Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 ... 6 7 [8] 9

Author Topic: Mineral Occurrence Setting  (Read 43495 times)

white_darkness

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #105 on: March 08, 2011, 01:26:49 pm »

*snip*
Attention interloper - heed this recorded message! This drone vessel speaks with the voice and the authority of the Ur-Quan. You are trespassing within Ur-Quan space.

Pkunk wasn't the strange looking birds? and Ilwrath wasn't that fucking spiders who became invisible and kicked your butt in the beginning of the game?

This drone now leaves to inform the Ur-Quan of your transgressions. You are commanded to remain here and await the arrival of the Ur-Quan. Disobedience... Will be punished.

man, one of the best(if note THE one) games I played, ever. And the Ur-Quan voice, and the overall voice acting was awesome.

The Ilwrath were the evil spiders who worshiped Dogar and Kazon, and yes the Pkunk were the strange hippie type birds and the unloved non-scotish accent offshoot of the Yehat race.

I loved those games and it was so easy to get lost and never actually finish the game and run out of time and then have the races of the galaxy wiped out...

But then, the galaxy must be cleansed...
« Last Edit: March 08, 2011, 01:34:59 pm by white_darkness »
Logged

Lemunde

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #106 on: March 08, 2011, 02:02:59 pm »

...

I may graph/plot this later, to see if I can find some sort of curve along the points, but going this far gave me a headache.

My results...

...


I plotted your data:
Spoiler (click to show/hide)

Hmmm, it looks like a natural logarithm curve rather than a 1/x curve.

[fake edit]

ha ha, I just tried generating a pocket world with scarcity 100000, which should have nothing, and it actually had a lot of metals. From the metal amounts and types (159927 and 19), using my equation, as well as some binary math, it looks like Toady made a programming error. If you gen a world with scarcity 100000, it comes out the same as scarcity 1696. I think this is because the scarcity value is stored in a 16 bit binary, which has a maximum value of 65535. See http://en.wikipedia.org/wiki/Integer_overflow

What have other people seen happen for large values of scarcity?

I guess the obvious test would be to generate a few worlds at 65500, 65600 and 100000 and compare the results. If your theory is correct, 65500 would have almost no minerals, 65600 would have an overabundance of minerals and 100000 would be comparable to your previous test.
Logged

Jeoshua

  • Bay Watcher
  • God help me, I think I may be addicted to modding.
    • View Profile
Re: Mineral Occurrence Setting
« Reply #107 on: March 08, 2011, 02:09:55 pm »

MEthinks Toady is going to have to dial the maximum back a bit.

Seriously, 100,000? He should have just had it be 1-255, and had an off switch at 0.  Do we really need THAT fine grained of control over how little resources we get?
Logged
I like fortresses because they are still underground.

Psieye

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #108 on: March 08, 2011, 02:42:00 pm »

We've established that it's very much not linear, but a parameter to go into the random number dice game. Fine grained 'control' means we get to re-roll the dice for pretty much the same odds in the hopes we get something more favourable at a given site.
Logged
Military Training EXP Analysis
Congrats, Psieye. This is the first time I've seen a derailed thread get put back on the rails.

arkhometha

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #109 on: March 08, 2011, 03:58:39 pm »

The Ilwrath were the evil spiders who worshiped Dogar and Kazon, and yes the Pkunk were the strange hippie type birds and the unloved non-scotish accent offshoot of the Yehat race.

I loved those games and it was so easy to get lost and never actually finish the game and run out of time and then have the races of the galaxy wiped out...

But then, the galaxy must be cleansed...
Yes, the first time I played I was taking my time, and I wasn't even mid game in the history when I saw the Kzer-za sphere of influence start moving and making another races spheres disappear. Visited one of the planets of this races and I found only ruins. I was like "FUUUU"
I was accustomed with games that says "hey, they are battling for the destine of the galaxy, but take your time, they won't make their move until you actually defeat them."
Losing this way was quite a surprise, discover that times matters too. I started a new game with great pleasure!
Logged

Supercharazad

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #110 on: March 08, 2011, 04:18:44 pm »

What I don't get is... if regular setting (2500) is like .18, then what on earth is 100,000 like?  It would be like a stone-age scenario where everyone fights with bone and rock weapons, except for the god-like people with the magical shiny copper axe.

I actually did 100000 because I thought it would mean many minerals. It's not that bad, most civs have metal.
Logged

white_darkness

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #111 on: March 08, 2011, 05:08:03 pm »

The Ilwrath were the evil spiders who worshiped Dogar and Kazon, and yes the Pkunk were the strange hippie type birds and the unloved non-scotish accent offshoot of the Yehat race.

I loved those games and it was so easy to get lost and never actually finish the game and run out of time and then have the races of the galaxy wiped out...

