Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: [1] 2 3 ... 5

Author Topic: From Iron to Steel: A survey of v0.43.05 combat [Long] *VERSION UPDATE*  (Read 10697 times)

fragfish

  • Bay Watcher
    • View Profile

I haven't played since v0.34, and for my new forts started to think about how to arm my citizens. Since most combat threads I found on the topic had some years on them, I decided to launch a series of tests to see how things look in v0.43. Exact version used was v0.43.05.

Disclaimer: I originally posted this for v0.43.03. Only afterwards was I made aware of some import changes from v0.43.03 to v0.43.05. I have thus rerun the tests and updated the results. The discussion up to page 4 is on the previous version.

The major differences in numbers from 0.43.03 to 0.43.05 were:
* helmets are no longer such a huge problem for iron weapons because they are now removed from helpless opponents
* combat seems to have gotten harder, with dwarves faring worse across the board
* undead removing armour has made them a lot stronger


The main question for this post is: How do the different damage types compare to each other, and how does the step from iron to steel influence them?

I will focus on my three main weapons:

  • Battle Axes (slashing)
  • Spears (piercing)
  • War Hammers (bashing)

Some Teaser Results

  • Hammers have become more efficient, no longer breaking every bone before killing the opponent
  • There seems to be a bug with undying undead heads
  • Hammers are very inefficient against inorganic enemies
  • It is now possible for reanimated arms to remove your helmet and use it to crack your skull

The Setup

I ran groups of 10 dwarves against 7 different enemies:

  • 10 Goblins with Pikes
  • 10 Goblins with Mace & Shield
  • 10 Goblins with Short Sword & Shield
  • 10 Goblins with Whip & Shield
  • 50 Human animated dead and one human necromancer
  • 1 Bronze Colossus
  • 10 Iron Men

Each test was repeated 20 times for each weapon as well as a mixed squad, both in iron and steel variant, for a total of 1120 tests.

A note on margin of error: I computed standard deviations, though I left them out for most tables so as not to overload them. I realise this is not the best model for numbers which are bounded from below and above, so take it with a grain of salt.

The tables that have deviations should give you a good idea of the range of error we're dealing with - it's rather substantial. These results should not be considered final, but they do show clear trends in most places.

More details on setup, skills, equipment
Spoiler (click to show/hide)

Squad Effectiveness Overview

Listed here are, for each squad of dwarves, average survivors per combat with standard deviation in square brackets, and number of won combats in round brackets.

Squad/Enemy   Goblin Pikemen   Goblin Macemen   Goblin Swordmen   Goblin Lashers   Undead and Necro   Bronze Colossus   Iron Men   
Iron Axes6.2[2.4] (18)2.2[2.5] (11)2.4[2.9] (9)2.6[3.2] (9)8.7[2.4] (19)0 (0)0 (0)
Steel Axes9.8[0.4] (20)5.4[3.0] (17)7.1[1.9] (20)6.9[3.1] (18)2.3[3.3] (7)8.6[2.3] (19)5.8[1.6] (17)
Iron Spears4.3[3.1] (14)2.2[3.3] ( 8 )2.0[2.7] ( 8 )2.4[3.0] ( 8 )0 (0)0 (0)0.5[1.5] (2)
Steel Spears9.7[0.5] (20)8.4[1.1] (20)9.2[0.8] (20)9.2[0.8] (20)0 (0)8.6[1.9] (20)0.2[0.7] (1)
Iron Hammers8.8[0.9] (20)5.3[3.6] (15)6.2[2.9] (18)4.5[3.8] (12)0 (0)0 (0)0 (0)
Steel Hammers8.5[1.3] (20)6.1[2.9] (17)8.1[1.6] (20)7.3[2.1] (19)0 (0)0 (0)0 (0)
Iron Mixed6.4[2.5] (18)1.6[2.3] (7)3.4[3.0] (12)4.4[3.6] (13)0.3[1.3] (1)0 (0)0.1[0.4] (1)
Steel Mixed8.6[1.3] (20)4.9[2.9] (17)8.0[1.1] (20)7.9[1.3] (20) 0 (0) 7.8[3.4] (17)0.7[2.1] (2)

Excluding for a moment the Undead column (more about that later on), this paints a pretty clear picture:

  • Steel Axes and Spears are top, winning most combats with little to no losses. Steel Axes are the only weapon to regularly beat Iron Men, putting them ahead of the Spears.
  • Both kinds of Hammers form a middle ground, with Steel Hammers slightly ahead.
  • Iron Spears Axes are dead last.

