Print

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

[Megaman3321]

Armor and weapon !!SCIENCE!!

THE HYPOTHESIS

      A question that's been bugging me for quite some time is this: How exactly do quality modifiers affect armor and weapon effectiveness?
   Specifically, is an ☼iron breastplate☼ as effective as a +steel breastplate+ when deflecting attacks from a *Steel spear*? Does an adamantine
   low boot perform as well as an +iron high boot+ in terms of damage and penetration calculation? To find these answers, I have devised this scheme:

Code: [Select]

         |+|
|^^^|
|^^^|
|^^^|
___
      Effectively, it is a set of 168 danger rooms, with 24 for each dwarven weapon (NOTE: I am excluding crossbows from the test due to !!SCIENCE!!
   already existing for them.), with each weapon trap having either 1 or 2 weapons each (I'm not decided on which would be easier to calculate.).
   These will all be linked up to either a repeater mechanism powered by a basic minecart logic system, or an U.R.I.S.T type A.I.

THE GOALS

   Here are the questions that these tests will hopefully answer:

   1. By how much does quality make up for poor material?
   2. How much more protection is afforded with each layer of armor?
   3. Just what exactly does the to-hit modifier do?
   4. How much does weapon quality matter when fighting against different types of armor?
   5. What are the To-hit modifiers for the armor intermediate qualities?
   6. What are the To-hit modifiers for the weapon intermediate qualities?


TESTING STAGES

      The initial phase of testing will be a simple proof-of-concept to prove what we already know: That a no-quality iron spear on a no-quality iron
   set of armor should miss about 50% of the time, assuming I'm reading the wiki correctly. Each test will involve at least 50 strikes per weapon type.

      The next stage will be the testing proper: no-quality against no-quality, all the way up to ☼masterwork☼. Expect to see many an instakill
   during copper versus ☼adamantine☼ testing.

      The third and final stage will be the drop-testing. This stage will test the elasticity of the armor and how well it holds together dwarven
   bodies.


THE MODS

   As this is to result in massive casualties, I will be making the following modifications to my game:

   1. Dwarves will reach adulthood by age 1, and give birth in litters.
   2. Invasions will be deactivated.
   3. Dwarves will have [SPEED:0] applied to them so that testing will proceed at a reasonable pace (my computer sucks and can only handle ~10fps
      for a 100 strong fort). This tag will be disabled during the actual tests.
   4. All dwarves will have a natural ability with medicine (naturally proficient in all fields).
   5. There will be one layer each of all metals and one layer of marble for steel.
   6. I may enable [NO:EAT] and [NO:DRINK] on the dwarves, but I don't know if I should.
   
   I realize that these mods go against the grain of DF, but otherwise these tests would be unfeasible.

THE CRY FOR HELP

      If there's already been !!SCIENCE!! done in this field, then just point me to it. If someone more knowlegable than I in basic DF
   mechanics can answer a few of these questions that I've posed for myself, then plese tell me. If you wish to help, then please by all means
   do so; many heads thinking together is better than just one. I readily admit that I'm out of my depth here, so if I make a mistake and you
   catch it please tell me.

THE CONCLUSION

      Basically what I'm going to be doing here is answering some of the unanswered questions floating around the wiki, with some good-ol'
   bloodshed thrown into the mix. It's an ambitious project for sure, but if I can't handle it then I'll ask you guys to pick up the torch
   for me I'll try and have my first update on the matter before Black Friday. For now though, I'm off to bed. Wish me luck!

[Megaman3321]

Reserved for future updates.

[Megaman3321]

Also reserved for future updates.

[Drazinononda]

A tip: check the wiki to compare coverage and rigidity of each armor piece. For example, you mentioned in the OP, "Does an adamantine low boot perform as well as an +iron high boot+," but a high boot covers the lower leg as well as the foot, while a low boot does not. If you use high boots and greaves, attacks against the lower leg will register inaccurately with regards to your research: both pieces of armor will get a roll to block, but only one will report.

I assume you'll only be using dwarves with zero Armor User skill, as a control?

[Nezumi]

You may consider using only dwarves with maxed armor user skill so their skill won't be changing as you're trying to test.

Also, is there a way to mod the game so certain skills won't train?  If so you might consider disabling dodging for similar reasons.
Although I guess you can filter dodges out of the combat reports by hand?

[Trif]

Won't [SPEED:0] result in better dodging chances? I haven't tried it, I'm just assuming that faster creatures should be able to attack and dodge faster since the combat / movement split hasn't happened yet.

[Drazinononda]

You may consider using only dwarves with maxed armor user skill so their skill won't be changing as you're trying to test.

Also, is there a way to mod the game so certain skills won't train?  If so you might consider disabling dodging for similar reasons.
Although I guess you can filter dodges out of the combat reports by hand?

Ah, your Armor User suggestion is better than mine.

There is a tag that can be applied to creatures to cause them to learn skills faster or slower (or to rust faster or slower). I can't recall what it is exactly, but careful browsing of the tokens listed on the wiki should turn it up. That along with the "natural skill" token (again, don't recall the exact form) should allow you to make all your dwarves into permanent Legendary Armor Users and Not Dodgers, which should speed your research considerably.

[expwnent]

Won't [SPEED:0] result in better dodging chances? I haven't tried it, I'm just assuming that faster creatures should be able to attack and dodge faster since the combat / movement split hasn't happened yet.

