v50 Steam/Premium information for editors
  • v50 information can now be added to pages in the main namespace. v0.47 information can still be found in the DF2014 namespace. See here for more details on the new versioning policy.
  • Use this page to report any issues related to the migration.
This notice may be cached—the current version can be found here.

Difference between revisions of "40d Talk:Item quality"

From Dwarf Fortress Wiki
Jump to navigation Jump to search
(Are preferences really more helpful with novices?)
m (moved Talk:Broken/40d\x3aItem quality to 40d Talk:Item quality: Fixing talk page name (352/738))
 
(39 intermediate revisions by 13 users not shown)
Line 7: Line 7:
 
:Curious.  Makes me wish Toady let out a little bit more info than he does about his game (we still don't know how combat works).  But then again, it adds another level to an already deep game.--[[User:Draco18s|Draco18s]] 17:39, 4 November 2007 (EST)
 
:Curious.  Makes me wish Toady let out a little bit more info than he does about his game (we still don't know how combat works).  But then again, it adds another level to an already deep game.--[[User:Draco18s|Draco18s]] 17:39, 4 November 2007 (EST)
  
 +
:The consoler skill allows a noble or whoever the dwarf talks to to calm the tantrum dwarf down. --[[User:ZombieRoboNinja|ZombieRoboNinja]] 03:08, 4 May 2009 (UTC)
 
== Trading masterpieces. ==
 
== Trading masterpieces. ==
  
Line 23: Line 24:
 
Wear is similar to quality, in that it adds a symbol notation to the item description - "x" "X" or "XX".  I would also be very surprised if it didn't impact the value and effectiveness of an item - in particular, dwarves notice when their clothes start to decay, and give unhappy thoughts.  Should we incorporate that material here, or just link to it? [[User:Doctorlucky|Doctorlucky]] 00:25, 19 November 2007 (EST)
 
Wear is similar to quality, in that it adds a symbol notation to the item description - "x" "X" or "XX".  I would also be very surprised if it didn't impact the value and effectiveness of an item - in particular, dwarves notice when their clothes start to decay, and give unhappy thoughts.  Should we incorporate that material here, or just link to it? [[User:Doctorlucky|Doctorlucky]] 00:25, 19 November 2007 (EST)
 
: And while we're at it, for completeness, things can also have decorations «», be imported () or forbidden {}.  Any other designations? [[User:Doctorlucky|Doctorlucky]] 00:32, 19 November 2007 (EST)
 
: And while we're at it, for completeness, things can also have decorations «», be imported () or forbidden {}.  Any other designations? [[User:Doctorlucky|Doctorlucky]] 00:32, 19 November 2007 (EST)
 +
: Any idea how wear and tear is calculated?  I've have a long-time miner whose clothes are still rather pristine but what my mayor, who did next to nothing, wore was tattered. --[[User:FJH|FJH]] 00:54 20 February 2009 (EST)
  
 
== Alcohol Quality ==
 
== Alcohol Quality ==
Line 73: Line 75:
  
 
:AFAIK, it's in memory the same quality as masterpiece. --[[User:Bartavelle|Bartavelle]] 09:16, 1 September 2008 (EDT)
 
:AFAIK, it's in memory the same quality as masterpiece. --[[User:Bartavelle|Bartavelle]] 09:16, 1 September 2008 (EDT)
 +
 +
::Can you provide concrete confirmation of that?  It would go a long ways towards...well, towards exactly this, which would help several pages. [[User:LegacyCWAL|LegacyCWAL]] 15:18, 11 December 2008 (EST)
  
 
==Two Quality Levels==
 
==Two Quality Levels==
Line 87: Line 91:
  
 
[[User:Gairabad|Gairabad]] 19:47, 6 December 2008 (EST)
 