Living up to their name, the mixed squads had mixed results: better than the worst weapon in the squad, worse than the best.

Weapon Efficiency

Listed here are numbers on how efficient the weapons were: How many hits do they need to kill, what kind of injuries do they inflict?

These numbers do not include shots blocked or parried - see further below for those. Included are only attacks that specify the weapon. Averages here are over series (one combat per weapon).

A note on attack types:

  • Hammers deal only blunt damage.
  • Axes have a blunt "slap with flap" and a blunt "pommel strike" attack. Both occur at about 3% and usually achieve nothing, except for a very rare crushed hand or foot.
  • Spears have a blunt "bash with shaft" attack that occurs at about 4% and is mostly inefficient. It does have a somewhat higher "crush rate" than the blunt axe attacks.

vs Goblins

In this table,

  • Minor injuries are bruises (not to the brain), tearing the fat or similar mostly harmless things
  • Major injuries take time to heal, but are not lethal - e.g. shattered/fractured bones or torn muscles.
  • Severe injuries are permanent, but not immediately lethal: severed extremities, broken lower spine, crushed feet
  • Fatal injuries are just that: Crushed Heads, Torn brains and the like.

To compute the averages and deviations, I grouped together the goblin enemies, so the numbers are computed across 20 sets of 4 combats each.

Weapon/Injury  Total Hits  Deflected  Minor      Major      Severe    Fatal       Total Kills
Iron Axe854[137]51.2%33.7%0.3%14.6%0.2%33[4]
Steel Axe480[43]10.7%2.9%44.7%36.1%5.6%39[3]
Iron Spear1315[585]34.6%43.6%14.0%7.3%0.5%29[6]
Steel Spear483[35]0.8%5.2%37.7%50.0%6.2%40[0]
Iron Hammer626[61]2.9%49.5%26.2%15.5%5.9%37[3]
Steel Hammer711[55]2.4%48.4%27.3%16.2%5.7%40[1]

Listing the amount of hits needed per kill, and the percentage of deflection/minor injuries in parantheses, we get a list that conforms well to the squad efficiency found above:

  • Steel Spears: 12.1[0.9] (6.1%)
  • Steel Axes: 12.3[0.6] (55.4%)
  • Iron Hammers: 16.9[1.2] (52.4%)
  • Steel Hammers: 18.0[1.4]  (50.8%)
  • Iron Axes: 26.2[3.7] (84.9%)
  • Iron Spears: 47.7[23.1] (78.4%)

The advantage of spears at steel level is well within margin of error. Further testing would be required to establish if they are truly better against organics with some armour.

As for the hammers, the results are somewhat surprising. I suspect it's down to deviation, but either way there seems no great advantage of steel hammers over iron ones. Time to test the silver hammers, I guess.

vs Undead

Listed here are the statistics for weapon attacks against the undead. Killed Zombies refers to kills on the initial 50 undead. The kills on reanimated corpses/parts are also given in relation to killed initial zombies to account for the effect that weapons killing less zombies overall produced less corpses to reanimate.

The numbers are averages per Combat, computed over 20 runs. I only listed deviation for efficency (in square brackets), so as not to overload the table.

WeaponTotal Attacks  Killed Zombies  Killed reanimated Bodyparts (per killed zombie)  Killed reanimated Corpses (per killed zombie)  Efficiency in strikes/kill 
Iron Axe3725069 (1.4)49 (1.0)2.4 [0.9]
Steel Axe45848151 (3.2)144 (3.0)1.3 [0.2]
Iron Spear361235 (0.2)21 (0.9)8.9 [3.5]
Steel Spear417286 (0.2)15 (0.5)9.9 [3.8]
Iron Hammer529181 (0.07)0 (0)30.7 [8.6]
Steel Hammer564181 (0.06)0 (0)34.2 [10]

Sometimes for steel axes, there is a reanimated human head which just won't die. 100 Pages of Combat Logs listing its brain being torn apart, and it still is not struck down. When the steel axes won, I twice aborted after only one undying head was left, surrounded by steel axedwarves happily chopping away at it.