I have definitely noticed combat advantages to having SPEED:0.

[Megaman3321]

Won't [SPEED:0] result in better dodging chances? I haven't tried it, I'm just assuming that faster creatures should be able to attack and dodge faster since the combat / movement split hasn't happened yet.

I have definitely noticed combat advantages to having SPEED:0.

Oh, I completely forgot about that. I'll update the OP accordingly

You may consider using only dwarves with maxed armor user skill so their skill won't be changing as you're trying to test.

Also, is there a way to mod the game so certain skills won't train?  If so you might consider disabling dodging for similar reasons.
Although I guess you can filter dodges out of the combat reports by hand?

Ah, your Armor User suggestion is better than mine.

There is a tag that can be applied to creatures to cause them to learn skills faster or slower (or to rust faster or slower). I can't recall what it is exactly, but careful browsing of the tokens listed on the wiki should turn it up. That along with the "natural skill" token (again, don't recall the exact form) should allow you to make all your dwarves into permanent Legendary Armor Users and Not Dodgers, which should speed your research considerably.

Thanks for the info. I'll search the wiki today for the tags.

I'm going to be genning the world tonight after I finish doing the mods.

[Azated]

I'll be interested to see your results.

[knutor]

Armor cannot be compared in a Danger Room.  Nice scientific method, displayed in the first post, however.  Shows real preparation and thought.  Except the test cannot account for density or layering.

How are you going to design a weapon trap on a repeater?  I thought automated traps came in one flavor, upright spear traps?  Interesting. 

How exactly do quality modifiers affect armor and weapon effectiveness?

DwarfFortressWiki has a nice table, with ratios.

Specifically, is an ☼iron breastplate☼ as effective as a +steel breastplate+ when deflecting attacks from a *Steel spear*?

No, ☼iron breastplate☼ is worse.

Does an adamantine low boot perform as well as an +iron high boot+ in terms of damage and penetration calculation?

Yes, as well and better.  Additionally, its far lighter, and increases flight of foot on the dorf wearing it.

Sometimes in DF, survival is all about timing, skill level, and attributes, and nothing to do with item quality.  Don't forget about the stances some of these dorfs take, as a result of their attributes.  There is a considerable variable, in number of enemies attacking and their position, rear vs front, which traps do not account for. 

Also there is size to compare.  Some of the armored foes are large, and that has to be a problem for smaller area slicing weapons to cope with.  I would think a larger foe, would be a benefit to a piercing weapon, more area to hit, less chance of missing, but I dunno.  Curious of your results, thanks for doing the testing.  Sincerely, Knutor

[knutor]

Could you limit this just to things being dropped on them, to test material and quality on helmets alone vs the material being dropped?  I've always wondered if support traps could be automated.  So far I've only been able to automate upright spear traps.

[Megaman3321]

Hmmm, I forgot about the fact that you can't do upright weapons with anything other than spears. Solution? I'm thinking that we could design multiple types of spears/spikes that have the qualities of hammers, swords, etc. but use the spearman skill.

Armor cannot be compared in a Danger Room.  Nice scientific method, displayed in the first post, however.  Shows real preparation and thought.  Except the test cannot account for density or layering.

How are you going to design a weapon trap on a repeater?  I thought automated traps came in one flavor, upright spear traps?  Interesting. 

How exactly do quality modifiers affect armor and weapon effectiveness?

DwarfFortressWiki has a nice table, with ratios.

Specifically, is an ☼iron breastplate☼ as effective as a +steel breastplate+ when deflecting attacks from a *Steel spear*?

No, ☼iron breastplate☼ is worse.

Although I realize that the DF wiki has a table, it doesn't go into detail about how -breastplate- and +breastplate+ compare, unless I'm not looking in the right place (which is entirely possible, mind you ). I wanted to find those values.

Sometimes in DF, survival is all about timing, skill level, and attributes, and nothing to do with item quality.  Don't forget about the stances some of these dorfs take, as a result of their attributes.  There is a considerable variable, in number of enemies attacking and their position, rear vs front, which traps do not account for.

I realize that there is more to survival in DF than just armor quality (I've made that mistake before ). These factors, though interesting, are irrelevant to this test; it is purely an excersize in the way the game handles quality.

[knutor]

What I was thinking... test for armor protection vs blunt falling material.  Since there seems to be a trend in danger room design, with falling refuse stacks.

Make a stack of coins, one for each type of combat metal.  Make a helm, 6, or 7 if ya can land a lucky artifact one of the metal in question, lets say iron.  Then drop 1 coin onto the head of of a dorf wearing

iron coin [1], silver coin [1], copper coin [1]... etc
iron block [1], silver...
iron chunk [1], silver.. *

With h, height varying.  Maybe use the mod that allows ya to write your own dwarf attributes, to keep the ducks in order, with dodge, evasion, etc..  To avoid erroneous data collection as a result of a super herculean dwarf.

iron helm, -iron helm-, +iron helm+... etc

*Drop chunks maybe lethal to the lab mice.  I'm not claiming responsibility for the safety of any pixels.

[Megaman3321]

I'll have to try that during the drop-testing phase And why wouldn't you want to claim casualties? This is !!SCIENCE!!, not science. 

Incidentally, I'm gonna be starting work on the mods in about an hour or two and get the world all genned up. Hopefully they'll work the first time through.