[[User:Gairabad|Gairabad]] 19:47, 6 December 2008 (EST)
 +
 +
:Well, what's really needed is formal testing of a dwarf working on preferred materials and non-preferred.  Might just be one of those DF myths.--[[User:Maximus|Maximus]] 22:40, 6 December 2008 (EST)
 +
 +
::From the article: "Although the exact adjustment is unknown, preliminary testing indicates that they can boost effective skill (not quality) by one to three levels (if the craftdwarf likes the item type, subtype and material), and have thus greatest impact when the dwarf is still relatively unskilled."  If this is true, then you can derive from the "Jump" column that gaining subsequent levels does not bring diminishing returns.  [[User:Gairabad|Gairabad]] 13:24, 7 December 2008 (EST)
 +
 +
:::Sorry, I'm totally confused about what you're trying to say here.  Higher skill tends to produce higher-quality goods.  A preference for the materials being used supposedly also produces higher-quality goods.  What's your claim?--[[User:Maximus|Maximus]] 22:47, 7 December 2008 (EST)
 +
 +
::::I'm trying to say that this bold part is false:
 +
::::>Preferences for particular materials and objects make a considerable difference. Although the exact adjustment is unknown, preliminary testing indicates that they can boost effective skill (not quality) by one to three levels (if the craftdwarf likes the item type, subtype and material), and '''have thus greatest impact when the dwarf is still relatively unskilled'''.  [[User:Gairabad|Gairabad]] 23:53, 8 December 2008 (EST)
 +
 +
:::::Oh, okay.  Well, you could just drop that part of the sentence.  By some thinking, though, it may still be true: goods made by a competent artisan are 81% more valuable (on average) than those made by a dabbler, but goods made by a legend are only (5.8 / 4.8 =) 21% more valuable than those made by a master.
 +
:::::I still think the "jump in value" column is still of very limited use.  It shows a trend -- value peaks at each of the levels where a lower quality grade is no longer produced (Proficient, Accomplished, Legendary) -- but dwarves are gaining skill continuously, so it's more of a statistic than anything else.  A more interesting statistic is that from dabbling to talented, item values triple; from there to Legendary, they only double.  To put it more generally, skill increases produce large value increases at low levels, small increases at high levels.--[[User:Maximus|Maximus]] 12:19, 9 December 2008 (EST)
 +
 +
::::::"To put it more generally, skill increases produce large value increases at low levels, small increases at high levels."  No, they only produce larger ''percentage'' increases at low levels.  I find the jump column useful to know when to stop leveling up my dwarves and start actually working on stuff.  For example, Proficient and Accomplished are both good levels to start actually working.  [[User:Gairabad|Gairabad]] 13:03, 9 December 2008 (EST)
 +
 +
:::::::"Working on stuff" and "leveling up" generally happen simultaneously.  I guess if you're doing an armorsmith or something and don't want to make crap steel items... but I say make 'em when you need 'em, train continuously otherwise.--[[User:Maximus|Maximus]] 13:12, 9 December 2008 (EST)
 +
 +
