Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Dwarfu]

Way to kill a thread, Roxorius...

[darkflagrance]

World gen battles have messed up math. It's exceedingly common for a single survivor in a town to kill hundreds before being the last one conquered, making this thread somewhat a contest of "how badly did world gen model the battles for your world" rather than "awesome, here's a really rare occasion."

I don't think it's necessarily a result of messed-up math, depending on your definition of messed-up.

It appears that world-gen basically inputs the available weapons, armor, materials, and built-up stats of individuals into the battle sim and then rolls for victory, survival, casualties, injury, etc. (The game does appear to use available weapons and quality when calculating battles: civs with swords get messages of slashing or stabbing, and a civ with [METAL_PREF] - which implies steel weapons - does better than that same civ with the same seed without it) Because world-gen models the results of duels between individuals, I assume it must calculate each one individually based on these criteria.

Thus, potential champions are eternally striving against the fate meted out to them by the RNG. It seems they can skill-up by gaining kills, but it changes not the fact that they are fighting the odds; the more they fight, the more opportunities they have to die by the RNG regardless of skill. This is why it's rare to find individuals with more than a certain number of kills. Is there anything wrong, then, with discussing how the RNG occasionally throws up a truly legendary figure in generated history? It's not as if the RNG constantly spits out heroes with over 1,000 kills in vanilla.

In addition, when a civ is on its last knees, whether its few remaining warriors get slaughtered or become heroes depends on the skill-ups they previously had. I've seen more human cities and goblin fortresses perish under the rule of some random nobody who was the only one left to rule after everyone else died than I have seen hold-out fortresses guarded by a few century-old champions.

Furthermore, you do have to bear in mind: elves have arrows, which I've noticed strike down enemies in world gen regardless of skill level. In my experience, it is extremely common for heroes who should have been powerful to perish due to bow fire. I've seen a human swordsman who fended off an entire siege of werewolves single-handed get killed by the first arrow fired in the next fight. Thus, it is an exceptional instance for someone to defend herself against the arrow-wielding elves, alone, for ten consecutive years, until death of old age.

The question is, then, if world-gen is messed-up, what should happen? If the last survivor is indeed bad-ass, why shouldn't he be allowed to survive? And, is there anything wrong in discussing the epic legends that the DF world-gen creates?

Ultimately, this question is shrouded in obscurity by the fact that we only have a vague idea of how world-gen works. I personally wish I knew more about it so that I could regen Tholtig.

[Roxorius]

I don't think it's necessarily a result of messed-up math, depending on your definition of messed-up.

It appears that world-gen basically inputs the available weapons, armor, materials, and built-up stats of individuals into the battle sim...

Much less is going on than you think in the above quote. The math is rather more simplistic and prone to the results in this thread. It does seem to factor in overall experience of the armies, perhaps individually or more likely as a group average, but I doubt it's going down to materials, armor, wound status, etc.

I just generated a world in which dwarves had their usual equipment, then re-generated it without all their armour and all weapons except spears... so, lots of stabbing? No. They were still smashing, shooting, tearing away, etc. They did pretty much the same in battle as well, although I got different civs on the same map each time despite the same history seed.

So, they didn't have mail or any weapons but spears... and yet were doing as well as before, and even giving messages appropriate to different weapon types. I think the weapon messages are randomly generated without actually refering to any equipped status. In fact, the same dwarf can shoot, smash, and stab all in the same battle.


A human, on the other hand, has his hand torn off and goes on to kill 129 more people. Does combat even roll attacks between individuals? Maybe it just takes the average skill levels for the groups, and weights the amount of casualties it generates for a given battle, then assigns names as needed to fill out the casualty reports.

There's much less there than you think. It's mostly random messages painted over a simplistic algorithm. I call it "messed up" because kill ratios are usually obscenely high for city defenders, and skirmishes in the open tend to be even trades between novice groups, even if one side outnumbers the other overwhelmingly, like 50 or more to one.

If dwarven defenders fight their first battle in a mountainhome, as complete novices, the results are fairly predictably the following:

If they are outnumbered 450-to-one, each will generally kill at a 110-to-one ratio.
If they are outnumbered 250-to-one, each will generally kill at a 70-to-one ratio.
If they are outnumbered 150-to-one, each will generally kill at a 43-to-one ratio.
If they are outnumbered 125-to-one, each will generally kill at a 35-to-one ratio
If they are outnumbered 97-to-one, each will generally kill at a 24-to-one ratio.

