Foundation and Empire

[Amao]

I haven't played noble in a long time so I'm not that familiar with what trades are available. But my original point a few posts back was that a civil service beeline going through alphabet was a good idea. It was pointed out that going through maths was another good option. Either way though, a Civil Service beeline is my recommended strategy. I just got a little side-tracked, .

If CS was the priority, it's normally better to do the oracle bid through the Myst line. I'll take CoL as the free tech and self teching CS. It's more failsafe than use oracle for CS, because you have to research both CoL and math. By doing this line, you missed the ability to build cash, but once you have CoL, you can use Scientist/Merchant to offset this. Personally, I often get Monarchy early and grow capital as much as possible to max the benefits from Civil Service.

 

[coanda]

As I understand it, they seem to value techs right--on a marginal basis.

If you're considering trading a tech, there isn't one value for that tech. Trading Scientific Method to first-place Mansa Musa who's leading in tech, land, and score is a lot riskier than trading it to last-place Tokugawa who's isolated on his own island with just 3 cities. Trading Rifling to Montezuma your neighbor is a much worse idea than trading it to Alexander, Montezuma's neighbor. You may not want to trade Theology to Huayna Capac the Buddhist if you're hoping for a Hindu religious victory, but you might not mind trading it to Ragnar the universally-hated Hindu. In multiplayer, you might demand much more favorable terms from the guy who won the last 10 games in a row than from the guy who's never played Civ4 before. And you might be more willing to trade generously with the guy who has a huge army camped out near your borders and could always just invade and wipe you out if you piss him off too much.

It's also hard to apply generic evaluations to specific techs and game situations. If you're busy building The Colossus you probably don't want to trade Metal Casting away unless several other civs already have it, or you're confident you can finish Colossus first. If you're never planning on building Colossus, there's no reason to keep it as a monopoly tech - you might as well just go ahead and trade it as soon as you get it.

Plus there's a bit of the "that which can wait, must wait" in play (see how I tied that back in to the earlier advice? I thought it was very elegant). A human might reject an offer of Aesthetics for Alphabet simply because he doesn't happen to have any use for Aesthetics at the moment, and sees no reason to trade yet (maybe he'll get a better offer if he waits; even if he doesn't, he's slowing his opponent down at no cost to himself by not immediately accepting the offer but instead waiting a few turns). The AI will simply assign a value to Aesthetics (and not a value of 0, which is roughly what it's really worth right now), and compare that to the value of Alphabet. Conversely, if a human is about to start researching Engineering himself then he'll probably check first to see if someone is willing to trade it to him; I've never figured out what prompts the AI to offer trades but those sorts of consideration certainly don't seem to.

As I said, a very complicated assessment to make - so I don't really blame the developers for the AI being bad at it, but I do recognize that it is quite bad at it compared to any reasonably skilled human player.

 

[MarigoldRan]

That, and you end up haggling with dice. Yay.

I got soured on the whole tech trading thing back during Adventure 4. When the smoke had cleared, I found Sirian's arguments to be pretty compelling.

Really, the only supporting argument I'll make for Alphabet is that it allows you to compare your research to that of the AI.

Tech trading is powerful. But it's part of the game, and you use what you're given.

It's like saying you shouldn't use "queens" in chess because it's broken.

 

[Um the Muse]

compared to any reasonably skilled human player.

Ooh, burn!

Alright, you caught me; I'm terrible at gauging how badly I want to trade for a tech.

You definitely taught me some perspective on it, though, so thank you Is this why people research things themselves, instead of stealing? I was recently playing a game where I got a GSpy early in the game. I was able to nab three big techs that I really wanted, and another half dozen through trades. I thought it was awesome, in a getting something for nothing way, even though the smaller opponents gave me +4s for our "fair" trades. I wanted to trade them out before the tech leader could. Should I have waited?

 

[PeterWolf]

Tech trading is powerful. But it's part of the game, and you use what you're given.

It's like saying you shouldn't use "queens" in chess because it's broken.

Doesn't your opponent have a queen also? It comes down to who uses the piece more efficiently.

Not like tech trading - the AI will always analyse tech trades on a tech-to-tech basis, never considering that he'd be better off maintaining his tech lead, not feeding the human player who's trying to catch up and will eventually kill him.