::::::::Not in my experience.  If I want to level up a mason, I make blocks for constructions.  If I want to level up a carpenter, I make bins, barrels, and buckets.  With glassmaking, you can make raw glass.  Mechanics can start by using common stone and move on to economic stone.  Etc.  Often things that you "need" can be put off a bit.  For example, it takes a while for immigrants to get sleepy... plenty of time to level your carpenter so he can make higher-quality beds.  [[User:Gairabad|Gairabad]] 13:31, 9 December 2008 (EST)
 +
 +
::::From a fortress perspective, the expected increase in value for a level up is much more important than a percent increase. When a preference makes a novice act like a competent, it's not as good as one that makes an expert act accomplished. [[User:VengefulDonut|VengefulDonut]] 14:06, 9 December 2008 (EST)
 +
 +
:: Just yesterday I noticed that one of my seven embarking dwarves preferred alder (wood) and beds, so I made him my carpenter/woodcutter and maxed his skill in carpentry at embark.  In the first batch of eight beds I made, I only had one made of alder... but it came out at masterpiece quality.  I ''could'' have gotten lucky and rolled a masterpiece bed (approx. 1 in 400 chance), but I suspect that preferences played a role.  Since he preferred alder and beds, I let him have the room with the masterpiece in it.  He was ecstatic right up until the werewolves ate him.--[[User:Jurph|Jurph]] 15:44, 11 December 2008 (EST)
 +
 +
:::Possibly.  An unambiguous result would be if a dabbling dwarf who liked the materials/etc. made a specific-quality item.  Though it's possible all skill modifiers are inactive for dabblers.  The effect should still be most noticeable in a Novice, though... which is almost the exact assertion Gairabad removed from the article.--[[User:Maximus|Maximus]] 19:13, 11 December 2008 (EST)
 +
 +
:I just saw this for my first time.  I was training my cooks/peasants/etc. to dabbling in armorsmithing and weaponsmithing, and a no-skill farmer made a +iron bucker+.  When I checked, he had a preference for bucklers. --[[User:LegacyCWAL|LegacyCWAL]] 19:32, 24 March 2009 (UTC)
 +
 +
== Quality Probability Formulae ==
 +
 +
Could someone who understands that code tell us what it means? --[[User:Savok|Savok]] 20:14, 15 March 2009 (UTC)
 +
:The assembly code is there to prove my assertion that the dice rolls are independent.  There should be just enough commenting that someone skilled in the art can follow along.  If you're not skilled in the art, it is essentially gibberish.  Which is why I hid it inside a comment.<br/>&mdash;[[User:0x517A5D|0x517A5D]] 06:41, 16 March 2009 (UTC)
 +
::I don't want to go into a paternity war for something I could not have fathered, but I believe I documented this first, and the table is actually copy-pasted from my page (http://www.dwarffortresswiki.net/index.php?title=User:Bartavelle&oldid=15474), and that's why I changed the discoverer name (actually I'm not sure it is that important to have this name in this page anyway). I'll try to produce a more readable description of the algorithm, but it would be nice if somebody checked the numbers. [[User:Bartavelle|Bartavelle]] 08:06, 16 March 2009 (UTC)
 +
:::My bad, this table is not the same table that I produced, and it seems I was right to ask for number checking as mine are probably wrong :) [[User:Bartavelle|Bartavelle]] 08:09, 16 March 2009 (UTC)
 +
::::This table too is wrong. For novice, quality == 0 is P(d5 >= 1)*P(d10 >= 1)*P(d15 >= 1)*P(d20 >= 1)*P( (d25 >= 1) || (d3>1) ) = 4/5 * 9/10 * 14/15 * 19/20 * (1 - P( d25<1 && d3==1 ) ) =~ 0.63. Or I failed again ... [[User:Bartavelle|Bartavelle]] 08:55, 16 March 2009 (UTC)