But then, the galaxy must be cleansed...
Yes, the first time I played I was taking my time, and I wasn't even mid game in the history when I saw the Kzer-za sphere of influence start moving and making another races spheres disappear. Visited one of the planets of this races and I found only ruins. I was like "FUUUU"
I was accustomed with games that says "hey, they are battling for the destine of the galaxy, but take your time, they won't make their move until you actually defeat them."
Losing this way was quite a surprise, discover that times matters too. I started a new game with great pleasure!

Got to love the fact that no topic around here will stay on topic for long.

Aye, it was fun, I think I got wiped out by the Slyrandro drones my first time.  Stupid drones.

On on topic notes, I think I'll produce a 4.5k 6k, 9k, and 12k scarcity and see what I come up with next.  While it might be a case of wraparound, it might also be something a little bit more interesting, you can never know.  I've seen some very strange formulas before.
Logged

white_darkness

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #112 on: March 08, 2011, 08:06:20 pm »

Fired up a few more along the way, learned a few nice new pieces of detail.

#1:  Some forms of "crap rock" (I'm looking at you Orthoclase).  Are not counted in the shallow/deep metals even though they're part of the stone minerals raw.  Good to know that.  This may only be true for .21 and not earlier.

#2:  Raw Adamantine is not a deep metal.

At about 6k my shallow/deep metal 2 biome site didn't have any metal in some situations, but it did have Orthoclase.

I unfortunately had to skip 4.5K 24K, and36K since a dwarf civ was genned right on top of it.  They needed to move.  Regenned the worlds 5 times with no luck.

Spoiler: Mineral Scarcity: 6000 (click to show/hide)

Spoiler: Mineral Scarcity: 9000 (click to show/hide)




And this one I was hoping would be particularly interesting since it's past the theoretical wrap-around point.  Unfortunately it's not. Most of the map is lit up as shallow/deep metal just like 48000 but my location, I've been recycling over and over is...utter crap.


It'd require more testing to delve further into this, but at least one thing seems to be conclusive now.  Stuff isn't getting flagged as shallow/deep metal in .21 if you can't smelt it.  Which is a general relief compared to the .19 situation where people were reporting cobaltite or something for a metal.

A little more learned and a little fun from it.  :)
Logged

Miuramir

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #113 on: March 09, 2011, 11:40:25 am »

i don't see any raw candy there :(

I think that is possible in a 2x2 embark. You need a 4x4 at minimum to be guaranteed candy.

I saw this and went back to check with reveal, and there is no candy in my 2x2 embark.  I don't think that is a bug though, because I know before that I have seen 3x3 embarks with only 1 spire in a corner, which would leave a 2x2 section with no fun times.

The above is not exactly correct; please see this post for more info.  In short, you need the smallest dimension of your embark to be either 2 and aligned to the grid, or at least 3, to guarantee a cyan pillar; a random 2x2 embark has a more than 1/5 chance to have none. 
Logged

Shandra

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #114 on: March 10, 2011, 12:37:06 pm »

I unfortunately had to skip 4.5K 24K, and36K since a dwarf civ was genned right on top of it.  They needed to move.  Regenned the worlds 5 times with no luck.

Couldn't you just randomize the history seed - AFAIK that one should have no control over the results we are looking for, but may result in worlds where that spot ain't occupied by any civ ?

Edit: Just tried playing with some new world-gens. In my previous runs with my world-profiles changing the mineral occurrence hasn't changed the morphology of the world - maybe just some internal map layouts (as sometimes on the exact same spot there where cave-ins upon embarks and with other mineral settings not), but the world always looked the same (except for distribution of good/evil and of course the civs (using all constant seeds)) and the embark spot had always the same surface morphology. But with my current wg-profile I am getting all different worlds with just altering the mineral setting. I don't have any idea what may be going on here, only thing I can test for is the Mesh-Size settings, as in previous times I always used fixed settings there, but for this profile I left the Mesh Sizes on their default Ignore.

Grmpf, confirmed and somehow that makes sense - defining the mesh size preserves the world/side appearance throughout the various mineral occurrence changes - sigh.
So what was the profile for all those tests in this thread again?
« Last Edit: March 10, 2011, 01:20:18 pm by Shandra »
Logged

white_darkness

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #115 on: March 10, 2011, 01:38:48 pm »

The one's I did?  Spoilered in the first set of them a few pages back.

Re-spoiler'd here.


Which I had thought about changing the history seed, but I didn't want to add any additional variables, no matter how unlikely they might be to have an impact.

