Print

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

[thistleknot]

well, I disagree whole heartedly.

Yes, a legendary dwarf does build faster.  However, in my head.  A dwarf who has higher strength when masonry uses strength http://dwarffortresswiki.org/index.php/v0.31:Mason to do masonry, I imagine the higher the strength, the faster it goes (I have no reference for this other than the wiki has attributes for certain skills).

So I could have a legendary person doing the job, or I could have a BETTER dwarf in terms of mason attributes doing the job as well in hopes that he would be a better fit when he becomes legendary.

I tend to assume that certain traits, which control a dwarf's behavior fit to certain jobs.  CMD was a great example.  However, there are other labors, such as feed prisoner that relies on traits as well.

Now this is all subjective.  And there is a solution for it if you don't like it.  Disable the weights.  But I disagree that it is misleading.  Because if Dwarf A has better attributes than legendary Dwarf B, then I can train Dwarf A up to be legendary.  Traits affect a dwarf's thoughts, that can not be modified in game, hence there inclusion.

Now with legendary, just because a dwarf is legendary in a skill, DOESN'T mean he likes the job.

I think "satisfied with work lately" might have something to do with preferences as well as traits.

we know some attributes train skills faster & some traits give dwarf's happy thoughts from doing that job.

Where is this information? Wiki is useless as usual and I haven't been able to find any conclusive research on the forum.

see wiki
http://dwarffortresswiki.org/index.php/DF2012:Skill

    Traits:
        cannot be modified in-game.Verify
        affect which social skills gain experience (if the dwarf has X trait it will not gain experience in X skill) at all.
        have other in-game effects that can be useful for certain professions.
        give thoughts when performing certain activities.

No. A way to search for a particular trait or material / item preference or sort by them would be useful. Forcing bogus "Related Roles" based on some arbitrary weights is counter-productive. Those values mean nothing and people trust them, so they are misleading. I was mislead by them!

For your CMD example, at the very least you want those weights set per-skill and not globally, but a way to sort the list of dwarves by "finds helping others very rewarding" and "devoted to duty" would be better.

that used to be the way preferences were done, via built in filter.

[Maklak]

wow you're really going off on this stuff eh?

Yes, I tend to feel strongly and overreact about some things. Especially when they are wrong and it is so obvious that they shouldn't be wrong in the first place. In case of Dwarf Therapist I respect your work on maintaining it, but not the decision to port the roles from dwarven guidance councillor and make them so important.

alright so first, turn off the role sorting in the roles tab and see if that helps.

I deselected "Include role ratings when sorting labour columns" and "Include role ratings when sorting skill columns" and it didn't help. Well, after changing the weights it works slightly better.

if you've got any suggestions for changes to the default roles, by all means let me know and i'll make adjustments.

Other than gutting it out completely because the system is inherently wrong and adjustments won't help much? Well, these are some ideas, but they will make comparing suitability to different roles impossible:

* For most civilian labours, like Masonry use just XP and skill rates, I've already given you the formula.
* For stone detailing creativity may be important, but not to me.
* For military I guess at least average physical stats and recuperation and disease resistance are important, but I just don't see a way of making a weighted average based on them. Different things are important to different people.

My way of doing this would be:
* Assume that using a skill gives constant amount of XP, regardless of stats and traits and sort all skills and labours in grid view by XP and learning rates only.
* Include a sortable grid view with stats. (The one we have works for sorting, but not displaying stats.)
* Include a sortable grid view with traits. That way people can sort by helpfulness, choose a dutiful dwarf near the top of the list and make her the CMD.
Give warning to any tooltips with the roles that the values there are arbitrary and should not be relied upon.

what exactly is your problem with roles that makes them so bogus? is it the default weights? the calculations?

