Turn 2, 3960BC

[Dominae]

The blue circle probably indicates Furs or Silver, or maybe Game. It's too bad we're going to have to settle down there to access any of them though.

As far as I know the site B (Flood Plains) supporters are in the minority.

If we get a Scout from the Tribal Village, the Hill site becomes less attractive since one of the main reasons to favor it is for the faster early-game scouting potential (how many Scouts do we plan to build?). For this reason perhaps we should move the Warrior before settling.

[notyoueither]

Originally posted by Dominae
If we get a Scout from the Tribal Village, the Hill site becomes less attractive since one of the main reasons to favor it is for the faster early-game scouting potential (how many Scouts do we plan to build?). For this reason perhaps we should move the Warrior before settling.

We can't get a tech if we do that.

[Blake]

I don't think there's any such thing as too many scouts. Well, 3 is as many as we can effectively use, assuming one or two directions are basically dead ends. But we wont be training more than 2 scouts, possibly only one.

[nbarclay]

Let's not be in a hurry to play this. I just noticed that the flood plains blue circle is in a different place from where we've been settling in our flood plains experiments. That different place, after border expansion, provides access to three tiles that are currently hidden by the fog of war. If the reason the circle's there is that there's a bonus on one of those tiles, settling on the hill could prove to be a big mistake. And if we're going to be relying heavily on the whip for production, not having access to as many mined hills in our capital won't be as big a deal.

The fact that the flood plains blue circle is in a different place from in NYE's simulation lends added point to the possibility that a flood plains start there could be better than what we've been getting in NYE's simulation. The difference may be trivial, or even reflect the fact that the blue circle algorithm is sometimes a bit stupid. Or it might mean something important.

Also note that settling there would apparently provide access to one more forest than NYE's simulation included (a forest we can see only the fringes of). Assuming that's the case, I don't know whether it would affect health. But at the very least, it would let us chop and mine a forested hill without giving up a health point.

So I'm very strongly tempted to do one of two things. Either move the warrior to uncover one of the three currently hidden tiles a FP start on the blue circle would make available, or go ahead and settle on the blue circle FP and hope we get lucky in what's revealed.

[notyoueither]

Good point.

[Aginor]

I remain in support of the FP - settling at the blue circle brings another 2 Forests into the initial radius for yet another health point. Do you want to have one extra hammer in the city radius, or the equivalent of the Expansive trait in the capital? Plus we get a forested Plains hill we don't have to mine early on that we can still exploit for 3 hammers. I think this is a pretty easy call.

We're going to be running Bureaucracy starting in the mid game...I know which one I'd choose.

Another thought in that vein - IF we get religion, it may make sense to set up the conditions for alternate slingshot (Meditation, Poly, CoL, no Masonry, lightbulb CS with GP) in order to further turbocharge this production center in the long run.

[nbarclay]

Looking at the tiles more closely, and bearing in mind that I'm definitely not an expert on reading the tea leaves of what tiles are like based on their fringes, here's what I can come up with for the tiles that a blue circle flood plains city could work but that we can't see yet. If anyone else wants to offer alternative speculations, I'd love to hear them.

3-2 from the flood plains blue circle is clearly some kind of tundra. It also happens to be in the city radius of the tundra blue circle, which may increase the likelihood of its being better than would normally be expected of a tundra tile. That's a tile we can check out with our warrior before deciding where to settle if we want to.

6-6 from the flood plains blue circle looks like it's probably either a grassland forest or a plains forest. (If it were a hill forest, I'm pretty sure we'd see it the way we do the tile south of it.) Offhand, I can't think of any bonus that a grasslands or plains forest might have that would make much difference in the early game. But it would give us either an extra health or an extra chopping opportunity without sacrificing health. (I don't know the mechanics of health calculations well enough to know which.)

6-9 from the flood plains blue circle looks like probable desert If it has incense, that could throw the blue circle algorithm off, since it's useless until Calendar and of dubious value to work even after a plantation is built, and since incense can provide happiness even if it's not in range for any city to work the tile as long as it's in the civ's borders. On the other hand, deserts have been known, on very rare occasions, to have oases, or even stone or marble. It may also be conceivable, albeit only barely, that the tile could be a flood plains with its associated river hidden out of sight.

Anyhow, those are my best guesses. As I said, I'd be happy to hear any other thoughts.

[nbarclay]

I did some experimenting in the WorldBuilder, and it looks like based on what we see, the blue circle algorithm views the two flood plains starting positions as almost equal. If I plug in my guesses (above) at what's hidden, without any bonuses, the blue circle ends up due south of the hill we're on. (I tried a few different variants of tundra, and both grassland and plains forest, but didn't try every possible combination of the two parameters.) But if I so much as change my presumed desert to a plains, that shifts the blue circle to where we see it in the game. Obviously, a bonus would shift the calculation more decisively in favor of what we see.