There are two winners here:

  • Hammers are really good at making sure things stay dead. They take a while to kill stuff, though - in fact, they lost the combats because they just could not kill the undead fast enough and were worn down by cumulative damage.
  • Axes are really good at ripping through undead fast. Steel axes have a strike:kill ratio of 1.3 - that's almost a kill per strike. The iron axes reliably tore through the undead and dismembered the Necromancer in record time. Steel axes were even more efficient in dismembering stuff - to their own detriment, as they then got swamped by undead parts. However, axes produce a lot of parts for reanimation. In addition, mutilated (chopped down) corpses can be resurrected, while mangled (crushed heads/bodies) corpses can not. So axes are good for getting to a revealed Necro through his hordes or dealing with undead where there is no reanimation, but quite dangerous if corpses are immediately reanimated by the environment or an invisible necromancer.

Spears seem to be a bit in the middle ground - somewhat less severed limbs, but still too slow to get to the Necro reliably. It seems piercing attacks are just not very good against undead.

vs Inorganic

The fights against colossus and iron men went rather differently, so the results are listed separately.

Colossus

The results here can be pretty easily summed up: All iron weapons, as well as steel war hammers, have a 100% deflection rate. The Colossus won all those combats without a scratch. For efficiency, the numbers are thus only relevant for the two weapons that could acutally damage it (standard deviation in square brackets):

Steel Spears took an average of 148.5 [21.7] attacks to kill the Colossus. 1.6% [1.2] of attacks glanced away.

Steel Axes took an average of 27.0 [14.6] attacks to kill the Colossus. 4.6% [4.3] of attacks glanced away.

The advantage of axes comes from 7% decisive injuries (removed parts/outright kills) for spears vs only 54.3% for axes. Spears often removed hands/feet/leg/arms while axes went directly to cutting apart the torso.

For the survivor ratio, it seems Spears might be slightly better - maybe due to removing hands and feet first, making the colossus safer overall. However, the difference is small enough to be within margin of error, so more tests would have to be run to see if this is indeed a trend.

Iron Men

This appears to have been the most difficult combat. In this table,

  • Minor damage is broken gas, dents, chips. This amounts to little.
  • Major damage is fractures. They add up to enable destroying a part.
  • Severe damage destroyes parts. If not instantly lethal, they at least make the combat easier by hampering the opponent.

The numbers are averages over 20 combats. I listed deviation only for strikes, kills and efficiency, so as not to clutter the table.

Weapon Total Strikes  Deflections  Minor     Major     Severe    Killed Iron Men
Iron Axe59.5 [22.2]29.5%29.1%7.1%33.4%2.2 [1.5]
Steel Axe35.8 [7.3]1.7%1.5%0.2%93.0%9.6 [1.2]
Iron Spear82.1 [43.9]2.3%28.5%52.4%16.6%2.5 [2.9]
Steel Spear82.1 [33.5]0.7%32.9%48.0%18.0%3.9 [2.9]
Iron Hammer70.6 [24.8]0%55.7%32.3%11.8%1.4 [1.4]
Steel Hammer76.8 [29.4]0%55.1%33.6%11.1%1.1 [1.3]

This gives the following list for efficiency (strikes/kill):

  • Steel Axe 3.7 [0.5]
  • Steel Spear 27.8 [19.4]
  • Iron Axe 33.6 [23.7]
  • Iron Spear 43.1 [22.1]
  • Iron Hammer 46.8 [18.8]
  • Steel Hammer 61.7 [23.0]

The large deviation for all but steel axes is due to the fact that the dwarves lost the combat and kills are usually via cumulative damage - meaning some very damaged iron men did survive.

Steel axes were the only weapons that could reliably win the combat. Note how Hammers have moved down here in comparison to their performance against goblins.

In conclusion, inorganic enemies seem best killed by tearing them apart until you severe something crucial. Axes excel at this as long as they manage to penetrate the armour. Hammers are simply horrible, taking forever to smash their way through the layers until parts are crushed.

Threat Assessment: Enemy Weapon Effectiveness

Goblins

Listed here are the statistics for each goblin weapon, in similar fashion to the above dwarven weapon table. The averages are computed over 20 sets of 8 combats each (against steel and iron axes/spears/hammers/mixed).

Weapon        Total Hits   Deflected   Minor     Major     Severe    Fatal    Killed Dwarves
Iron Pike622 [151]24.7%49.0%15.8%9.4%1.1%18 [5]
Iron Mace819 [135]4.6%59.8%17.7%13.7%4.0%44 [9]
Iron Short Sword785 [103]39.9%37.5%5.1%17.0%0.6%34 [5]
Iron Whip573 [125]0.1%81.0%2.5%11.5%5.0%35 [9]

Iron Whips completely ignore armour: The deflections all glanced off eyelids. (A lot of attacks glance of eyelids, even steel axes.) However, they usually only chip a bone or tear some skin; their lethality is mostly from hitting the spine or head.

