AU 202: Analysis, Solutions, and Stories (spoiler)

[Dominae]

Another twist

Well well, things are looking up! What is widely regarded as the worst civ actually has a secret weapon that makes them formidable: the English are lucky!

I send a few more Horsemen to grab the relevant Iroquois cities, then turn my attention to the Germans, who are getting angrier by the minute (someone really should tell Bismarck to calm down...). For the first time in recent memory, I use the Iroquois as a "punching bag" to generate a steady stream of Elites (a strat from the Theseus when he was known as rpodos). Killing the Iroquois for sport is certainly fun (if not completely un-PC), but it has yet to be profitable; I haven't gotten on Leader from them yet.

Just as I'm celebrating my good fortune, I look toward the German cities and begin pondering why they all have City Walls. I don't recall the AI even thinking Walls were all that great...must be a "minor tweak" in 1.14. Then it hits me: the Germans got the Great Wall off a cascade from the Great Library! ARGH! What was once looking like a really good game is now surely going to be long and arduous. Worse yet, I contact the rest of the world at around this time (~300AD), and they're behind our continent in tech! Bismarck is the one with the tech lead, and he's not trading, so the Great Library is crap (although I admit it got me up to 4000 gold, edit: more like close to 3000). Out of fear of German superiority (they've got Chivalry and Theology over me), I begin doing my own research, with an unexpired Great Library collecting dust.

[Dominae]

Some tough choices

I realise that trying to conquer German cities with Horsemen is suicide, so I begin devising plans to obtain Chivalry. Do I research it myself? Nope, that would take too long. Do I buy it from the Germans? Nope, that would give them money, which they could use for Horsemen upgrades. Do I steal it from the Germans? Nope, too expensive, as I wouldn't have any money left for my own upgrades. Do I wait for someone else to research it? Everyone else is basically in the Stone age, so that would also take too long.

The Germans and I have been fighting pretty much ever since my war with the Iroquois ended, but with no real progress so Bismarck is ready for "peace" anytime. In order to buy myself some time to mull over how to get Chivalry, I sign an alliance and a right of passage with the Greeks. When our alliance is over, I declare peace with Germany, but keep the ROP in place. For about ten turns there's some major fighting going on between those two, right in the core of my empire. The Germans have a silly amount of Horsemen, so I help out the Greeks by doing some "interference".

Here's a very messy screen of the situation:

[Dominae]

That's it for now, although I've already passed this "hump" in the game and it appears to be smooth sailing until the end. I got a pretty lucky break against the Germans. If you've played the game, you may know what I'm referring to. Hint: the answer is in my last screenshot...can you find it?




Dominae

[Tharsonius]

hmm, has it something to do with Bremen?

[Catt]

Originally posted by Dominae
That's it for now, although I've already passed this "hump" in the game and it appears to be smooth sailing until the end. I got a pretty lucky break against the Germans. If you've played the game, you may know what I'm referring to. Hint: the answer is in my last screenshot...can you find it?

Well-played and a good recap (and great screenshots! ). I want to follow your game to the end!

Originally posted by Dominae
Catt, I was thinking about this very discussion last night, and came up with almost exactly the same analysis! I still hold by my different conclusions, however. Here's why (sorry for the math!)

Good analysis! Exactly what I was thinking, but was too lazy to lay out coherently.

But I think we'll just end up being on opposite sides of the fence on this one (unless we confirm that a "settler" RNG result will always produce a city if a settler is prohibited -- in which case, I'll never close off the settler option!).

There are simply too many variables in the formulas, as applied to a given game circumstance, to produce an ironclad rule, IMHO. For instance, I think our personal experiences (or our personal game outlooks in terms of aggressiveness / conservatism) dictate different probabilities on a settler dying or being captured during a 20-turn, unescorted walk home. My views might dovetail with yours if I were playing sedentary barbs and/or with a "promising" mix of AI civs -- but as of now I too often experience aggressive AI opponents and/or pretty crafty barbs that will target a settler (and if they have horsemen, watch out!). I think it is also dependent on difficulty level -- my view is that early AI aggression is necessarily more frequent the higher the level -- the AI starting bonus units means the relative power level almost dictates an early war of aggression (provided there is a moderately aggressive AI civ nearby). The prospect for AI aggression may decrease over time if the human works to close the gap (as human unit totals grow to match AI levels). But the 30 - 40 turns it takes to produce a settler from a hut and get him back (assuming a 20-turn walk home) provides an awful lot of time for aggression to get bubbling. In sum, I am pretty skeptical of much settler value 20 turns away from safety without an escort; I think the value is pretty darn low, absent some very favorable circumstances. And I think the ongoing investment of 1 gpt while the settler is making its way home decreases the settler value further -- whereas the value of, say a tech, is immediate and without further investment. No "time value of money" costs associated with the other goody hut possibilities.

And if even if the v-coefficients on the other possibilities remain unchanged and the p-coefficient moves only a bit upwards across the board, my gut still tells me that trading a very small settler value (v * p on the distant settler) for a slightly increased p' on certain alternatives (principally techs) may be a good trade. Of course, this still all devolves to our differing view of settler value from a distant hut!

Catt

[Dominae]

Originally posted by Tharsonius
hmm, has it something to do with Bremen?

Yup! Bremen is their only source of Iron, and they have yet to hook it up to the rest of their trade network!

MUWAHAHA!


