Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 [2] 3 4

Author Topic: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}  (Read 7161 times)

Tharwen

  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #15 on: September 19, 2011, 06:53:00 am »

Wait... did you say you gave one side adamantine greaves and the other side [control metal] greaves?

If so, I wish to have words with you about your experimental technique.
Logged
[Signature]

AWdeV

  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #16 on: September 19, 2011, 09:03:38 am »

So copper chainmail may actually be a decent starting armor for your dwarves?  Just asking, because that would make it much easier to outfit my new militia without having to wait until I can get steel or leather.

It's definitely a good choice if you have nothing better. Waiting for steel is fine and all but your foes may not be so considerate.
Logged
Teenage Bearded Axelord Turtles
Teenage Bearded Axelord Turtles
Urists in a half shell (Turtle Power)

Urist Da Vinci

  • Bay Watcher
  • [NATURAL_SKILL: ENGINEER:4]
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #17 on: September 19, 2011, 09:09:54 am »

Please note that the
   [IMPACT_YIELD:xxxxxx]
   [IMPACT_FRACTURE:xxxxxx]
   [IMPACT_ELASTICITY:xxx]
material values are also used in blunt attack protection.

acetech09

  • Bay Watcher
  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #18 on: September 19, 2011, 09:16:56 am »

Well - it apparently seems those values have little effect in mail protection against blunts, but undoubtedly protect against blunt attacks in rigid plate armor.
Logged
I challenge you to a game of 'Hide the Sausage', to the death.

nightwhips

  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #19 on: September 19, 2011, 11:03:09 am »

Hey all-

I'm too new at this to be running lots of tests myself, but I can give input on a bunch of experimental design considerations, which you more experienced types can decide are useful or not.

If you want to test only chain vs. breastplates, or chain of different metal types, it seems to me there are a number of factors to control.

First, you don't want all these death by skull-crush, since that doesn't reflect the variable you're testing. Give everybody addy armor all over EXCEPT for the pieces that you're testing.

As was mentioned, it seems to me that the first kill of a 5v5 match then biases the results to the team with the first kill. Do many (more than 6) tests of 1v1 for each variable. This should cause your statistical signal to settle down, esp. if the properties of the two metals you're testing are similar.

You might also throw out lucky one-hit killshots, since these also don't reflect what you're testing.

I'd be really curious to know more about different metals. All I've got now is copper on my map, so if breastplates vs chain is a big difference, I'd love to know. For some reason I assumed breastplates were better...
Logged
: Miner dwarves? In my volcano?

:I put childs into danger room...
They die, and their parents care nothing because legendary dining room.

OcelotTango

  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #20 on: September 19, 2011, 12:46:46 pm »

What seems like a good plan would be to mod in metals with very high values in  Impact yield, Impact fracture, and Impact elasticity, as well as combinations of the three, as well as a baseline metal that is moderate in all regards. Then perform these tests to isolate what the key factor in chain mail mitigation is.
Logged

Melissia

  • Bay Watcher
  • Confused noob having lots of fun!
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #21 on: September 19, 2011, 01:16:29 pm »

So copper chainmail may actually be a decent starting armor for your dwarves?  Just asking, because that would make it much easier to outfit my new militia without having to wait until I can get steel or leather.

It's definitely a good choice if you have nothing better. Waiting for steel is fine and all but your foes may not be so considerate.
I have iron.  Lots and lots and LOTS of iron.  But making pig iron and steel seems to be a problem for me, they keep asking for 150 coal...
Logged
"If a job is worth doing, it's worth dying for."
-- Warhammer 40,000: Space Marine

Roraborialisforealis

  • Bay Watcher
  • [HATEBILLION:UNTHINKABLE]
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #22 on: September 19, 2011, 01:17:50 pm »

As per this reply and following responses, there was the question brought up about how density affects chainmail armor's effectiveness in protecting against blunt force attacks.

