73 réponses à ce sujet

[Texasmotiv]

20 promotions / +10 % attack power equals an additional 3.58% damage. Boooo. Willpower stinks.

 

Yeah 40 hours of grinding for 100 more damage. ugh...

 

Willpower is definitely not the end all-be all of damage. That designation belongs to cunning(and crit damage). I think Yarpen had a formula that he used to calculate a smoothed out damage increase from Cunning. I may compare and contrast this to willpower. Also it would be interesting to see if you get more DPS overall from 20 willpower, 20 cunning or 10 willpower/10cunning. Of course with varying levels of +Crit damage. This seems like it will be much more complicated to calculate.

 

One other thing to look at/think about is that willpower will be more important for classes that commonly attack from flank, like assassin or alchemist. Especially with talents like sneak attack, which will invalidate the need for cunning to get high crit. With flanking your crit damage and flanking damage add together to multiply your attack% gains from willpower, making it more effective per point. I can try to illustrate that as well when I have some time to theorycraft.

 

 

Edit: Due to my variables being off in my calculation its actually 50 damage. Much more 'Ugh-worthy'.

[Denrok1]

Yeah 40 hours of grinding for 100 more damage. ugh...

 

Willpower is definitely not the end all-be all of damage. That designation belongs to cunning(and crit damage). I think Yarpen had a formula that he used to calculate a smoothed out damage increase from Cunning. I may compare and contrast this to willpower. Also it would be interesting to see if you get more DPS overall from 20 willpower, 20 cunning or 10 willpower/10cunning. Of course with varying levels of +Crit damage. This seems like it will be much more complicated to calculate.

 

One other thing to look at/think about is that willpower will be more important for classes that commonly attack from flank, like assassin or alchemist. Especially with talents like sneak attack, which will invalidate the need for cunning to get high crit. With flanking your crit damage and flanking damage add together to multiply your attack% gains from willpower, making it more effective per point. I can try to illustrate that as well when I have some time to theorycraft.

 

Cunning can only get you so far though. Once you reach 200 you have no where to go with it. Just make a damage function with cunning and willpower and cut cunning off at 100%. I am guessing the graph goes willpower --> cunning --> willpower once cunning reaches around 75%.

 

I am a math nubtard so I could be wrong.

[Drasca]

Cunning can only get you so far though. Once you reach 200 you have no where to go with it. Just make a damage function with cunning and willpower and cut cunning off at 100%. I am guessing the graph goes willpower --> cunning --> willpower once cunning reaches around 75%.

 

I am a math nubtard so I could be wrong.

 

It takes around 500 hrs of playing to get there though from nothing, so that is no trivial amount of time invested

 

Edit: In addition, the only people I know with anywhere close have easily over 1500 hrs invested, and usually over 2k to top the leaderboards with more balanced stats.

Modifié par Drasca, 26 juin 2015 - 03:38 .

[Denrok1]

It takes around 500 hrs of playing to get there though from nothing, so that is no trivial amount of time invested

 

Edit: In addition, the only people I know with anywhere close have easily over 1500 hrs invested, and usually over 2k to top the leaderboards with more balanced stats.

 

Good point.

[Medallian]

Wow I like these pretty graphs but as some one with god stats that are well to be honest off the charts... lol well off of yours lol I will say your data is wrong Willpower is nothing with out crit, crit is nothing with out willpower from what I have observer in my growth as a player and over my 2500+ hours I''m going to say your  wrong and there's some other hidden variables at work.

[Medallian]

It takes around 500 hrs of playing to get there though from nothing, so that is no trivial amount of time invested

 

Edit: In addition, the only people I know with anywhere close have easily over 1500 hrs invested, and usually over 2k to top the leaderboards with more balanced stats.

yeah wrong..... there are more hypothesis that I have made about cunning since I have over 200 cunning 

[Courtnehh]

yeah wrong..... there are more hypothesis that I have made about cunning since I have over 200 cunning 

[Gya]

[Texasmotiv]

Wow I like these pretty graphs but as some one with god stats that are well to be honest off the charts... lol well off of yours lol I will say your data is wrong Willpower is nothing with out crit, crit is nothing with out willpower from what I have observer in my growth as a player and over my 2500+ hours I''m going to say your wrong and there's some other hidden variables at work.