Logged

Shandra

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #116 on: March 10, 2011, 02:43:06 pm »

Thanks, must have missed that spoiler in my brief scan of the whole thread. So you are working with the default frequency tags. Was my above observation then also true for your tests and the same spot appeared each time in a different shape - or is my experience just a coincidence for that particular parameter set of mine? Because if it is each time different I thing about conducting a few tests of my own with fixed mesh sizes to see if the results would be any different towards your observations.
As Arkhometha has just mentioned something about just the internal side layout - but for my quick test it was the whole world and side layout that changed. But then you always used the default embark side, right? As that spot changes AFAIK for each gen then it may not have been your goal to always observe the exact same spot?
Mh, for my last Embark I only sampled data for the mineral settings where at least 3 (including the 2 mayors) of 5 biomes also listed flux stone and had at least shallow and deep metal(s) - wich where the minority of all the numbers I tried. And sadly that is what I am looking for as player, but I have no idea how to incorporate the biome information for representation of the data (at least the -ab parameter helps in that part, newer known that before). Ok, but focusing on the pure data. Plotted so far was Y1 total number/Y2 total types vs. parameter value, right?

P.S.: Is there a quick way to replace the " : " in the prospector output with tabs, or is it really needed to do that via a text editor for quick copy&paste into excel/calc/whatsoever ?

Edit: And would it be better to just look at a side covering just one biome?
« Last Edit: March 10, 2011, 02:56:14 pm by Shandra »
Logged

white_darkness

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #117 on: March 10, 2011, 04:14:38 pm »

Base world layout never shifted in the slightest at any level.  It was only the actual embark locations that would vary in layout, often by very little, and there was the oddness I had at 3k.

Pretty much yes, parameter value vs. total and types.  The later ones raised some questions for me as to what it classes as a metal since the shallow/deep metals disappeared in higher number gens at the location.  Though the "stuff" there had been classed that way in .19.  Which leaves me wondering if the classifying has been hard-coded to specific types or if it's somehow tied to the "Flux" not appearing in the summary of the location but you can still find it with the finder bug, or maybe the two are connected...who knows? but Toady.

The initial ones I did from 100 to just under 3k (whatever that number was now) were all at the same spot.  At that point, I'd become so familiar with the map, that I could literally eyeball and head back there.

To be semi-exact, it was about 4 squares up from where the mountain range on the west side bowed out, or about 7 squares south of the north boundary.  Then in the regional view, it was pretty much the southern edge.  As long as a settlement didn't take the spot, it was always there.

Heh, I've still got all the saves except the ones where a settlement was on my spot.  When i first threw some data together I offered to upload them all to the depot.  Some of them have "active" "play now!" embarks and some are abandoned.

Eliminating the ":"?  I didn't bother myself, but did my maths manually.  But there should be a text to columns option in most spreadsheet software under "Data" (good luck dealing with the ribbon Excel users).  Just highlight the target data, select text to columns, specify ":" as the delimiting character, and it splits it into two columns.

No idea about one biome.  I have thought about rerunning everything at a different location, but we got some initial data and a theory and most people seem to have moved on.  Getting more data and getting into the nuts and bolts of it would be great in some aspects. 

I'll admit freely I wouldn't mind digging into all the info.  If it wouldn't drive me mad from boredom, I'm tempted to carefully experiment with every single parameter until I understand their impacts fully, particularly since the wiki info on advanced world generation parameters is kind of meh...

Right now, I'm just enjoying playing .21 and digging in and collecting cobaltite.  Don't know yet what I'm going to do with it and a major river, but I'm going to do something (unless another update rolls along or I screw up so horrifically I abandon in disgust).  So far those two items have been the only things that cause me to abandon.
« Last Edit: March 10, 2011, 04:26:15 pm by white_darkness »
Logged

arkhometha

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #118 on: March 10, 2011, 05:32:19 pm »

Awesome piece of work, white_darkness. A good research direction.
"If it wouldn't drive me mad from boredom, I'm tempted to carefully experiment with every single parameter until I understand their impacts fully, particularly since the wiki info on advanced world generation parameters is kind of meh..."


I understand this, as I was messing with advanced world generation to reduce the aquifer number and didn't get any good results, not a clue in the wiki or with the parameters.

As I suspect and Khift said, the mineral seems to be biome tied, but I found some odd situations as to find iron ore on a shore, and I'm sure this is not the kind of biome for it.
Logged

white_darkness

  • Bay Watcher
    • View Profile
Re: Mineral Occurrence Setting
« Reply #119 on: March 10, 2011, 06:36:18 pm »

Problem is it'd be a sanity sapping research direction.

163 settings, from simple and obvious like "Reveal History" to unobvious like mineral scarcity.  Testing all those at various settings one at a time and seeing their impact would probably drive an individual bonkers.
Logged
Pages: 1 ... 6 7 [8] 9