Overall Efficiency, as hits/kill and total deflections + minor injuries:

  • Iron Whip 16.6[1.8] (81.1%)
  • Iron Mace 18.8[2.2] (64.5%)
  • Iron Short Sword 23.5[3.0] (77.3%)
  • Iron Pike 36.0 [7.4] (73.7%)

Interestingly the goblin mace, while being the deadliest iron weapon like the hammer, was no more efficient than its slashing counterparts. I cannot currently say if that is due to Hammer vs Mace or some other factors.

Inorganic

Inorganic enemies sometimes took weapons, shields or armour from the dwarves. This is usually a good thing: A colossus hitting you with a wooden shield is far less lethal than a colossus pushing you around.

In the rare case were they get a steel axe this could be more problematic - then again, the guy holding the steel axe is usually cut apart by the axedwarves before he can do a lot with it, given he has no skill wielding it.

Attacks statistics here exclude those made with captured weapons.

The numbers are calculated over 20 sets of 8 combats (one per squad).

Enemy   Total Hits   Deflections Minor     Major     Severe    Fatal      Killed Dwarves 
Colossus237[54]2.3%33.8%8.5%38.0%17.4%55[4]
Iron Man1081[625]28.8%22.4%20.3%20.7%7.8%73[4]

This gives the colossus an attack efficiency of 4.3[0.9] hits/kill with 36.2% minor injuries/deflections, while the iron men need 14.8[8.2] hits/kill with 51.2% minor injurise/deflections.

The large variation for iron men is due to some freak fights where only one ot two remained standing on both sides: The dwarves unconscious, the iron men too damaged to get anything done. The logs are then full of "no force" and "deflected" messages. Without these combats, effiency is closer to 11[1].

Undead

Finally, the statistics for the undead attacks show how little iron-clad dwarves have to fear from individual unarmed attacks. Numbers are again averages over 20 sets of 8 combats.

Enemy      Total Hits   Deflections Minor     Major     Severe    Fatal      Killed Dwarves 
Undead18707[3304]37.6%54.6%7.3%0.4%0.189[22]

This sums up to 221 hits/kill, with 92.2% deflections and minor injuries.

Note that on 43.03, the number of hits/kill was ten times that. It seems the new "deflections can cause injury on nearby parts/joins" and "remove armour from unconscious" mechanics really benefit the undead.

Avoiding Harm

Armor Effectiveness

Listed here are the percentages of attacks deflected for each piece of dwarven armour, against goblins, colossus and iron men. I left out the undead - basically everything deflects here.

As said, whips completely ignore armour. They don't even list what they are ignoring in their hit description (such as "through the iron chain leggings").

Again, the table contains average and standard deviation computed over 20 sets of 8 combats.

Enemy      Helm        Mail Shirt   Leggings   Gauntlet    High Boot   
Pikemen49.5 [8.9]1.0 [1.0]23.0 [3.9]63.5 [7.3]73.1  [6.1]
Macemen9.9 [3.3]02.8 [1.4]15.4 [6.2]27.2 [8.0]
Swordmen91.3 [3.0]1.1 [0.9]36.8 [4.0]94.2 [3.4]94.4 [2.1]
Colossus1.2 [2.6]001.4 [2.3]1.0 [2.5]
Iron Men27.6 [16.4]47.5 [19.9]46.3 [19.8]36.2 [20.3]17.7 [18.2]

Interesting to note here is that against the goblin weapons, the "plate" armour pieces (Helm, Gauntlet, High Boot) fared better than the mail ones (Shirt/Leggings), while against Iron Men the mail parts did better.

Also note that this chart only lists deflections. For blunt weapons in particular, the amount of hits lessened from broken bones to bruises would probably be more interesting.

Lastly, the Iron Men values again show huge variation due to the reason mentioned above: In a few freak fights, neither side was able to finish the other off: The Iron Men were too weak to do anything, the dwarves unconscious. Actual deflections against vital Iron Men are around 10 % less.

Defence

Listed here are avoided attacks (with standard deviation in square brackets) against the Goblins, Colossus and Iron Men.