Well, I did some arena testing, and here are my findings:

------------

It was originally stated that addy chainmail was the worst protection against blunt force attacks, compared to other metals, due to density. To determine the effect, I ran eighteen 5 vs 5 hammerdwarf matches, where one side would always have adamantine mail shirts 'n pants, while the other side was either copper-, bronze-, or iron-clad. I did the testing in 3 phases - Phase 1 which did six rounds with addy vs bronze, since bronze was a median of density in armor. I then tested copper, then iron.

Testing Phase 1 - Addy vs Bronze
   Addy won 5 matches, bronze won 1. It was pretty obvious that the bronzers were dying a lot faster than the addy dwarves were. The addy dwarves won with minimal injuries.

Testing Phase 2 - Addy vs Copper
   Copper 4, addy 2. I was slightly surprised, but then again - it fit the hypothesis perfectly. The denser copper pulled an advantage over adamantine in the competition.

Testing Phase 3 - Addy vs Iron
   Iron 3, Addy 3. The results were nearly even, with both sides winning 3 matches with similar injuries. I was expecting iron to do poorest, perhaps with no victories, since it was the lightest of the three typical metals.

As brought up by OcelotTango, he asked if a even denser metal, (silver) would be even better, so I tried:
Testing Phase 4 - Addy vs Platinum
   Platinum 3, Addy 3. Now the results are getting confusing... the densest bar metal in game preformed as well as iron... hrm...

Results:

    Results were slightly odd - after the first 2 tests I figured that the denser copper did better, the lighter bronze did worse, I expected iron to be the absolute worst because it was lighter than iron - but it turns out that the orders of density are different than the orders of effectiveness, but the theory had some proof in that the densest did the best, but alas, the lightest did not do the worse.

   After the platinum test, I discovered that density might not be the factor in these tests, since platinum is denser than copper but did worse. The key is to try to find a property that orders copper > iron,addy,platinum > bronze.

-------

Detailed Notes:
I took detailed test notes while experimenting, so if you want to read a dry and boring and not-proofread test log
Spoiler: It's right here. (click to show/hide)

EDIT: Just want to let you know that I just duplicated the test, and verified the gross win results, I didn't catalogue injuries, just wanted to increase the test pool - results were near identical this time around.


IT SEEMS that softer (Why candy wasn't the best,) and the lighter (Why plat wasn't best) metals give better blunt trauma protection. I would like to see gold and lead tested now........
Logged
Spoiler (click to show/hide)

EveryZig

  • Bay Watcher
  • Adequate Liar
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #23 on: September 19, 2011, 02:09:14 pm »

IT SEEMS that softer (Why candy wasn't the best,) and the lighter (Why plat wasn't best) metals give better blunt trauma protection. I would like to see gold and lead tested now........
If softer and lighter are better, try aluminum.
Logged
Soaplent green is goblins!

Xen0n

  • Bay Watcher
  • Took joy in ‼SCIENCE‼ lately.
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #24 on: September 19, 2011, 02:23:52 pm »

So copper chainmail may actually be a decent starting armor for your dwarves?  Just asking, because that would make it much easier to outfit my new militia without having to wait until I can get steel or leather.

It's definitely a good choice if you have nothing better. Waiting for steel is fine and all but your foes may not be so considerate.
I have iron.  Lots and lots and LOTS of iron.  But making pig iron and steel seems to be a problem for me, they keep asking for 150 coal...

For jobs requiring bars (Metal, coal, ash, soap, etc.) a job cancellation saying '150' just means '1 bar.'  The game tracks each bar as having 150 pieces (Because the hospital can use pieces of a soap bar without using up the whole bar).  So I believe you just need 1 bar of coal/coke.  To be safe, put the coal/coke near the smelter, and be careful of having the smelter in a burrow that doesn't contain all the needed materials (there appears to be a bug of smelters not displaying jobs for items if the smelter is a burrow that doesn't cover all the materials).  You can also try issuing the job through the manager (j>m>q, I think?).
Logged

Steelgeek

  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #25 on: September 19, 2011, 02:36:49 pm »

