Print

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

[greycat]

• Dwarf population will be capped at 20 to prevent the issues with migration experienced in the first round.  The hope is that this bug is not affected by dwarf population (it does not appear to be).

If you set the STRICT_POPULATION_CAP to 7, that should prevent all dwarf immigrations and births (unless one of the starting 7 dies).  I'm not 100% positive that will prevent domestic animal immigrations, but a pet sheep arriving without any dwarves would strike me as a bug.

[Tacomagic]

I set both caps to 20.  I want some immigration to test whether immigrant annimals are effected by the same bug.  Since I'm working with such a low population cap and 4 groups of annimals, hopes are that I have enough bases covered to test both "uncontaminated" embark animals and those that immigrate into standing populations.

The issues I was talking about mostly had to do with that vampire which showed up in the big spring migration.  That guy set my population on track for a very nasty tantrum spiral.  Ideally I'd like the populatin content enough to continue using the same experimental setup if I need to run it longer or want to run additional trials using the same embark.

But, if I get too much contamination, I'll definitely tweek down the numer to 7 to assure the purity of the sample.

EDIT:  Now that I think about it, it's an init.ini setting so I should be able to set it initially to 7, and then scale it up to 20 when I want to start allowing immigration testing.  That's actually a much better way to go about testing.  I get my pure initial sample and then can introduce the immigrants and see what happens.  Good suggestion!

[Button]

Yes, that's what I read, but when I did [0:0:1] for each caste, I ended up with 0 romantic relationships, which has been confirmed by Quietust (not a problem in vanilla, is a problem with altered ORIENTATION tags), so there is probably a bug in there somewhere, but it needs to be narrowed down somehow as to where exactly the bug is.

For anyone interested, the ORIENTATION bug report on mantis can be found here: http://www.bay12games.com/dwarves/mantisbt/view.php?id=7905

-Dame de la Licorne

Madame de la Licorne, do you know if anyone has tested if the 0s are the issue here, as was suggested on that ticket? With for example [ORIENTATION:MALE:1:1:999]. If no one has, I may make that my next effort.

[Tacomagic]

Yes, that's what I read, but when I did [0:0:1] for each caste, I ended up with 0 romantic relationships, which has been confirmed by Quietust (not a problem in vanilla, is a problem with altered ORIENTATION tags), so there is probably a bug in there somewhere, but it needs to be narrowed down somehow as to where exactly the bug is.

For anyone interested, the ORIENTATION bug report on mantis can be found here: http://www.bay12games.com/dwarves/mantisbt/view.php?id=7905

-Dame de la Licorne

Madame de la Licorne, do you know if anyone has tested if the 0s are the issue here, as was suggested on that ticket? With for example [ORIENTATION:MALE:1:1:999]. If no one has, I may make that my next effort.

If 1:1:999 works for Dwarves, let me know.  If I can narrow down the bug in sterility to being set per-creature rather than per species, then my next move would be to add explicit orientation tags to the domestic animals to see if that fixes the problem.

[Dame de la Licorne]

Hiya,

Madame de la Licorne, do you know if anyone has tested if the 0s are the issue here, as was suggested on that ticket? With for example [ORIENTATION:MALE:1:1:999]. If no one has, I may make that my next effort.

That's what I was planning to run tests on, but as I said before, I'm on vacation and won't have access to my DF computer until next week (~Aug 20th-ish), so if you want to run tests feel free!  I can always corroborate/fill in gaps/do additional tests when I do get back to DF, depending on what needs doing to pinpoint this bug. 

If 1:1:999 works for Dwarves, let me know.  If I can narrow down the bug in sterility to being set per-creature rather than per species, then my next move would be to add explicit orientation tags to the domestic animals to see if that fixes the problem.

Great minds think alike.

-Dame de la Licorne

[ToadChild]

I set both caps to 20.  I want some immigration to test whether immigrant annimals are effected by the same bug.  Since I'm working with such a low population cap and 4 groups of annimals, hopes are that I have enough bases covered to test both "uncontaminated" embark animals and those that immigrate into standing populations.

The issues I was talking about mostly had to do with that vampire which showed up in the big spring migration.  That guy set my population on track for a very nasty tantrum spiral.  Ideally I'd like the populatin content enough to continue using the same experimental setup if I need to run it longer or want to run additional trials using the same embark.

But, if I get too much contamination, I'll definitely tweek down the numer to 7 to assure the purity of the sample.

EDIT:  Now that I think about it, it's an init.ini setting so I should be able to set it initially to 7, and then scale it up to 20 when I want to start allowing immigration testing.  That's actually a much better way to go about testing.  I get my pure initial sample and then can introduce the immigrants and see what happens.  Good suggestion!

