A study of Hut Outcomes

[samson]

samson - nope it was rivered grass - hence our amazement

Can you please post a save? Either from before or soon after the hut tip?

[Scouse Gits]

samson

The hut was on rivered grassland. I have enclosed the 4000BC save. (180 x 180 giga map) The hut in question is at 76, 82.

One (untested) aspect of huts appears to be your position on the Powergraph.

Results seem less likely to be Barbs if pathetic to moderate. As the Powergraph place affects tech gifting it's possible this comes into play here as well.

---------------------------

SG(2)

[samson]

Originally posted by atawa


Its my observation from playing 1000+ games.

Not that I get an archer every time from a forrest but I get much more archers from forrest then horsies or chariots.

If I have time this weekend I'll test this.

Please do. And report the exact conditions of your tests as well as the results. As I learned from the great nomad dispute, there may be some obscure factor which creates different results under different conditions. For now, I stand by the statement that no direct correlation between terrain type and mercenary unit type exists.

[samson]

Originally posted by Scouse Gits
samson

The hut was on rivered grassland. I have enclosed the 4000BC save. (180 x 180 giga map) The hut in question is at 76, 82.

These is no hut at 76,82.
The location of your chariot in the snapshot is 73,81 and there is a hut there. That square, however, is rivered Forest, not Grassland. Which explains the occurance of a nomad at that location.

Originally posted by Scouse Gits

One (untested) aspect of huts appears to be your position on the Powergraph.

Results seem less likely to be Barbs if pathetic to moderate. As the Powergraph place affects tech gifting it's possible this comes into play here as well.

That's possible. I haven't concluded my testing of the Barbarian outcome yet.

By the way, I seem to remember a post about the pattern of Barbs from huts (the number and location of barb units) being related to year-cycle. Was that one of your (SGs) studies by chance? Do you still have a link to it?

[DrSpike]

It is my prior belief that Atawa's claim has some merit, but I appreciate that is it is easy to be fooled by other factors that are not controlled for in a game environment. Test away.

STYOM: Were your results due to a small sample then?

[samson]

DrSpike -

Yes, the idea of terrain influencing the resulting unit of a mercenary outcome does have merit. That's why I specifically tested it. But my conclusion after examining the results of those tests is that it isn't true.

[DrSpike]

Yeah I got that. More testing is never bad though......a Popperian falsificationalist approach is the best approach here.

BTW Did you see STYOM's results before? Did you test the same thing?

[samson]

More testing is definitely good.

I did see some results STYOM posted under a thread called "Huts and Scrolls of Wisdom" where he got a very large number of Tribes in 3850BC. But there isn't enough information to tell whether or not another factor was at work. I suspect the Continent Rule was in play, because that would explain the large number of Tribes outcomes.

And yes, I did test Oedo years. No correlation to outcome as far as I can see. The only year-related effects I have found so far are to the specific years: 1550BC (50 turns) and 1000AD.

[Scouse Gits]

So we WERE more drunk than we thunk!

The SGs in contrition

[Six Thousand Year Old Man]

Originally posted by DrSpike

STYOM: Were your results due to a small sample then?

50 hut tips, reloaded. Quite different results in 3850 BC than in 3800.

The continent rule was NOT in play. I had put down 1 city - obviously in 3850 BC I was still on the same continent. The results in 3850 were 9 units, 7 barbs, 25 techs, 9 gold. In 3800, it was 13 units, 13 barbs, 11 techs, 13 cash. I'm willing to chalk that up to statistical aberration, but at the time it seemed meaningful.

Here's the 3850 save:

[Six Thousand Year Old Man]

and the 3800 save:

[Six Thousand Year Old Man]

I know you were referring to the 3850 save with a lot of tribes... still looking for that save file!

[Six Thousand Year Old Man]

Here's the 3850 BC save that for me, produced a LOT of tribes.

[DrSpike]

Samson?

[samson]

Originally posted by Six Thousand Year Old Man


50 hut tips, reloaded. Quite different results in 3850 BC than in 3800.

The continent rule was NOT in play. I had put down 1 city - obviously in 3850 BC I was still on the same continent. The results in 3850 were 9 units, 7 barbs, 25 techs, 9 gold. In 3800, it was 13 units, 13 barbs, 11 techs, 13 cash. I'm willing to chalk that up to statistical aberration, but at the time it seemed meaningful.

Here's the 3850 save:

Yes, the continent rule is not in play here, but the Nomads Rule is. The hut square you are tipping produces Nomads not Advanced Tribe even though it is Grassland. Yet, you already have one Nomad, so nomads are suppressed in the situation and the Tribes chance is added to Scrolls.

Thus, the high number of techs you saw on turn 3850BC is expected. Scrolls should occur twice as often as each of the other outcomes.

It is the results from the year 3800BC that are aberrant. However, when both results are added together to produce 100 chances, you have:

22 units, 20 barbs, 36 techs, and 22 gold.

This is nearly what is to be expected from Scrolls getting the benefit of Tribes' chance. Techs are a bit light here, but not too much. A couple hundred more tips and the results should even out.

An interesting hut location: a peninsula with only grass and ocean squares, no specials, and it produces Nomads when your NONE settler is disbanded. Out of curiousity, I went into Cheat Menu and created a Japanese setter on the square where the hut was. It moved two squares SW and founded a city there. Another confirmation of the connection of Nomads/Tribes to AI city site selection.