From UESP:


Swords having a 20% faster swing speed is a pretty big deal. A Daedric sword has a base damage of 14, where a similar Daedric mace hits for 16. Even in that state, completely unimproved, the sword is superior (sword DPS is 14, while mace DPS is 12.8). Factor in that improvements will decrease the relative damage difference, and it's not even close.

Unless power attack swing times are standardized for all One-Handers, I don't see how maces can ever be make up the difference, especially given the significantly higher weight and stamina drain of maces.

Yes. It increased the value without improving the armor. I just re-equipped it.



Yes. Smithing, Archery, Enchanting and Alchemy are maxed out on this char and I created a new bow (not sure if this might be the reason though) since I don't have the arcane blacksmith perk and improved it wearing my crafting gear and using a 130% Smithing Potion. The Result was a 593 Damage Bow. Arrow Damage will not be affected though. I would test one-handers and two handers but since my one-handed skill is ~20 with no perks this might not help a lot. Maybe someone with max skills can test this.

I actually translate that as the other way around. 20% Damage increase is nothing compared to bypassing 75% damage reduction. Assuming physical damage resistance cap of 80%:

Sword:
14 damage * 1 .0 weapon speed = 14 dps
14 dps * 1.5 = 21 damage on crit
(0.8 * 14) + (0.2 * 21) = 15.4 average dps (80% of the time doing normal damage, 20% doing critical damage which is 1.5x)
15.4 * 0.2 = 3.08 dps (80% damage reduction on late game enemy)

Mace:
16 damage * 0.8 weapon speed = 12.8 average dps
12.8 * 0.95 = 12.16 dps (80% damage reduction on late game enemy - 75% damage reduction bypass from mace perks)

Early game, with lightly armored opponents, you won't care because the dps difference is small. Late game heavily armored opponents won't notice a sword, but they'll fear a mace. The weight is a concern, yes, but Daedric Sword @ 16 vs Daedric Mace @ 20, so it's not a big difference. Stamina doesn't matter either, since you only need 1 Stamina to Power Attack. A single Absorb Stamina enchant negates Stamina dependencies entirely. This actually makes it sounds like Mace wins outright, even if Power Attack speeds are dependent on weapon speeds.

Or am I missing something?

No need; you got the necesary info. Thank you.

Ancient Knowledge perks increase your heavy Armor skill, not just your Dwarven Armor skill. They also increase your Improve Weapons and Armor skill, as oposed to making you learn Smithing faster. Two bugs in one perk; very entertaining.

its so nice to be able to understand this after a little college algebra. I also really appreciate the work, coming from a non-min-maxer who wants to get the most out of perks.

Bladesman wouldn't make much difference against an armor capped enemy. The problem is that you're assuming a best case scenario for maces. Does any enemy in the game have 80% physical mitigation? If so, how many? If it's just a few boss-type mobs, that's a pretty weak case for maces over swords. But if lots of enemies have non-negligible physical mitigation at higher level, maces could still win out.

So an enemy armor rating of 145/ 20% mitigation isn't the inflection point for mace superiority, because it also has to overcome inferior DPS which (correct me if I'm wrong) get worse as the quality of your weapons increases.

You are correct, but I'm still having trouble.

Best case Swords vs Maces says 3.08 dps vs 12.16 dps.
Worst case Swords vs Maces says 14 dps vs 12.8 dps.

That still appears to be heavily in the Maces favor. You're right that there's a tipping point, but it's just as faulty to assume no armor as it is to assume maxed armor. Unless the tipping point is greater than 284 Armor Rating, maces feel superior. If there's something I'm missing, I'm open to being proven wrong.

Regarding improvements.. the difference in weapon damage (percentage-wise) actually gets smaller as you apply better improvements. (I need to verify this again).

edit: Thinking about Weapon Improvements again. It would actually be beneficial to the mace if the weapon damages diverge with larger improvements. If they got closer, then yes, the sword has a better chance of keeping up. But in my mind, it's still the sword trying to catch up, not the mace. I don't like basing end-game gear decisions on early-game enemy stats. :-p There is also the question you mentioned earlier... how many enemies actually have armor, and how much do they have?

Assuming 3/3 Bladesman or Bone Breaker and unimproved Daedric weapons, the numbers look a little different.