I believe you can turn off vampiric curses in world gen.

[Tacomagic]

I so rarely get vampires in my forts (usually because I actually WANT a vampire) that having one show up when I didn't want one came as a surprise.

But that's the way the RNG works.  Don't see something until you don't actually want it.

The RNG is a harsh mistress.

[Tacomagic]

So far the second test is looking promising.  I'm about six months in and it's all but confirmed in my mind that full population sterility is caused by sterile males.  With 4 males and 8 females per group, I have no group that is entirely sterile (but I'm still seeing individual sterile animals).

I'm not going to make any conclusions yet since I've still got 18 months of dorf testing, but it's looking like I'm going to have some good data to make a few guesses at what is going on.

Unfortunately I forgot to lock the doors to the peahens hatching rooms, so their data isn't in yet.

So far the take-home advice I have for anyone trying to raise a domestic animal population is bring at least 2 of each gender if you want them to breed.  3 each would give better certainty of fertility, but 2 each had a very good chance of producing at least 1 fertile couple.

EDIT: OH GOD THE PEACHICKS!

*FPS DEATH*

[dwarf_reform]

http://www.bay12forums.com/smf/index.php?topic=140479.90 Some info on the orientation tag here (on page 7 if the link takes you to page 1..), hopefully it'll shed some light

[ImagoDeo]

The RNG is a harsh mistress.

Indeed.

EDIT: OH GOD THE PEACHICKS!

*FPS DEATH*

'Twas likely to happen sooner or later.

Is there any way to increase the population cap of individual animal types? A split of 20/20 M/F would provide better data to evaluate the 75/20/5 theory, but with only 10 additional slots for children, I doubt the numbers would line up.

Toady said something about the orientation tags a few weeks ago. Let me see... ah... this thread, page 10. (direct link to post)

And here is a good explanation of the mechanics behind orientation tags.

Spoiler: Full Quote (click to show/hide)

Here are my calculations. They might be a bit different.

First the game rolls a 100-sided (75+20+5) die for attraction to the same gender. If the result is 1-75 the dwarf is not interested in the same gender, if it's 76-95 the dwarf can get a lover of the same gender, if it's 96-100 the dwarf can get a spouse of the same gender.

Then it rolls another 100-sided die for attraction to the opposite gender, where the ratios are flipped.

This gives nine possible categories for what relationships the dwarf is willing to get into:

1. No homo + no hetero (3.8%)
2. No homo + hetero love, no marriage (15.0%)
3. No homo + hetero marriage (56.0%)
4. Homo love, no marriage + no hetero (1.0%)
5. Homo love, no marriage + hetero love, no marriage (4.0%)
6. Homo love, no marriage + hetero marriage (15.0%)
7. Homo marriage + no hetero (>0.0%)
8. Homo marriage + hetero love, no marriage (1.0%)
9. Homo marriage + hetero marriage (3.8%)

The total orientation ratios become:

3.8% of all dwarves are asexual
1.0% of all dwarves are homosexual
23.8% of all dwarves are bisexual
71.0% of all dwarves are heterosexual

(edited to use the actual numbers)

(Ignore the date. I had to manually quote the post since the original thread is now locked.)

[Tacomagic]

Luckily I eventually managed to get the peachicks wrangled into a cage and then queued up for butchering.

Before that, about 30 of them found a pet-impassable door and used it to attack my FPS.  Tanked it down to under 15 for a while.

I've finished with tonight's SCIENCE and will do the writeup in the morning; though the result is not much different at 2 years than it looked like at 6 months. I did find some extra information about breeding that I'll share in the report since I don't see it on the wiki anywhere.

Until tomorrow, lab partners!

[Dame de la Licorne]

Hiya,

(Spoilered for length)

Spoiler (click to show/hide)

The RNG is a harsh mistress.

Indeed.

EDIT: OH GOD THE PEACHICKS!

*FPS DEATH*

'Twas likely to happen sooner or later.

Is there any way to increase the population cap of individual animal types? A split of 20/20 M/F would provide better data to evaluate the 75/20/5 theory, but with only 10 additional slots for children, I doubt the numbers would line up.

Toady said something about the orientation tags a few weeks ago. Let me see... ah... this thread, page 10. (direct link to post)

And here is a good explanation of the mechanics behind orientation tags.
[spoiler=Full Quote]

Here are my calculations. They might be a bit different.

First the game rolls a 100-sided (75+20+5) die for attraction to the same gender. If the result is 1-75 the dwarf is not interested in the same gender, if it's 76-95 the dwarf can get a lover of the same gender, if it's 96-100 the dwarf can get a spouse of the same gender.

