. But in the meantime, we've lost whatever benefit having the commerce available sooner would have provided.

Then, starting in 1390 AD, the colosseum-first strategy starts another burst of pulling ahead again, and will continue to pull ahead for the rest of the game. With a bank added to the city and still none of the commerce bonus lost to corruption, it would gain at a rate of four commerce per turn until 1640, and a stock exchange or research lab would add a little more. I won’t try to project past 1640, but suffice it to say that the colosseum-first strategy would keep pulling ahead.

If the game is going to end in domination before 1250, the colosseum-first strategy would almost certainly be a mistake. If it is going to end somewhere in between 1250 and 1390 (or not long after), the question becomes whether a little less commerce earlier or a little more later is more valuable. And past 1390, the “more later” starts playing a bigger role.

In the face of that kind of mathematical complexity...

If you've ever participated in a democracy game into the Medieval era, you know that such mathematical complexity already exists in the stock version of the game (spreasheets, oh my!). Yet players ignore it and do reasonably well (do you require them to do perfectly?), relying on intuition. There is nothing special about Tourism.

This should be a debate about whether it is overpowered or not, not about whether it is too complex. I doubt anyone is fooled by these complexity arguments, when what is really going on is that you do not want to incorporate something new (and potentially useful) into your standard strategy.

[nbarclay]

Dominae, I am getting really, really tired of your trying to tell me what my motives are. In effect, you are calling me a liar, and that makes me angry. The reality is that your mental picture of me is a caricature: there are elements of truth to it, but you exaggerate those elements so out of proportion that the end result looks very different from the real me.

In case you missed it, there was a fairly long gap between the time the tourist attraction idea came up and the time I took a side regarding it. Why? Because initially, my conservative streak was counterbalanced by another part of me that saw interesting strategic potential in the idea.

The thing that caused me to strongly oppose the idea was when I started considering how I might use the new feature and, in connection with that, considering how I would tell whether I was using it well or poorly. Before long, I realized that with cause and effect so far separated for the tourist attraction bonus, I couldn't do much more than guess at whether my decisions are good or bad without pulling out a calculator. That creates a learning curve that bears a remarkable resemblance to the side of a cliff, and makes the use of the feature more guesswork than genuine strategic thinking from my perspective.

If we lived in the Star Wars universe, "Use the Force, Luke," might sound like a valid way of determining what strategy to follow. But in the univese we actually live in, I prefer to base my strategy on something more sound than just intuition. And if my only real options for using a proposed feature are to pull out a calculator or to put essentially blind faith in my intuition, that looks to me like a serious flaw.

As for the kinds of calculations you describe being done in democracy games, so what? Such calculations might provide a small incremental advantage in certain types of situations, but I can feel comfortable that I'm making good (if not always absolutely perfect) choices without them. Why? Because I can look at how the benefits of different choices compare in the relatively near term and use that as a pretty reliable approximation for how overall benefits of prioritizing each item compare. (After all, I can presumably build the other possible items later.)

But the concept of using near-term benefits as an approximation for overall benefits becomes useless when a key benefit does not kick in for a thousand years. That makes it dramatically harder to get a clear picture without relatively complex mathematical calculations.

[Rommel2D]

Conservatisim is a term that has little meaning out of context. In the instance you mention, an American 'conservative' viewing the scope of the US's 200+ year history could view the century-old industrial model of school-based education as an anomoly and see little harm in pursuing better educational options. Someone with greater or lesser scope might find a different meaning in the term.

There's no confusion here. I'm just pointing out that constant use of what has become a hot-button term in our vernacular invites an emotional reaction from the reader and clouds the discussion. When its used to raise a point of concern no big deal, but I'd say its a poor reason to use in support of an argument.

Halving the maintenance cost (not to mention the earlier change) makes what is an ancillary improvement in stock to a more ubiquitous one. This seems drastic to me compared to the Tourist Attraction, which I would not normally alter my research pattern or building priorities much for. If no one can demonstrate that it is a disproportianate exploit, why should we expect masses of anxious players chomping at the bit to scrap their normal strategies and rush after these phantom cash cows?

