Start Bias

[KahunaGod]

Appreciate it.

I would like to post some screen shot of what I am talking about.

Anyone tell me who to do that with the .tga files I get from print screen? (sorry, not really literate in such things)

 

[PawelS]

I don't know what you exactly mean by "surrounded by desert". There are quite a lot of deserts on a typical map (the amount depends on the map settings, of course), so you'll have some desert close to you most of the time when you start a new game.

 

[lschnarch]

Appreciate it.

I would like to post some screen shot of what I am talking about.

Anyone tell me who to do that with the .tga files I get from print screen? (sorry, not really literate in such things)

The easiest would be to convert them via IrfanView (freeware, check the net for the version which fits your system and OS) into jpg.

 

[LegioCorvus]

Are you changing some other settings, like rainfall or world age? Those both also effect the presence of jungles. I've played three (maybe four?) games as Aztec and I've had Jungle every time.

Most recent game with Aztecs. (see spoiler).

Spoiler :

No jungle anywhere close. Actually didn't see jungle this entire playthrough. I already gave up on this game, as I had basically won at this point. I tried gifting India's capitol to Rome to make it interesting, but... anyway, digression.

Just used the normal menu of "Pick Civ, Pick Map, etc." not the advanced one.

I'm not upset or anything by not seeing jungle as the Aztecs. I just think there are other factors beside start bias that come into play during map creation. Because I'm not seeing it in a lot of my games.

 

[SevenSpirits]

The start bias does not guarantee you get a jungle start. It's just bias.

What it does mean is that if there's a predominantly jungle area, you'll get first pick of it.

 

[KahunaGod]

I don't know what you exactly mean by "surrounded by desert". There are quite a lot of deserts on a typical map (the amount depends on the map settings, of course), so you'll have some desert close to you most of the time when you start a new game.

Well, i mean being around 5-7 hexes deep on each side. Sometimes I am near the edge of 30+ long/wide desert. Its really getting unplayable.

 

[bite]

Honestly, I think there is a bug/bugs with the civ placement on huge maps at the moment. I think that's the problem. Would it be satisfactory to play a Large map instead? Large maps are still pretty big.

Can I suggest running the game in debug mode and generating a map a few times (with same settings), and having a look at the results? Certainly when I did this on the Earth map on huge, I kept seeing weird things like city states starting 3 tiles from civ capitals, or a bunch of city states being in a strangely straight line (though it may have been pure chance).

I have an interesting City State placement 'bug' I have 4 city states appear right next each other (in norther egypt and western saudi arabia for those playing at home) and a fith which got killed off before being able to spawn. I assume it was a placement bug, but it has resulted in some interesting gameplay as they act as a buffer in and out of africa

 

[PieceOfMind]

Appreciate it.

I would like to post some screen shot of what I am talking about.

Anyone tell me who to do that with the .tga files I get from print screen? (sorry, not really literate in such things)

There's a program called IrfanView (google it) that lets you open the files and then save them as jpeg (so you can upload them to imageshack or whatever).

Well, i mean being around 5-7 hexes deep on each side. Sometimes I am near the edge of 30+ long/wide desert. Its really getting unplayable.

As I said earlier, as far as I can tell there is a bug with huge maps at the moment. Moreover, it's unlikely the game will work as well as you want when you load up the map with more civs than recommended. This is not an excuse, but just an attempt at being realistic. Did you try my suggestion before and run the game in debug mode? Do you know how to? It really helps in judging how good a map is, with some 'practice' rolls, before you actually jump in and play one, because getting into a huge map is a big time investment - one that you wouldn't want to get half way in then discover an annoying bug.

 

[DalekDavros]

If you flip a coin 100 times, and you get heads every single time, you know what the odds are going to be that you get heads on your next flip? The exact same as the first time.

While technically true, if you're suggesting that this probability is 50%, I'd disagree. If you flip a coin 100 times and get heads every single time, it means you're flipping a coin with two heads.

 

[Maxor127]

I'm not sure if I play with Start Bias on or not, but I've found as Rome, I usually start near rivers with marble nearby.

 

[Licinia Eudoxia]

While technically true, if you're suggesting that this probability is 50%, I'd disagree. If you flip a coin 100 times and get heads every single time, it means you're flipping a coin with two heads.

No, it doesn't, but thanks for demonstrating my point.

Probability does funny things to people's heads and makes them think silly things when their brains start trying to deduce patterns and anticipate results. You can explain someone the exact mathematics of something as simple as a coin flip, but some people will still resist it on instinct, even if they know it's factually true.

It's even worse with video games and such where people can't see all the mechanical moving parts.

I have seen some of the moving parts, and I have no reason, from my experience, to believe that start bias is messed up. KahunaGod, from his experience, strongly believes that start bias is messed up and causing Rome to start near deserts.

Now, without looking at individual lines of code and spotting errors, neither of us can really know for sure (and even then, maybe not!)

It's the same as the coin flip. If I take the coin in a room and flip it 100 times and get heads. And he takes the coin into a room and flips it 100 times and gets tails, then we both meet up and flip the coin, his gut will tell him it's going to be tails, and my gut will say heads.

 

[SevenSpirits]

No, it doesn't, but thanks for demonstrating my point.

Probability does funny things to people's heads and makes them think silly things when their brains start trying to deduce patterns and anticipate results. You can explain someone the exact mathematics of something as simple as a coin flip, but some people will still resist it on instinct, even if they know it's factually true.