 +
 +
== Experimental values ==
 +
Here are values recorded in a test fortress. The skill value is AFTER the bonuses are applied.
 +
 +
{| border="1" cellpadding="5" cellspacing="0" valign="bottom"
 +
|-
 +
!Skill  !!No quality!!Well-crafted!!Finely-crafted!!Superior quality!!Exceptional!!Masterful
 +
|-
 +
| 0|| 317(100.00%) || 0(0.00%) || 0(0.00%) || 0(0.00%) || 0(0.00%) || 0(0.00%)
 +
|-
 +
| 1|| 180(56.25%) || 119(37.19%) || 20(6.25%) || 0(0.00%) || 1(0.31%) || 0(0.00%)
 +
|-
 +
| 2|| 142(38.59%) || 157(42.66%) || 53(14.40%) || 14(3.80%) || 2(0.54%) || 0(0.00%)
 +
|-
 +
| 3|| 90(22.22%) || 149(36.79%) || 126(31.11%) || 37(9.14%) || 3(0.74%) || 0(0.00%)
 +
|-
 +
| 4|| 16(4.68%) || 128(37.43%) || 131(38.30%) || 55(16.08%) || 12(3.51%) || 0(0.00%)
 +
|-
 +
| 5|| 0(0.00%) || 73(21.16%) || 171(49.57%) || 87(25.22%) || 13(3.77%) || 1(0.29%)
 +
|-
 +
| 6|| 0(0.00%) || 61(16.85%) || 146(40.33%) || 106(29.28%) || 45(12.43%) || 4(1.10%)
 +
|-
 +
| 7|| 0(0.00%) || 28(8.89%) || 112(35.56%) || 122(38.73%) || 48(15.24%) || 5(1.59%)
 +
|-
 +
| 8|| 0(0.00%) || 9(3.90%) || 67(29.00%) || 91(39.39%) || 57(24.68%) || 7(3.03%)
 +
|-
 +
| 9|| 0(0.00%) || 3(1.28%) || 47(20.00%) || 121(51.49%) || 55(23.40%) || 9(3.83%)
 +
|-
 +
| 10|| 0(0.00%) || 0(0.00%) || 30(12.00%) || 125(50.00%) || 85(34.00%) || 10(4.00%)
 +
|-
 +
| 11|| 0(0.00%) || 0(0.00%) || 29(10.94%) || 118(44.53%) || 111(41.89%) || 7(2.64%)
 +
|-
 +
| 12|| 0(0.00%) || 0(0.00%) || 11(3.86%) || 119(41.75%) || 126(44.21%) || 29(10.18%)
 +
|-
 +
| 13|| 0(0.00%) || 0(0.00%) || 11(3.67%) || 103(34.33%) || 155(51.67%) || 31(10.33%)
 +
|-
 +
| 14|| 0(0.00%) || 0(0.00%) || 4(1.27%) || 73(23.17%) || 203(64.44%) || 35(11.11%)
 +
|-
 +
| 15|| 0(0.00%) || 0(0.00%) || 0(0.00%) || 80(23.88%) || 203(60.60%) || 52(15.52%)
 +
|-
 +
| 16|| 0(0.00%) || 0(0.00%) || 0(0.00%) || 54(15.43%) || 240(68.57%) || 56(16.00%)
 +
|-
 +
| 17|| 0(0.00%) || 0(0.00%) || 0(0.00%) || 40(10.96%) || 248(67.95%) || 77(21.10%)
 +
|-
 +
| 18|| 0(0.00%) || 0(0.00%) || 0(0.00%) || 21(7.45%) || 199(70.57%) || 62(21.99%)
 +
|-
 +
| 19|| 0(0.00%) || 0(0.00%) || 0(0.00%) || 6(4.51%) || 93(69.92%) || 34(25.56%)
 +
|-
 +
|}
 +
 +
I'm updating the table with the proper stats. <small>&ndash; [[template:unsigned|unsigned]] comment by [[User:Bartavelle|Bartavelle]]</small>
 +
 +
== Masterpiece destruction and effect on mood ==
 +
 +
Suppose I somehow manage to have a dwarf who's made 10000 masterpieces. Then if one of them gets destroyed, his happiness should decrease by 1000/10000 = 0.1 points. Does this round to 0, or will happiness always decrease by at least 1 point if a masterpiece gets destroyed? [[User:DaPatman|DaPatman]] 23:26, 22 April 2009 (UTC)
 +
:It rounds to 0, even if all 10000 get destroyed.
 +
:I coincidentally tested this today, while testing whether destroying mass quantities of objects speeds up the game.  (It does.)
 +
:Many thousands of masterpieces were atom-smashed.  All the legendary workers who had made them stayed ecstatic.  The only worker who became unhappy was a proficient metalcrafter who must have made just a few masterpieces.<br/>&mdash;[[User:0x517A5D|0x517A5D]] 03:02, 23 April 2009 (UTC)
 +
 +
== Effects of quality ==
 +
 +
I've rearranged the last section a little. Can someone please add information for what quality does with different kinds of items? --[[User:Ar-Pharazon|Ar-Pharazon]] 12:34, 9 February 2010 (UTC)