I already explained that you cannot just make a weighted arithmetic average out of skills and stats, because it is meaningless. Even if you could, the weights are unknown. For things like probability of a disease based on blood pressure, temperature, weight, age and so on you can use optimisation or data mining to find something that works, but here there doesn't seem to be any data, just people and their preferences. The entire way of thinking that lead to introducing these roles is fundamentally wrong and based on wrong assumptions. A dead give-away was that you wanted to include skill rates with a weight, instead of tying them with current level of skill, as they no longer matter after Legendary +5.

you can find dwarfs by preference by using the preference window and clicking on a preference.

Lol, this is kinda fun for jokes. I never explored that part of DT, because I never needed it for anything.

[Maklak]

well, I disagree whole heartedly.

Likewise.

Yes, a legendary dwarf does build faster.  However, in my head.  A dwarf who has higher strength when masonry uses strength http://dwarffortresswiki.org/index.php/v0.31:Mason to do masonry, I imagine the higher the strength, the faster it goes (I have no reference for this other than the wiki has attributes for certain skills).

Keywords: "imagine" and "in your head". You have no evidence to back it up. I've seen skill levels do wonders to how fast something is done. Mining and cutting wood are the easiest to notice, but a legendary crafter will often barely enter the workshop when he is done. If stats have any influence on how fast the work is done, it is overshadowed by the skill anyway.

So I could have a legendary person doing the job, or I could have a BETTER dwarf in terms of mason attributes doing the job as well in hopes that he would be a better fit when he becomes legendary.

I can agree with you on this: a very strong and agile worker is better at the same skill level, because he walks faster both when and when not carrying heavy materials. Even if something is done slightly faster at legendary skill thanks to high stats, it still won't matter as much as say, efficient fort layout.

see wiki http://dwarffortresswiki.org/index.php/DF2012:Skill

You think I haven't seen it? This is useless. There are no formulas or numbers or links to research with them.

True, traits cannot be modified in game, influence which social skills can be learned (like liar) and can decide how likely a dwarf is to take a break or ignore a patient. The things like "likes tradition" can give good or bad thoughts when talking to nobles. So what? It only really matters for doctors and they are an exception rather than the rule.

Stats do matter for the military, I have to admit that much. Strength and Agility give better walking and swinging speed, Endurance means fighting longer before getting exhausted and Toughness, Recuperation and Disease Resistance increase resistance to harm. "Prone to anger" trait may lead to rages which can help too.

I think I see why we disagree. When selecting a Dwarf for a job, I think you ask "Which one of you will work the best what Legendary +5?" For me the question is "Which one of you will get to Legendary +5 the fastest and wasting the least materials?"

[thistleknot]

Keywords: "imagine" and "in your head". You have no evidence to back it up. I've seen skill levels do wonders to how fast something is done. Mining and cutting wood are the easiest to notice, but a legendary crafter will often barely enter the workshop when he is done. If stats have any influence on how fast the work is done, it is overshadowed by the skill anyway.

well, lets disagree then.  I see the wiki, the wiki has attributes for skills.  We used them.

But you can adjust their attribute weight, which you already have to compensate for this.

you acknowledge some attributes are useful for speed (such as walking).  I believe they are useful for the workshops as well.  I have no info to tell me they are not, I have info telling me they are used for skills on the wiki.

[Intrinsic]

Found a crash bug in 19.3. When i go to open the Skill Legend the 1st time the program just crashed, the 2nd time it openned the window but upon resizing it gets sort of corrupted, then i clicked the X to close it and bam another crash. Just now it didn't crash but again the window still got corrupted(not really corrupted, but i guess you'll see the issue when you try)

Edit: Just to say i don't know how long this bug has been present as i've been away for a while, the last version i had is 17 and that works fine.

[splinterz]

I deselected "Include role ratings when sorting labour columns" and "Include role ratings when sorting skill columns" and it didn't help. Well, after changing the weights it works slightly better.

if you have the role stuff off of the labor columns the weights don't affect anything. it's purely sorting on the rating and the skill rate. the problem is that it's combining those two elements, which you disagree with. i'll run some tests to compare with your suggested method and consider changing it, since i think it was you who asked for it in the first place.

hey i've got no problem gutting the default roles if people think they're downright awful. you're the first one to complain so vehemently, and i'm willing to look past the blatant insults to listen to what you've got to say.

