Announcement

**nbarclay** · February 19, 2004, 20:08

I decided to do some playing with the map generator and see how likely islands really are to contain industrial and/or modern resources. (Earlier resources don't factor in as much with long-range exploration and settling because by the time you get Navigation or Magnetism, you know whether you have horses and iron and very likely whether you have saltpeter, and you can look at an island you're considering settling to see whether it has the resources you know about.) I won't guarantee that these figures are exact because I didn't exercise the greatest level of care humanly possible, but they should give a good idea.

The statistics indicate how many islands larger than one tile contianed at least one industrial or modern resource out of how many such islands total. An island for this purpose was defined as any land mass that none of the civs starts on.

Large, Archipelago, 70% water, Temperate, Normal climate, 4 billion:

Map 1: Three of eleven
Map 2: Four of nine
Map 3: Four of ten

Large, Archipelago, 80% water, Warm, Arid, 5 billion:

Map 1: Seven of ten
Map 2: Nine of sixteen
Map 3: Ten of twelve

Large, Archipelago, 60% water, Cool, Wet, 3 billion:

Map 1: One of nine
Map 2: One of ten
Map 3: Three of seven

More generally, large islands are likely to have an industrial or modern resource somewhere on them, while small ones most likely will not. Terrain type also has a lot to do with it: islands that have only grassland and/or plains have no possibility of containing industrial or modern resources.

In any case, while it might be true that most islands have a resource on them if you count horses, iron, and saltpeter, it is clearly not true that most have an industrial or modern resource hidden on them. Claiming remote islands can improve the odds of having industrial and modern resources a bit, but since the resources on islands are often redundant with resources available on continental holdings, the overall improvement in access to resources is generally smaller than the chance that a resource will be on an island in and of itself.

There is certainly nothing wrong with a strategy that tries to get and hold lots of islands in the hope of claiming additional resources. But such a strategy is normally not anywhere near as important or as clearly advantageous as Dominae made it in this particular game.

Nathan

**nbarclay** · February 19, 2004, 20:18

Originally posted by Dominae

As I mentioned elsewhere, the sea lane to the English, Mayans and French was probably a mistake. I did not even consider that with the Great Lighthouse you could have access to all the Luxuries in the Ancient era. Silly me.

The fact that other civs had to build harbors before they could trade luxuries delayed the ability to get access to all luxuries somewhat. It wasn't just a matter of, "Once you have the Lighthouse, you can trade for a civ's luxuries the moment you meet them." I was in the medieval era before I switched to Republic, and at that time, not all of the trade routes were open yet.

**nbarclay** · February 19, 2004, 21:04

Originally posted by Dominae

Nathan, I think we've reached the point where we need to "agree to disagree" regarding this entire issue. As a final point, I would like to ask exactly how many C3C games you've played out to completion (or, at least to the mid-Industrial era). It is my suspicion that your reaction to the resource distribution in this scenario is based on little experience regarding C3C resource distribution in random map games.

Somewhere a bit over half a dozen, but I'm not exactly sure. Not enough to get a really fantastic feel for how resources work under the new rules, but enough to get a very definite feel that resources are harder to come by. I'm almost sure I've even commented in at least one of the resource scarcity threads.

And yes, I've gotten some annoying situations. In the Chasqui Scout game you posted, I had to fight for coal and wouldn't have had saltpeter without fighting if it weren't for a culture flip. In another game, I had to launch an amphibious assault for Saltpeter. (And that's just what I remember off the top of my head.) Then there are the games where if I'd REXed any less aggressively, I would have had to do without a resource until I conquered it, or where if I hadn't conquered some territory earlier, I wouldn't have had a resource later. So I very definitely know that resources are harder to come by in C3C.

What I'm trying to say is this: learning that the resource distribution in C3C sometimes "sucks" and how to deal with it when it does occur is a far more important lesson than how to estimate where resources will lie given a certain land mass and tile distribution. C3C challenges our previously-gained Civ3 knowledge, and it's a good idea to learn to adapt.