Essentially my point with this was to highlight the effect willpower has on your damage in a vacuum. Basically showing how your damage increases when the only stat you change is willpower. You are correct in that willpower without crit is next to useless beyond the magic defense it provides. Crit interacts with willpower quite nicely as it then introduces another multiplier (crit damage). I will disagree with you that there is no benefit to crit without willpower.

[Angelus_de_Mortiel]

Wow I like these pretty graphs but as some one with god stats that are well to be honest off the charts... lol well off of yours lol I will say your data is wrong Willpower is nothing with out crit, crit is nothing with out willpower from what I have observer in my growth as a player and over my 2500+ hours I''m going to say your  wrong and there's some other hidden variables at work.

yeah wrong..... there are more hypothesis that I have made about cunning since I have over 200 cunning 

 

Excuse the harshness here, but could you please provide some proof here? What hidden variables? I see you correcting Texas and making bold claims, but with no evidence to back it up. If you are going to tell someone they are wrong on a very detailed analysis, please be equally detailed in your rebuttal. It's just common courtesy. 

[Medallian]

Excuse the harshness here, but could you please provide some proof here? What hidden variables? I see you correcting Texas and making bold claims, but with no evidence to back it up. If you are going to tell someone they are wrong on a very detailed analysis, please be equally detailed in your rebuttal. It's just common courtesy. 

watch my twitch look at my damage you will see .... i rather not write some **** out because I am not the most elegant writer but i will discus it all day over some type of chat app maybe I sjhould make a youtube or something IDK

[Drasca]

 maybe I sjhould make a youtube or something IDK

 

Wouldn't hurt.

 

Also would not hurt to become a better writer either.... ok I lied, it generally does hurt--as does anything worth doing that is non-trivial. Totally worth the pain and effort to be a better writer though.

[sonofbarak]

Texas my boy but kinda agree with Medallian. No graph just experience talking.

[Angelus_de_Mortiel]

Wouldn't hurt.

 

Also would not hurt to become a better writer either.... ok I lied, it generally does hurt--as does anything worth doing that is non-trivial. Totally worth the pain and effort to be a better writer though.

 

Hahahaha that is not only the most applicable and funniest statement I've heard in a while, but bloody hell is it true.

 

 

 

Seriously, though, Medallian: You have earned quite a reputation for grinding hard hours for your rank on Xbox One, and being a generally cool person to play with from what I've heard. If you want to step into the ring of theory crafting, however, it's a whole different world. 

 

If you want to make YouTube videos about it, I would say that is quite a fair way to go. Need any help with that, I would happily help any way I can. Seriously. I know that I, and many others, would love to hear what you have to say in detail.

 

I was not intending rudeness. You may very well be correct. All I am saying is that it's a tad insulting to tell someone they are wrong, and then not tell them why, especially when they put at least a fair amount of work into their findings. Shilling for your Twitch channel doesn't quite cut it.

[Campout King]

Good job Texas... that calculation is pretty much spot on imho. Thank you for sharing! Time to griiiind!!!!

[yarpenthemad21]

watch my twitch look at my damage you will see .... i rather not write some **** out because I am not the most elegant writer but i will discus it all day over some type of chat app maybe I sjhould make a youtube or something IDK

It's theory crafting thread, in which ppl use old forbidden art of math. I know it can be hard, all those numbers...
It's not the place to advertise your own twitch channel.

If you want to prove something, pretty much as you said tell us that damage formula is all wrong, that we have some "hidden" variables all over the place, show us proofs of it.

[Texasmotiv]

Medallian,

I mean no disrespect but I kind of feel like you did not read my post or understand my methodology. Your numbers will be higher because in addition to willpower you have other stats like %100 crit chance, additional crit damage%, possible flanking damage and attack power. Your crit damage will multiply your attack power gains from willpower. If you give me all your stats and gear loadouts I would gladly crunch the numbers for you and see the same numbers you see in game.

The experiment here was to manipulate the numbers in an unrealistic way to isolate certain variables. Not to say "people with high willpower don't do good damage".

[sonofbarak]

I give you my number lol

[Texasmotiv]

I went ahead and crunched the numbers with the stats sonofbarak provided me in PM.

