playing vs. random civs - bugged? (all the time same civs)

[Nicol.Bolas]

in civ3, there was a bug that if you picked "random opponents", it wasn't random
it wa always the same mesoamerican civs (which were also the most annoying)


now it seems to be happening all over again. I ussually play standard - all random. (6-7 opponents IIRC)
and considering there is 30+ civs, I really seem to be seeing the same civs over and over :/

so is there a similar problem in civ4, did someone else experience this?

these civs seem to be in every game:
Greece, zulu, celts, native americans, holy roman empire, byzantium, dutch

......and i also see them in public games, but that might be intentional by gamehost

well.... i think I really should be seeing more variety among the civs.
among 30+ games among 3 difficulties, there are still some civs I never saw, and some that I saw maybe once...... os opposed to the above ones, that are in almost every game

 

[TheMulattoMaker]

I see the same leaders/civs over and over if I hit the map regen button, is this what's going on with you?

 

[Nicol.Bolas]

no I don't use map regeneration.

I just use "play now" and use the map i am given

 

[Ataxerxes]

IIRC the Civilizations are supposed to be selected randomly and then the leader. Therefore, you'll see leaders that are the only leader of their Civ (HRE, Aztec, Zulu, etc.) more often than other leaders. However, in some stretches it does seem like you see the same leaders over and over again. For a time mine was Ragnar.

 

[TheMeInTeam]

With 34 civs and random 7 a repeat from one game to the next is actually pretty likely...

 

[Wheatandoats]

When I had Warlords, I felt like I saw the Warlord civilizations and leaders most often..When I got Beyond The Sword....I felt I saw those leaders and civs the most....

 

[totalepicness]

i guess the best thing to do would go to custom game and choose each leader yourself.

it does seem like kind of a hassle

 

[wolfblue]

i had not thought about it but your right. I tend to see some leaders much more than others... I see isabella all the time... i cant remember seeing Louis, Pacal, or Joao more than once if at all in the last 20 games or so.

 

[totalepicness]

ive never seen greece in a single one of my games.

 

[Nicol.Bolas]

hmm yeah, i gues handpicking is allright....

after all, it doesnt spoil anything...... the nearest civ gets rushed, the other conquered or space raced...

as long as you dont keep handpicking the 7 weakest civs, but go thru them in order, it doesnt affect gameplay really.....

 

[Kaosprophet]

in civ3, there was a bug that if you picked "random opponents", it wasn't random
it wa always the same mesoamerican civs (which were also the most annoying)

I never noticed that. I did notice a tendency towards 'cultural-blocking' according to my civ. IIRC there was an option to turn that off as well, but it never seemed to make a difference.

Haven't done any testing, but I wouldn't be surprised if there's a similar culture-block skewing in Civ4.

these civs seem to be in every game:
Greece, zulu, celts, native americans, holy roman empire, byzantium, dutch

......and i also see them in public games, but that might be intentional by gamehost

I never noticed that myself, in random games.
Public games... there's a couple leaders that are 'popular' opposition so you'll see them more frequently than others.

 

[s.bernbaum]

in civ3, there was a bug that if you picked "random opponents", it wasn't random

now it seems to be happening all over again. I ussually play standard - all random. (6-7 opponents IIRC)

so is there a similar problem in civ4, did someone else experience this?

these civs seem to be in every game:
Greece, zulu, celts, native americans, holy roman empire, byzantium, dutch

(I trimmed the quote down to what I'm replying to.)

I think that it is entirely random. I always play random leaders. There do seem to be some civs that I see more often than others, chiefly Qin and Huana Caypac. However, that is probably an artifact of the random process. Since something that is entirely random will seem to have a pattern, if you don't consider a large enough number of cases. My current game, which is a large fractal map with 9 civs, has none of the civs that you listed, although it does have both Qin and HC. I have never yet seen the Celts nor Byzantium in any of my games. I assume, as more games accumulate, they will show up.

