Random civ truly random?

[tsf4]

Are random civs truly random or are they weighted? I’d like to play with them truly randomized, but it’s painstaking time to roll a dice for each slot and then finding what number civ it is.

any info or advice?

 

[Snarf054]

I always play random everything possible. It may just be me but it does seem weighted. You tend to see a lot of the same civs and it seems to be related to your game settings. If I start 2 games in a row with the exact same settings quite often I will get the same civ. So I try to vary the number of city states every game, change the map up occasionally, ect.

 

[Pfeffersack]

Are random civs truly random or are they weighted? I’d like to play with them truly randomized, but it’s painstaking time to roll a dice for each slot and then finding what number civ it is.

any info or advice?

I don't feel that there is any weighting done. Random is random. My d60 sees still occasional use though, if I want to exclude certain civs.

 

[Sostratus]

In advanced settings you can type in a game seed. This is distinct from the Map seed. Automatically generated seed pairs only differ by 1, but you are free to type in any number you want. This controls the civs and CS in game. You may end up in a different location on the same map for differing game seeds!

I cannot recall off my head if gameseed affects the player’s random choice, though. It definitely controls everyone else.

 

[tsf4]

Would the seed always be different? Or would it be say same group of civs?

 

[Sostratus]

Would the seed always be different? Or would it be say same group of civs?

If you put the same seed in it’s the same group of civs and CS afaik. On the same map it’s always the same PROVIDED you have the same number of civs / number of CS obviously. I believe it’s always the same group controlling for that.

 

[UWHabs]

I always find it weird when I do random and get the same civ twice in a row, but there's still like a 1/50 shot of that happening every time, so the odds of that coming up here or there is pretty high.

 

[Aurelesk]

I always find it weird when I do random and get the same civ twice in a row, but there's still like a 1/50 shot of that happening every time, so the odds of that coming up here or there is pretty high.

If you play Duel, this is true: 1/48 chance of seing the same opponent again, so ~2%.

For other map size, it is way higher than that:

• Tiny (4): ~18%
• Small (6): ~43%
• Standard (8): ~69%
• Large (10): ~87%
• Huge (12): ~96%
• Huge max (20): ~99,9998%

(Math was 10 years ago: I hope I get the formula right).

Try to see it with an other angle: what is the probability to not see at least one civilization again?
If there is 48 civilizations, on a Tiny map, the first opponent have to choose respectively 45 civilizations out of 48 (there are 48 civilizations, but 3 of them can't be picked), the second 44 out of 47 (47 civilizations left, and 3 of them can't be picked), and the last 43 out of 46. So the probability of not seeing the same is 45/48 × 44/47 × 43/46 (~82%) so the probability of seing at least one of them again is the opposite (100%-82% → 18%). And so on for all the map size.

Starting to Standard / 8 players, it will be twice more likely to see at least one same opponent again (69%) than none (31%).

 

[tsf4]

Lol I just want no weighting!

 

[Kwami]

I would be very, very surprised if there was any weighting at all. So far, there's no evidence of weighting.

 

[Uberfrog]

It is the curse of humanity always to look for patterns in everything. So what is truly random does not appear random, and confirmation bias means you always remember that Civ who is always in your games, but forget cases when they don’t show up.

I think it doesn’t help that sometimes people interpret “random” as “different every time”... which is very much not what randomness means.

 

[tsf4]

Good point. I just aim for essentially a dice roll by the game choosing what civ to put in the game. Equal odds across the board +/- the one civ which gets the .02 better chance...