DefenceGoblinsColossusIron Men
Block29.3% [1.2]52.9% [5.2]35.4% [9.1]
Parry3.0% [0.2]1.2% [0.3]0
Dodge11.6% [1.1]7.1% [1.0]7.2% [2.2]

The low block rating against goblins is partly due to the fact that dwarves going down (from exhaustion or wounds) would often receive a number of further hits before they died, none of which they could block. In contrast, the bronze colossus did not have a problem dispatching dead dwarves.

This is substantiated by block rates for weapons varying from 24.9% for iron spears to 45.6% for steel spears, in accordance with the general success of the weapon.

I did not investigate parry and dodges further. Note that no dwarf had any experience dodging.

Removal

The new mechanics giving sentient creatures the ability to remove armour from helpless foes have done the most to shake up some numbers from v0.43.03.

Dwarves and Goblins removed little armour beside the occasional helm, in particular iron spears/axes/short swords/pikes.

The undead, though, removed 98% of all helmets, 76% of mail shirts, 73% of leggings, 44% of gauntlets and 43% of boots. They have become far more deadly in numbers, as they will quickly remove armour and use it as a blunt weapon.

The inorganic enemies only did any removing was when they had gotten hold of a dwarven weapon to use.

Wear

With the new combat wear mechanic, establishing a well-equipped military has finally gotten somewhat harder. Listed here are, for each attacking weapon or creature, number of armour pieces.

Tracking wear is difficult: Sometimes, items skip a level of wear, going from x directly to XX or from undamaged to X. Other times, they take the normal progression.

Listed here are the numbers of armour pieces that were listed as worn (x), damaged (X), tattered (XX) or broken per combat. A number in parantheses indicates total amount, where the number was too small for percentages to be meaningful.

Note that, because of the tracking difficulty described above, categories are not inclusive. An item may be listed in all four. To get more reliable numbers, 1-on-1 testing would be needed.

Undead are not listed because they only inflicted wear when using iron armour as a weapon.

Attacking Weapon Iron Helm    Iron Gauntlet  Iron Mail ShirtIron Chain LeggingsIron High Boot
Steel Axe (3)/(1)/0/0  2.9/0.6/0.1/(2)  5.5/2.9/1.3/0.5  6.0/3.7/1.9/1.0  2.0/0.3/0.1/(1) 
Steel Spear 7.5/0.3/0/0  4.0/0.9/0.2/(2)  7.3/5.8/5.4/5.6  7.4/5.9/4.0/2.5  3.5/0.9/0.2/(4) 
Iron War Hammer 5.7/2.8/1.0/0.6  5.3/1.3/0.3/0.2  2.4/0.5/0.1/(3) 
Steel War Hammer 6.2/3.2/0.6/0.7  5.8/2.2/0.6/0.2 2.5/0.9/0.2/0.2 
Iron Mace 2.6/1.2/0.4/0.2  0.8/0.1/(1)/0 0.5/(4)/(1)/0 
Iron Whip 2.6/1.1/0.5/0.6  4.1/1.8/0.8/0.2 1.7/0.7/0.3/0.3 
Iron Men 5.4/0.3/0.5/0.6  2.5/1.0/0.4/0.1 1.4/0.4/0.1/0.1 

Number differences, in particular for the blunt weapons, are at least partly down to Goblins loosing more fights and thus having less time to destroy armour.

Not listed are iron axe, iron spear and iron short sword because they did not inflict any wear. Iron Pikes managed to break one helm and wear down about a single-digit number, so I did not include them either. The colossus did inflict only little damage to equipment, so I didn't list it as well.

Also not listed are shields - shields don't seem to suffer wear from blocking, but from attacking with them. That makes wood shields somewhat less useful, as they tend to break quickly when bashing armoured foes. Average numbers of shields destroyed per combat were 3-6 in the goblin combats (depending on how fast the weapons finished the combats) and marginal against the inorganics.

Of particular interest is that blunt attacks seem not to damage the "flexible" armour pieces.

Fun fact: Only right gauntlets were ever damaged. Guess you'll have a lot of left gauntlets in good condition when using shields.

Conclusions/tl;dr

The tests allow some conclusions about roles and effectiveness of the different weapons:

Hammers have two major roles: Give you a weapon that can do something even if you have far worse material, and making sure stuff stays dead. They are quite bad against inorganic enemies. Against a bronze colossus, they seem to not do anything at all!

Axes are heavily reliant on material, but as long as they have the edge, they are absolutely lethal, far outclassing anything else. Their only trouble is stuff that just won't stay dead.