Dominae

[Dominae]

Catt, you're right in saying that it depends on personal experience concerning the chances that a "far out" Settler actually makes it home. In my games it's definitely around or over 50%, but I'm probably just lucky. Given that you've had bad experiences with losing "far out" Settlers, I can see how increasing your chances of getting a better "sure thing" (preferably tech) would be appealing.


Dominae

[Jawa Jocky]

Damn Germans

After getting a settler, cerimonial b. and bronze working I got 3 warriors, 2 mini-maps, and a lot of empty huts. The barbs have also been given my weak civ a lot of problems. I wasted to much time before pruining and now Germany is being a real pain. I just finished the acient era and I'm hoping for a GA to help me out of the hole I have dug for myself.

This is also my 1st attempt not to overlap my cities so much. I can't find the right balance. I either space them too far like this game and have problems early, or they are too close and I hurt myself in the 2nd half. Oh well its only been a year I might figure this out yet.

[Dominae]

Hm, your cities are a lot more tightly-packed than mine. Are you playing Deity?


Dominae

[Jawa Jocky]

Originally posted by Dominae
Hm, your cities are a lot more tightly-packed than mine. Are you playing Deity?


Dominae

No just Emperor


Okay, here is my moment of truth. I've held off the Germans and I'm actually losing more cities to the Ir. to due culture flips. I hope I manage my GA better than the rest of this game.

[Nor Me]

Depriving the Germans of iron. Now that would have been a good idea:

[notyoueither]

Hmm.

Playing Yanks. Didn't know it going in, but their traits are ideal for REXing on this map. It's almost not fair. At Monarch, anyway.

I am at 900bc. Have the entire continent scouted. Mostly peaceful, but I told Otto to stuff an ultimatum. He declared war, then gave me 30 gold some turns later for peace.

The scouts did great for me. I got the free settler from hut 1 (I won the game right there). Then I proceeded to get many, many techs. The last hut popped gave me Polytheism.

The Iroquois just built the Oracle for me (nice of them) and started the Pyramids. Only fly in the ointment so far is that Washington was supposed to finish the Colossus a turn or 2 ago and then switch to units to start the conquering. Salamanca would build the Pyramids in 40. It would have taken NewYork longer than that, so I switched Washington to them. Oh well. Iroquois will most likely switch to the GrLib so then I can take that at the same time as the Oracle.

[Aeson]

On the issue of Settlers from huts, I agree with Catt. There comes a time when a tech is more useful. I like your formula Dominae, but it needs to be modified to account for the non-static nature of various game situations. Namely, number of cities, tech level (research time), desireability of location and/or distance to desirable location.

I look at huts as a turn advantage. A Settler when you have only one city is going to be worth 14-20 turns (depending on terrain). Distance plays a factor as well, as the turns that you would save/lose from having a Settler 'produced' at the hut location vs. in your city is just as much a part of turn (dis)advantage. That makes an early Settler very useful, as it basically puts everything ahead of schedual (population, research, gold, production) by at least 14 turns, and is usualy going to be nearby where you wish to settle. Normally a tech in that time frame is worth little to nothing, as it will be a starting tech, which you can trade for anyways.

Once you get past the starting techs, a tech from a hut is worth a lot more. On higher difficulties, or larger maps, it may be up to a 40 turn advantage in commerce/research, plus whatever you can get through diplomacy for it. Now compare it to a Settler in the same time frame, where you probably have 3 to 4 cities, or fewer with Granaries. Now a Settler can be produced on an average of every 4 to 7 turns. Distance again plays a part, but is very much game dependant.

A Settler's turn advantage is in several areas of course, and if one of those areas is at a critical juncture (claiming territory especially), then the long term benefits might tip the scales back in favor of the Settler. That's something the player has to judge, as it can vary wildly from hut to hut.

For example, in a recent game I had already built to over 10 cities. This makes a Settler value very low in turn advantage (I could produce a Settler in my empire just about every turn) from a production sense. In contrast it was a Deity/Huge game where I had already traded or gotten from huts all the starting, and most of the second tier techs. The remaining techs were all 40 turns regardless of research.

At each hut I had been making sure I was blocking off the Settler from being a result because of these factors. Then I ran into a hut nearby a 'chokepoint' (basically a grassland alley between large expanses of jungles) that had 2 cows and 3 luxuries and was about 30 tiles from my nearest city. That made the decision very easy. To make sure a Settler was available from the hut, because I wanted to settle there for certain. I ended up getting a tech anyways.

Later on when I was hitting huts, trying for both Monarchy and Republic from them (researching either construction or currency... not sure which), I made sure to keep Settlers out of the equation. Both those tech's turn advantage was about 20 turns, while I was producing Settlers at least 1 per turn (which makes it very easy to block off Settlers ).

[Dominae]

Catt and Aeson, I'm starting to see the situations when you would want to block off Settlers. The problem is, they do not happen too often in my games. In AU202 for instance (hey! this is the right thread!), once I had all the huts popped, I had under 5 cities and was barely at the level 2 techs. This situation is standard in Expansionsist games that I play. However, if Barbs were running wild, I might think twice about blocking Settlers. Similarly if was further ahead in the tech tree, or at the end of my REX phase.

All in all, it's least nice know that it's possible ('Settler blocking'). Knowledge is power.


Dominae

[Dominae]

nye, I believe the Americans are the best Expansionist civ for this map. The combination of Scouts, large continent with room to expand and Industrious Workers is just dominant.

Have fun!


Dominae