 

[TheMeInTeam]

To play 's advocate... tech trading is one of the few features of Civ4 which is rubber-banding instead of snowballing (especially in multiplayer). The problem really lies with the AI, which is terrible at judging when a tech trade is worth accepting. To be fair to the developers, that's a very complicated assessment to make, and I don't think I've seen any game AI which does a really good job at it. But it means it's a huge weak-spot for the AI; as you go up in difficulty levels it becomes more and more critical that the human player exploit weaknesses like that one in order to stay competitive with the absurd bonuses the AI starts getting - which is not necessarily smart play so much as it is cheesy play, but it's undeniably necessary as well.

First we need a convincing argument as to why rubber-banding is a good thing; even crap BS like blue shells and lightning bolts in mario kart are inexcusable, but here in a strategy game are you SURE you want to argue in favor of a mechanic that allows weaker play to catch up? There are ways to catch up with NTT.

Tech trades also allow players to form alliances or de facto alliances that completely shatter any other aspect of play; literally getting on the wrong side of a trade alliance in MP, if tech trades are allowed, is game over. When playing with AI, this happens at semi-random, with the player having some but far from complete influence. It can hurt the player, but often enough it helps him tremendously and can disguise bad play. It also requires the AI to have more bonuses to be effective against top players; why is this a good thing?

Tech trade has some strategy to it, but quickly patterns emerge and the amount of actual thinking behind tech trades becomes trivial, at least compared to what you have to sacrifice in terms of the rest of the game to allow them.

 

[Kaosprophet]

As I understand it, they seem to value techs right--on a marginal basis. It's like they can't target techs, they just go for the best immediate choice. If there was a way to get players to not know the tech tree, that's how we would behave, too.

There is a way: semi-random, semi-blind trees. Sword of the Stars does a form of this, and it's definitely interesting.
But we're not working with that situation here, so it's a bit tangential.

If CS was the priority, it's normally better to do the oracle bid through the Myst line.

Probably, but I think that's beyond the scope of this particular exercise.

 

[VoiceOfUnreason]

Fourth checkpoint - our first satellite city has been settled, and we're getting ready to gear onto the next.

Note: because of lost data, the T60 checkpoint is going to be skipped in this demonstration. So this update is a blur of T45 ideas, and T60 ideas. Sorry - I know what went wrong now, and I'll take steps to prevent it in the future.

Overview: I trained a settler and a worker in Washington to get that next city started, then followed that with an Axeman in the city - one of the problems to solve in the capital is "what do I build when I need to grow?". Library and Granary, once they are ready, but you can rarely go wrong with units.

The capital will usually serve as a strong worker/settler pump right up through the discovery of writing, and then there's a small tactical fork. The capital is usually generating all of your research in the early going - so the 25% boost from the library is a big deal; you'll want to get it on-line quickly.

What's quickly? If you think a great city location is at risk, wait. If you think the last good location is at risk, wait. But if you've got the great locations, or there are enough good locations to share, then get it in.

Once the Library is on line, there is a light fork in the development of the capital. In one branch, it continues to act as a worker/settler pump, until the satellite cities are online. Alternatively, production priority drops, and we start looking to the city for research and commerce - still contributing units, but taking some citizens off the hills to hire specialists or invest in cottages.

(In more cramped space, we'd also consider putting it on war footing - the available land tells us that War with Mansa is a ways off yet. "A neighbor's fair share of good land" comes later.)

First worker went with the settler to New York. Because there's no culture readily available there, my worker sequence was (a) chop the monument (b) farm the corn (c) pasture the horses (d) road the horses. Until the city gets up to size, chariots are the build of the day.

It's in this time period I'm beginning to think about replacing my warrior sentries with something stronger - you can see the first axeman moving into position. Warriors will normally garrison the nearest city location once it's settled.

Note the blue squares - each unit is preventing barbarian units from spawning in its square. I'm sliding those tiles around to keep the distractions to a minimum. Eventually, cultural borders will clear out those areas, but it's a good idea to know where the barbs can come from.

The map - there are still a lot of city sites to fill in. Large cities focus on workers and settlers, smaller cities contribute units. At this point in the game, you are trying to minimize infrastructure builds. At most: monuments in new cities, and a single library in the capital, until the contested area of the map is settled.

After Writing, there's usually a research fork - a number of different ways that you can advance into the early game. Best is to review your circumstances, and choose the classical age technology that is going to best improve your situation.

