Hut luck

[Jvstin]

Originally posted by Addled Platypus
majority is gold 30-52 gold

then:
local map
unit
tech (1 out 10 if then)

So gold is the most common result? And maps are less common and thus "higher value"?

Hunh. I would rather have the gold than the local map, any day of the week.

[Yosho]

I got 3 techs in a row one time, and basically started out the game with horeseback riding, which was pretty awesome. Now THAT was a rush the AI never saw coming, heh heh.

[martin mcmartin]

"what are the odds of one hut? like 1:10 at emporer? That's make it 0.1 to the eighth power, if my math skills are correct anyway. That'd be 1 in 100,000,000"...

"only happens once for all civ games ever played by all player in the world - so: thanks rah... it wont probably ever happen to any of us ever, since it was you who hit that jackpot"..

this math is not correct though! everybody will still have the same chance as Rah: put 5 balls in a bowl, 4 black and 1 red. take one out. chance for red is 1/5. Rah took red, but before we have a go, Rah's red ball is put back in the bowl! other way of putting it, is that we all have the same bowl with balls in our pc. the chance that we hit the jakpot is as big as it was before Rah hit it....



MC

[Unimatrix11]

i remember there has been a thread about this. It explained it all quite detailed. IIRC its like a list with ten or so items, one for each possible event. And a list for each difficulty. Or something to that effect. Spawning barbs is tagged ´negative´ and those items of the list are ignored when using a scout. I dont remember the details anymore, so i suggest to anyone interested in them to search for that thread...

But a list of item could look like:
free scout, experience, much gold (2x), a little gold (2x), map, tech, barbs (2x). That means with a non-scout you´d have a 20% chance of popping barbs and a 10% for a tech. With a scout it would be 12,5% for a tech (1/8). On a lower difficulty one of the item would be settler, and maybe just one barb-item. On a higher difficulty, you´d have maybe three barb-items and only one much-gold-item... The total number of items for a list and what item appears how often i pulled from thin air, but in general i think it works something like this. So there is not really an internal value to the poppings, as Jvstins statement suggests, apart from barbs being negative and the rest positive. Out of all the positives it is entirely a matter of difficulty what you are likely to get. (tho i dont know if using scouts has also an effect apart of eliminating the negatives from the list)... well-again: search for that thread if you want relyable and detailed info on it...

[martin mcmartin]

I reload, just to get techs. yes, that means sometimes (always!) 8 techs in a row. and still I get out-teched form 500 BC onward!

MC

[Lancer]

Yeeesh rah, I thought two techs was a thrill ride, man. Nice luck. You should have run right out and bought a ticket.

[LzPrst]

in a MP test game (marathon speed, diplogame) my warrior popped a hut 5 squares from my capital and got 3 barb warriors. they killed my warrior and marched straight to my undefended capital that needed 8 more turns building a warrior. in year 3100 BC...

earliest wipeout ever. the shame!!!

[Unimatrix11]

Originally posted by martin mcmartin
"what are the odds of one hut? like 1:10 at emporer? That's make it 0.1 to the eighth power, if my math skills are correct anyway. That'd be 1 in 100,000,000"...

"only happens once for all civ games ever played by all player in the world - so: thanks rah... it wont probably ever happen to any of us ever, since it was you who hit that jackpot"..

this math is not correct though! everybody will still have the same chance as Rah: put 5 balls in a bowl, 4 black and 1 red. take one out. chance for red is 1/5. Rah took red, but before we have a go, Rah's red ball is put back in the bowl! other way of putting it, is that we all have the same bowl with balls in our pc. the chance that we hit the jakpot is as big as it was before Rah hit it....



MC

I somehow doubt that. I think it´s a matter of perspective: If the chances of hitting a tech is 1:10, then hitting 8 techs in a row mathematically indeed is 1:10^8, even though each time it still is 1:10 when you look at it seperately, be it the first time or the 8th. (With your example of the balls, putting back the ´tech-ball´ after each draw just means, that the probability to have it to be drawn even just twice is >0)

Same goes with my statement: Of course Rah having had that luck, doesnt really reduce the chances of me having it as well, since there is no real connection between the two chances (that original statement of mine was not really meant to be taken serious). Yet it is obvious, that this indeed does not happen all that often and if you look at it ´globally´ it is a lot less likely to occur twice than once. So since it occured once already, it is kind of a valid statement to say that it probably wont happen another time ever for anyone. The math for the actual probability of this happening among all games ever played would be a bit more complicated, since we´d have to know, how many games of civ will actually ever be played - which we will never do... With the 100.000.001st game played (disregarding difficulty levels and such), it would become more likely to happen once than never. But it is as likey to happen in the 1st game as it is in the 32,242,324th or the 100.000.001st, no matter if it happened before or not. Hence, to my understanding, as i said, wether the statements above were true or not, depends a great deal on perspective, and it is things like that, that lead me to doubt the existence of truth itself.

