The Immortal Challenge 1: Apocalypto

Unconquered Sun said:
You get +20% research bonus for every prerequisite you have, required and optional. There was (is?) an exploit to that, not sure if it had been addressed in Warlords.
Click to expand...

Suppose that you have researched Preisthood and AH. Does that mean you have a 40% research bonus on Writing? And if, in addition, you have Pottery, does this meam the Writing reserch bonus is 60%?

Just wondering if the bonuses accumulate in this way.
 
Snaaty said:
When you work the oasis right from the start (I still tend to settling on the spot), it will take:

12 turns for poly
Click to expand...

That's 100 * 130% (map size) * 125% (difficulty) = 162 / 14 = 12? With our hypothetical +0 research opponent researching at 12 beakers per turn hits on turn 11. For Meditation, +0 hits turn 9, where we hit on turn 10. So I guess we only get the religion if no one else wants it. OK, that makes Mysticism a bit less fun as a starting tech.
 
Yes, the bonuses are cumulative and your example is correct.
 
Godel said:
Suppose that you have researched Preisthood and AH. Does that mean you have a 40% research bonus on Writing? And if, in addition, you have Pottery, does this meam the Writing reserch bonus is 60%?

Just wondering if the bonuses accumulate in this way.
Click to expand...

Pretty much - it's a fairly easy test to run in world builder.
 
I ran a test to gauge the probabilty of founding Buddhism/Hinduism.

Settings: Immortal, standard size, normal speed, continents, random opponents, playing Monte

I allowed city governor to manage tiles since concensus is that if an opponent researches Polytheism off the bat he/she will beat us too it most times anyway.

There were 20 trials.

Results below:
Spoiler :

Out of 20 trials:
* 7 times I would have been beaten to Buddhism
* twice I was beaten to Hundiusm
* 3 times a civ starting without mysticism founded Buddhism before I researched Polytheism. They must have researched Mysticism and got Meditation in a hut (or possibly, in Liz's case, researched both by working a lake tile). On these occasions I would have beat them to Buddhism had I researched it first since you can research Meditation in 10 turns (no commerce tile worked).

20 trials is not really enough to give a convincing estimate for the probability of founding Hindiusm, but I think the results strongly advise that researching Polytheism first is best.

Full results:

Trial; Bud Founded; Hin Founded; Opponents starting with Myst
-------------------------------------------------------------
1; x; x; Isa, Wang, Gan
2; Sal; x; Sal, Wang
3; Gan; x; Gan, Sal, Bre
4; Liz (after 10 turns); x; none
5; x; x; HC, Sal
6; Aso; x; Aso, Sal
7; x; x; Sal, HC
8; Aso; x; Aso, Sal
9; Sal; Isa; Isa, Sal
10; x; x; Sal
11; x; x; Gan
12; x; x; Wang, Isa, Bre
13; Wang; x; Sal, Wang
14; x; x; none
15; x; x; Bre, Aso
16; Bre; x; Bre, Sal
17; Hat (13th turn); x; Isa
18; x; x; HC, Bre
19; x; Gan; Gan
20; Ragn (after 10 turns); x; Wang

Note: x means that no one founded the religion before I research Polytheism


What was scary was having a look in WorldBuilder after 13 turns. For example Ragnar had 5 farms (including 2 FP farms!) by this stage. It is not even enough time for us to build a worker!
 
Unconquered Sun said:
There's ~25% additional bonus when defending cities vs barbs. Warriors can hold the capital no matter what. However, they can't protect the irrigation or the mine.
Click to expand...

Indeed,

If you fail with your first atempt for copper/horses you have to start a good force of warriors, you can't let yourself be pillaged on this level. Furthermore you need more than one warrior to guide the route to the second city because otherwise the settler will be delayed or might even die. Building these warriors is not a huge problem but if you get resources connected before the barbarian border rush you can go for much more useful axes/chariots meaning you get to barracks/libraries quicker.

On emperor i could almost always found hinduism first and still i found out that the problem with delaying key techs was so big that i gave up on it. On immortal the chances of founding a religion are worse and the problems with barbarians bigger. Also having this religion doesn't mean you have +1 happiness everywhere automatically, meditation or monotheism has to researched, missionaries and in the case of meditation monasteries have to be build. There are useful other things that can be researched and build instead of these.

If there has to be an attempt at religion is should be poly i think for Literature.
 
Godel said:
I ran a test to gauge the probabilty of founding Buddhism/Hinduism.
Click to expand...

I guess that blows my whole 'wait for aelf' thing. Meh, I guess there's no harm in this kind of parallel test.

I ran a similar set of tests to you, but my results were slightly different.

Spoiler :

I played eighteen different starts with aelf's settings and various combinations of civs, rerolling on each occasion if I didn't start with a tile with at least one commerce available.

I used the best commerce tile in each case (didn't get any oases though), and nailed Hinduism on every occasion, even on the six starts in which all 6 opponents started with Mysticism.