I'm still inclined to think there's a good chance that one of the three invisible tiles a blue circle flood plains city could work is better than it appears from looking at its fringes, and possibly a lot better. But with the two flood plains city sites apparently so close in value as calculated by the blue circle algorithm, and with no more than I know about exactly how the blue circle algorithm works, aamong other factors, it's hard to be anywhere near as sure as I wish I could be.

[Aginor]

You are correct in all your territorial surmises under the FOW as best I can tell - the named squares (in order) appear to be Tundra, Forest and Desert.

Here's my thinking on the subject:

Where we are situated we get one special resource, guaranteed. Not an ideal situation. Plus we've got the massive health problem.

On the blue circle, we get a *chance* of additional resources only. Nothing says that the possible resources in question are even visible yet. Could be Uranium in the Desert for all we know.

But the point is this: the FP site is better for long run production, regardless of what is or is not in the squares we cannot see. At some point we're going to be running Bureaucracy in this town, and we'll want to bleed every drop of possible efficiency out of it in the midgame. Plus, we'll grow more rapidly early on, which *in and of itself* is going to be worth more than one hammer per turn. That additional production is just going to come online later, meaning fewer early Warriors and Scouts.

I'll take that trade.

[Dominae]

Let's assume for the moment that there are no additional resources within reach of any of the potential locations.

Looking more closely at the two FP sites:

Site B (2 from the Settler's current location):

3 Plains Hills
1 Forest Grassland Hill
6 Flood Plains
6 Forest
16 River tiles (wow!)

Site C (3 from the Settler):

2 Plains Hills (one forested)
1 Forest Grassland Hill
1 Desert Hill
6 Flood Plains
16 River tiles
7 Forest
1 Desert (probably)
1 Tundra

So, just looking at the tiles themselves, I think site B is the better Flood Plains site. Just about the only advantage site C has it the extra Forest.

There might be some hidden resources site C, but then again the blue circle would be in the same place if there were none.

Based on this information I favor site B.

We could move the Warrior 3 this turn to uncover one of site C's mystery tiles. That would leave our Tribal Village for our first border expansion if we do end up settling site C.

[nbarclay]

Dominae, where do you get that the blue circle would be the same place regardless of whether or not there is anything better than what we seem to see now in the three hidden tiles? In my experiments, adjusting NYE's latest scenario to match my speculations of what the hidden tiles are based on what we can see of their fringes, it wasn't in the same place. It's possible that I goofed in my experimentation, and it's possible that something could be different in the real game even if I didn't goof. But do you have any specific reason for contradicting what my experimentation seems to indicate?

As for the comparison of the two sites, Site B only offers a near-term advantage if we mine more than one hill. In contrast, Site C's forested plains hill provides a tiny bit better gold efficiency if we trade food for production before Site B has enough mines to gain the advantage. If we want scouts out reasonably early, and especially if we want more than one, such a trade will be necessary. So I don't view Site B as definitively, significantly better even if the hidden tiles aren't any better than what we can see.

[Dominae]

Dominae, where do you get that the blue circle would be the same place regardless of whether or not there is anything better than what we seem to see now in the three hidden tiles? In my experiments, adjusting NYE's latest scenario to match my speculations of what the hidden tiles are based on what we can see of their fringes, it wasn't in the same place.

No specific reason, just intuition. I think that the algorithm weighs Forest quite heavily, but I could be wrong. I must have misread your post, sorry!

As for the comparison of the two sites, Site B only offers a near-term advantage if we mine more than one hill. In contrast, Site C's forested plains hill provides a tiny bit better gold efficiency if we trade food for production before Site B has enough mines to gain the advantage. If we want scouts out reasonably early, and especially if we want more than one, such a trade will be necessary. So I don't view Site B as definitively, significantly better even if the hidden tiles aren't any better than what we can see.

I'm advocating the long-term perspective. I think site B is superior enough (barring hidden resources) to justify losing minor early-game efficiencies. This is our capital we're talking about, and in Civ4, capitals matter!

[nbarclay]

The long-term difference between the two flood plains sites really isn't all that big. Site C trades a plains hill for a desert hill, costing one hammer before modifiers. Then, sometime in the far future when Site C grows to a point where it's stuck with nothing left to work but the desert and tundra, Site B has a couple more tiles with higher values available. Those differences have only a very tiny chance of making a significant difference in the course of the game - especially since other cities could work the tiles and lose only whatever bonus the capital is providing at the time. In contrast, even if the chance of finding something nice in the fogged-over tiles is small, the chance of its making a significant difference in the course of the game if we do settle Site C and find something nice is pretty significant. So overall, I think Site C (the blue circle flood plains) offers a greater chance of providing us with a large enough advantage to make a meaningful difference.