For most civilian labours, like Masonry use just XP and skill rates, I've already given you the formula.

according to the wiki, these labors don't exclusively use only skill rates/xp. yes it's assumed that skills use the attributes associated with them. not every skill may be accurate. it's true that most, if not all the roles are based on wiki/forum information found because i've got no other data. if you're saying the wiki is full of shit, well i guess i'd like a few other people to back that up. again it sounds like you're have a major issue with the default roles right? have you overridden most of them, or you're just upset because you feel it misleads new users?

Include a sortable grid view with stats. (The one we have works for sorting, but not displaying stats.)

what stats aren't capable of sorting? do you mean attributes?

Include a sortable grid view with traits. That way people can sort by helpfulness, choose a dutiful dwarf near the top of the list and make her the CMD.

sure, this isn't an issue, you can already do this yourself. i can include a view in the default set though, that's a good idea.

I already explained that you cannot just make a weighted arithmetic average out of skills and stats, because it is meaningless. Even if you could, the weights are unknown. For things like probability of a disease based on blood pressure, temperature, weight, age and so on you can use optimisation or data mining to find something that works, but here there doesn't seem to be any data, just people and their preferences. The entire way of thinking that lead to introducing these roles is fundamentally wrong and based on wrong assumptions. A dead give-away was that you wanted to include skill rates with a weight, instead of tying them with current level of skill, as they no longer matter after Legendary +5.

you don't need to know the weights, the weights are whatever a user decides. the system doesn't care. you love your skills, go ahead, weigh them as 10x that of everything else. maybe you want to override that on a specific role, or leave skills out of the next role. what's so wrong with that? what's wrong with the design that a user should be able to decide what to value more for a particular job?

speaking of wrong assumptions, in the role calculations, the raw level and experience are never used. it's capped at legendary +5, since for a job in DF higher than 20 doesn't matter. they are tied to the current level of skill, they're just valued equally (which i think i clearly mentioned may need adjustment).

Give warning to any tooltips with the roles that the values there are arbitrary and should not be relied upon.

really now, if you can't compose yourself enough to have a decent discussion i'll be hard pressed to keep taking you seriously.

Edit:

I think I see why we disagree. When selecting a Dwarf for a job, I think you ask "Which one of you will work the best what Legendary +5?" For me the question is "Which one of you will get to Legendary +5 the fastest and wasting the least materials?"

i think this is key. you don't want attributes, traits or preferences in your roles at all. so is the system totally flawed in it's concept, or are you just unable to figure out how to make it work for you?

[Lycaeon]

Though I disagree with Maklak's argument of the concept, I do see his point about caste skill rates being weighted more highly than attributes and traits. Over the course of a fortress's lifetime, most dwarves, barring the rarity of strange moods, will reach legendary through sheer effort, repeating dozens of reactions to produce goods in order to gain enough experience. This is the most time-consuming portion of training a dwarf to legendary, and it is to this portion that skill rates have the highest contribution, and so should be duly considered when assigning unskilled dwarves belonging to a caste-system with differing skill rates.

Of course, attributes, traits, and preferences have significant roles as well, and so must also be weighted (Especially as the majority of DF games played have no differing skill rates). As such, perhaps an option to weigh/not weigh skill rates would serve as a compromise?

In the end, it is your effort splinterz, and so it is entirely up to you how you want Dwarf Therapist to work. Regardless of what happens, I will greatly appreciate the end result. Thank you!

[thistleknot]

I think I see why we disagree. When selecting a Dwarf for a job, I think you ask "Which one of you will work the best what Legendary +5?" For me the question is "Which one of you will get to Legendary +5 the fastest and wasting the least materials?"

yes.  That's why skill ratings (before being combined into a role rating) take into account the caste's skill rate learning.  I believe it's something like skill rating * % of skill increase rating.  Example, if a wood savant (not sure if wood savants learn carpentry faster) has a 130% increase in skill learning, than the skill % * 130% would be the new skill rating before being combined with attributes and traits and preferences % before being combined into an overall percent.

