Print

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

[Rekov]

I'm making this topic after a discussion on Discord about what specific stats matter when it comes to how good a metal is for weapons/armor.

Here are the stats for the vanilla weapon/armor metal materials



For blunt weapons, SOLID_DENSITY appears to be most of the story. It explains why silver is good and adamantine is bad. It doesn't, however, explain why steel is better than copper, bronze, and iron, so it can't be the whole story.

Here is where stuff gets interesting:
Green cells, 'better' metals have higher values in a consistent pattern. Red cells are where this pattern is broken.

So what does this mean when it comes to iron and adamantine?

The _AT_YIELD stats refer to the amount of 'give' once it has reach the YIELD point of the corresponding type of force. Why does this number generally increase, but not for iron, and is 0 for adamantine?

Is iron really better than bronze for weapons and armor? If it is, it means the non-red statistics must matter a lot more than the ones that break the trend.

[Klisz]

For adamantine's strain-at-yield values being 0, I think it's because the yield and fracture values are the same - so if you manage to apply 5000000 of that force to adamantine, it'll just break anyway - thus making the strain-at-yield value irrelevant.

[Starver]

I take it (after a few glances) that your green/red expectations are purely on whether the values rise in the cycle of Copper, Bronze, Iron, Steel, Adamantine? (And you haven't placed Silver, yet.)

I think (based on those same glances) you'd have fewer reds if you started with Cu, Fe, CuSn, Fe++. Mere iron is second in line of that subset more than it's third, and that gives an element, element[1], alloy (of first element) and alloy[1] (of second element). Which doesn't 'explain' anything, but is more logical.

Spoilermetal does indeed spoil the pattern (because it's mystical, probably, and inclines to physical rules with far less of a nodding acknowledgement of our reality than the others). Set that aside for the moment.

And Silver's odd, too, as it looks like you discovered, not botheting to fit it in. I haven't done a full tot-up of its optimal ranking (for least overall hiccoughs in the series) but individual qualities ask it to appear anywhere from before Copper to even after Steel. Likely it sits most comfortably/least uncomfortably immediately before or after iron in my suggested relocated position.

But that's in overview. There should be no all-rounder ranking all different physical qualities. If you pare it down to "light and deformation-resistant", for proof against certain types of weapon damage, or "blunt-forceful and not fracturable" for use in certain attacking implements, then you can perhaps come to individually different rankings for each use. As others have already done, IIRC.


Some games do treat it as a basic sequential ranking. Wood<Stone<Iron<Gold<Diamond or somesuch, ignoring that gold armour is probably stupid (or, at best, better restricted to decorative filligree to a more substantial iron-based main mass, just to show off) and I'm not sure I'd trust a diamond helm (more than most of the others) against a warhammer and wood of the right sort definitely has its uses (best stuff for a longbow, I suggest, and often not that bad at all for a pure blunt-force attack that doesn't rely upon edges or points specifically vs stiff metal protection).

But there's a rock/paper/scissors nature as well, here. Not entirely balanced/absolute, like a true RPS-derivative, yet you're not left with a totally third-rank material (especially if you include availability and production capability to your metrics) that you should just forevermore ignore once you have started to be able to use second- and/or first-rank materials, or are encountering enemies who have.

At the very least, if you're planning on deciding your own good/better/best system, first split it into separate weaponry and armour categories, before then considering sub-dividing further, and calculating what values are better-high vs better-low for each subset you'rev considering. Or just straight crib from the work I'm surs is already posted in the Wiki.


[1] Ok, not exactly, with how iron and steel are, IRL, but it suffices for this case when we don't fully know if its wrought or cast iron, and there's steel and then there's steel, and apart from the slagging off of impurities we don't know (nor do the dorfs, I imagine) if we're talking about chrome/moly, austenitic stainless, etc...