Latest revision as of 21:47, 8 March 2010

With regard to dwarves throwing a fit if a masterwork item is destroyed; I accidentally wrecked a masterwork engraving by having my miners dig out the other side of the wall where it was engraved. Sure enough, the engraver proceeded to make a fuss.

Then he scheduled a meeting with my broker/manager/leader dwarf. After the meeting I checked the engravers moods and saw the message "He recently yelled at a superior, but this only made him more angry" (not a direct quote). I'm wondering, does this mean that if the appropriate noble (I'm guessing the leader) has decent social skills they can actually get a tantruming dwarf to calm down and get back to work.

Might be worth further investigation.--WyldKarde 17:22, 4 November 2007 (EST)
Curious. Makes me wish Toady let out a little bit more info than he does about his game (we still don't know how combat works). But then again, it adds another level to an already deep game.--Draco18s 17:39, 4 November 2007 (EST)
The consoler skill allows a noble or whoever the dwarf talks to to calm the tantrum dwarf down. --ZombieRoboNinja 03:08, 4 May 2009 (UTC)

Trading masterpieces.[edit]

As far as I can make out, masterpieces can be traded without a tantrum. Can anyone confirm this? Is it worth adding? Runspotrun 16:38, 8 November 2007 (EST)

I traded away a half-dozen masterwork stonecrafts to the elves this Spring. So far, no reaction from my craftsdwarf, and the elves have left the map (it's been 2 months). So I'd say, conservatively, that you can at least trade stonecrafts. If this is true in general, then trading might be the safest thing to do with your masterworks. Doctorlucky 02:42, 20 November 2007 (EST)
I've traded various masterwork crafts and masterwork prepared food so I'd tend to agree. My first response to most masterwork items is to try this as it saves a lot of hassle :) --Shades 08:44, 20 November 2007 (EST)
So how bad is it to have Masterwork or Masterwork adorned clothing on your dwarfs? Chrispy 22:14, 18 March 2008 (EDT)
The stuff eventually rots and gets thrown away, causing the dwarf who made it to tantrum. --Savok 11:49, 19 March 2008 (EDT)

Wear and Tear[edit]

Wear is similar to quality, in that it adds a symbol notation to the item description - "x" "X" or "XX". I would also be very surprised if it didn't impact the value and effectiveness of an item - in particular, dwarves notice when their clothes start to decay, and give unhappy thoughts. Should we incorporate that material here, or just link to it? Doctorlucky 00:25, 19 November 2007 (EST)

And while we're at it, for completeness, things can also have decorations «», be imported () or forbidden {}. Any other designations? Doctorlucky 00:32, 19 November 2007 (EST)
Any idea how wear and tear is calculated? I've have a long-time miner whose clothes are still rather pristine but what my mayor, who did next to nothing, wore was tattered. --FJH 00:54 20 February 2009 (EST)

Alcohol Quality[edit]

Is this confirmed? I've never heard of it before, and as fasr as I know, the "had a wonderful drink lately" happy thought just comes from consuming one's favourite drink ("prefers to drink dwarven rum when possible" etc.) --Furiousfish 10:40, 19 November 2007 (EST)

Multiple quality modifiers[edit]

Does anyone know how to determine the value of items with multiple quality modifiers on them? Like bridges (design quality and construction quality) or clothing (weaving & clothesmaking). VengefulDonut 14:20, 20 December 2007 (EST)

I thought that at least bridges have only design quality. All walls constructed with masonry/carpentry/metalsmithing don't have any quality at least. As for clothing - I believe that I've seen well clothing, description of which states that it was made from fine cloth and the thread was dyed with a specific plant extract. The general rule is imho that every multi-staged job uses only 1 skill to determine the quality of the output (always architecture in any construction) but the quality of the materials is preserved (if any at all). One needs to look closer at the values of various multi-quality clothing to determine the value rule in that case. --Another 19:13, 20 December 2007 (EST)
In fact bridges have two quality modifiers. The design modifier is based on the architect's skill, the construction modifier is based on the builder's masonry skill (or carpentry, or metalsmithing, whichever is appropriate to the given bridge). It's quite difficult to determine the actual value of the bridge, since the value and number of materials used factors in as well. It would probably be easiest to just halt all mining and engraving, build the bridge, and see how much the architecture value of your fortress raises. As far as clothing values go, I believe that dye is counted as a decoration, and the value of the dye, multiplied by the quality of the dye job, is simply added to the value of the clothing. I believe the quality of the cloth is also considered a decoration, using the quality and the base cloth value to calculate the end value. Both dye and cloth quality values are added to the end value of the clothing. Most of this comes from tests done in the old 2D version, and so the formula may have changed, although I doubt it. With skilled weavers and clothiers you can make some ridiculously valuable cloth items out of giant cave spider silk. --Mzbundifund 19:01, 20 December 2007 (CST)

See Also[edit]

Item designations
Perhaps a merge might be nice? --Chrispy 22:15, 18 March 2008 (EDT)

Although Quality creates item designations, they are not at all the same topic. So no. --Savok 11:51, 19 March 2008 (EDT)

Masterpiece Meals[edit]

It says in the article that rotting masterpiece meals gives the maker an unhappy thought, but i have had many masterpiece meals rot without any repercussions. I do remember however that in the 2D version chasming rotten meals would cause the creator to tantrum. Curudan 16:03, 5 April 2008 (EDT)

I've just had a masterpiece meal rot away, and it did indeed give the creator an unhappy thought. Wasn't enough to bring her down from ecstatic, though, so I've no idea how powerful it is. --Gh3yz0r 19:25, 27 April 2008 (EDT)
Masterpiece destruction is no longer the headache it once was. Used to the loss of a masterpiece would absolutely crush the soul of a dwarf. Removing 1000 mood points. Now it removes mood based on how many other masterpieces they have. So legendary chefs hardly notice when one of their thousands of masterpieces rot. --Ikkonoishi 01:17, 28 April 2008 (EDT)
This seems like a REALLY important piece of information; someone may fear a legendary cook without this knowledge. --Bombcar 17:21, 17 October 2008 (EDT)