I could've got Buddhism instead in about 50% of cases, though never with all six myst-civs on the map.

I also noticed that the AI has a tendency to switch between techs on occasion (changed priorities or just easily confused?), and to prioritise worker techs over religion when there are resources to be improved.


And yes, the AI's speed of development is quite terrifying. :eek:
 
Winston Hughes said:
And yes, the AI's speed of development is quite terrifying. :eek:
Click to expand...
Which is just yet more reason not to despair when you find yourself behind early in the game. With the AI's bonuses at the higher levels, it's normal.
 
Interesting results WH. I don't see the problem in doing this kind of test, particularly since there is so much discussion about early religion at present. It can only help the civ community and help aelf make a more informed decision.

Still, when I read your post I thought I'd put my results in a spoiler just in case.

The more tests/trials that people do under similar settings, the better (Central Limit Theorem and all that.)
 
Godel said:
Interesting results WH. I don't see the problem in doing this kind of test, particularly since there is so much discussion about early religion at present. It can only help the civ community and help aelf make a more informed decision.
Click to expand...

My only concern was that the results might get discussed before aelf decided whether he wanted to know in advance. But I guess I should have more faith in the readers of this thread...

I'm glad you did a similar test; more evidence has got to be a good thing (though I have no idea what Central Limit Theorem is :dunno: )
 
Godel, I think one of your assumptions was backwards - if a civ without Mysticism scores an religion startling early, that ought to mean that it popped Mysticism from a hut, then researched the religion. (bGoodyTech is zero for both of the religions, but 1 for Mysticism - see CIV4TechInfos.xml).
 
flamingzaroc121 said:
but in warlords, poly is favored because it unlocks 2 wonders so be careful of that
Click to expand...

Yes and no....

A month ago I modified in some code to report the average tech value when the AI started researching, because trying to work out the math seemed an inefficient use of time. I then ran a test where I put each civ on a football field (green, flat, no resources), and wrote the tech scores of the opening turn out to the log file.

What that test told me is that (1) almost all of the civs think Polytheism is the better tech, but (2) 50% of the time they will research Meditation first - the tech selection algorithm weighs the tech choices by research time, and since Meditation requires less time to research, it normally ends up ahead.

The extra wonder definitely makes Polytheism more popular than it used to be.

Also, keep in mind that the dicerolls are pretty significant in scale - the raw scores for the techs fall into the 3000-5500 range, and the first die roll is +/- 1000
 
as you probably know some leaders like wonders because they have a shiny factor to them and if aelf has a bunch of those they will probabaly go for Poly over Meditation
 
flamingzaroc121 said:
as you probably know some leaders like wonders because they have a shiny factor to them and if aelf has a bunch of those they will probabaly go for Poly over Meditation
Click to expand...
Quite true. One of the few good things about having Montezuma around is that he is very unlikely to beat you to any wonders.
 
Sisiutil said:
Is he the only one, or are there others? VoU, care to wade knee-deep into the XML yet again for us? :cool:
Click to expand...

Louis XIV,50
Ramesses II,50
Asoka,40
Augustus Caesar,40
Cyrus,40
Hatshepsut,40
Huayna Capac,40
Bismarck,30
Catherine,30
Kublai Khan,30
Peter,30
Qin Shi Huang,30
Stalin,30
Victoria,30
Alexander,20
Elizabeth,20
Frederick,20
Isabella,20
Julius Caesar,20
Mansa Musa,20
Mehmed II,20
Roosevelt,20
Tokugawa,20
Wang Kon,20
Washington,20
Brennus,10
Churchill,10
Gandhi,10
Genghis Khan,10
Hannibal,10
Mao Zedong,10
Napoleon,10
Ragnar,10
Saladin,10
Shaka,10
Montezuma,0
Barbarian,0

The value indicates the magnitude of the random number used to bias in favor of wonders - so the average impact of Ooh! Shiny! is the number above divided by two... no, I lie, subtract one half from the quotient. To give a sense of scale, the "Let's not build the same buildings the same way every time" random number has an average value of 12....

For those playing at home: Ooh Shiny is LeaderHeadInfo.iWonderConstructRand; the die rolls are "Wonder Construction Rand" and "AI Best Building".
 
You must log in or register to reply here.

Similar threads

AcaMetis
Nobles' Club 351: Sitting Bull of the Native Americans
Replies
19
Views
1K
Conquistador 63
Conquistador 63
AcaMetis
Nobles' Club 350: Hatshepsut of Egypt
2
Replies
27
Views
2K
Gumbolt
Gumbolt
Implo
  • Poll
Deity Cathrine Challenge!
2 3
Replies
48
Views
3K
CarpoolKaraoke
C
AcaMetis
Nobles' Club 347: Boudica of the Celts
2
Replies
26
Views
2K
pigswill
pigswill
AcaMetis
Nobles' Club 352: Joao II of Portugal
Replies
16
Views
812
antimony
A
Top Bottom