Best case for Maces (armor capped opponent) is 3.22 dps v. 12.16.
Worst case for Maces (opponent with no mitigation, which might not even exist) is 16.1 dps v. 12.8 dps.
Maces compare look great there, but there are other complicating factors that I'll discuss below.


That favors swords over maces and pushes the inflection point to even higher armor rating levels. For instance, an unimproved Elven sword has a base damage of 11, while a similar Daedric sword has a base damage of 14; that's a 27% difference. But we know from Domilasa's testing that at maximum improvement, the difference is only 5%. An unimproved Daedric mace starts out dealing 14% more damage per swing than a Daedric sword, but at maximum improvement, that difference is negligible, thereby granting the Daedric sword a significant DPS advantage.

Thus, at the highest levels of improvement, Maces have to overcome both Bladesman's +15% damage boost (which is universal, regardless of mitigation) and significantly inferior DPS. It's kind of a moving target, so I'm not sure how to calculate it.

Ultimately, we need to know how much armor the average enemy has around level 50, and whether that's affected by difficulty level. If it's significant, Maces are probably better, but my intuition at this point is that Swords are better early and mid game, and may even be superior late game against all but the most heavily armored opponents.

I'm disturbed that our numbers keep coming out differently. Is Bladesman 20% or 30%? Crit damage is 1.5x, yes?

The 5% difference in Domilasa's numbers are between materials on the same weapon type. We don't have numbers on the difference between weapons of the same material, after improvement. We don't actually know which way it goes. I will go dig those up momentarily.

We both keep using the word significant. :-p We should probably stop, and just find some real numbers.

But I think we can both agree that this boils down to enemy armor values at various stages of the game. So where do we want to start, in terms of determining enemy armor values? Is it possible to use the console to look at enemy Armor Skill? Using that + whatever armor they're wearing, we can assume the player model, and reverse engineer their mitigation. Other thoughts?

You're right; it's 20%. I had read that Bladesman crits are 1.5x, and that according to the guide, 3/3 Bladesman equates to a 15% increase in damage. That would all make sense if 3/3 Bladesman provided a 30% chance to crit, but it doesn't. I can't find any reliable info on how much bonus damage Bladesman crits offer. A 20% chance to do 1.5x damage is only a 10% DPS increase over time, which is a pretty crappy return on 3 perk points.


My point was that at maximum improvement, differences in damage seem to decrease significantly. Some aspects of the crafing/ improvement loop must be additive, because if it was all purely multiplicative, initial % differences in unimproved damage would remain the same after improvement as well. It should definitely be tested, but I'd be very surprised if Maces and Swords somehow diverge further in damage after improvement given that damage values all seem to converge after improvement.


I have no clue. A console solution would be ideal, but I don't think it's possible to check the armor rating of an enemy that way. We could always go with Domilasa's tried and true measure of whacking various mobs with a Mace without Bone Breaker, doing it again with Bone Breaker, and inferring armor rating that way.

I went forth and spammed 'getavinfo damageresist' on a number of stuff, with my lvl 55 character, and got a lot of funny numbers.

*updated*

Housecarl in fully improved ebony + shield w. ebony mail artifact:
Current Value: 1001
Computated base: 1001
Reference Base Value: 0
Derived Value: 1001
Modifiers: Temp: 0,00 Perm: 0,00 Damage: 0,00
Armor Perks: All 0es
Heavy Armor: Current Value / Computated Base: 89 Reference Base Value: 100
Gear enchanted with Fortify 1h/bow or magic/elemental resist (no fortify armor enchants).

City guard:
Current Value: 126/89 (shield/no shield)
Computated base: 126/89
Reference Base Value: 0
Derived Value: 126/89
Modifiers: Temp: 0,00 Perm: 0,00 Damage: 0,00
Armor Perks: All 0es
Light Armor: Current Value / Computated Base: 60 Reference Base Value: 60
Heavy Armor: Current Value / Computated Base: 55 Reference Base Value: 55
Wears light imperial armor (studded chestpiece) and a heavy imperial shield.

Ancient Dragon:
Current Value: -448,54
Computated base: 0
Reference Base Value: 0
Derived Value: 0
Modifiers: Temp: 0,00 Perm: 0,00 Damage: -448,54

We have no evidence of convergence or divergence. We've both been confusing difference between materials or weapon classes (one hand, two hand, bow) with differences in improvement between weapon types within the same material and class. I'll get those numbers momentarily, and edit this post with the results.