[Rekov]

Found this on the wiki:
http://dwarffortresswiki.org/index.php/DF2014:Metal#Weapon_and_armor_quality

Density: Used in conjunction with other factors - heavier weapons (higher numbers) hit with more force, light weapons tend to have less penetration. Denser armor absorbs more force from being transmitted through it as well, though low elasticity is much more reliable protection. Value shown here is g/cm3, which is the raw value divided by 103
Impact yield: Used for blunt-force combat; higher is better for weapons, but lower is better for armor.
Impact fracture: Used for blunt-force combat; higher is better.
Impact elasticity: Used for blunt-force combat; lower is better.
Shear yield: Used for cutting calculations in combat; higher is better.
Shear fracture: Used for cutting calculations in combat; higher is better.
Shear elasticity: Used for cutting calculations in combat; lower is better.

So that would appear to answer the question for those stats at least. Elasticity presumably refers to the STRAIN_AT_YIELD.

There are a few more notes here on what matters for what types of weapons:
http://dwarffortresswiki.org/index.php/DF2014:Material_science#Material_and_item_properties

And yeah, I left silver aside and was mostly talking about the generally accepted ranking of materials for edged attacks.

[Putnam]

For adamantine's strain-at-yield values being 0, I think it's because the yield and fracture values are the same - so if you manage to apply 5000000 of that force to adamantine, it'll just break anyway - thus making the strain-at-yield value irrelevant.

No, strain-at-yield is linear. If something has a tensile strain-at-yield of 100000, it'll stretch to 1.5x its length when you apply 1/2 its TENSILE_YIELD. FRACTURE is actually completely irrelevant to strain-at-yield.

Elasticity is strain-at-yield, yes. Strain-at-yield is in units of parts-per-100,000, and is equal to YIELD / relevant elastic modulus, which you can see in the comments of the code.

This information is all on the wiki, though in various places.

[Dwarf_Fever]

Is iron really better than bronze for weapons and armor?

Who said iron is better than bronze? They're basically same tier. That's been the accepted placement for quite some time as far as I know.

Arguably, you shouldn't be making any end-products out of iron anyway. Iron is for steel.

Not that I don't appreciate the science, mind you.

[Mohreb el Yasim]

There seems to be a misconception widely spread about iron being better than bronz. Even IRL they are sane-ish if bronze not being better (but for iron it depends much on the way of forging too)
I think it is due to the bronze age being followed by iron age.
But wasn't because iron > bronze. It is mostly because iron is much easier to produce than bronze.
I love that DF reflects this (and other details) so carefully...

[Putnam]

Is iron really better than bronze for weapons and armor?

Who said iron is better than bronze? They're basically same tier. That's been the accepted placement for quite some time as far as I know.

Arguably, you shouldn't be making any end-products out of iron anyway. Iron is for steel.

Not that I don't appreciate the science, mind you.

With the way the game's properties work, bronze is actually only slightly better than copper for edged weapons and worse for blunt weapons. It's also the worst metal to make armor out of in the game, despite what people may intuit--it has the worst blunt defense of any armor material, is only better than copper for edged defense, and has lower density than copper, which also matters for blunt defense. Brass is actually strictly better than bronze for both weapons and armor.

[Alastar]

If the Material Science page is correct, bronze is rather disappointing.

Armor:
Against edged, shear yield is dominated by shear fracture in calculations (shear yield irrelevant for large contact areas, still much less important for small ones); bronze worse than iron but slightly better than copper.
Against blunt, high impact fracture and low impact yield is desirable, bronze worse than copper (!) or iron.

Weapons:
For edged calculations, shear yield is dominated by shear fracture, iron preferable even for spikes (small contact area, trap component so momentum scales linearly with density rather than degressively).
For hammers, maces and other blunt weapons with a small contact area, the only thing that matters is density.
For blunt hits with a large contact area, mostly from non-penetrating edged hits but possibly mauls/flails, impact yield may also matter.
This is a very nichey pass/fail situation that should rarely matter (weapon with large contact area hits target with contact area near to a cut-off), but bronze has a theoretical advantage here with decent density and high impact yield.