{Edit: to put it more simply, even if Dominae is right and Site B has a higher probability of being the better site, I think Site C has a better chance of being enough better a site to make a meaningful difference.}

Of course the human factor could throw a huge monkey wrench into what the odds would normally be. Whoever set up the game could have deliberately put us in what would be the best starting location around, or could have deliberately put us in a starting position that would not be best. But that's a guessing game I'd rather not play unless someone is in a position to offer a pretty high level of certainty regarding what whoever set up the map would have done.

[Dominae]

:crosses fingers:

[Dominae]

All right, I've been toying with a few openings. As far as I can tell, Blakes "Mine the Sheep" special looks the most promising, at least for a Flood Plains start.

Here's my complete log: please try to improve on it, or show why your sequence is better on the same date. I stopped at 1440BC right after I founded our second expansion.

Stats:

3 Cities
8 Population

2 Workers
2 Warriors
2 Scouts
1 Skirmisher

All visible Flood Plains improved with Cottages
2 Mines

Up to Writing, Priesthood and Bronze Working, five turns from Monarchy
Still without Fishing, Animal Husbandry, Meditation or Masonry

What I especially like about this start is that, with the two Mines available, we have the flexibility of quickly producing some extra units (or a building) if the need arises, without having to completely rely on poprushing. As you can see, my sim is slightly light on the military units, but that can be fixed almost instantly with the setup.

Given the Forest to the South there are no Health issues, but Happiness could be a factor if we fail to get a Religion. What I did in the sim was poprush Settlers when at 10/100 Shields for 3 pop. This requires size 6 which is our first unhappy if we do no have a Religion, or we if already poprushed within the last 10 turns. Making it any tighter would not be fatal, but certainly less efficient Commerce-wise.

Kudos to Blake for figuring it out!

Any here it is:



3960BC: Settler 3, found Timbuktu, emphasize Food, begin Warrior, research Hunting
3760BC: Borders expand
3720BC: Discover Hunting, begin Mysticism
3640BC: Timbuktu size 2
3480BC: Discover Mysticism, begin Polytheism
3400BC: Timbuktu size 3
3360BC: Complete Warrior, start Worker
3080BC: Discover Polytheism, convert to Hinduism, begin Agriculture
2960BC: Complete Worker, start Scout, Worker 57 onto Sheep Plains Hill
2920BC: Worker builds Mine
2840BC: Discover Agriculture, begin Pottery
2800BC: Timbuktu size 4, Worker finishes Mine
2760BC: Worker moves 2 onto River Plains Hill
2720BC: Worker build Mine
2600BC: Timbuktu size 5, Worker finishes Mine, Timbuktu works both Mines and three Flood Plains
2560BC: Discover Pottery, begin Bronze Working, Complete Scout, begin Granary, Worker moves 4 and builds Cottage
2400BC: Worker finishes Cottage
2360BC: Worker moves 2, builds Cottage
2320BC: Timbukta works Sheep Mine and four Flood Plains
2280BC: Complete Granary, start Scout, Timbuktu works five Flood Plains
2200BC: Timbuktu size 6 (1 icky/yucky), Worker finishes Cottage, Timbuktu switches from Scout to Settler working both Mines
2160BC: Discover Bronze Working, begin Archery, adopt Slavery Worker moves 6, builds Cottage, poprush Settler (11/100 - 3 pop)
2120BC: Complete Settler, continue Scout with overflow, Settler moves 44
2080BC: Complete Scout, start Barracks, Settler 7, found Djenne (shares Flood Plains with Timbuktu)
2040BC: Timbuktu size 4
2000BC: Discover Archery, begin Writing, Worker finishes Cottage, Timbuktu switches to Worker working both Mines
1960BC: Worker moves 9, builds Cottage, poprush Worker in Timbuktu (10/60 - 2 pop)
1920BC: Complete Worker, start Skirmisher (18/25 with overflow), Worker2 joins Worker1 on Cottage
1880BC: Workers finish Cottage4
1840BC: Worker move 6 begin Cottage5
1760BC: Discover Writing, begin Priesthood, Timbuktu size 4, Djenne size 2
1720BC: Workers keep on building Cottages...
1680BC: Timbuktu size 5, steals Flood Plains from Djenne, Djenne works Sheep Mine
1640BC: Discover Priesthood, begin Monarchy
1600BC: Complete Skirmisher, start Settler working both Mines (frees up Flood Plain for Djenne)
1560BC: Switch from Settler to Barracks working all Flood Plains
1520BC: Timbuktu size 6 (1 angry, 1 yucky), poprush Settler (11/100 - 3 pop), rearrange laborers once again
1480BC: Complete Settler, continue Barracks with overflow, Settler moves 99, Djenne size 3
1440BC: Timbuktu size 4, steals Flood Plain from Djenne, Djenne works Sheep Mine, found Kumbi Saleh