If a battle is a meeting engagement between dwarves and goblins in open ground, even if the goblins are 9000, the dwarves and goblins will kill similar numbers (a dozen or so, at 18 dwarves) and retreat. This was a constant no matter how large the goblin army.

Try generating different populations of dwarves and goblins. I've done from 9000 to 1000 goblins and 18 dwarves with the above results a dozen times each. The game favors the defenders particularly the more they are outnumbered, counter-intuitively. Meeting engagements in open ground between imbalanced numbers usually end up with a few of each dying, no matter how overwhelmed. These and other aspects lead to a reasonable definition of "messed up."

The question is, then, if world-gen is messed-up, what should happen? If the last survivor is indeed bad-ass, why shouldn't he be allowed to survive? And, is there anything wrong in discussing the epic legends that the DF world-gen creates?

Even novices can kill at high ratios. in fact, the more outnumbered, the more they kill. Bad algorithm, IMO. Discuss if you like, though.

[Ampersand]

It's the Law of Inverse Ninja Skill. The more ninjas in a room, the weaker they are individually. There may perhaps be a reasonable explanation behind this. Imagine you have a gun, and are surrounded on all sides by hundreds of people with guns. While you can freely turn and open fire in any direction and hit -someone- every time, the probability of hitting any specific enemy is relatively low. Like wise, the probability of any specific one of them hitting you is low, but compounded by the sheer number of them, it approaches 100 percent. But none of them can actually shoot at you. Since the probability of any one of them hitting you is low, that means that the probability of a miss is high. Since you are surrounded on all sides, a miss must equate to friendly fire. Thus, until the lone gunman in the middle thins the ranks sufficiently, he is free to fire with impunity.

I don't know how, if at all this is relevant to DF though.

[nil]

I can definitely confirm that type and materials of weapons makes a difference in worldgen combat.  Metal_pref is a huge, huge advantage in world gen survivability while minor_metal is a significant disadvantage.  I had a civilization who only used wrestlers that had a lot of trouble surviving worldgen.  I've added whips and blowguns to make them they were much more resilient, and when I added a bludgeoning weapon into the mix I had to tune their fertility way down to prevent them from wiping out lesser beings.  SEVERONBREAKS applies in worldgen, and I'm nearly positive the values of natural attacks are included as well.  DAMBLOCK applies, although not as much as it ought to (in my opinion).  I'm not as positive about armor, but I think the elves' frailty is a good case in point.

It affects battle messages, too.  I just read through a war between gnomes (spears and crossbows) and werewolves (scimitars, flails, and scourges).  Werewolves are getting shot and stabbed, while gnomes are getting stabbed, smashed, and occasionally burned by the werewolves' demon ruler.  It's actually more sophisticated than I thought it was--throughout the course of the entire war, I saw two cases cases where the werewolves shot gnomes.  One was a hunter before the battle, the other a ranger--they were using crossbows that had been created for their civilian jobs.  I've also got elves modded to only use bows, and there's almost nothing but shooting in their worldgen battles.  Did you clear data/objects during your experiment?  I've learned from painful experience that deleting stuff usually requires that.

While looking through the battle logs I found my own entry for this thread:

Spoiler: a bad-ass gnome (click to show/hide)

Saneb Spattershoveled the Silken Speachlessness of Meals was born in Ironstopped in year 11.  She was a gnome (size 3) living in the violent world of the Relentless Assault mod (where humans are babysnatchers and werewolves and ogres are CAN_CIV creatures of sizes 9 and 12 respectively).  While she was still a baby, her parents were killed by the demon leader of the werewolves who lived in the tundra to the north .  While the attack that killed Saneb's parents only resulted in a pillaging, the werewolves soon returned to destroy Ironstopped.  Still only a child, she fled to the newly founded town of Boltedcaskets, but the same pattern repeated there.  Saneb fled a fifth time to Wonderpick.