I don't disagree that learning how to deal with resource scarcity is important, but that was not the goal of this particular AU exercise. Further, the way you scattered extra resources over remote islands seriously devalued this game even for that purpose. Resource scarcity was a serious issue for players who did not make settling distant islands a priority, but someone who settled enough of the remote islands would not have had resource problems at all (aside from perhaps having to rush a city improvement or two here and there). And for those of us with resource problems, solving the problems by capturing islands rather than by striking into the heart of an enemy was a significantly more practical option than it would normally be with the level of resource scarcity you stuck us with. Even as a training exercise in dealing with resource scarcity, this map was skewed enough from what would normally be expected to undercut its value.

**ducki** · February 20, 2004, 11:19

that was not the goal of this particular AU exercise.

I think what Dom is getting at is that resource scarcity is an inherent issue in any C3C game and just because the game is _themed_ - not it's sole goal, but it's _theme_ - is Seafaring does not mean other, and particularly inherent, C3C issues should not be a significant factor. If we're learning better strategy and we don't want to deal with resource scarcity, then Conquests is not the game we should be playing - that would be PtW. Arguing that resource scarcity should not be part of a C3C game is like arguing that having to deal with a weakened Forbidden Palace should not be part of the course - it's an inherent part of Conquests and we have to deal with it unless we want to go back to PtW - in any game we play.

That's just my opinion and my guess at what Dom was trying to point out. I could be wrong.

**nbarclay** · February 20, 2004, 18:05

Originally posted by ducki

I think what Dom is getting at is that resource scarcity is an inherent issue in any C3C game and just because the game is _themed_ - not it's sole goal, but it's _theme_ - is Seafaring does not mean other, and particularly inherent, C3C issues should not be a significant factor. If we're learning better strategy and we don't want to deal with resource scarcity, then Conquests is not the game we should be playing - that would be PtW. Arguing that resource scarcity should not be part of a C3C game is like arguing that having to deal with a weakened Forbidden Palace should not be part of the course - it's an inherent part of Conquests and we have to deal with it unless we want to go back to PtW - in any game we play.

There is a difference between having to deal with the normal, natural resource scarcity that tends to result from the map generator and having to deal with artificial resource distribution patterns deliberately set up by the scenario designer in ways that defy probability. And whether or not the resource distribution Dominae set up can be considered plausible, I think it has to be considered improbable.

**Theseus** · February 20, 2004, 21:01

Originally posted by Dominae
I was expecting many of the more experienced players to enjoy a little late-game challenge (like the elusive massive inter-continental invasions Theseus craves for), but apparently you like to win in the early-game, and coast in the late-game.
Dominae

I'm writing up my final AAR, so I finally started reading this thread. Ummm, yes, massive IC invasions were involved.

Loved it. Absolutely loved it.

More than you imagine...

[hint hint, teaser teaser

]

**ducki** · February 20, 2004, 22:12

in ways that defy probability. And whether or not the resource distribution Dominae set up can be considered plausible, I think it has to be considered improbable.

In my opinion, your entire argument rests on this, and considering the actual definition of probability, plausible, and improbably, not to mention the phrase "defy probability", I just don't see eye to eye with you.

Your argument is that it's a plausible outcome, although not necessarily the most likely one, therefore it shouldn't be in an AU course. That seems preposterous to me.

It's improbable that any given tiny island(fewer than 4 tiles) would have a resource, and yet, it happens to me game after game. Your argument just doesn't hold water in my opinion - to me, the quoted sentence just sounds odd, especially in the context of a pseudo-random number generator.

You played the odds and lost. I don't see why it's such a big deal and why you think it's going to drastically alter the way newcomers to the University will approach their C3C games. This was a plausible map - meaning it could have come from the RNG. Why would that actually change anyone's general strategies?

**nbarclay** · February 21, 2004, 01:33

[QUOTE] Originally posted by ducki

You played the odds and lost

One of my biggest objections is that Dominae's manipulations not only deprived me of all genuine opportunity to play the odds but turned my knowledge of the odds against me. Imagine playing blackjack, making a bet that nineteen will win, and then finding out that the only cards in the deck were cards with a value of ten and a single nine. What looked like it was likely to be a good bet never had a chance because the game was deliberately rigged.

