3920 - Kgb

[Aginor]

Edit: I'm now using this space to keep track of what we can be certain of. The process of figuring all this stuff out, as well as some suppositions I cannot support by pure deductive logic, are in the full thread. Thanks are due to Mudhut for figuring out and explaining to me the rival average algorithm, and to Blake for determining how tech impacts army size. Thanks are due to both Mudhut and Blake for correcting logical inconsistencies along the way.

Explanations for where some of this stuff is coming from are in the thread. Reading the thread, I suspect, is like watching someone make sausage - it's an ugly process, but the final product tastes pretty darned good.

Ok, looks like everyone but the Horde has settled. Let's take stock of what we know at this point:



Rival food, based on in-game allocation of Workers and calculations from rival averages:
3960: 5,4,4,0,0,0
3920: 5,5,5,5,4,0

Rival hammers:
3960: 2,2,1,0,0,0
3920: 2,2,2,1,1,0

Rival Commerce:
3960: 10,10,9,0,0,0
3920: 12,11,10,10,10,0

7 tiles within rival city radii as of 3920 are water. At least one of those tiles belongs to civs that founded in 4000, and the one tile that we know about is one space away from that civ's capital.

No one has popped military techs, a Scout or a Warrior at this time.

Two of our rivals founded on Plains Hills and are working 3/0/x tiles. This can be deduced from the impossibility of an unimproved 3/1/x tile and the food/hammer distribution of our rivals. One of these rivals on a Plains Hill MUST be AC or Sara.

At least one of our Creative rivals moved from working a 2 food to a 3 food tile when their borders expanded. Both of those rivals may have, in which case either Sara or AC only has 4 food.

Two rivals currently research faster than us. This may throw a bit of a wrench in the Buddhism works.

We will gain additional information at the following times:

3880 BC: A rival whose borders expanded that was auto-switched by the AI governor to a 3 food tile may not wish to work this tile for some reason, in which case that rival's food (and perhaps commerce) production will decrease and its hammer production will increase. This is unlikely, but it could happen.

3760 BC: The civ that has had 5 food since 4000 BC will grow to size 2, and its food, hammer and commerce totals will be adjusted accordingly. Sara and AC will experience a border expansion, and their production totals may alter as well. We can also use this border expansion to try and figure out whether at least one of Sara or AC is coastal.

3720 BC: Any rival that founded in 4000 BC that went from 4 food to 5 food in the border expansion of 3920 BC will grow to size 2 on this turn. This should answer the mystery of who is currently producing 5 food, and who is producing only 4, if the 3760 BC border expansion cannot answer this question.

3680 BC: An AC/Sara presently working 5 food will grow to size 2 on this turn, and food/hammer/commerce totals will change. Our own Mysticism research should finish this turn, so this is the deadline for understanding what our opponents are up to and determining what our second tech should be.

3640 BC: Any rival that founded in 4000 BC that worked 4 food tiles from 4000 BC on will grow on this turn, and totals will adjust.

These dates will have to be adjusted for whenever the Horde founds, of course. Wonder why they're still wandering the earth - hard to conceive of a start THAT bad. Perhaps their Scout found the Promised Land? (Flood plains valley, 3-4x Gold tiles sort of situation?) If you come up with an explanation for that one, please let me know!

[Hot Mustard]

Re: 3920 - Kgb

Originally posted by Aginor
Looking at the demographics, it appears TWO civs have us out-commerced. Last turn, the max anyone had was exactly what we had (10). Now Liz AND Sara apparently have 12.

This may throw a bit of a wrench in the Buddhism works.

Ugh, yeah that could blow.

[notyoueither]

No, it won't.

How many teams have mining and fp's and therefore do not need techs to develop their starts?

We'll be fine, and if we do Hunting first along the way, we won't waste much before we find out that another team decided to throw away their start.

[Mudhut]

Anyone smarter than me who can explain how the health calculation works?

We have life expectancy of 70, max is 85 -> does this tell us anything about others being on flood plain starts (or not?)

I would have thought it likely that at least one team out of seven would have a coast/fishing start -> is this likely to be one of the two teams with 12 commerce?

[Blake]

Hmmmm.... interesting....
There are two ways to get +12 commerce, +1 commerce city tile (default) and +3 commerce tile (financial + lake, oasis, or comm special on river). The other way is +2 city tile and +2 worked tile - this is founding on a commerce special on a river.
I also believe it is possible that there's a +12 and a +11 out there, that would still put us at rank 3, am I missing something for why they'd have to be two others with +12?

I don't think even the +12 team(s) can give themselves an assured chance of getting the religion of choice, since they founded a turn late. I guess it's remotely possible a turn1 founder could be the guilty party, and they chose to work a 0 commerce tile on the first turn to be deceptive or something. Heck, anything is possible.

