Fight Club (or how to test builds)

[KDubya]

Skip to the bottom for conclusion  if you are lazy

 

For testing builds I started (in a separate save file to not screw with main game) hiring the highest level adventurers (currently lvl 9) and building various classes to compare what they can be like with various stats and what not.

 

Then I thought - what if I have them fight it out? With force attack I can make them bash each other with auto attacks. Being out of combat the endurance regens between swings but the health damage accrues until there is an auto pause due to low health and that shows the victor.

 

First question is, is this a valid way to compare two melee builds? Spell use or anything that requires combat won't work but melee attacks seem to be fine. The fights last a long time as only the accrued health damage matters which seems like it would balance out any weird lucky string of crits that could quickly overcome the much lower endurance.

 

I set up identical builds stats wise to compare weapon types.

 

The base build is Boreal dwarf Fighter: with the goal of making a high DPS fighter that can hold 3 engagements, and be durable enough to stand in the front. Instead of sword and board for maximum tanking with minimal damage, this is trying for enough tanking combined with good damage so that the frontline contributes more to dropping the enemy rather than being a damage sponge wall.

 

Might 20

Con    8

Dex   18

Per    12

Int       3

Res    17

Talents/abilities - knockdown, weapon focus, defender, wary defender, weapon spec, weapon style feat, weapon mastery, armored grace, critical defense

 

1st up was dual flails vs dual war hammers - with all fine weapons and fine hide armor (DR 7/ 9 vs pierce) dual flails  won 8 lost 2 - reason would be to some combination of the faster attack speed and the inherent graze to hit of the flail.

 

2nd up was dual flails vs dual war hammers but this time in fine full plate (DR14/ 18 pierce) dual hammers  won 5 and lost none - reason is the armor DR overcame whatever speed and hit conversion that the flails had. Somewhere in between DR 7 and DR 14 the higher base damage of the warhammers take over.

 

3rd up was a rematch of dual flails and warhammer guy in full plate but this time gave dual flail guy a fine estoc. - dual flail using an estoc (same weapon group as flails but missing the two handed style) won five and lost none - Even without specializing in two handed weapons, the Adventurer weapon group optimized for dual wield beat the optimized dual wield soldier.

 

4th up was dual flail guy against estoc guy specialized in two handed weapons  in fine hide since it is obvious that in full plate the estoc will win. - estoc two hander won five and lost none.

 

5th up was dual flail guy vs pollaxe guy in fine hide - this was pretty much a push, with two hander winning six and losing 4 to dual flail.

 

6th was two hand guy vs two hand guy but with one guy with 11 dex, 18 per and 18 res, both same abilities. - The faster guy won 5 times to zero versus the more tanky guy.

 

7th was identical two hander guys with one in fine plate, one in fine leather - the plate guy won five to zero

 

8th was two identical two handers with one in fine plate, one in fine robe - the plate guy won five to zero

 

9th was arquebus versus warbow - arquebus won five to zero in both robes and in full plate

 

CONCLUSION

Two handed does as well versus dual wielding at low DR and does much better with high DR when both are spec'd for damage while also having a modal choice for defender for the extra engagements. I had thought the +20% attack speed from dual wield would surpass two handed being a multiplier and all but I was mistaken.

 

Adventurer with flails and estocs did better than soldier with war hammers and great swords. Adding in the ranged options of warbow and arquebus swings it back towards soldier so overall fairly equal if you make use of initial ranged volley before melee.

 

In melee where you will get hit more armor is always better than faster with less armor. I had thought leather made a good compromise for faster with less armor but I was wrong, full plate is the way to go for front line.

 

More dex is better than more tanky stats.

 

Final analysis is that two hander with full plate with max might and dex is the way to go for a frontline fighter trying to maximize damage while maintaining enough tankability.

[mahe4]

what about dual sabre? they do nearly as much damage as a two hander each... i bet you get more damage by dual wielding them...

[peddroelm]

need to repeat each fight a few times (5000 + ) to get statistic relevant results,  because of  the huge amounts of RNG involved (to hit and damage rolls) ..

[Climhazzard]

Seems like there's been a lot of numbers crunched already, I can't take credit for this and can't give credit for this because I don't know who did it but anyways...

 

 

https://docs.google.com/spreadsheets/d/1SHdaQ8yNSvJhxcqFeP9ykXdua_EsjSAb9ypzPA8Oink/edit#gid=1848365110

 

 

Dual wielding sabers is the clear winner atm, but that estoc wins against targets with very high armor.  Consider running your tests with vulnerable attack instead of two weapon style, I don't know which benefits dual wielding more but I can tell you that vulnerable attack benefits a dual wielder a lot more than it does a two hander user.  Used with stilettos you have 8 DR bypass, with the bleak fang stilleto you would have 11, which basically negates the huge penalty fast weapons get from armored opponents.

Edited April 15, 2015 by Climhazzard

[KDubya]

The guys in the attack speed forum post are doing real testing and real math. This is just sort of a thing to do when you are bored, at an inn and about to log for the night. and want to have some sort of a test to back up your theory crafting.

 

In the one sided fights the losing side never had it even close, not sure hundreds of tests are needed. With the way the guys were made deflection > accuracy so there where no crits to wildly affect the outcomes, and with being out of combat only health damage mattered as endurance regenerated between each swing. In a real fight any crit or even higher than average damage rolls would sway the results as endurance is much lower than health.

 

A few times I ran one guy without defender modal which lowered his deflection to be below accuracy. The crits were telling and beat down the guy much faster.

[Epsilon Rose]

I don't think this will give you much information beyond "Who does auto attack better if that's all they're doing", because it ignores all of their other abilities and ignores how they interact with the rest of the party, both of which are incredibly important.