Again, I'm not entirely sure, but that was my general idea of it when talking with Splinterz.

[feelotraveller]

Have I missed the point?

Creativity (in my experience, repeatedly, but I have not kept hard numbers  leads to much higher quality works in Masons/Carpenters/Craftsdwarf's/Weaponsmiths etc., and Analytic Ability to higher quality meals and mechanisms.  At least for moderately skilled dwarfs.  Perhaps this effect disappears at Legendary+5?

I find attributes important, traits well worth considering and preferences 'can' be make or break.  Sure it is comparing plump helmets to prickle berries but that they are all factors in the equation of the best dwarf for the job seems beyond doubt.

[splinterz]

Though I disagree with Maklak's argument of the concept, I do see his point about caste skill rates being weighted more highly than attributes and traits. Over the course of a fortress's lifetime, most dwarves, barring the rarity of strange moods, will reach legendary through sheer effort, repeating dozens of reactions to produce goods in order to gain enough experience. This is the most time-consuming portion of training a dwarf to legendary, and it is to this portion that skill rates have the highest contribution, and so should be duly considered when assigning unskilled dwarves belonging to a caste-system with differing skill rates.

Of course, attributes, traits, and preferences have significant roles as well, and so must also be weighted (Especially as the majority of DF games played have no differing skill rates). As such, perhaps an option to weigh/not weigh skill rates would serve as a compromise?

In the end, it is your effort splinterz, and so it is entirely up to you how you want Dwarf Therapist to work. Regardless of what happens, I will greatly appreciate the end result. Thank you!

personally i agree completely with skills and their rates being worth more than attributes, traits and preferences. i usually set the skill weight very high and drop the others much lower to achieve this. the nice thing about applying the skill rates with a weight is it's easy to allow the user to set a weight themselves. the bad thing is that it adds more ui clutter, especially if the skill rate weight is allowed to be overridden for every skill! however Maklok's suggestion may give better results for the ratings, i need to test it out.

there's been a lot of effort (myself and others) put into the branch, but the direction is always community driven. regardless of the amount of tact when providing feedback, i still appreciate it and try to make whatever changes are necessary.

I think I see why we disagree. When selecting a Dwarf for a job, I think you ask "Which one of you will work the best what Legendary +5?" For me the question is "Which one of you will get to Legendary +5 the fastest and wasting the least materials?"

... skill rating * % of skill increase rating.  Example, if a wood savant (not sure if wood savants learn carpentry faster) has a 130% increase in skill learning, than the skill % * 130% would be the new skill rating...

no it's not doing that, and that's one of the major problems he's got. i used an equally weighted average between the skill experience and the skill rate, because i figured i wouldn't want a low level with high skill rate jumping up over a higher skill level with lower rate. this may not have been the best decision, which is fine! that's the reason there's patch notes and i asked for feedback, so it can be adjusted, because i don't get to play to know, i'm always coding

[Nkosi_SW]

Thanks for the latest build, splinterz.  Highest moodable skill now working beautifully for my dwarves with raw skill levels above 20.

Malak, I think you need to pull your head in a bit mate, as we say here in Australia.  Coming in to a thread demanding that the volunteer developer change his code to what you want, questioning why nobody has noticed this obvious bug before (and therefore implying everyone else in the thread is an idiot), and generally using confrontational, aggressive and overbearing language is no way to win friends and influence people.

The real answer to your demands isn't within unknown formulas buried in DF.  It's that everyone plays DF differently and therefore uses DT differently too.  The reason us poor idiots haven't noticed this "obvious bug" before you came along in all your brilliance and revealed it to us is that for some of us, it's not a bug at all.  When I play, I actually have the role skill weighting set to zero, and other roles weighted at various levels to achieve what I want in DF.  Others can do differently if they want.  We don't all do things your way.