I support continuing straight to Meditation, with fingers crossed. We are in a pretty research-rich position so even if we don't get meditation, we wont really suffer. By the time the sheep is mined, agriculture arrives, by the time we've trained a couple of skirmishers (or a 2nd hunter and a skirmisher), pottery arrives. Our city isn't lacking things to build at any point, nor is our worker.

If we fail to get Buddhism, we can make CoL a high priority.

[Metaliturtle]

I agree with Blake, he's my kind o Bloke!

[Mudhut]

Hmmm. How does the math work for commerce calculations? Because I kind of feel like we should be able to work this out.

Before we settled (in 3960 bc) stats were:
Rival Max: 10
Rival Worst: 0
Rival Average: 4
Settled: 3 teams (Vox/Bunch/George von F)

So an average of six teams is 4, knowing that three of them are at 0. Since we know that the other three teams have to be at LEAST 9 commerce (eight from Palace, plus one from city square) then you CANNOT get an average of 4 with actual math.
The minimum possible data set is [10,9,9,0,0,0] which leaves us with 4.667 as an average, which would round up to 5 using normal math. This tells us they must be using integer math (not surprising, all of Civ does) and are simply discarding the remainder. We can say, however, that the dataset is NOT [10,10,10,0,0,0] because the average there would be exactly 5.

Conclusion: Dataset is either [10,10,9,0,0,0] or [10,9,9,0,0,0]

At 3920 BC, after six teams have settled, we get this info:
Rival Max: 12
Rival Worst: 0
Rival Average: 8
Settled: 6 teams (all bar horde)
Our score: 10
Our rank: 3

If we assume that the first three teams haven't changed their tile working (can't imagine why they would!) then their numbers should be the same. Now, the integer math leaves open the following datasets:
1 - [12,12,10,10,9,0] ave=8.83. Worst case scenario for us (ie maximum commerce out there in the world)
2 - [12,12,10,9,9,0] ave = 8.67.
3 - [12,11,10,10,9,0] ave = 8.67
4 - [12,11,10,9,9,0] ave = 8.5 (best case scenario, minimum commerce out there in the world).

We need another data point -> but I'm not sure if one is coming. Seems to me whatever the horde end up with, the rival average will be 10 (using integer math).

Anybody see any other way of cracking it?

[Mudhut]

Of course! One way to get more insight is to simply go in to EOTS and start working a non-commerce tile to see what happens to our ranking.

In 3960, if commerce is 10 then our rank is 1. If our commerce is 9, then our rank is 3.

In 3920, if our commerce is 10 then our rank is 3. If our commerce is 9 then our rank falls to 6.

*How are ties ranked?*

If we assume that teams on the same score both get the upper rank (which would seem to be the "normal" way to do it to me) then this means that:
At 3960, the rival dataset is [10,10,9,0,0,0]
At 3920, the rival dataset is [12,11,10,10,10,0]
(note that it can't be [12,12,10,10,10,0] because the rival average would then be 9)

Now, this doesn't seem to make sense to me -> why would the team that was 9 commerce in 3960 change their tile allocation to be 10 commerce in 3920? What am I missing?

Could the ranking be based on turn order somehow?

[Blake]

Of course! One way to get more insight is to simply go in to EOTS and start working a non-commerce tile to see what happens to our ranking.

Ummm I'm pretty sure the demographics are generated between turns to avoid exploits like that.

[Mudhut]

Oops. Hope I didn't do anything wrong !?

But changing our worked tiles (to reduce our commerce by one) definitely seems to change our ranking. So it must be calculating on the fly, yes?

If we are allowed, can someone else check and see if the same thing happens on their demographics screen?

[notyoueither]

You've done nothing wrong.

The guideline is to do nothing with the save that is not reversible without reloading, and which is not what the team actually did in game.

Do not move units where they were not moved.

Do not disband units.

Adjusting workforce to measure demographics is just fine.

I think you are right. I have been able to examine demographics in PBEMs.

[vulture]

Originally posted by Mudhut
Now, this doesn't seem to make sense to me -> why would the team that was 9 commerce in 3960 change their tile allocation to be 10 commerce in 3920? What am I missing?

Could the ranking be based on turn order somehow?

It's possible that they are optimising for the early production of a warrior or scout, whilst aiming for high research. They want to maximise research as an early game goal, but realise that working e.g. a forest for 1 turn will give them a scout 1 turn earlier without affecting how soon they get their first tech. If they are like us, they may have simmed out their research strategy in depth and realised they get their religion tech at the same time despite lower GNP for one turn, and so go for the benefits of the extra early scouting unit turning up 1 turn sooner. One extra headstart on scouting withouth hurting your time to get a religion tech is worth a little micromanagement.