I've dug up some console commands:

• Select your target while in the console
  
• "getAVinfo DamageResist" from the console should tell us their natural armor.
  
• "getAVinfo LightArmor" from the console should tell us their light armor skill.
  
• "getAVinfo HeavyArmor" from the console should tell us their heavy armor skill.
  
• "getAVinfo ArmorPerks" from the console should tell us their perk usage.
  
• The gear in their inventory after death should tell us their base armor stats

I have a character at level 53, so I'll go start my Beastiary I guess. I think I can write this into a script, and just call it from the console during all my fights. Since I'm high level, this should give a reasonable approximation of enemy armor values. This character is also a stealth type, so I can get close to a lot of other enemies without starting the fight and check armor before and after Alteration Armor spells are cast.

Nicely done. I'll be doing something very similar shortly, but with the whole list of commands I listed above. The Dragon IS confusing.. looks like a variable type mismatch.

A Beastiary is a great idea! No rest for the weary, eh?

I should be able to help with testing of this sort soon. Thanks for your work, Cheshyr.

Found a frost dragon.

0 armor all around, 0 armor perks, light/heavy armor skill at 15.
Dragons seem to be health-sponges.

Updated my earlier post with other values.

I have negative motivation to create a real beastiary, but I will do the bare minimum needed to complete my research. :-p


Thanks for the assist. Your help is most welcome, especially after we get a more formal data collection procedure worked out.

I'll post a list of commands to run later. Health is another good option, but I feel like we're missing something entirely on dragons. If I can get a script together, I'll make sure to post it as well, so we're all gathering the same data.

So, results are in, and...

Daedric Sword vs Mace, 100 skill, No Improvements, from 0/5 to 5/5 Armsman

• 21 : 24
  
• 25 : 29
  
• 29 : 34
  
• 34 : 38
  
• 38 : 43
  
• 42 : 48

Daedric Sword vs Mace, 100 skill, Max Improvements, from 0/5 to 5/5 Armsman

• 90 : 93
  
• 108 : 112
  
• 126 : 130
  
• 144 : 149
  
• 162 : 167
  
• 180 : 186

The difference between Sword and Mace is exactly the same regardless of improvement level. That means improvement scaling favors swords. Let's look at some real numbers now:

Sword: 180 damage * 1.0 weapon speed = 180 dps | 180 dps * 1.5 = 270 damage on crit | (0.8 * 180) + (0.2 * 270) = 198 average dps

• 198 * 0.9 = 178 dps
  
• 198 * 0.8 = 158 dps
  
• 198 * 0.7 = 139 dps
  
• 198 * 0.6 = 119 dps
  
• 198 * 0.5 = 99 dps
  
• 198 * 0.4 = 79 dps
  
• 198 * 0.3 = 59 dps
  
• 198 * 0.2 = 40 dps

Mace: 186 damage * 0.8 weapon speed = 149 average dps

• 149 * (0.9 +0.75 max 1.0) = 149 dps
  
• 149 * (0.8 +0.75 max 1.0) = 149 dps
  
• 149 * (0.7 +0.75 max 1.0) = 149 dps
  
• 149 * (0.6 +0.75 max 1.0) = 149 dps
  
• 149 * (0.5 +0.75 max 1.0) = 149 dps
  
• 149 * (0.4 +0.75 max 1.0) = 149 dps
  
• 149 * (0.3 +0.75 max 1.0) = 149 dps
  
• 149 * (0.2 +0.75 max 1.0) = 141 dps

It appears your 25% is still correct; the turning point is around 145 Armor Rating. The minuscule data we have so far suggests City guards have 126 armor rating, which means swords have an advantage against them. We'll need more data on other creatures before we can be certain.

Before anybody throws a hissy fit.. this is not true Damage Per Second. If you want to get technical, it is Damage Per (Miscellaneous Time Unit), or more accurately Damage Relative to Sword Speed.

Went and cleared a few bandit camps.

The armor value shown are after all calculations from skills and such. I found two bandit highwaymen in identical fur armor, but with different light armor skill. The one with the highest armor skill had the highest armor value shown.