Ok so not knowing how Toady designed this, here's a view on chain armor from a RL blacksmith/armorer/blademaker. (15 yrs+ now)
We were also curious about how chain really worked, so we made some medieval equivalent armor, blades, maces/hammers etc.
We did our testing with meat and poplar wood to represent bone(lets just say our steaks were REALLLY tender that night). We did this all to the best of our abilities and time allowed, and mostly just expense.
riveted chain vs. 1 handed 42" steel longsword - as expected, against slashes and draw cuts it was really good. Most solid thrusts pierced, causing large amounts of damage, possible instakills.Hacking? Well, well placed solid hacks SHEARED the mail. The rivets gave first (not surprising, weakest part) but also many links were simply cut in half, in addition to them being bent and that causing the other mail links to split. We're only going to discuss hacking then. Hacks against just thin meat (i.e guts) were, well, impressively destructive. Joints  were broken, but not destroyed. Other bone areas (ribs, shoulders)were damaged and broken but viable. Against thick meat(legs, other muscled areas), definitely wounded, but in our best estimate, no instakill.
Axes 1hand, 3pounds, 7" cutting edge- Oh my dear god, the carnage. Generally only about 3" of the blade hit, increasing penetration. At best, the mail caused glancing blows. If it hit solid, it sheared and you had several inches of penetration/damage. Joints did however, not shear, mostly because they were simply destroyed by the blunt force.
Blunt mass weapons - Yeah, it was ugly. Joints and bone were simply splintered. Muscle had the links broken and driven into the wound. Guts, though, were not instakills.
Crossbows, 125 lb pull - Slightly reduced damage, but not noticeably.
Crossbow, 500(estimated) lb pull- No difference. As a matter of fact, it also went through the brick at the back of the garage, glanced off a tree, and buried itself in the ground. BTW - for the time, that was considered a hunting bow, or a light military weapon.
So then we added gambesons, padded under armor, which would have been standard, stuffed with hair. All damage was SIGNIFICANTLY reduced, except for crossbow, who still laughed on the way through, but at least stopped at the 16" of 4" x 4"s we hastily erected as a backstop. A lot of killing damage was then survivable, at least for the fight anyways. The hair stuffing was surprisingly tough. 
Ok so whats the point of all this?
1 - Tensile and shear strength of the metal used for the mail was really really important.
2 - Gambesons were absolutely vital, for reasons beyond comfort.
3 - Real medieval chain was STILL a hell of a lot better than ours. They used a tighter weave density, with thicker wire. We believe this would account for the difference in thrusting damage data. Real chain was simply better against it.

My best guess as a blacksmith with an engineering degree? All things being equal, the density shouldn't matter, well except to make it even heavier to wear. (25-50lbs depending on weave density and wire thickness) The amount of deflection is at best 2" - 3" around a 4" circle (approx). Reducing the area of contact would make density have even less to do with it, and dramatically increase the importance of yield, fracture, and elasticity. Some quick napkin math says a 4" circle would be about 1.3 lbs of steel, and 1.5 lbs of copper, and no that takes none of variables into account, just a quick show of approximates.

Now - the biggest question is how realistic is Toady's programming? His simulations are amazing, but he would still go by guesses and assumptions for how combat works, and how weapons work. Hell, even the scholars still just go by guesses and assumptions.
So, in other words, going by raws AND an understanding of how it really works, adamantine chain should be the best way to go, by a huge margin, so why isn't it?

 
Logged

Tharwen

  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #26 on: September 19, 2011, 02:52:51 pm »

Don't ignore me!

Did you or did you not give the non-candy side greaves of their chosen metal?
Logged
[Signature]

Melissia

  • Bay Watcher
  • Confused noob having lots of fun!
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #27 on: September 19, 2011, 05:46:28 pm »

So copper chainmail may actually be a decent starting armor for your dwarves?  Just asking, because that would make it much easier to outfit my new militia without having to wait until I can get steel or leather.