Spears are somewhat of an off choice, beat by axes if the material is better, by hammers if equal or worse. I guess they make for a good progression from iron upwards if you plan to do a lot of fighting with iron weapons (slow acquisition of steel). They also seem good at shredding lesser armour, if that's what you want.

Overall, Goblins seem to be mostly on par with dwarves. Even with steel axes, a combat 10v10 vs iron-equipped Goblins will leave permanent injuries and likely some dead. Steel armour or superior skill can presumably offset this.

A very important conclusion is that you should absolutely bring a shield. The goblin pikemen, the only goblins without a shield, fared far worse than other goblins because they were missing that crucial layer of protection.

So, that's what I got. If hope found it even half as interesting to read as I found it to assemble : )

Any feedback or comments are welcome.
« Last Edit: January 26, 2017, 08:46:42 am by fragfish »
Logged

PatrikLundell

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43 combat [Long]
« Reply #1 on: January 10, 2017, 05:49:19 pm »

I didn't read the whole thing, but I still have a few comments:

You didn't specify the version number, and it's of importance as some changes were made during the 0.43.X releases:
- Armor and weapon damage was introduced in 0.43.05. This should have some significant effects.
- When enemies were armored up properly some new mechanics were introduced, so joints can be twisted, and a blow that fails to penetrate a helmet may still break the neck.
- Sapient enemies now have the smarts to remove the helmet of a defenseless opponent to be able to kill it.

The standard recommendation on the forum is to use silver war hammers rather than steel.
Logged

fragfish

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43 combat [Long]
« Reply #2 on: January 10, 2017, 05:58:36 pm »

I didn't read the whole thing, but I still have a few comments:

You didn't specify the version number, and it's of importance as some changes were made during the 0.43.X releases:
- Armor and weapon damage was introduced in 0.43.05. This should have some significant effects.
- When enemies were armored up properly some new mechanics were introduced, so joints can be twisted, and a blow that fails to penetrate a helmet may still break the neck.
- Sapient enemies now have the smarts to remove the helmet of a defenseless opponent to be able to kill it.

The standard recommendation on the forum is to use silver war hammers rather than steel.

I added the minor version, it was on 0.43.03. I was not aware that there were such big changes : (
It sounds like that might change a lot, in particular the helmet thing ... I might just have to rerun the tests and see how the numbers change. If I can find the time again, that is.

I also ran some tests with silver hammers, and it did not seem to change much. In particular, the Colossus still was completely unaffected.
Logged

Ironfang

  • Bay Watcher
  • Cats adopting dwarves isn't annoying, it's realism
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #3 on: January 10, 2017, 07:19:39 pm »

Hmm, no wonder my first Forgotten beast killed 20% of my fort. I only had silver maces, obsidian short swords, and bone bolts.

I was fighting an aluminum quadruped, the thing killed over 30 dwarves before dying to a woodcutter with a decent axe.

I would also suggest possible tests using crossbow bolts. 
Logged
Human King: "So, how was your travel to dwarven lands?"
Human Diplomat: "Never piss off the dwarves, they have an army of hydras"

Dwarf Cook: "Another baby hydra meat roast coming up!"

MorsDux

  • Bay Watcher
  • If you only have a hammer, all problems seem nails
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #4 on: January 10, 2017, 07:32:30 pm »

awesome work!!
Logged

muldrake

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #5 on: January 10, 2017, 07:43:35 pm »

Hmm, no wonder my first Forgotten beast killed 20% of my fort. I only had silver maces, obsidian short swords, and bone bolts.

I was fighting an aluminum quadruped, the thing killed over 30 dwarves before dying to a woodcutter with a decent axe.

I would also suggest possible tests using crossbow bolts.

This really makes mixed choose-your-own-weapons squads look pretty bad for all around purposes.

I have read that bismuth bronze is the best metal for crossbows, but what about the bolts?  Metal is often scarce and spending it on bolts that then lie around the fort (and/or re-melting them) seems pretty inefficient.  (Also I rarely find much bismuthinite.)
Logged

Ironfang

  • Bay Watcher
  • Cats adopting dwarves isn't annoying, it's realism
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #6 on: January 10, 2017, 08:50:49 pm »

For bolts, you run into the mix of hardness and weight, I suggest copper for bolts.