 

[Dhoomstriker]

I tend to see some leaders much more than others... I see isabella all the time... i cant remember seeing Louis

Well, it is random Civs, not random Leaders, right?

So, you'd only see Louis if the French Civ got picked AND he was the 1 of 3 Leaders that only then got picked to lead the French Civ. Isabella has no competition so she will always win when the Spanish Civ is randomly selected.


Then there is the possibility of observer bias. When you're used to seeing Leaders, you don't make note of them (i.e. if you see Hatty once yet again, you'll just ignore her), but if you see a new BtS AI, it will seem strange and new to you (or at least relatively newer, now that you've played a lot of BtS games) and thus you'll be more likely to remember seeing the number of times that you've seen the new AI. Seeing Hatty "once yet again" will just get clumped together in your brain with the countless other number of times that you have seen her, so you won't make note of that fact.


Then, there's the point that TMIT made, where seeing a repeat is actually probably relatively statistically likely. How would the statistics go on that one? "34 choose 7" or something like that? Maybe someone with a good statistical background can jump in here.

However, even if we take a look at it not-too-mathematically, you'd think that it would take some pretty long odds to come up with 4 games in a row with zero repeats across any of the Civs (selecting 28 unique Civs out of 34 Civs).


Who's to say, though, that it really is random, unless you look at the code? Maybe there are some non-random elements to it. So what? Have you, after a decent number of games, had a chance to play against every other Civ? Probably. So what's wrong with seeing some repeats? Maybe you'll actually be better prepared that way (hey, if Monte backstabbed you last game, he's probably the type of AI that might do so again--and lo and behold, he is! ).

 

[ruff_hi]

I did put together an excel based random leader picker (see my downloadable content). You can restrict or force leaders based on traits and / or techs too.

 

[wolfblue]

Well, it is random Civs, not random Leaders, right?

i should have pointed out i always play custom game, random civ - random leader - unrestricted leaders (I pick my own civ and leader.)

 

[Hortulanus]

I always play with six random opponents, and I've been lax in cleaning up my saved games directory, so I decided to catalog which leaders have appeared in my games. These are emperor level games, epic speed, standard size maps, mostly fractal. I don't play any maps, such as Earth, that predetermine the opponents.

There have been 196 randomly generated leaders in 41 games. (Some games were aborted before all six opponents were met). The mean of the sample was 3.8 occurrences of a leader with a standard deviation of 2.3 occurrences. The 95% confidence interval (mean plus or minus two standard deviations) ranged from -0.7 to 8.3 occurrences.

Only three leaders fell outside the 95% confidence interval. Justinian and Mansa appeared 9 times and Joao 10 times. In addition there were two no-shows, Stalin and Victoria, but an occurrence of zero is not unexpected in this sample.

Unfortunately for the usefulness of my study, I failed to account for the fact that the leaders aren't a random sample, since some civilizations have more than one leader. It's noteworthy that Justinian, Mansa and Joao are the sole leaders of their civilizations. The no-shows, Stalin and Victoria, share their civilizations with two other leaders, and those leaders were well-represented.

Also, these games include my first attempts at emperor level, and when I move up a level, I choose a financial leader for myself, usually Hannibal, Pacal or Victoria. Those leaders should have been underrepresented. Victoria was, but Hannibal occurred four times as an opponent and Pacal six times. Both are a little above the mean, which is not what I would have predicted. I don't know what to make of this.

Overall, I can't conclude that there is obvious bias towards any particular leader, but I'm not convinced it's entirely random either. I don't think the game favors cultural neighbors. (Pacal and Huayna occurred six times each, but Monty only twice). I wonder if it's programed to generate a mix of different peace-weights.

Finally, I did learn that my memory is sometimes a poor indicator of who I have played against. I have no recollection of Sury as an opponent, but he's been in four games. I would have sworn that Alex has been a plague to me. He's only been in three.

Hortulanus