It's definitely a good choice if you have nothing better. Waiting for steel is fine and all but your foes may not be so considerate.
I have iron.  Lots and lots and LOTS of iron.  But making pig iron and steel seems to be a problem for me, they keep asking for 150 coal...

For jobs requiring bars (Metal, coal, ash, soap, etc.) a job cancellation saying '150' just means '1 bar.'  The game tracks each bar as having 150 pieces (Because the hospital can use pieces of a soap bar without using up the whole bar).  So I believe you just need 1 bar of coal/coke.  To be safe, put the coal/coke near the smelter, and be careful of having the smelter in a burrow that doesn't contain all the needed materials (there appears to be a bug of smelters not displaying jobs for items if the smelter is a burrow that doesn't cover all the materials).  You can also try issuing the job through the manager (j>m>q, I think?).
That makes sense!  Thank you for interpreting the game's madness for me.

For now though I'm using iron armor and weapons, and I'll deal with making steel later after my first gobbo invasion, building up a sizable cache of iron bars first.
Logged
"If a job is worth doing, it's worth dying for."
-- Warhammer 40,000: Space Marine

acetech09

  • Bay Watcher
  • Bay Watcher
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #28 on: September 19, 2011, 06:49:26 pm »

Don't ignore me!

Did you or did you not give the non-candy side greaves of their chosen metal?

I thought you were talking to the guy who also posted his tests in this thread...

if you can read, I on multiple scenarios specifically mentioned chain leggings.



What seems like a good plan would be to mod in metals with very high values in  Impact yield, Impact fracture, and Impact elasticity, as well as combinations of the three, as well as a baseline metal that is moderate in all regards. Then perform these tests to isolate what the key factor in chain mail mitigation is.

I would have, but I was mainly testing effectiveness of chain armor against blunt attacks overall, not specific factors. It has occurred to me, though, that mods with different metals might benefit from these tests, so I will probably do a second, more thorough test run this evening, since I was not in the mood for a 5-hour deeply analytic test last night (was around 0200 local when I finished the testing).


Hey all-

I'm too new at this to be running lots of tests myself, but I can give input on a bunch of experimental design considerations, which you more experienced types can decide are useful or not.

If you want to test only chain vs. breastplates, or chain of different metal types, it seems to me there are a number of factors to control.

First, you don't want all these death by skull-crush, since that doesn't reflect the variable you're testing. Give everybody addy armor all over EXCEPT for the pieces that you're testing.

As was mentioned, it seems to me that the first kill of a 5v5 match then biases the results to the team with the first kill. Do many (more than 6) tests of 1v1 for each variable. This should cause your statistical signal to settle down, esp. if the properties of the two metals you're testing are similar.

You might also throw out lucky one-hit killshots, since these also don't reflect what you're testing.

I'd be really curious to know more about different metals. All I've got now is copper on my map, so if breastplates vs chain is a big difference, I'd love to know. For some reason I assumed breastplates were better...

I, in my initial tests, did have extremity armor, but that, for some reason, ended up increasing the variance in the tests.

I didn't use some solution to prevent head caveins, because 90% of the headshots (I counted) were on unconscious enemies, which would be the case in a battlefield environment against most biological entities (your dwarves).

And just fyi - best to do breastplates AND mail shirts. Breastplates do better against blunt damage - use the best metal you have for that, but they only cover the breast, not the back like mail shirts do.
Logged
I challenge you to a game of 'Hide the Sausage', to the death.

Corneria

  • Bay Watcher
  • [TORTURE_FOR_FUN:ACCEPTABLE][TRESPASSING:SHUN]
    • View Profile
Re: Chainmail Material Blunt Force Protection Variance !!Science!! - {Spoilers}
« Reply #29 on: September 19, 2011, 06:52:44 pm »

Extra armor isn't needed. Larger sample size, only 1v1s.
Logged
Oh yeah: Nazis.  Now I lost the argument.  So you can't argue with me anymore.  Nyah.
Pages: 1 [2] 3 4