Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 [2]

Author Topic: stone detailing needs to be much faster.  (Read 2613 times)

NW_Kohaku

  • Bay Watcher
  • [ETHIC:SCIENCE_FOR_FUN: REQUIRED]
    • View Profile
Re: stone detailing needs to be much faster.
« Reply #15 on: February 19, 2010, 10:02:58 pm »

Constructions don't build experience (unless it is something that requires architecture, like bridges, but walls don't do it), so you're actually killing your XP gain with megaprojects most of the time.

(Also, engraving serves a very good purpose early on - an egraved dining hall massively cuts down on unhappiness.  Even if your statues are made of nothing but microcline, a statue garden in a fully engraved hall can easily have enough value to give strong happy thoughts.)

I only have one legendary engraver because he's the only one who I've let engrave.  The new guy is taking absurdly long to get up his first five levels, but once you get it that far, it's very fast, and once you hit legendary, it's probably the fastest XP gain you can do.

This is because the experience you gain is fixed - it's probably 30 xp per tile, although I haven't explicitly checked that.  However, it takes a couple game days for a low-level engraver to get that job done (and they often cancel to get a drink), so they get a very tiny amount of experience... this, incidentally, makes it a poor job for off-duty military if you only want a quick layover, better to use bookkeeping for that.  If, however, you actually manage to clock the several year it takes of constant engraving to get past the first few years, the rest of the levels are significantly quicker - the time it takes to complete the job swiftly approaches zero.  This means that, although there is a sub-geometric growth rate in the requried experience to gain levels, your growth in terms of how much XP you can acquire per given period of time effectively increases what appears to be geometrically.  This makes engraving possibly the fastest attribute-building skill, and as the only bottleneck on engraving time is movement speed, the more attribute gains you have, the more agility you get, the faster you even move! 

As a cherry on top of all this, you can do what I did - make one of your initial miners your initial engraver.  When he reaches legendary status in both, he will be a virtual superdwarf in stats, and be capable of transmuting mountainsides into long tunnels and huge collections of stones with a wave of the hand, then go on to smooth the whole cavern on the trip back, and, if you don't expect to ever need to knock down any of the walls or dig channels through the floors, you can send him back once more for engraving the whole thing.  That's likely 90 xp per tile for a job that gets done almost instantaniously.

This is why you need to bring the legendaries back down a bit, and dabblers should probably be made a bit faster.  The disparity is too extreme, to the point where the higher levels are, contrary to the sub-geometric experience requirement growth, actually much easier to obtain than the lower levels.
Logged
Personally, I like [DF] because after climbing the damned learning cliff, I'm too elitist to consider not liking it.
"And no Frankenstein-esque body part stitching?"
"Not yet"

Improved Farming
Class Warfare

profit

  • Bay Watcher
  • Finely Crafted Engravings... Or it didn't happen.
    • View Profile
Re: stone detailing needs to be much faster.
« Reply #16 on: February 20, 2010, 11:25:10 am »

I am afraid I do have to agree... The mining and engraving times of dwarfs who are dabbling and proficient seems to be about dead on, but once they reach about competent.... They simply do the task way too fast.

I think there should be a little alteration done on the speed gains given for better skills.

Right now if my eyes do not fail me... Currently if Dabbling is a dwarf running at 100% of speed, then a legendary+5 dwarf would be running at approximately 10,000%  Speed. A hundred fold increase is kinda extreme.

Perhaps lowering the speedup rate of dwarves will go a long way to making this more balanced...

Lets say in this fictitious fix,  a dabbler smoothss at 100% and a Legendary+5 will smooth at 300% (of course scaling smoothly from one step to the next) He will still be 3x as fast as the dabbler, AND when he chooses to do engraving, it will on average be significantly more valuable... But the disparity between the two sides will be reduced.

Then with a baseline to work with, and a semi-stable time frame as a point of reference, we can tweak the time a smoothing job requires.
Logged
Mods and the best utilities for dwarf fortress
Community Mods and utilities thread.

Randall Octagonapus

  • Bay Watcher
  • We gota get some medicine for the computer machine
    • View Profile
Re: stone detailing needs to be much faster.
« Reply #17 on: February 20, 2010, 04:30:37 pm »

I remember toady saying that a legendary engraver can smooth 25 tiles in the time it takes a dabbling engraver to smooth 1
Logged
would you like a slice of one humped camel cheese or two humped camel cheese on that elf burger?

Impaler[WrG]

  • Bay Watcher
  • Khazad Project Leader
    • View Profile
Re: stone detailing needs to be much faster.
« Reply #18 on: February 21, 2010, 02:27:57 am »

I completely concur Kohaku and I think I have a method that would be flexibly enough to do what you describe along with a great variety of other speed profiles.

The time necessary to complete a task should follow an exponential decay based on level plus a small minimum value.  In other words the Dabblers take a very long time to complete a task but the speed increases very rapidly in the first few levels then starts to level off as the minimum value begins to dominate. 

A good way to think of this is to say that when a task is performed their is a minimum dwarvenly-possible speed in which it can be done if it is done with absolutely no wasted time or effort, The Dabbler naturally wastes a huge amount of time quite literally 'dabbling' and as skill increases this wasted time is eliminated, each skill level gain reduces the wasted time by a percentage but it can never touch the minimum time.  If a dwarfs stats are going to be considered then it acts to reduce the minimum time component, again imagine a blacksmith hammering Iron when the Iron is actually being hit a strong dwarf gets more done with each strike but the dabbling blacksmith will still waste a lot of time and their stats will not make much difference until their skill improves markedly.

For example lets say the minimum is 10 and the starting exponential decay value is 1000 for the Dabbler and the decay is 50% per level.  The time units to perform the task per level are...

Dabbling         10 + 1000 = 1010
Novice         10 + 500   =  510
No label         10 + 250   =  260
Competent      10 + 125   =  135
Skilled         10 + 62     =   72
Proficient         10 + 31     =   41
Talented         10 + 15     =   25
Adept         10 + 7      =    17
Expert         10 + 3      =    13
Professional    10 + 1      =    11
Accomplished  10 + 0      =    10 (for all higher levels as well)

This is of course a very extremely rapidly speed decay reflecting roughly ware the game is now. By varying the three values which I call the Minimum (10), Waste (1000), and Learning (.50) you can very finely control both the speed of tasks but also the rate of improvement.  So for example a better progression might be produced by Min-200/Waste-800/Learn-.30.  Generally the more sophisticated & complex a task is the higher the Waste value should be relative to the Minimum value along with a low learning value to reflect the need for reflection/examination/memorization when learning the complex task.  A simpler task would be the reverse so that it more quickly decays to that minimum time.
« Last Edit: February 21, 2010, 02:49:28 am by Impaler[WrG] »
Logged
Khazad the Isometric Fortress Engine
Extract forts from DF, load and save them to file and view them in full 3D

Khazad Home Thread
Khazad v0.0.5 Download
Pages: 1 [2]