[martin mcmartin]

"If the chances of hitting a tech is 1:10, then hitting 8 techs in a row mathematically indeed is 1:10^8, even though each time it still is 1:10 when you look at it seperately, be it the first time or the 8th. (With your example of the balls, putting back the ´tech-ball´ after each draw just means, that the probability to have it happen even just twice is >0)"



yes, you are right, there is no "global counter" that changes the chances for all players. so, every time the chance is 1:10. if you want 8 techs in a row, the math is:

1:10 x 1:10 x 1:10 x 1:10 x 1:10 x 1:10 x 1:10 x 1:10

but! "it is a lot less likely to occur twice than once" is not correct though. that is what I tried to explain with the balls. we all have the same bowl with balls in our computer with the same chance of generating this combo. Rah's bowl with balls does not have any influence on our bowls. he had his combo, the chance that we get it has not changed. I didn't see Rah enter my house and take any ball out of my bowl, that is for sure! my chance is as big as Rah's... even Rah has an even chance of doing it again. in a new or even the same game the chance is the same again. easiest example is the coin. Rah flipped it and got heads. now I try, my chance is also 50% to get heads... Rah's flipping his coin at one end of the globe has no effect on my flipping of the coin. it did not drop to 49% all of a sudden.

if you really wanna know, call my old math teacher, his name is Willem, I can get his number if you like.

MC

[Unimatrix11]

As much as doubt lingers in my head concering this, nobody, no matter what´s his name, profession or phone number, can tell me that something that has a low probability of occurance is as likely to occur 1000times as it is to occur just once (in a given number of tries).

This reminds me of a long discussion i once, years ago, had, at the "web space alliance" forum. There someone claimed to be one of the best math-students of all of romania, and we had an argument concerning some stochastic problem. He took his points from a book, i sat down and figured it myself. In the process, i happened to draw a graph, which looked somehow familiar. I pulled out a 10-mark-bill, which had Friedrich Gauss on one side and his bell-curve on the other. And that is what it was. Later it turned out that this romanian guy had applied the wrong formulae to the problem. That tought me to trust my own thinking rather than what somebody, no matter what his title (or claim), tells me.

Of course rah´s luck doesnt influence mine - but i said that. And still it´s more likely to happen just once, than twice or even more often, given a limited number of tries. It´s not like a one-time coin toss. It´s a hundred-time-always-face-up coin toss-sequence. Tell a hundred people to try it once. Say everyone who happens to always toss face-up, wins something. Now if one happens to toss it like that, he is lucky, but it might happen. But it´s rather unlikely, that 2 make it. And even still less likely for 3 to make it. As you increase that number, it gets less and less likely to actually happen, even though no one´s luck influences the other ones luck. If You were to give a price to everyone making 100-face-up tosses in a row, you could actually calculate a reasonable price/participation-fee-ratio based on this logic (tho it will never gurantee 100% that you wont end up loosing some money with your little gambling game). If it wasnt like that, Vegas would just not exist...

[Jvstin]

Originally posted by Unimatrix11
The total number of items for a list and what item appears how often i pulled from thin air, but in general i think it works something like this. So there is not really an internal value to the poppings, as Jvstins statement suggests, apart from barbs being negative and the rest positive. Out of all the positives it is entirely a matter of difficulty what you are likely to get. (tho i dont know if using scouts has also an effect apart of eliminating the negatives from the list)... well-again: search for that thread if you want relyable and detailed info on it...

I'll have to look for that thread--but isn't assigning a probability to the positives across difficulty levels a way of giving an internal value, just because of relative frequency?

[Theben]

Originally posted by Ming
My latest emp game (using a scout), I got 4 techs from a total of 9 huts... I was quite impressed.

Clearly Rah and Ming tweaked the files.

[Ming]

Why bother... if you want to cheat, you can just use the reseed option, and keep resetting until you get what you want

[snoopy369]

Originally posted by Unimatrix11
As much as doubt lingers in my head concering this, nobody, no matter what´s his name, profession or phone number, can tell me that something that has a low probability of occurance is as likely to occur 1000times as it is to occur just once (in a given number of tries).

The above is of course false (it is certainly less likely for someone to hit any odds-determined probability<1 thing twice in a row as compared to once), but the wording is important.

Without knowing what has happened yet, you can accurately say that it is very unlikely for rah and someone else to hit the 1/10000000 chance; but once you know rah hit it, the chance is still 1/10000000 that you or anyone else will hit it.

[snoopy369]

Originally posted by Unimatrix11
i remember there has been a thread about this. It explained it all quite detailed. IIRC its like a list with ten or so items, one for each possible event. And a list for each difficulty. Or something to that effect. Spawning barbs is tagged ´negative´ and those items of the list are ignored when using a scout. I dont remember the details anymore, so i suggest to anyone interested in them to search for that thread...

Not quite... there is a list of 20 (I think) for every difficulty level in the XML file (Civ4Handicaps.xml in GameInfo), and things that are more likely just have more entries