It will usually be one of these
(a) Code of Laws - large open areas of land to be settled, and we have to worry about how to pay for it all
(b) Construction - when we've run out of "our fair share of land" too quickly.
(c) Calendar - when we are surrounded by plantation resources (this one often couples with Iron Working)
(d) Monarchy - when we need a happiness boost from Hereditary Rule.

For this map, the target I recommend is Code of Laws. Big land, seemingly limitless open space to the west, that beautiful target city that's forever and a day from the capital. Maintenance costs are going to suck.

Next checkpoint: a belated T75

 

[vranasm]

if the yellow rings are future city spots, that southeast of capital violates the rules of the game

btw spectacularly good land... is this any way doctored?

 

[VoiceOfUnreason]

if the yellow rings are future city spots, that southeast of capital violates the rules of the game

Yup - side effect of using an external graphics tool. 1S.

btw spectacularly good land... is this any way doctored?

Only the choice of level, no world buildering.

 

[vranasm]

Yup - side effect of using an external graphics tool. 1S.



Only the choice of level, no world buildering.

btw one of the big mistakes newer players do and need to learn is city placement.

Maybe it would be worth the thought to explain why settle the cities the way they are marked...

oh and btw why no city for the marble? the pigs+corn+fish seems like overkill on itself, it's nice to have tripple food city, but imo it's better to get there 1 2-food city and 1 1-food+marble city ;-)

 

[learner gamer]

As an aside VoU, how do Mansa’s EP compare with yours?

Given the theme - do you think I'm paying any attention at all to EP? Really?

. Espionage might not even be a deity level tactic. I've seen it used sometimes to great effect, most recently by AbsoluteZero in his Kossin shadow. But I play pretty much without it and do alright up to immortal.

Just wanted to pop by and say thanks for completely failing to read my post. FWIW, I quite deliberately used the phrase “as an aside” to reflect (i) that I was completely aware that the question that I was asking was likely above the target audience and (ii) I wanted to know the answer because my interest in the game was rising. However, instead of a simple reply to the effect of “ours versus Mansa’s EP is x/x but that’s way above our intended audience” or similar, I get mocked for asking a very simple question. Needless to say, I won’t commit the same mistake (ie. being interested in the game) for the remainder of this series. Have fun writing your guide...or, to put it another way gents....go and do one!

 

[Don Promillo]

Hi, following this thread with great interest!

oh and btw why no city for the marble? the pigs+corn+fish seems like overkill on itself, it's nice to have tripple food city, but imo it's better to get there 1 2-food city and 1 1-food+marble city ;-)

<- Would be interessting for me as well!

Another question: when is it worth to violate the SIP-Rule in order to move 1 or 2 tiles to be adjacent to a river?

I can't wait to find out, what our gurus suggest for building in your cities (especially the capital) after around 60 turns. (less infrastructure, more workers/settlers, or more military???)

very interessting discussion, let's keep it up, and hopefully / please keep fokussed on the main topic!

 

[VoiceOfUnreason]

oh and btw why no city for the marble? the pigs+corn+fish seems like overkill on itself, it's nice to have tripple food city, but imo it's better to get there 1 2-food city and 1 1-food+marble city ;-)

Well, because Marble isn't food....

Or more specifically, because the neighborhood is relatively food poor. The green tiles to the east are covered by the existing city; the tiles that don't overlap with the placed cities are plains or tundra.

If the pigs "belong" to the city in the east, our marble city can use borrow them to grow - without the pigs, it's basically capped out at size 5 (assuming it wants to do anything useful).

Although I'm not planning to put city specialization into the syllabus, my play is definitely influenced by that style, so I'm saving the pigs for the "GP Farm".

If you don't need wonders, then you don't need wonder resources, so I think avoiding that area helps subtly reinforce the "no wonders" message. Totally different story if that marble were gold or silver. Or food.

Even were it on the map, it wouldn't be high on the priority list. Perhaps even with the pig city I misplaced on the east coast? At least that one has the excuse of a food resource that can't be used anywhere else.


All that said, I did look at it - one dot mapping lesson that I think we fail to emphasize strongly enough is the notion of making sure that we have cities working all of the good land. Having a few published examples of cases where dividing food between cities is stronger than loading it all into one would be a good thing.