Lower than expected item quality[edit]

I have just made an observation that appears to contradict the information given on this page. I have a legendary mason (I believe she's at the lower legendary level) who I set to constructing statues. However, one of the statues ended up being merely well-crafted, when according to the table, the lowest possible quality should have been superior. Does anybody know what could explain this occurence, or is the table simply wrong? --User:Morlark

Are you certain that no other dwarf could have made the statue? --Savok 00:12, 25 June 2008 (EDT)
Absolutely sure. Not only was the workshop set so that only that dwarf could use it, but I was specifically watching the dwarf the whole time. --Morlark 01:38, 25 June 2008 (EDT)
Was the dwarf in question unhappy/hungry/thirsty/having-a-negative-status-effect? According to one guy, those cause skill levels to be effectively lowered. --Savok 08:49, 25 June 2008 (EDT)
Nope, she was ecstatic, with no negative status effects. --Morlark 15:37, 25 June 2008 (EDT)
It should be noted that the provided table only takes into account a quite small part of the code. There might be external effects. Does the guy dislike the material / item being made? -Bartavelle 09:17, 1 September 2008 (EDT)

Artifact quality modifier?[edit]

I can't find the damage / to-hit / armor modifier values for artifact level items anywhere. Can anyone shed some light on the subject? --PrettyGrizzly 11:16, 10 July 2008 (EDT)

The current status of this question is "nobody knows, but we think it's a lot higher than masterpiece." --Savok 15:06, 10 July 2008 (EDT)
I've heard somewhere that it's about 500%. I'm not sure about it in any way though. --Someone-else 20:51, 11 July 2008 (EDT)
I guess I should've been a bit more specific. The current status of this question is "nobody knows, but we think it's a lot higher than masterpiece. We're going with 500% as an estimate for now." --Savok 21:45, 17 July 2008 (EDT)
AFAIK, it's in memory the same quality as masterpiece. --Bartavelle 09:16, 1 September 2008 (EDT)
Can you provide concrete confirmation of that? It would go a long ways towards...well, towards exactly this, which would help several pages. LegacyCWAL 15:18, 11 December 2008 (EST)

Two Quality Levels[edit]

If a clothier were to make a finely crafted image on a piece of clothing, then a bone carver makes a masterpiece decoration on the same thing, how do they determine the quality level? Is it the "average" of the quality levels because, my cook made a roast that had one exceptionally minced ingredient, then two finely minced ingredients and one well minced ingredient it became a fine quality roast which is the average of the qualities.--Stinhad Limarezum 02:29, 21 October 2008 (EDT)

Prepared meals seem to have a different quality formula than other objects, although I don't know if that formula is an average of all qualities or something else. For all other objects, decorations have a separate quality from the item's inherent quality (i.e., a *<<+bolt+>>*, where the bolt is + and the decorations are *).--18:50, 21 October 2008 (EDT)

Are preferences really more helpful with novices?[edit]

I just added the "Jump" column to the table on this page. Based on it I think this statement is false:

>Preferences for particular materials and objects make a considerable difference. Although the exact adjustment is unknown, preliminary testing indicates that they can boost effective skill (not quality) by one to three levels (if the craftdwarf likes the item type, subtype and material), and have thus greatest impact when the dwarf is still relatively unskilled.

Gairabad 19:47, 6 December 2008 (EST)