Your argument is that it's a plausible outcome, although not necessarily the most likely one, therefore it shouldn't be in an AU course. That seems preposterous to me.

My argument is that even if the map is considered something that would be pluasible from the map generator, which I don't agree that it is, it is improbable enough to skew the lessons people take away. Consider your own inaccurate impression (at least assuming my tests were valid) that islands practically always contain a useful resource, and how Dominae's map design encourages that inaccurate impression.

It's improbable that any given tiny island(fewer than 4 tiles) would have a resource, and yet, it happens to me game after game.

Have you ever studied the mathematics of probability? The effects are essentially exponential in nature, which causes chances of something happening at least once to grow dramatically in multiple "trials" but also makes the chances of several coincidences happening at once very small.

If the chance that an island that is eligible to have an industrial or modern resource will actually have one were 50-50, games with four such islands would have a resource on at least one of them fifteen games out of sixteen. That sort of phenomenon is likely where your exaggerated (according to my tests) impression of the value of settling small islands for resources comes from.

But the same statistical forces that make having some small islands with resources likely make having resources on practically all eligible islands unlikely. Then multiply that unlikelihood with the unlikelihood that so many late-game resources will be absent from the home area, and multiply that with the further unlikelihood of having resources follow such a strong pattern of having getting farther away from home the later in the game you get (horses and iron at home, saltpeter and coal a bit farther away, and all the later resources except oil farther still), and the result is something that looks far more like deliberate design than like random chance. Someone looking at each of those items in isolation might think the end result looks within the bounds of random chance. But for someone who understands the mathematics of probability and is looking closely, I think Dominae's fingerprints are pretty clear.

The irony is that Dominae could have given us some pretty significant resource challenges without being so obvious (at least to someone with a suspicious mind and the right mathematical training) about it. With coal or rubber one more readily available, the difficulty obtaining the other would have been less suspicious. Similarly, with aluminum or uranium one readily available, having to chase after the other would not have been especially surprising. And unless the rate of industrial and modern resources on remote islands started out unusually high, some could have been added without pushing things to the extreme of having one or another industrial or modern resource on practically every island.

That would have produced a map where we had to deal with some pretty serious late-game resource issues but kept things a lot closer to what can normally be expected in a C3C game. And if it threw my attempts to play the odds a bit off, it at least would not have rubbed my nose in it.

Nathan

**Aeson** · February 21, 2004, 01:53

There are 4294967296 possible seeds for a map. (I think values above that wrap around, but not sure) Then the features of mapsize, water level, landform, climate, and barbarians all make significant differences in the outcome of the map. Age, temperature, and climate don't affect the shape of landmass, but do affect the tiles that comprise it.

Given the number of permutations on map generation, I don't see the problem. What Dominae gave us certainly wasn't average, but is it supposed to be? I'll bet among the full set of possible map generation results you could find thousands, if not millions, of maps that had many of the same characteristics. Not a large percentage of the maps (which number in the trillions), but certainly within possibility.

I've seen sea passages like that before. Started out on a two tile island on a Huge Pangaea map... Weird things happen with the map generator, and I would say that if anything, this map Dominae gave us is a good representation of the discrepancies that you run into if you play enough.

The other thing that applies to all games... the more land you claim, the better your odds of holding a given resource. But claiming a set amount of land doesn't guarantee you will have it. If there is a free island (or any area of unsettled land), you should always claim it. Always, always, always...

**nbarclay** · February 21, 2004, 03:31

Aeson, do you have any views on the possibility that claiming remote islands, especially if they are not defended strongly, increases the risk of being attacked by an AI? In C3C, I've run into a situations where an AI broke a trade deal to attack me and by the time the war ended, it was already selling the luxuries I wanted to someone else. That makes me view anything that increases my risk of getting involved in an unwanted war as at least partly a liability. Which, in turn, makes me less than entirely enthusiastic about claiming remote islands if I have plenty of land closer to home, especially during a phase of the game when I'm operating with a lean military.