What would be the point of me doing playtesting on it if I don't plan to alter my approach more than an early one in my capital? If someone produces a game that looks like the TA was the deciding factor in, I'm sure someone or two will step forward to test out other possibilities from early on in the game. [I hope punkbass's recent silence means he is playing out his Diety game diligently, and not feeling neglected from the conversation ]

I have a personal bias in favor of this mod- it carries a certain poetic justification outside the goals proscribed by the AU philosophy: Colloseum->spectacle->indefinite benefit. Although you can't produce a mathmatical thesis on why it should be built at a given time, one can imagine the possible lure of such a build to a civilization mired in the murky fog of ancient times. Although this doesn't come from reasoning based directly on the AU mod's stated philosophy, it does address the original reason to change it in the first place- a reason to consider the normally no-brainer no-build- without the 'drastic' change of removing it's unique status (low man on the happiness totem-pole, I believe someone once said...).

Its dificult to unravel the chronology of the mod with all the revisions, but it looks like the colloseum change was added in the same version as the electronics requirement removal for MT, so it was indeed 'passed over'.

Any personal feelings aside, I completely agree with Dom's point that this should be a discussion about overpowering potential, not complexity. As alexman pointed out earlier, there are dozens of tough decisions every turn in the game. Applying the level of analysis you demonstrated to a generic build decision is a huge waste of time. Have you considered that getting hung up on details at this level might have something to do with why you've given up on Diety level?

BTW- we are in the Star Wars universe, merely seperated by a great deal of time and a few galaxies...

[Moonbars]

In your example nBarclay, you could just as easily have gotten hung up on granaries & pop grow versus barracks and conquest, or library & marketplace. Imagine doing population growth forecasts based on the opportunity cost of just one more warrior, or worker/settler/temple, before granary

It seems to me that the only time a player is able to build an unusually large (i.e. more than 2) super-early cols, the game is going his way anyhows and he doesn't need the TB to swing the game - in fact he might win before he gets the benefit at all!

On a tangent, wouldn't it be great if the AI would make cost/benefit calculations like this for it's cities.. that should be something it is better at then us guesswork based humans.

[nbarclay]

Originally posted by Rommel2D

Halving the maintenance cost (not to mention the earlier change) makes what is an ancillary improvement in stock to a more ubiquitous one. This seems drastic to me compared to the Tourist Attraction, which I would not normally alter my research pattern or building priorities much for.

If it is true that the tourist attraction bonus has such a minor impact as to have essentially no effect on research or building decisions, what is its benefit in terms of adding interesting strategic choices? Where is the meaningful strategy if the difference between one choice and another is too small to matter?

I see a catch-22 here. To the extent that the difference between one choice and another is too small to make a meaningful difference, the change would add complexity to the game without adding genuine strategic depth, and therefore would violate the philosophy of the AU Mod. Similarly, replacing a situation where not building early colosseums is almost always a no-brainer with a situation where not building them is a no-brainer sometimes and building them is a no-brainer sometimes would not add any meaningful strategic depth.

But to the extent that the choice players make matters enough to be strategically interesting without the correct choice being completely obvious, that raises the question of how players are supposed to go about determining what choice to make and how they are supposed to figure out whether a decision they made was a good one. Can anyone show me a situation where the tourist attraction bonus would create a strategically meaningful choice for players, yet where players could feel comfortable that they know what they're doing without pulling out a calculator?

[alexman]

From the discussions in the past week, it seems that there are now three alternatives for the Colosseum:

• A: No change from version 1.07: Colosseum is Militaristic, and costs 120 shields and 1 gpt for maintenance.
• B: Turn Colosseum into a tourist attraction and restore maintenance back to stock level of 2 gpt.
• C: The same as B, above, but also remove the militaristic flag.

The debate has already been going on (in circles) for two weeks, so the panel has 48 hours to vote by ranking the choices A, B, and C.

[nbarclay]

My vote is ACB. In my view, the Militaristic flag is too powerful if the tourist attraction bonus is implemented.

[pvzh]

Am I on the panel? If not, disregard my vote.

A - by the large margin
CB

[nbarclay]

Considering that you've been nominated, numerous people endorsed the nomination, and no one objected, I'd say you're on the panel. (Don't you love how informally we do this? )