Why not be more constructive to get things working the way you want, such as learning more about the program (you apparently weren't even aware about the Default Rating settings under options), modifying these and the settings in the config files, and then if if all that fails, politely suggesting some changes to the code that will still allow others to do things the way they want too?  If you still aren't happy with the outcome, well, you are quite free to go and fork the code for DT and make it work exactly the way you want.

[thistleknot]

i used an equally weighted average between the skill experience and the skill rate, because i figured i wouldn't want a low level with high skill rate jumping up over a higher skill level with lower rate

can you give an example?  This intrigues me, as I too wouldn't want a higher skilled dwarf to be outranked by a lower skilled dwarf with a higher learning rate.

In my eyes, two equally skilled dwarfs would be outranked by the higher learning rate

[Sutremaine]

I wouldn't mind a pre-made option for Raw XP in the same vein as the options for caste, migration wave, squad, etc. I liked the old behaviour because it allowed me to see who could be nudged from Dabbling general handydwarf to a Novice able to be taught the job as a trade (it just bugs me seeing old first-year skills hanging around at Dabbling. I wish Toady would make them rust away to nothing). Mostly this was used for Masonry, since it's one of those jobs that's both useful on everyone and better done by a skilled dwarf.

Speaking of workshop skill limits, have you made any headway on determining skill rust? There's pretty good evidence that at higher skill levels, rust doesn't work properly. The game doesn't display a (Rusty) note, but the skill is still treated as Dabbling for the purposes of workshop assignments. I don't know if this translates into quality loss the way 'proper' rust does, but it does make a mess of workshop profiles.

[splinterz]

i used an equally weighted average between the skill experience and the skill rate, because i figured i wouldn't want a low level with high skill rate jumping up over a higher skill level with lower rate

can you give an example?  This intrigues me, as I too wouldn't want a higher skilled dwarf to be outranked by a lower skilled dwarf with a higher learning rate.

In my eyes, two equally skilled dwarfs would be outranked by the higher learning rate

yeah i explained it poorly, it works more like a bonus now if it helps to think of it that way. since the rates can range from 0 to 500% i was tentative in applying it directly to the skill level/xp by doing something like current xp/29k * skill rate. you can easily have situations where a lower level with a high skill rate will rank higher than an almost maxed skill with a normal skill rate.

it's also important to note that what's used in the roles doesn't have to be used for the sorting. they're calculated separately.

currently if the levels/xp are equal the higher rate will end up with a bonus and thus a higher rating.

[splinterz]

I wouldn't mind a pre-made option for Raw XP in the same vein as the options for caste, migration wave, squad, etc. I liked the old behaviour because it allowed me to see who could be nudged from Dabbling general handydwarf to a Novice able to be taught the job as a trade (it just bugs me seeing old first-year skills hanging around at Dabbling. I wish Toady would make them rust away to nothing). Mostly this was used for Masonry, since it's one of those jobs that's both useful on everyone and better done by a skilled dwarf.

Speaking of workshop skill limits, have you made any headway on determining skill rust? There's pretty good evidence that at higher skill levels, rust doesn't work properly. The game doesn't display a (Rusty) note, but the skill is still treated as Dabbling for the purposes of workshop assignments. I don't know if this translates into quality loss the way 'proper' rust does, but it does make a mess of workshop profiles.

for the option were you talking about another way to group dwarves? did i break one of them?

as far as i can tell the rusting is totally broken. the only thing that seems to work is checking for skills actually degrading and losing xp. at higher levels i believe what happens is the rust level tops out. what it does is it increases a last used counter to the skills unused limit, then it increases a rust counter. once the rust counter hits it's max, it increases a demotion counter. once that hits it's max, it adds a 'layer' of rust.

i tested with fast rusting, and the rust on a skill won't go over 6 'layers'. when it attempts to, the tag is removed from the skill (rusty, v.rusty) as you described, and from that point on, rusting seems to be completely ignored. no new layers, no counters increase or anything.