It's not impossible that the times AIs have attacked island posessions I held, they would have attacked me more directly if I didn't have holdings they could get to more easily. But occasional assaults on island posessions when I hold them are enough to give me a feeling (whether valid or not) that controlling islands is a mixed blessing, good for possible resources and an extra specialist or two if nothing else but not so good for trying to stay at peace.

Nathan

**alexman** · February 21, 2004, 05:09

Re: AU501 - Post-game comments

1. What did you learn about the power of Seafaring?

Archipelago maps are easier than other types of maps, and the Seafaring trait exaggerates the human advantage on such maps. The AI suffers because, although improved over PTW, it still can't mount a serious invasion. The human also benefits from the AI's lack of aggressiveness in seeking communications with all other civs.

Dromons are awesome.

2. About Civ3 in general?

Communism rocks in C3C 1.15, especially in archipelago maps, where the distances from your capital are greater than usual for the number of cities. When I switched from Monarchy, Communism was almost as good as Democracy in research, and much better in production. I'm sure that after I built my infrastructure and the SPHQ, Communism was better than Democracy even in research. Communism was also more cool because without the ability to rush-buy, I had more cash than I could spend. The money went to spy missions, which was lots of fun because I had not used them a lot before. I had never actually spent 3,500 gold to steal plans before, but I could afford to do that when I had 40,000 in the bank. Another nice thing about Communism is that the luxury slider affects all your cities, as opposed to only your core cities, so you don't have to rely as much on luxuries. When I attacked the Maya, I lost two luxuries, but most of my cities were still in WLTKD with a 20% luxury setting plus MP.

I like the new corruption model, although I was a bit surprised at how close the FP needed to be to the Palace to get an effective second core. The standard-map OCN on a large map and the 60% Deity multiplier exaggerated this effect though.

The new specialists made me spend a huge amount of time micromanaging my cities. Communism made it worse, because all your cities are productive, so you don't have any totally corrupt cities to forget about.

3. Was there anything you would have done differently?

Up until the very end, I couldn't decide on how I wanted to win. I would have reached Domination much sooner if I had abandoned research after Synthetic Fibers, and if I had produced units instead of Power Plants, Offshore Platforms, Research Labs, Commercial Docks, et cetera.

4. Which civ gave you the most trouble, and why?

Japan and the Hittites were surprisingly tough for their size and technological level. Fortunately, I got to choose when to attack each civ (except Egypt, but they were weak) so that made it easier to deal with the other civs. The Maya were the biggest baddest civ all game long, becuase they conquered France early. That actually worked in my favor though, because they could supply me with crazy amounts of gpt.

5. What did you think of the Plague?

I don't like the Plaque, I don't like disease, I don't like pollution, I don't like Volcanos. I don't even like the RNG!! What can I say, I'm a control freak!

6. How did the AU mod affect gameplay?

First of all, sorry about the caravel/Frigate/Galleon movement bug. It was annoying. But overall, I was pleased with the AU mod.

The AI research choices worked out more or less as expected - more unpredictability. I even had to research Communism and Espionage myself in my game!

The increased attack of Infantry, along with the reduced attack of Cavalry, and the archipelago map with the seafaring trait, made my artillery+slowmover stacks more poweful.

The SoZ was not a factor at all because the tech pace was so fast and the Japanese were so backwards.

I had to think hard before deciding to research Philosophy.

The movement restrictions of Galleys and Curraghs delayed contacts, which is a good thing. I had to think before sending my precious Dromons off to suicide missions instead of helping with bombarding Tokugawa. Oh yeah, I actually saw the AI build tons of Curraghs!

The only happiness buildings I built were Colosseums!!! With access to all the luxuries, I really didn't need more happiness, and this building is quite efficient for non-religious civs, especially in Communism. Much better to build 4 units for MP than a Temple and a Cathedral.

I still didn't reseach Advanced Flight because I got Stealth (which offers better lethal bombard airplanes) relatively soon after Flight. The AI didn't come close to missing the lethal bombardment from bombers, but I missed it for sure. Good change.

