Print

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

[IndigoFenix]

It bears mentioning that [EVIL] is no longer a valid entity tag, so it has nothing to do with goblins employing trolls.  I don't think it ever did, actually, despite what the wiki says - I've added sapient pets to neutral races years ago - though the exact criteria for what allows a sapient creature to be added to the pet list is still a bit of a mystery to me.

[FantasticDorf]

It bears mentioning that [EVIL] is no longer a valid entity tag, so it has nothing to do with goblins employing trolls.  I don't think it ever did, actually, despite what the wiki says - I've added sapient pets to neutral races years ago - though the exact criteria for what allows a sapient creature to be added to the pet list is still a bit of a mystery to me.

I mean [SLOW_LEARNER] and [EVIL] reacting together for valid picks to use for [USE_EVIL_ANIMALS] particularly. Including creatures like Ogres & Blizzard men in native vanilla. When you embark with such races (accompanying your starting 7) they're unmanagable as livestock and can be drafted into the military, so all in all not very fun in their base form.

Last time this happened was with the Tiger men for sale by elves for no value, adding ontop of trolls with more tokens (like a assigned pet value) or differentiating them from a unintelligent unit is often the most direct way of handling it.

[Shonai_Dweller]

So, my current world has about 1.3 million trolls in its outside populations, a whole load at various sites throughout the world too.
And yet, Legends mode tells me that "Quelge Lastwarns the Impunity of Warmth was a troll. He was one of the only ones of his kind".
How can this be?

Only civilization-joined troll, I guess.
" [LOCAL_POPS_PRODUCE_HEROES] --new with 0.42.xx: wild creature can join a civilization. "

It didn't join a civ. It followed regular troll behaviour since at least 40.x in killing stuff, gaining a name and moving to a cave/catacomb. Being made a priest is new, but weird stuff with new features is kind of expected.

And all 50 historical figure trolls in this world are described as "one of the only ones of his kind" whether they got themselves a home or not (can't find any more priests).

Blizzard people histfigs in this world (11 of total pop 3k) are all "the first of its kind" or "of unknown parentage. Which is what you'd expect. Gorlak's too (17 histfigs of 2.7 million pops).

So what's going on with trolls?

[Urlance Woolsbane]

I am reminded of a scenario I witnessed in an earlier version, in which members of a kobold group in the catacombs of an abandoned city assumed the roles of its resident religion. One of the priests went on to desecrate a temple, resulting in that singular rarity, a kobold vampire.

[Toady One]

Is it possible to create experimenters that don't make the mutant freaks, or do EXPERIMENT interactions require the NIGHT_CREATURE_EXPERIMENTER tag?

Not currently.  I also posted over on your bug report a bit about the particular issue with the mod.

Would it be possible to create a secret that frequently resulted in experiments, by modifying the unit's personality in some way?  What about a secret that had a greater chance of being passed on through writing or apprenticeships?

Having a historical one in an attacking army that destroys/takes over a site should do it 50% of the time (more if they can animate too).  The site take over/destroy guarantees them a place to put the experiments.  Setting up an administrator doesn't do it.

I'm finding lots of rumours of "bone-chilling horror" pointing to nearby forts. And these forts are populated by regular fort dwelling groups. Isn't "bone-chilling horror" a reference to necromancer dwellings?

Without looking at it, I agree that this is almost certainly just a site type issue.

So, my current world has about 1.3 million trolls in its outside populations, a whole load at various sites throughout the world too.
And yet, Legends mode tells me that "Quelge Lastwarns the Impunity of Warmth was a troll. He was one of the only ones of his kind".
How can this be?

This looks like it might be a vestige of the historical-figure-only days, before we had entity populations.  It uses "first of" only if there are historical figures of that kind that have been born after history starts.  So I guess it didn't have time to elevate any born ones later in gen?  Trolls can live a thousand years, so it probably takes a while for elevated trolls to get positive birth years.

[Shonai_Dweller]

Ah, that would probably explain it. The blizzard men and gorlaks I checked were mixed, some born before time, others later so got either "first of its kind" and "unknown parentage".

Will bug report it for future reference. Only troll apart from the million other trolls isn't right. 

[IndigoFenix]

Is it possible to create experimenters that don't make the mutant freaks, or do EXPERIMENT interactions require the NIGHT_CREATURE_EXPERIMENTER tag?

Not currently.  I also posted over on your bug report a bit about the particular issue with the mod.

Ah, thanks, I saw the comment and was able to fix it.

I guess that even generated necromancers very rarely create ghouls on a per-case basis, the only reason why some end up doing so is because there are so many generated necromancer secrets in each world.  Meaning that any given modded secret has a very low chance of actually seeing results in a given world.  A tag to control this chance would be nice, though in the meantime I guess I'll try making a lot of copies of the secret and see if that helps.

Would it be possible to create a secret that frequently resulted in experiments, by modifying the unit's personality in some way?  What about a secret that had a greater chance of being passed on through writing or apprenticeships?

Having a historical one in an attacking army that destroys/takes over a site should do it 50% of the time (more if they can animate too).  The site take over/destroy guarantees them a place to put the experiments.  Setting up an administrator doesn't do it.

Good to know.  Now the next thing to figure out is how make a unit more likely to join an attacking army...:hmm:

A more in-depth explanation of how personalities and values impact unit behavior overall, especially in worldgen, would be nice.  We've had the values and the ability to modify them for ages, but apart from speculation very little is actually known about what they do.  I've been using DFHack scripting to scan average personality trends of units that have been involved in certain events, but so far I've found no statistically significant impact of personalities or values...but generated necromancers have modified personalities, which suggests that personalities do something.

