Print

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

[Malroc The Valiant]

I feel like I've seen a similar suggestion to this before, but after some googling I couldn't find it, so I'll make it again if someone else already has. I think I'd be cool if you could assign a dwarf with a legendary (or master I guess) crafting skill an apprentice. I find it slightly annoying to be training up a second weapon smith and have all their shitty weapon in with my high quality ones, and even more annoying micromanaging steal n' candy weapon production. The apprentice could stand next to the master while they work, boosting the master's work speed as the apprentices skill increases (I'd be fine with it increasing slower than normal). The apprentice could also fetch things for the master. The apprentice could automatically switch to working on their own once they were a high enough level.

I swear I've seen basically the same suggestion somewhere before. Maybe some old df talk where Toady was talking about having children be apprentices? I don't know.

[Starver]

I'm sure it has been suggested/discussed fairly recently, but slightly different from your balancing features.

My suggestion as to implementing your version might be to make an apprentice/master pair a 'unit worker', in effect, at least during workshop job assignment (perhaps guild-based activities, could expand to meal-taking, not sure it should extend to sleeping) where if both are available, they are assigned as a pair to the workshop(/wherever). The apprentice takes up the hauling duties (can go wandering for the next material just before the prior item is finished), the master stays working. The time the former spends 'idle' in the workshop counts towards skill-education, as per demonstrative learning in other contexts.

It could be an expansion of the "attend a meeting" code, limited to when a given Relationship entry (1:1 only, allowed?) combined with availability (immediate, or option to abort a 'lesser' task) allows the grouping to activate. The relationship can come from Overseer activation, perhaps, with only the requirement that the expertise of the master be above a certain threshold, but the rest must happen organically.

I'm wondering whether to suggest the possibility that, at a certain point and balance of master/apprentice skills, the latter can raise their skills (by chance) to higher than their still-developing leader. I'd quite enjoy the (slight) chance of an Announcement that "...the apprentice has now become the master."

[Malroc The Valiant]

Ha, I love it, that would be great. Depending on the personality of the master they could get a happy or negative thought. "Urist Hammerwave was jealous of their former apprentices accomplishments"

[Pillbo]

I think it would make more sense if the master worked slower with an apprentice. Teaching is a burden, so trade the work speed of a master to boost skill of an apprentice. Seems like a more balanced and realistic deal.

It could train the master's teacher skill and in turn increase their speed with apprentices.

[Shonai_Dweller]

I think it would make more sense if the master worked slower with an apprentice. Teaching is a burden, so trade the work speed of a master to boost skill of an apprentice. Seems like a more balanced and realistic deal.

It could train the master's teacher skill and in turn increase their speed with apprentices.

Depends on how complex the task is. Having someone to run around the workshop fetching things, doing the mundane tasks as they learn them can speed them up. That's part of the reason masters take on apprentices, after all.

[Azerty]

Apprenticeship should be managed by guilds, who would accept apprentices they thought to be able to be taught (for exemple, blinds would never be accepted in a painting guild and illiterates would have some issues being accepted in some guilds) and then would sort them to various established guildmembers to be taught.

[Red Diamond]

Apprenticeship should be managed by guilds, who would accept apprentices they thought to be able to be taught (for exemple, blinds would never be accepted in a painting guild and illiterates would have some issues being accepted in some guilds) and then would sort them to various established guildmembers to be taught.

There is no strict requirement this be so.  No actual reason an individual could not take on an apprentice without any guilds being involved. 

[Starver]

Thinking about the speed advantage/penalty:

While apprentice is present, work progress and master-skill improvement is is slowed directly proportional to the apprentice-skill gain. i.e. a random amount of skill-points normally due to the crafter[1] are assigned instead to the assistant, and a relative number of 'ticks to completion' are added to the time it will take (or 'ticks progressed' negated, however it is best modelled). This represents the expert going through a technique they know well with care and attention to demonstrate, at the expense of some personal opportunity for discovering a new finesse.

Spoiler: Detail... (click to show/hide)

If it were me, I'd use an S-curve function to the the way the apprentice goes from zero to master's level (a moving target, but here 'proportion of target') with non-zero gradients at each end (always possible to learn a little as no-skill novice, always possible to go above the target when got to equal skill... for aforementioned reasons, best learning rate when not completely clueless nor lmost up-to-speed) and maximum gradient in the centre, or (in derivative terms, which is how it would he calculated, per-instant) effectively a standard distribution of gain, pegged to maximum (near, but not 100% benefit-leaching in the centre-X, both tails located at "round down to non-zero" values at the 0..complete range of X). (Dis)aptitude to the skill could shift this, somehow. Perhaps the effective Teaching skill of the trainer (though I was thinking more along the lines of greater Teaching aptitude mitigating the trainer's skill-gain loss, only, like only a frac^teaching of the skill granted to the learner is lost by the trainer).

This is meant to reflect the totally clueless assistant not worth much time taken to explain, and the onebclose to 'graduation' not needing/benefitting much from it, but a half-decent apprentice has earned some attention.


OTOH, some might argue the proportion-deferred-to-learner should go from 100% at utter novice to (nearly?) zero at utter-equal (straight gradient between?) because of the near constant hand-holding required at the start, or something.

A compromise could be the mean of the two lines (normal dist and negative gradient) giving a downward (plateau-ish centred) S-curve of derivative gain (path of XP from zero to max therefore being a sort of 'r' curve). Or perhaps master-XP 'lost' works as the -1ish gradient but Apprentice-XP gain as the normal dist, with different amounts of adjustment due to each party. But that needs to be checked not to give more net skill-gain than without an apprenticeship relationship (even/especially after any Teaching Ability adjustments).


These graphs are all in my head, right now. I imagine that nobody cares for me to put them into graphics, to better explain the words, but I could if anyone is bothered...

But when the Apprentice is not present, the master works as normal, gaining personal XP to the skill as if without (nothing skill-related going to the student-worker). The clincher being that the assistant (when free to Assist), the closer in skill to their tutor they are, will anticipate the next material needed in the job-queue/R in advance of task completion and go off and get it, to bring it back and *TSK* it into the given workshop(/zone?), so that they do that legwork and their senior does not have to, to speed up the transition time (would have wait, rather than get their own candidate material(s), if the assistant was still bringing the thing(s) back - could be a detriment if the strength and therefore haul-speed was sufficiently different between the two, or burrow/whatever differences made it delayed).  Lack of material-to-get would prevent any prefetch (message, as per "Job Cancelled, No MAT available", but no delay until prior work completed, by which time maybe a precursor stage had created/freed-up the required MAT.


(Simple Matter Of Coding, etc, although I'm trying to base this off modifications/expansions to the current code, insofar as it can be implied to be.)


[1] Includiing, if applicable, "skill overflow" points that already do nothing to raise them above Guru-level expertise, if they're basically a Yoda-of-the-Forge, or similar.

[Aelwen]

I think children should be able to help with hauling things around and bringing food and water. Simple tasks. Gathering fruit, for example.

Not babies but kids starting with 4? years old.

[Starver]

Well, at least until child labour laws (or societal norms) don't reach 19thC levels, maybe...

(Familial apprenticeship, i.e. a spare pair of hands for the craftsman that can be shown how to do the fiddly bits but nothing too dangerous or strenuous, would be the obvious first step, before actually sending the kids down somebody else's mine to push/pull wagons. I don't think dwarves are 'industrialised' to the extent Britain was during the dark-ages of child labour, and long may they not be. Use a baby as a shield, yes, but work them to exhaustion in a dark hole, no...)