General armor values, bandits
Fur/studded armor: 30-50
Iron armor: 60-80
Steel armor: 90-110
Steel plate armor: 140-180

Higher tiers of bandits (thugs, marauders) had at most 10 extra armor due to higher skill. The biggest variation came from shields.
Steel/Steel plate are generally found only on Chiefs.
So, don't use maces if you fancy yourself the Scourge of Bandits.

Edit: Nevermind.. you're level 55.

So, this makes a good case for Swords are better. With a break-even point of 145 armor rating, they are certainly better against Bandits.

Wow. After improvement but pre-mitigation, swords do nearly 33% more DPS than maces. Thanks for collecting the data.


My calculations were based on displayed armor rating. Does getAVinfo DamageResist report the 100 "hidden" armor you get from being fully clothed?

My initial inflection point came out to be 144 displayed armor rating (which may actually be 244 including the hidden 100) which should be roughly 20% mitigation. I assumed it would be much higher given the large DPS difference between swords and maces post improvement, but I could buy 25%.

The fact that city guards are well under the inflection point at high levels does not bode well for Maces.

My only gripe with using low-armored opponents as our deciding factor is... they're low armored opponents. You're going to one-shot them regardless of what weapon you use. Faster attack speed is moot if you're only swinging once. That said, if there are almost no reasonably armored enemies in the game... then yeah, swords are better. The other half of this is your hidden 100 points. If those are additive to the displayed value, then a mace is superior against even the lowliest bandit.

Unarmored NPCs doesn't seem to have the hidden 100 armor. Checked Falion, random town people and people at the College.
My own character seems to have 50 armor when wearing nothing (player.getavinfo damageresist).

e: And yes, I can't think of any enemies with high armor values. Monsters all seem to be health sponges, and bandits/imps/stormcloaks/thalmors all wear light armor for the most part.

50 Armor sounds like the Lord Stone to me. I just checked my crafting toon, and the console coincided with the inventory displayed armor. No hidden values here. Actual displayed value from the console is correct.

So it appears only the player is allowed to use armor, and Swords are superior DPS against the enemies we have tested. I will continue to gather data, since this disappoints me greatly, but I can't argue with the numbers. Swords are presently better.

I'm not sure it matters much. All that really matters is the mitigation %. 108 displayed armor rating = 208 "true" armor rating = 25% damage mitigation. If your math is correct, that's the inflection point.

The real question is this: what is getAVinfo DamageResist giving you? Displayed armor rating, or true armor rating? That should be easy to test just by going out and hitting stuff with Bone Crusher and then without.

I tested this by looking at myself, and changing my equipment. It reports Displayed Armor Rating. This hidden 100... it's based on what pieces we're wearing, right? Chest is 40, Gloves, Helm, Boot are 20 each? So if the enemy has all 4 pieces equipped, the Displayed Armor Rating should be increased by 100? I've never been quite clear on this hidden 100 stuff. Also, I thought the 567 assumed you had the entire 100 already... If we manually manage the hidden 100, isn't the cap 667?

D'oh, of course

All dwemer bots have 0 armor (from spiders to steam centurion masters), falmers have from 0 to 50, and chaurus reapers have 100.
At this point, I'd say that the only NPCs with more than 200 armor are the wandering non-hostile mercs.

That's right. Displayed armor cap is 567. True armor cap is 667.

So for enemies who are fully clothed, we need to add 100 armor to the getAVinfo DamageResist #. Is it safe to assume that anything not wearing armor doesn't have a hidden armor value?

Nothing else modifies it. Damage Resistance % = (Displayed + Hidden) * 0.12

Alright.. while you're gathering this data, if they have >= 106 armor, we need to know how many pieces of armor they're wearing as well, so we can determine the actual armor rating. Chest adds 40, gloves, boots, helm each add 20. Add this to what the console reports; we'll call it Final Armor Rating.

The exact break-even point is 24.75% Damage Reduction, which is 206 Final Armor Rating, or 106 Displayed Armor Rating assuming they're wearing all 4 pieces.

At this point, it appears nobody in Skyrim wears armor, so Armor Penetration is a useless perk set, and Maces aren't worth the effort. If this is indeed the case, then the 'Best Weapon' debate is now between Swords and Axes.