Well, what's really needed is formal testing of a dwarf working on preferred materials and non-preferred. Might just be one of those DF myths.--Maximus 22:40, 6 December 2008 (EST)
From the article: "Although the exact adjustment is unknown, preliminary testing indicates that they can boost effective skill (not quality) by one to three levels (if the craftdwarf likes the item type, subtype and material), and have thus greatest impact when the dwarf is still relatively unskilled." If this is true, then you can derive from the "Jump" column that gaining subsequent levels does not bring diminishing returns. Gairabad 13:24, 7 December 2008 (EST)
Sorry, I'm totally confused about what you're trying to say here. Higher skill tends to produce higher-quality goods. A preference for the materials being used supposedly also produces higher-quality goods. What's your claim?--Maximus 22:47, 7 December 2008 (EST)
I'm trying to say that this bold part is false:
>Preferences for particular materials and objects make a considerable difference. Although the exact adjustment is unknown, preliminary testing indicates that they can boost effective skill (not quality) by one to three levels (if the craftdwarf likes the item type, subtype and material), and have thus greatest impact when the dwarf is still relatively unskilled. Gairabad 23:53, 8 December 2008 (EST)
Oh, okay. Well, you could just drop that part of the sentence. By some thinking, though, it may still be true: goods made by a competent artisan are 81% more valuable (on average) than those made by a dabbler, but goods made by a legend are only (5.8 / 4.8 =) 21% more valuable than those made by a master.
I still think the "jump in value" column is still of very limited use. It shows a trend -- value peaks at each of the levels where a lower quality grade is no longer produced (Proficient, Accomplished, Legendary) -- but dwarves are gaining skill continuously, so it's more of a statistic than anything else. A more interesting statistic is that from dabbling to talented, item values triple; from there to Legendary, they only double. To put it more generally, skill increases produce large value increases at low levels, small increases at high levels.--Maximus 12:19, 9 December 2008 (EST)
"To put it more generally, skill increases produce large value increases at low levels, small increases at high levels." No, they only produce larger percentage increases at low levels. I find the jump column useful to know when to stop leveling up my dwarves and start actually working on stuff. For example, Proficient and Accomplished are both good levels to start actually working. Gairabad 13:03, 9 December 2008 (EST)
"Working on stuff" and "leveling up" generally happen simultaneously. I guess if you're doing an armorsmith or something and don't want to make crap steel items... but I say make 'em when you need 'em, train continuously otherwise.--Maximus 13:12, 9 December 2008 (EST)
Not in my experience. If I want to level up a mason, I make blocks for constructions. If I want to level up a carpenter, I make bins, barrels, and buckets. With glassmaking, you can make raw glass. Mechanics can start by using common stone and move on to economic stone. Etc. Often things that you "need" can be put off a bit. For example, it takes a while for immigrants to get sleepy... plenty of time to level your carpenter so he can make higher-quality beds. Gairabad 13:31, 9 December 2008 (EST)
From a fortress perspective, the expected increase in value for a level up is much more important than a percent increase. When a preference makes a novice act like a competent, it's not as good as one that makes an expert act accomplished. VengefulDonut 14:06, 9 December 2008 (EST)
Just yesterday I noticed that one of my seven embarking dwarves preferred alder (wood) and beds, so I made him my carpenter/woodcutter and maxed his skill in carpentry at embark. In the first batch of eight beds I made, I only had one made of alder... but it came out at masterpiece quality. I could have gotten lucky and rolled a masterpiece bed (approx. 1 in 400 chance), but I suspect that preferences played a role. Since he preferred alder and beds, I let him have the room with the masterpiece in it. He was ecstatic right up until the werewolves ate him.--Jurph 15:44, 11 December 2008 (EST)
Possibly. An unambiguous result would be if a dabbling dwarf who liked the materials/etc. made a specific-quality item. Though it's possible all skill modifiers are inactive for dabblers. The effect should still be most noticeable in a Novice, though... which is almost the exact assertion Gairabad removed from the article.--Maximus 19:13, 11 December 2008 (EST)
I just saw this for my first time. I was training my cooks/peasants/etc. to dabbling in armorsmithing and weaponsmithing, and a no-skill farmer made a +iron bucker+. When I checked, he had a preference for bucklers. --LegacyCWAL 19:32, 24 March 2009 (UTC)

Quality Probability Formulae[edit]

Could someone who understands that code tell us what it means? --Savok 20:14, 15 March 2009 (UTC)

The assembly code is there to prove my assertion that the dice rolls are independent. There should be just enough commenting that someone skilled in the art can follow along. If you're not skilled in the art, it is essentially gibberish. Which is why I hid it inside a comment.
0x517A5D 06:41, 16 March 2009 (UTC)
I don't want to go into a paternity war for something I could not have fathered, but I believe I documented this first, and the table is actually copy-pasted from my page (http://www.dwarffortresswiki.net/index.php?title=User:Bartavelle&oldid=15474), and that's why I changed the discoverer name (actually I'm not sure it is that important to have this name in this page anyway). I'll try to produce a more readable description of the algorithm, but it would be nice if somebody checked the numbers. Bartavelle 08:06, 16 March 2009 (UTC)
My bad, this table is not the same table that I produced, and it seems I was right to ask for number checking as mine are probably wrong :) Bartavelle 08:09, 16 March 2009 (UTC)
This table too is wrong. For novice, quality == 0 is P(d5 >= 1)*P(d10 >= 1)*P(d15 >= 1)*P(d20 >= 1)*P( (d25 >= 1) || (d3>1) ) = 4/5 * 9/10 * 14/15 * 19/20 * (1 - P( d25<1 && d3==1 ) ) =~ 0.63. Or I failed again ... Bartavelle 08:55, 16 March 2009 (UTC)