Ultimately, I decided I didn't like the land there enough to explore that lesson.

 

[VoiceOfUnreason]

Another question: when is it worth to violate the SIP-Rule in order to move 1 or 2 tiles to be adjacent to a river?

After you are good enough to beat Noble regularly while accepting the settle in place constraint....

a) When moving gives you an abnormally strong city tile. The most common example would be moving to a plains hill for the extra hammer (stone, marble, brown elephants). Less common, to settle on a sugar for the free food (you'll see the same trick with dry rice, but if it's on a river, the rice ain't dry).

b) One City challenge - the city you start with is the one you are stuck with for the rest of the game, and if you are expecting to research deep enough into the tree, Levees are pretty important.

Otherwise, I wouldn't bother - there are other advantages available, but they are pretty small, and the opening move is usually being made with limited information.

It should be possible to put a simple turn based guideline in place, like "don't move unless you see a benefit that will pay off in less than 50 turns". Haven't given it enough study to suggest an actual number of turns, though.

 

[VoiceOfUnreason]

You can see the beginning of the second push. Philadelphia probably should be training a worker, rather than constructing a granary. It also shouldn't be unhappy - that's the lack of a city garrison at work. If you squint very carefully with at the picture, you can find evidence that I simply haven't improved enough tiles.

I've used Writing to construct a library in Washington. I also opened up the borders to scout Mali and England. It's a relatively inexpensive way to observe how the AI is managing your land for you. I can't be bothered to click too many times up there, so I sent a 2-mover.

Still pushing out toward the target in the west. Boston will settle the sheep/horse/fish city to the SW (sorry about the lack of resource icons, there are just too many of them - they haven't moved from earlier pictures). Washington and New York are pushing out toward the wheat and the target.

Pretty much all of these cities were improved by chopping into a monument, then improving the food tile(s).

Still a lot of warriors. There are a couple errors visible here. A minor error is the Axe down in the tundra - there should be a warrior standing there, with the axe doing something useful. A more significant error is the lack of good units around the gem city to the west - that's a little peninsula of civilization in a sea of fog, and it will turn out that I lose time trying to dodge barbs.

Still gunning for code of laws, although you can see to the south west of Philadelphia I discovered Iron Working. In a great place if I had split the floods into two cities. Darn.

Next checkpoint: T90.

 

[sadmachines]

Are those extra units on intended future city locations?

 

[gspchamp999]

Are those extra units on intended future city locations?

I believe they are for spawnbusting. The future city sites were shown in post 68.

 

[sadmachines]

Ah, so in most cases, yes.

 

[VoiceOfUnreason]

9 Cities at this point, and still lots of room to the west. So I'm still in phase one here.

New cities get to start on infrastructure a little bit earlier than old ones do, but the warriors garrison the cities, so those units really need to be replaced to keep up my troop strength.

I'm feeling a bit of a squeeze on worker turns (because I'm sending a large work team out to Seattle. Atlanta is getting a library because I'm still feeling squeezed on tech rate. I think that's primarily a reflection of the fact that I haven't done anything to lift the happy cap, combined with the fact that I've been celebrating hammers over commerce.

OK, Time for what I think is the most important tip here. Review the demographics screen!

My goal at Noble is to be number one in all six of these categories by 1AD (T115). So somewhere between T75 and T90 I need to start sneaking a look to see how I'm doing, and where I need to shore things up.

For sure, the population and soldiers numbers are a bit twitchy - you can look up the mechanics elsewhere. I'm primarily considering each category in one of three bins - I'm behind, I'm ahead, or I'm ahead by a lot.

Here, we can see that I'm close to my goal in GNP, and quite a ways from my goal in Soldiers. In practice, in this map, that's safe - I can keep England and Mali under control. But the object isn't "safe", the object is dominant position - so in spite of how little infrastructure I've dropped in place, I still need more units!

Barracks would help here - I'd normally have at least one by this point, but with so much open land and such peaceful happy targets, I'm not ready to gun the guns just yet.

As for GNP: Seattle will be very helpful here, and further strides follow as I unlock Courthouses.

For the moment - the theme is vertical. I want to get these numbers boosted. I'm not worried about competition for the lands to the west: I know that England is blocked and Mansa is still playing small. So my focus is purely on the numbers.

Next checkpoint: T100.