There were 8 years of peace there, but in 25 the werewolves returned.  The initial siege was repelled with half of the 100-strong invasion force dead, but the 32 out of Wonderpick's 36 able-bodied gnomes died in the process.  The outlook for for Wonderpick was grim.  In year 26, another 100 werewolves invaded, and of the dozen gnomes left to defend Wonderpick, only one surived--Saneb Spattershoveled the Silken Speachlessness of Meals, who had came of age  that year and killed 25 werewolves, mostly via crossbow.  Until 32 Saneb fought in Wonderpick alone, repelling 7 werewolf sieges averaging about 80 werewolves and 7human sieges of 10-20.  Since there was no one to grow any food, Saneb would occationally devour a slain enemy--eating sapients is more of a personal matter in gnomish culture.  A few surviving gnomish children started to come of age in 33, but most didn't last long. In 36, Saneb led an poorly thought-out offencive campaign against the werewolves.  60 werewolves caught her and her companions on the open ground.  Three of the red-shirted gnomes were devoured, taking only 6 werewolves to hell with them. Saneb fled back to Wonderpick to find that the humans had overwhelmed the small garison she'd left behind.  She and the other surivior tried to found a new town, Paintglistens, but they only had time to construct a couple of hovels before the humans found them.  Without fortifications, the 2 gnomes surrendered.  The humans brought in a gnome they had kidnapped seven years previously to rule the town, but she was soon killed by a cyclops, who was then killed by Saneb.  Although Saneb would liver for more than another century, this was the last creature she killed; in only 11 years of fighting she had defeated 310 sapient entities, mostly werewolves with a few dozen humans for spice.

When Igniz, the gnome who had co-founded Paintglistened with Saneb, was killed by a cyclopes in 64, Saneb was the only gnome born under the Perplexing Tongs known to still live; she'd already began ardently worshiping a human god of birth and rebirth.  She would stay in Paintglistened for another 60 years, but in 125 a band of humans managed to drive out the werwolves and their demon leader from a werewolven city, Scourgeposts (note that this is the same demon that killed Saneb's parents).  Saneb and a few humans moved in, but the werewolves launched a devestating counterattack in 127, killing all defenders and retaking the town, although their demon was killed in a duel during the battle.  For some reason, Saneb didn't participate in the battle but was instead captured and taken to the old werewolven capital, Gurglelies.

She was imprisoned there for the next 80 years (I wonder if that means I could stop worldgen early and easily recruit her in adventure mode?).  In 206 she managed to escape, and returned to Paintglistened.  Perhaps aware of her impending mortality, she converted to a new religion (same god) and became a babysnatcher the following year.  She kidnapped six ogres, passing on her ways of birth-god-worshiping and werewolf slaying to one of them, Irne Tombsatin the Rhyming Deity of Tools.  She died of of natural causes in 214 at 203 years of age.

[Roxorius]

I've had mixed results in furthering testing, nil. Equipment does not really affect the outcomes of battles. Something in the tokens does, obviously, or else elves and humans would trade equal losses: but it is not equipment. On the other hand, equipment really does cosmetically change the kill messages. When I said that civs without bows were still shooting and killing, that proved to be the rangers of the civs, as you pointed out.

Normal human equipment led to these first battles:

Battle of Horns, meeting engagement Initial Size Losses Humans: 1116 342 Elves: 999 428 Fierce Sieges, elven defense Initial Size Losses Humans: 996 440 Elves: 1998 388 Incinerated Siege, elven defense Initial Size Losses Humans: 994 423 Elves: 1991 360 I took away all the equipment from humans, even their clothes and socks, and re-genned the history to look at the first battles. I was, in fact, able to note that messages involving archery or cutting disappeared except for rangers. Otherwise, the results were almost identical: Assault of Claws, elven defense Initial Size Losses Humans: 995 424 Elves: 1994 362 Clash of Martyring, meeting engagement Initial Size Losses Humans: 995 202 Elves: 1832 644 Sieges of Clobbering, elven defense Initial Size Losses Humans: 994 431 Elves: 1997 347 Notice the prominent patterns emerging even in so few samples. The elven defenses are all astoundingly similar in outcome. In like scenarios, very small variations occur in outcomes. If two groups meet in a first battle, based on numbers and certain factors that don't include equipment, the outcome is highly predictable. Some token or tokens influence elves being weaker, but I think that is a more macro-level token rather than weapons, perhaps your METAL_PREF or DAMBLOCK, which elves have set to -1. There probably aren't individual weapon attacks or armor defense rolls, though, since naked humans lose no effectiveness. Those messages are a cosmetic touch.

[nil]

Ha, no shit, you're right about the equipment.  Teach me to extrapolate from uncontrolled experiments.

[balath]

I think Ampersand was getting at this:
http://tvtropes.org/pmwiki/pmwiki.php/Main/ConservationOfNinjutsu

In any martial arts fight, there is only a finite amount of ninjutsu available to each side in a given encounter. As a result, one Ninja is a deadly threat, but an army of them are cannon fodder.

This is also known as The Law of Inverse Ninja Strength: Threat = 1/N, where N = number of Ninjas or other "Elite Adversaries".