Experimental values[edit]

Here are values recorded in a test fortress. The skill value is AFTER the bonuses are applied.

Skill No quality Well-crafted Finely-crafted Superior quality Exceptional Masterful
0 317(100.00%) 0(0.00%) 0(0.00%) 0(0.00%) 0(0.00%) 0(0.00%)
1 180(56.25%) 119(37.19%) 20(6.25%) 0(0.00%) 1(0.31%) 0(0.00%)
2 142(38.59%) 157(42.66%) 53(14.40%) 14(3.80%) 2(0.54%) 0(0.00%)
3 90(22.22%) 149(36.79%) 126(31.11%) 37(9.14%) 3(0.74%) 0(0.00%)
4 16(4.68%) 128(37.43%) 131(38.30%) 55(16.08%) 12(3.51%) 0(0.00%)
5 0(0.00%) 73(21.16%) 171(49.57%) 87(25.22%) 13(3.77%) 1(0.29%)
6 0(0.00%) 61(16.85%) 146(40.33%) 106(29.28%) 45(12.43%) 4(1.10%)
7 0(0.00%) 28(8.89%) 112(35.56%) 122(38.73%) 48(15.24%) 5(1.59%)
8 0(0.00%) 9(3.90%) 67(29.00%) 91(39.39%) 57(24.68%) 7(3.03%)
9 0(0.00%) 3(1.28%) 47(20.00%) 121(51.49%) 55(23.40%) 9(3.83%)
10 0(0.00%) 0(0.00%) 30(12.00%) 125(50.00%) 85(34.00%) 10(4.00%)
11 0(0.00%) 0(0.00%) 29(10.94%) 118(44.53%) 111(41.89%) 7(2.64%)
12 0(0.00%) 0(0.00%) 11(3.86%) 119(41.75%) 126(44.21%) 29(10.18%)
13 0(0.00%) 0(0.00%) 11(3.67%) 103(34.33%) 155(51.67%) 31(10.33%)
14 0(0.00%) 0(0.00%) 4(1.27%) 73(23.17%) 203(64.44%) 35(11.11%)
15 0(0.00%) 0(0.00%) 0(0.00%) 80(23.88%) 203(60.60%) 52(15.52%)
16 0(0.00%) 0(0.00%) 0(0.00%) 54(15.43%) 240(68.57%) 56(16.00%)
17 0(0.00%) 0(0.00%) 0(0.00%) 40(10.96%) 248(67.95%) 77(21.10%)
18 0(0.00%) 0(0.00%) 0(0.00%) 21(7.45%) 199(70.57%) 62(21.99%)
19 0(0.00%) 0(0.00%) 0(0.00%) 6(4.51%) 93(69.92%) 34(25.56%)

I'm updating the table with the proper stats. unsigned comment by Bartavelle

Masterpiece destruction and effect on mood[edit]

Suppose I somehow manage to have a dwarf who's made 10000 masterpieces. Then if one of them gets destroyed, his happiness should decrease by 1000/10000 = 0.1 points. Does this round to 0, or will happiness always decrease by at least 1 point if a masterpiece gets destroyed? DaPatman 23:26, 22 April 2009 (UTC)

It rounds to 0, even if all 10000 get destroyed.
I coincidentally tested this today, while testing whether destroying mass quantities of objects speeds up the game. (It does.)
Many thousands of masterpieces were atom-smashed. All the legendary workers who had made them stayed ecstatic. The only worker who became unhappy was a proficient metalcrafter who must have made just a few masterpieces.
0x517A5D 03:02, 23 April 2009 (UTC)

Effects of quality[edit]

I've rearranged the last section a little. Can someone please add information for what quality does with different kinds of items? --Ar-Pharazon 12:34, 9 February 2010 (UTC)