[FantasticDorf]

A more in-depth explanation of how personalities and values impact unit behavior overall, especially in worldgen, would be nice.  We've had the values and the ability to modify them for ages, but apart from speculation very little is actually known about what they do.  I've been using DFHack scripting to scan average personality trends of units that have been involved in certain events, but so far I've found no statistically significant impact of personalities or values...but generated necromancers have modified personalities, which suggests that personalities do something.

That's kind of bypassing though the acknowledgement that they are transformed beings, who used to have lives & personalities of their own and whatever the tower's own hardcoded faction has in place of upheld values. Currently in one of my fortresses there is a necromancer occupied (not destroyed) hamlet populated by precisely >1 unit ready to be investigated if i find any resistance with its own site-government and everything seperate to the tower.

[clinodev]

Admittedly not 47.03 release specific, but it would be great if we could get notifications for these release threads turned on, as with for instance the Future of the Fortress threads. Maybe I'm the only one who still uses email, but it's way handier than hitting f5 every couple hours. 

[iceball3]

With bug http://www.bay12games.com/dwarves/mantisbt/view.php?id=7444 fixed, does it also simultaneously fix the issue with Marksdwarves dodging through fortifications or is it a different cause for that bug?

A quick test also shows that fortifications can still be leapt through, for any dwarves trying to path through them for combat.

[Shonai_Dweller]

With bug http://www.bay12games.com/dwarves/mantisbt/view.php?id=7444 fixed, does it also simultaneously fix the issue with Marksdwarves dodging through fortifications or is it a different cause for that bug?

A quick test also shows that fortifications can still be leapt through, for any dwarves trying to path through them for combat.

Are you sure they're not being leaped over?
Not one of my marksdwarves in the last 5 sieges has "leaped through" the fortification walls of his bunker (engraved walls, not built fortifications in case it makes a difference).

[iceball3]

With bug http://www.bay12games.com/dwarves/mantisbt/view.php?id=7444 fixed, does it also simultaneously fix the issue with Marksdwarves dodging through fortifications or is it a different cause for that bug?

A quick test also shows that fortifications can still be leapt through, for any dwarves trying to path through them for combat.

Are you sure they're not being leaped over?
Not one of my marksdwarves in the last 5 sieges has "leaped through" the fortification walls of his bunker (engraved walls, not built fortifications in case it makes a difference).

It might. The specific test case I did was adventure-style control in arena: manually jumping into a fortification allows you to bypass it entirely. I wasn't able to the object arena AI to go through the fortification without climbing up a higher Z level yet, though. Will be testing a bit more of that here and there.

A quick test of pelting a fat stack of grand master dodgers into a fortified pillbox with little space to dodge, with a few thousand bolts, seems to indicate they won't dodge out of fortifications anymore, maybe.

[Shonai_Dweller]

Oh that was close, the damn tavern keeper just drunk himself into a nauseous heap again. This one didn't die fortunately. That's the second over indulging tavern keeper I've had in 47.x

Good proof that you don't need a tavern keeper to die of alcohol poisoning. His last thoughts would have been annoyance at having a drink without a goblet...

[Loci]

Good proof that you don't need a tavern keeper to die of alcohol poisoning.

Except, you know, for the part where you had a tavernkeeper both times. If you want to prove that, try disabling all your tavernkeepers and see how many of your dwarves drink themselves to oblivion. I'm around 0/200 so far, so your 2/200? is either extremely poor luck or proof of the exact opposite.

[IndigoFenix]

A more in-depth explanation of how personalities and values impact unit behavior overall, especially in worldgen, would be nice.  We've had the values and the ability to modify them for ages, but apart from speculation very little is actually known about what they do.  I've been using DFHack scripting to scan average personality trends of units that have been involved in certain events, but so far I've found no statistically significant impact of personalities or values...but generated necromancers have modified personalities, which suggests that personalities do something.

That's kind of bypassing though the acknowledgement that they are transformed beings, who used to have lives & personalities of their own and whatever the tower's own hardcoded faction has in place of upheld values. Currently in one of my fortresses there is a necromancer occupied (not destroyed) hamlet populated by precisely >1 unit ready to be investigated if i find any resistance with its own site-government and everything seperate to the tower.

It's not the best method, I'll admit, especially since worldgen seems to unfold through butterfly-effect-like event-based chaos.  I was hoping there would be something obvious, like position holders having above-average AMBITION or PRIDE that motivated them to reach that position or something, or book writers placing greater value on KNOWLEDGE, but a census of almost every significant worldgen role I could think of (position holders, necromancers, authors, army leaders, necromancers who raised ghouls, etc.) revealed an almost perfectly average overall distribution of personalities and values.  Of course, if there's a more complex process involving the interactions of multiple units and every personality trait had potential to cause a unit to ascent to power in particular circumstances, finding global averages wouldn't reveal these.  Sadly the details of how particular individuals rise to power, apart from the fact that it probably is related to social skills, still is very much a black box.

There is almost definitely some effect of personality and values when scanning an individual's deeds once they achieve power - there's a notorious law-giver I've examined who had been involved in ordering a number of religious persecutions and wars, and sure enough has low TOLERANT and high VIOLENT and VENGEFUL values.  But things like that are just anecdotal evidence - which traits are actually causing which effects?  What is the point of making all necromancers have lower TRUST and higher ANXIETY_PROPENSITY, and does it have anything to do with their tendency to wind up in positions of power?  Or are these events just random and I'm paying attention to the details that make sense while ignoring the ones that don't?  Without a better way of scienceing out general trends, or a statement from Toady explaining what all of these personality traits and values do, it's all just guesswork.

I crave data for my horrible experiments!