And the point of the squads I had against the FB, it was in fact the first thing I had to really fight in Dwarf Fortress. It was my first fort that I had stable, and I used it for learning stuff.  I was on an island, so no goblins. And the martials I had for the weapons were completely unsuited for fighting it. 
Logged
Human King: "So, how was your travel to dwarven lands?"
Human Diplomat: "Never piss off the dwarves, they have an army of hydras"

Dwarf Cook: "Another baby hydra meat roast coming up!"

fragfish

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #7 on: January 11, 2017, 02:41:36 am »

Thank for the Feedback so far!

Since my saves for the arena setups also work with v0.43.05, and I have the analysis in place anyhow, I'm willing to rerun this with the newest version if there is sufficient interest. So far, it seems like the "glancing off helmets endlessly" problem is really gone now.
Logged

Nilbert

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #8 on: January 11, 2017, 08:40:53 am »

This is a cool post and a great idea!  Two questions:

1.  If you rerun the test, would you consider running one with a mixed squad (e.g., 2 spears, 2 swords, 2 hammers, 2 maces, and 2 axes).  There may be some advantage to having a combination.  (it also tests muldrakes assertion directly; I am guessing a lot of us use mixed weapon squads and it'd be neat to know if it actually reduces effectiveness)

2.  Did you try the steel breastplate for defense?
Logged

fragfish

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #9 on: January 11, 2017, 09:19:17 am »

1.  If you rerun the test, would you consider running one with a mixed squad (e.g., 2 spears, 2 swords, 2 hammers, 2 maces, and 2 axes).  There may be some advantage to having a combination.  (it also tests muldrakes assertion directly; I am guessing a lot of us use mixed weapon squads and it'd be neat to know if it actually reduces effectiveness)

I might do that, yes. Though since each "weapon config" means 140 further tests (10 each all 7 scenarios and both iron and steel), I'll not go overboard with the mixing. One mixed squad should suffice. I'd also be hesitant to include weapons that don't exist on their own, as it would then be hard to tell whether the mixed squad did worse/better because it is mixed, or because it contains weapons that are better/worse than the "pure" squad.

From a Fortress Mode Point of View, I always suspected mixed squads would gather experience slower, as a "mace demonstration" probably does no good to 8/10 dwarves in above mixed squad.

2.  Did you try the steel breastplate for defense?

I don't know if I will. Presumably, steel armour will probably make goblin pikes and short swords mostly obsolete - they have enough trouble penetrating armour as it stands. What particular comparison would you be most interested in regarding steel breastplates?
Logged

oldmansutton

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #10 on: January 11, 2017, 10:29:15 am »

A very interesting read, thanks for your work!
Logged
I suggest using kilokittens. As cats are 10X the volume of kittens. That way, 50 cats would be .5 kilokittens.

100 cats would be 1 kilokitten.

Nilbert

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #11 on: January 11, 2017, 10:49:04 am »


I might do that, yes. Though since each "weapon config" means 140 further tests (10 each all 7 scenarios and both iron and steel), I'll not go overboard with the mixing. One mixed squad should suffice. I'd also be hesitant to include weapons that don't exist on their own, as it would then be hard to tell whether the mixed squad did worse/better because it is mixed, or because it contains weapons that are better/worse than the "pure" squad.

From a Fortress Mode Point of View, I always suspected mixed squads would gather experience slower, as a "mace demonstration" probably does no good to 8/10 dwarves in above mixed squad.



I agree that one mixed squad would suffice, and thanks for considering it!


I don't know if I will. Presumably, steel armour will probably make goblin pikes and short swords mostly obsolete - they have enough trouble penetrating armour as it stands. What particular comparison would you be most interested in regarding steel breastplates?

I have had a quandary in my fort for a while over steel breastplates and steel mail shirts.  Steel breastplates I always assumed were better, but they don't cover the upper arms (I believe).  In previous forts, I was always seeming to have legendary dwarves die from jumping out of the way and crushing or opening an upper arm artery when they hit an obstacle or skidded on the floor.  This I found out was due to using only steel breastplates and having exposed arms.  I then decided to use both a steel mail shirt and a steel breastplate, but it weighs a lot more and my army became less effective moving.  I am considering using mail shirts instead of breastplates or using leather armor and a breastplate.  For now, I have gone back to breastplate with no leather armor or mail shirt.

Long story short, I think an interesting test would be breastplate vs. mail shirt.
Logged

fragfish

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #12 on: January 11, 2017, 10:58:54 am »


I have had a quandary in my fort for a while over steel breastplates and steel mail shirts.  Steel breastplates I always assumed were better, but they don't cover the upper arms (I believe).  In previous forts, I was always seeming to have legendary dwarves die from jumping out of the way and crushing or opening an upper arm artery when they hit an obstacle or skidded on the floor.  This I found out was due to using only steel breastplates and having exposed arms.  I then decided to use both a steel mail shirt and a steel breastplate, but it weighs a lot more and my army became less effective moving.  I am considering using mail shirts instead of breastplates or using leather armor and a breastplate.  For now, I have gone back to breastplate with no leather armor or mail shirt.