I decided to follow the classic ToE-Hoover beeline, even though Motorized transportation did not require Electronics. I just wasn't ready to attack with Tanks even if I had them earlier, but I suspect that this might not be the case in other games. With the relatively low value of Hoover on this map, I did think about getting Communism as my free tech instead of Electronics, but I decided against it because of its low cost.

7. Did you find any Easter eggs?!

Dominae did a great job with this map in all areas.

I like the more scarce resources in C3C, and I loved the resource distribution in this game (although it didn't affect me much with Russia under control from relatively early on). I really don't see the reason for all the fuss. Catt and Aeson, among others, proved that in this game it was possible to trade for resources like rubber if you didn't have them in your territory.

**Aqualung71** · February 21, 2004, 11:37

Are Armies too strong?

I haven't started a new thread on this since some people are still playing, but perhaps we can open some discussion here.

I noticed that a lot of players (including myself) are building large numbers of armies, that make invasions relatively one dimensional (or two dimensional....as in, artillery + 2 armies = captured city). Conversely, I can't recall seeing one AI army in my AU501 game, and although clearly the AI does use them, it doesn't seem to place any higher importance on them given the huge boost in Army power in C3C.

So my question is, are Armies too powerful and if so, is it within the brief of AU to consider a mod to tone down this power (if possible, that is) which in my opinion gives the human player a significant and often game-winning advantage? I got the impression from reading DAR's that often the battle objective seemed to be generaing an MGL for building unbeatable Armies, rather than the overall defeat of enemy forces or capture of cities.

Comments?

**nbarclay** · February 21, 2004, 12:03

Re: Re: AU501 - Post-game comments

Originally posted by alexman
I really don't see the reason for all the fuss. Catt and Aeson, among others, proved that in this game it was possible to trade for resources like rubber if you didn't have them in your territory.

For some strange reason, any time I approached an AI to look into trading for rubber, the response was, "Rubber? What in the world is that?"

Keep in mind that the Dutch were still using riflemen when my TOW infantry invaded, and they were essentially on par with the other leading AIs in tech. I seriously doubt that I could have gotten a net advantage giving the AIs the techs they needed in order to trade me rubber because the upgrade to their combat capability would have been at least as great as the upgrade to mine. And with aluminum and uranium not available, I would have had to either give at least one AI most of an era's worth of techs so they could trade me those or invade someone anyhow.

Trading for missing resources can work great in a close-fought game, but the bigger a tech lead a player builds, the less of an option it becomes. And in this game, I had one of my biggest tech leads ever. Granted, the absence of rubber for me was no more than a nuisance. But I've found over the years that nuisances are, well, nusances.

**Dominae** · February 21, 2004, 13:02

Re: Re: AU501 - Post-game comments

Originally posted by alexman
I like the new corruption model, although I was a bit surprised at how close the FP needed to be to the Palace to get an effective second core. The standard-map OCN on a large map and the 60% Deity multiplier exaggerated this effect though.

Could someone please tell me why the OCN and research rates were different from a typical Large-size map in this scenario? Perhaps because there were only 10 civs instead of 12? This seems like a big deal and I would like to control for it in future scenarios.

Dominae

**Dominae** · February 21, 2004, 13:19

Re: Are Armies too strong?

Originally posted by Thriller
I noticed that a lot of players (including myself) are building large numbers of armies, that make invasions relatively one dimensional (or two dimensional....as in, artillery + 2 armies = captured city). Conversely, I can't recall seeing one AI army in my AU501 game, and although clearly the AI does use them, it doesn't seem to place any higher importance on them given the huge boost in Army power in C3C.

There is a major discrepancy between human use of Army units and AI use of them.

In the attempt to "balance" the MGL's lost ability of rushing Wonders, the Army was made too powerful. No one wanted to lose the power of their beloved Great Leaders from Civ3 and Play the World. Now quite the opposite effect happened: the MGL is arguably more powerful than the SGL (at least the first one).

Personally I wish they had left the Army unit alone (perhaps just improved the healing rate). They're more common now because it's no longer a decision between Army and Wonder. This would have been enough to insert them back into strategic gameplay without boosting their stats sky high.

The AI never used Armies much, and never well. Like mass-bombardment, this is another human-only toy.

Dominae