They could have changed from 9 to 11 or 12 commerce as well.

[vulture]

Question: our land area on that demographics screen is 9000 sq km, which since we have 9 tiles within our borders, works out at 1000 sq km per tile (tricky maths). However, our lead rivals in this regard have 21000 sq km. Do they really have 21 tiles? Looking at the average score, and assuming values are either 21k or 9k, it isn't possible to get the average value there. If we assume that a team that hasn't settled counts as 1k (is this true? can't check since I'm at work, and I never paid enough attention to demographics before settling to notice...) then with us in 4th with 9k, the data are [21, 21, 21, 9, 9, 1, 1] including us, giving the average of 12,333 sq km without us (which is what the demo data has).

All of which looks very strange, and makes me doubt the 'land area = no. of tiles * 1000 sq km' value. (Okay, I've ignored the possibility of coastal starts here, so maybe 5 of the 6 teams have settled rather than 4, but there are still 3 civs out there with very large land area scores for having completed 3 turns of the game. Does their higher culture count the extra tiles to make the 21 even before the border expands?)

Obviously, from the pop data, 5 teams have actually settled, not 4. So, any clues as to what is going on with this?

[Aginor]

Originally posted by Blake
Hmmmm.... interesting....
There are two ways to get +12 commerce, +1 commerce city tile (default) and +3 commerce tile (financial + lake, oasis, or comm special on river). The other way is +2 city tile and +2 worked tile - this is founding on a commerce special on a river.
I also believe it is possible that there's a +12 and a +11 out there, that would still put us at rank 3, am I missing something for why they'd have to be two others with +12?

The two teams that *didn't* found 1st turn are Sara and Liz. Both are Financial. +2 tiles are impossible for Financial civs due to the +1 bonus. QED.

It's possible, as you noted, that one of the non-Financial civs switched worked tiles after turn 1; however, I consider this an extremely low probability in a demogame, particularly where the first turn has been debated to death by all teams already.

So my read on the situation is that Liz and Sara both have 12.

I don't think even the +12 team(s) can give themselves an assured chance of getting the religion of choice, since they founded a turn late. I guess it's remotely possible a turn1 founder could be the guilty party, and they chose to work a 0 commerce tile on the first turn to be deceptive or something. Heck, anything is possible.

This is a relative matter. Doesn't matter whether either of 'em can guarantee religion of choice given that it's a race between the two of them. What matters is that they can beat us to the punch if they so desire. Further, those are the two civs with the most incentive to get there first (as the other 4 are Creative).

I'm not advocating a total bailout on the religion strat, but we're in a position where any of 5 civs could outrace us to Meditation if they started Myst->Meditation. We've only got position on the Horde. Further, there's the strong possibility that both of our main competitors for a religion are going for it, given that they apparently moved specifically to get 12 commerce starts.

We may need to consider more complex options than a pure Meditation rush given the circumstances. Since we're out of position on everyone, we're in a situation where we have to react to a blind guessing game as to whether or not Sara/Liz is gunning for Myst, and if so whether either/both are going to go for Meditation or Poly.

Picking up Hunting is clearly out, as our only hope IMO for Buddhism or Hinduism is to get lucky and pick the religion the other competitor(s) don't go for and get it before research can be swapped.

[Aginor]

Originally posted by notyoueither
No, it won't.

How many teams have mining and fp's and therefore do not need techs to develop their starts?

Unfortunately, our competitors are Qin and Liz, assuming that teams that worked tiles on turn 1 continued to work them on turn 2. Qin's Industrious/Financial with Ag and Mining. Liz is Philo/Financial with Fishing and Mining. Given that it appears Liz has 12 commerce, there's a STRONG possibility that Liz hit the land lotto and caught a coastal start. I don't know how Qin has 12 commerce, but my best guess is he's got an Oasis. Possible that it's a commerce special on a river as well, as Blake noted.

*IF* Liz has a coastal start, she probably has a special to toss a work boat at and can postpone Ag and Pottery. Qin doesn't need tech to develop rudimentary infrastructure out of any start; he's got Ag and Mining.

Also, these are the civs that are likely to be thinking the same thing we're thinking - we're not Creative, and we can't chop culture.

I'd say our probable competition doesn't need tech to get off the ground floor, and can pursue what their little hearts desire in the short run. The question is what they intend to do with their starts. They could be thinking:

1) Religion
2) Bronze out of a food-heavy start
3) River and Cottage spam

I ask: which is more likely? I'd say we've got at least one player in the religion game with us, and probably 2.