It's even worse with video games and such where people can't see all the mechanical moving parts.

I have seen some of the moving parts, and I have no reason, from my experience, to believe that start bias is messed up. KahunaGod, from his experience, strongly believes that start bias is messed up and causing Rome to start near deserts.

Now, without looking at individual lines of code and spotting errors, neither of us can really know for sure (and even then, maybe not!)

It's the same as the coin flip. If I take the coin in a room and flip it 100 times and get heads. And he takes the coin into a room and flips it 100 times and gets tails, then we both meet up and flip the coin, his gut will tell him it's going to be tails, and my gut will say heads.

This is actually an even less unlikely situation. Many thousands of people take the coin into a room and flip it less than ten times. Then almost all of them go on their merry way because they got normal results, but one of them - KahunaGod - got weird results so he inevitably posts on the internet about it. Of course, it was actually very likely that a few of those thousands of people would flip ten heads.

 

[PieceOfMind]

Exactly, and it's basically for that reason that we have seen a few people on the boards complain that they've seen mostly the same leaders in each of their few games so far. Some of them are quick to assume it means there's something wrong with the game.

 

[gunter]

I have played several games with Romans and the game ALWAYS picks up the same 5/6 opponents. ( 6 AIs )

Rarely only 1 of them changes, always the same 6 or 7 anyway.

Patethic Shafer's concept of randomness, pathetic bug indeed.

 

[PieceOfMind]

I have played several games with Romans and the game ALWAYS picks up the same 5/6 opponents. ( 6 AIs )

Rarely only 1 of them changes, always the same 6 or 7 anyway.

Patethic Shafer's concept of randomness, pathetic bug indeed.

I was playing Yahtzee once. 5 times in a row I rolled 2 of a kind (out of 5 dice).

Pathetic concept of randomness, pathetic bug indeed.

 

[LegioCorvus]

I was playing Yahtzee once. 5 times in a row I rolled 2 of a kind (out of 5 dice).

Pathetic concept of randomness, pathetic bug indeed.

I think he was being sarcastic. Hard to tell online. Poe's law may also apply.

 

[Helmling]

I've had some games where the closest civ is literally 10 hexs away, and others where I'm the only one on a continent at least 35 hexs by 25. Something must be bugged with civ placement. All on the same map type/settings BTW

Yeah, I had one epic win that was made simple by the fact that I started on a fairly large continent all to myself...even though it was Huge with 19 players.

I like it...you never know what to expect.

 

[PieceOfMind]

I think he was being sarcastic. Hard to tell online. Poe's law may also apply.

I am pretty sure he was not sarcastic.

 

[KahunaGod]

After more testing (science) I have determined that: the greater number of civs in a game the greater the likelyhood of a civ without startbias to be placed in less then ideal surroundings. I believe it is because all start bias and start avoid civs must be placed first to satisfy their coded start rules. After this is done, the other civs are placed at best possible random distance from each other and then forcibly given at least 2 starting happiness resources and a small amount of workable land in their Cap, but whatever is outside the Cap is ignored in the process.

When I play smaller games, I have normal predictable results with Rome. Grasslands, hills, rivers, plains, deserts, forests, mountains, jungles all random and varied from start to start. Once a larger number of civs is introduced, the percentages change drastically and the most likey start points are pre-claimed by other civs.

What I havent determined yet is if resources have a factor in this. As I said before, everyone automatically seems to be given 2 happiness resources in/next to the starting Cap plot. What I still wonder about is after civ land bias is finalized, does any civ resource bias take effect? Such as: Computer- OK, I have placed the bias civs correctly, now lets place Mongols, where are horses, ok there they are, I will put them there.......etc. This might further prejudice some civ starting locations. I have no idea though when resources are placed onto the map, if it is before or after civ placement. I suppose some mod type fellows could determine that.

In a nutshell, there seems to be a law of diminishing returns when playing maps with more and more civs. This is all done with Huge maps, which P.O.M. said he thought to be buggy anyway. Next I will test large and standard maps.

 

[Welnic]

While technically true, if you're suggesting that this probability is 50%, I'd disagree. If you flip a coin 100 times and get heads every single time, it means you're flipping a coin with two heads.

No, it doesn't, but thanks for demonstrating my point.

Probability does funny things to people's heads and makes them think silly things when their brains start trying to deduce patterns and anticipate results. You can explain someone the exact mathematics of something as simple as a coin flip, but some people will still resist it on instinct, even if they know it's factually true.

It's even worse with video games and such where people can't see all the mechanical moving parts.

I have seen some of the moving parts, and I have no reason, from my experience, to believe that start bias is messed up. KahunaGod, from his experience, strongly believes that start bias is messed up and causing Rome to start near deserts.

Now, without looking at individual lines of code and spotting errors, neither of us can really know for sure (and even then, maybe not!)

It's the same as the coin flip. If I take the coin in a room and flip it 100 times and get heads. And he takes the coin into a room and flips it 100 times and gets tails, then we both meet up and flip the coin, his gut will tell him it's going to be tails, and my gut will say heads.

If everyone on Earth, 6.8 billion people, flipped a 100 coins every two minutes, 24 hours a day for a billion years, the chances of someone flipping 100 heads in a row would still be over 700,000 to 1. In the face of those odds, I would go with a coin with two heads.