Then it rolls another 100-sided die for attraction to the opposite gender, where the ratios are flipped.

This gives nine possible categories for what relationships the dwarf is willing to get into:

1. No homo + no hetero (3.8%)
2. No homo + hetero love, no marriage (15.0%)
3. No homo + hetero marriage (56.0%)
4. Homo love, no marriage + no hetero (1.0%)
5. Homo love, no marriage + hetero love, no marriage (4.0%)
6. Homo love, no marriage + hetero marriage (15.0%)
7. Homo marriage + no hetero (>0.0%)
8. Homo marriage + hetero love, no marriage (1.0%)
9. Homo marriage + hetero marriage (3.8%)

The total orientation ratios become:

3.8% of all dwarves are asexual
1.0% of all dwarves are homosexual
23.8% of all dwarves are bisexual
71.0% of all dwarves are heterosexual

(edited to use the actual numbers)

(Ignore the date. I had to manually quote the post since the original thread is now locked.)

I was reading through the ORIENTATION part of the ethics thread (before the derail/flame war), and would have loved to see the results of Loud Whispers' tests.  However, at least part of ToadyOne's intent with the tag doesn't seem to work.  Given what he said here, and given that he's stated the values are: <caste>:<dis-interested probability>:<lover-only probability>:<marriage probability>; having the following tags: [ORIENTATION:MALE:0:0:100] and [ORIENTATION:FEMALE:100:0:0] for females, reversed for males, should result in only heterosexual relationships.  However, when I implemented said tags, I had NO romantic relationships.  Period.  No lovers, no marriages that didn't migrate in (Quietust confirmed this).  So something there is a bug too (also reported on Mantis), and one of the tests I'll do when I get back (assuming no one else beats me to it), is to see if that bug is related to the animal breeding bug.

Has anyone yet tried adding the ORIENTATION tag to an animal with custom values?

-Dame de la Licorne

[celem]

I dont follow what you tried there Dame.

Does each gender need both tags?  from what you posted it seems you tell males they want to marry females but are completely disinterested in males.  And then tell females the opposite, i thought each gender only got the one tag detailing their opinion toward the opposite sex, with one of the values then implying a homosexual preference.  I hadnt realised you could actually switch them off completely and remove their drive entirely, which of course the pair would allow.

We really use both for both?  This is made worse by the fact that I cant find the orientation tags anywhere, where do they live?  cant see em in entity or creature for dwarves and I havent really modded since before this tag.

[Dame de la Licorne]

Celem,

As I understand it, each caste needs to know how it reacts to its own caste, AND to each of the other castes.  Therefore, a male caste needs one ORIENTATION tag for MALE:#:#:# (how does a male react to another male) and one ORIENTATION tag for FEMALE:#:#:# (how does a male react to a female), and the female caste needs equivalent tags (with a different sequence of numbers, obviously).

This is based on ToadyOne saying that the tag works as follows: [ORIENTATION:<caste>:<probability uninterested in stated caste>:<probability interested in lovers but not marriage with stated caste>:<probability interested in lovers-then-marriage with stated caste>] (paraphrased) according to his posts here and here.  I believe that if a caste is missing one of the ORIENTATION tags (e.g. you only added custom numbers for [ORIENTATION:MALE] for females), the game uses the hidden default tag (you won't see it in the creature's raw entry unless you specifically want to change the numbers, in which case you "add" it in), so some females may still be homosexual/asexual.

Of course, my understanding may be wrong, given that I haven't been able to get my dwarves to act 100% heterosexually.  Which is why I need to do some in-depth tests once I get back to my DF computer (this one runs at ~50FPS on embark in a pocket world, and drops to ~20 after the first migrant wave, by the first winter it's down to ~5FPS and completely unplayable from a testing standpoint, let alone a real fort).

The point of my previous post was to wonder if we can't add ORIENTATION tags for animals to see if that fixes the breeding issues (at the very least it would give us some more information), but to do that we first have to understand exactly how to make the tags work for dwarves.

-Dame de la Licorne

Edited for clarity.

[celem]

Ahh, I follow your thinking now.  I was working on the opposite assumption that each caste had 1 tag detailing preference to the opposite gender, with the disinterest number then somehow generating a homosexual/asexual chance, but that second mechanism would then be hidden and mysterious since we run out of visible numbers.  While possible this is probably unlikely, in general toady has been drifting from hardcode toward raw for years.

The dual tags makes more logical sense as it allows better specification of what you want from the creature's behaviour.

Given that we arrive at an understanding of these tags then yes, they might have an effect on animals, we could indeed be seeing in animals the results of all default orientation tagging (due to absence).  With you now.

I'll dig around and find where the blasted tags are so I can have a proper look.