Long story short, I think an interesting test would be breastplate vs. mail shirt.

Hmm, I always use Mail Shirts and Breastplates on any serious military dwarves. If I use only one, I also go for mail shirts - having uncovered body parts is really not good. Maybe someone has an answer to this already? Otherwise, yes, it would be an interesting question.
Logged

Sanctume

  • Bay Watcher
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #13 on: January 11, 2017, 11:04:29 am »

I think I read somewhere in the update about dwarfs will remove enemy helmets when they are down / unconsicous.

Fleeting Frames

  • Bay Watcher
  • Spooky cart at distance
    • View Profile
Re: From Iron to Steel: A survey of v0.43.03 combat [Long]
« Reply #14 on: January 11, 2017, 12:14:17 pm »

- Well, Iron axes performed better than I expected. Though Iron spears having similar result versus macemen and pikemen while iron axes had completely different results versus those is weird, given that pikes are essentially longer spears - oh wait, you didn't give them shields. I see.

Their not causing heads might be specific to picked skill as well, as going by wiki higher skill hits harder. An interesting idea nonetheless.

- It seems spears can certainly penetrate equivalent material armor - but seem rather ineffective against equivalent material inorganics. If best you have is bronze, either hammer or spear would be reasonable, with axe to be avoided.

Also seems that spears tend to cause less severe injuries.

- Possible future consideration testing target would be something big and meaty, to simulate fleshy FBs.

An interesting future weapon to test would be dwarven pick, though this requires some messing around to give wielders proficent minining. It does have the x2 velocity multiplier, with contact area is between spear and an axe.

- Also, iron men are pretty kickass. I wonder where's the "whip apart" point for hit area? Would dagger's 5 be sufficient?

Bit of a moot point, admittedly.

- Surprising to me that mace had higher rate of fatal and severe injuries than whip. I have seen piles of bruising myself, so that wasn't surprising, but lower rate of fatalness for whip is.

- Also surprising was the high deflection rate on gauntlet and high boot, compared to the chain pieces - is this because they're rigid, I wonder?

Note that standard suggestion is to use greaves for leg coverage due that rigidness being supposedly more effective versus blunt damage, and I think steel breastplate with spoiler shirt.

- Seems shields have been nerfed? Bit of a standard deviation there, though.

The lower blocking rate versus goblins might also be due goblins' proficent skills - it is harder to block skilled combatants.

- On another weapon not featured, I have tested morningstars versus bronze colossi myself a little [43.03]. I found low odds for an omni-grandmaster dwarf clad in steel beat one when using steel or adamantine (3/25 on both in 28 tests), or failing that, at least paint completely red, and no chance at all  when using inferior materials such as copper or rose gold. Those weapons versus each other performed relatively equal in 1v1 (31-28-25 for steel-adamantine-rose gold).

Note for future setups:

Vanilla dwarves have [TRANCES] token, allowing them to get martial trances. This is rather random and an absolutely huge boost, to the point where battles between groups of dwarves can be more about who gets sooner/more trances.

For instances where it shows up with similar weapons here would be Goblin Pikemen versus Iron spears or perhaps goblin Macemen versus Iron Hammers.

As such, I personally default to goblins for my squad-based testing, should I not be able to do 1v1 tests. Dwarves are stronger, sure, but standard deviation should be smaller.

One that note, the snowball effect might be worth considering with larger squads; 100 times 1 goblin versus 1 bronze colossus can have different result than 1 100 goblins versus 100 bronce colossi - I'd expect the second option to help underdog kill more.



@bolts: Bolts are edged, so sharpness counts. See thread: Iron = useless crossbow bolts?

Furthermore, bismuth bronze has exact same stats to regular bronze, expect for the value.

@demonstrations and mixed squads: In fort mode, they're relevant mostly for bringing dwarves up to speed on a skill, i.e. when you increase the size of your military. Besides, once sparring, the weapon skill increases the fastest.

I don't recall whether multiples of train 2 minimum, preferred for encouraging sparring, allowed for larger demonstrations or not.
Pages: [1] 2 3 ... 5