 

200 Con, 200 Will, 205 Cun

Level 4 Keeper, Decloaking Blast

20% crit damage ring

Dragon Staff (assuming 81 fire from the equipment document but I think there are better variations)

Superb Spirit Rune

 

With the +/-5% randomizer the base damage min/max are:

100.7 - 111.3

 

This would be calculated:

Low:

100.7 * 10 * 1.6 * 2.36 = 3802.432

High:

111.3 * 10 * 1.6 * 2.36 = 4396.562

 

Also there was one thing that was not mentioned in the original calculations. I am not 100% sure you CAN flank with Fade Cloak but I don't see why not. Those numbers are even more impressive due to the multiplication.

Low:

100.7 * 10 * 1.85 * 2.36 = 4202.688

High:

111.3 * 10 * 1.85 * 2.36 = 4859.358

 

Let me know if this is close to what you see in game.

[Drasca]

I went ahead and crunched the numbers with the stats sonofbarak provided me in PM.

 

Let me know if this is close to what you see in game.

 

Those calculations are wrong. You don't use rune damage at all for most abilities. The weapon base damage remains the same.

 

final_damage = (base_damage = 81 * rand(0.95 to 1.05) - armor * (1 - armor_penetration magic ignores armor))
* (ability_multiplier = {fadecloak = 8} )
* (1 + critical_damage_bonus = {.4 +.2} + flanking_bonus = 0 or .25)
* (1 + attack_bonus = {.41 (staff) + .95 (willpower)} + damage_multiplier {zero here, but can be increased from other players} + type_bonus = 0)
* (1 - magic_resistance = 0)

 

So FC damage should be around...

 

81 * 8 * 1.6 * 2.36 = 2447 * rand(0.95 to 1.05) for non-flank crit.

81 * 8 * 1.85 * 2.36 = 2829 * rand(0.95 to 1.05) for flank crit.

 

That is at level 4 solo though. The damage ramps up around allies with damage bonuses on different parts of the formula. As a necromancer with lots of passives and critical damage, I've gotten 6k crits on normal resist enemies or double that vs spirit vulnerable enemies.

[sonofbarak]

Ok Texas how am going to get an level 4???

[Drasca]

Ok Texas how am going to get an level 4???

 

Routine run! Or kill zone one of FC . . . or just don't spend more than 3 ability points during testing.

[Texasmotiv]

Thank you for the correction Drasca I was working under the impression that rune damage will modify base damage attribute.

[Drasca]

Thank you for the correction Drasca I was working under the impression that rune damage will modify base damage attribute.

 

The following is why I say most abilities are not affected by rune damage. I am uncertain if this is still the case for MP though since Luke's been playing with how SB works. It still doesn't modify the base damage attribute. Afaik, only Mighty Offense potion in MP will affect base damage outside of swapping weapons. SP has vitaar for Qunari, but there is none in MP.

 

Spirit Blade

• Hits all enemies in a 180-degree arc in front of the caster.
• Unlike with most other abilities, rune damage is applied when an enemy is hit by Spirit Blade. Rune damage from Spirit Blade attacks is given an ability_multiplier value of 300%.
• For more information on Spirit Blade's attack sequence, see the Autoattack section.

 

http://forum.bioware...mbat-mechanics/

[yarpenthemad21]

I've got some observation about archer, bows and armor which I want to share.

I wanted to compare with my stats on archer different bows, mostly cruel redemption and dragon bow, probably also punch of the maker.
We all know that longshot is "different" and what is also very important every damage here counts, because it is really hard now to get enough damage on longshot for 1 shot kills on perilous.

So I've done a math, taking my stats, build and gear. Build in mind was "high end" so for archer around level 17, I tend to ignore passive which aren't 100% sure when it comes to damage bonus.

Promotions are: 38/56/43
Dex from passives 15
Cunning from passives 21
Armor: 16 cunning, 12 willpower
dexterity amulet, longshot ring and crit damage superior ring

Sum up stats for dragon bow was (I'm using 0.5% values also just to have "clean" and automated spreadsheet, it shoudn't matter that much):
attack: 0.825
crit dam bonus: 0.85
crit chance: 0.42
base dam: 176

cruel (t3 crit dam upgrade):
attack: 0.415
crit chance: 0.44
crit dam bonus : 1.66
139 base dam.

longshot stats from tooltip, so 191% and 5.73 bonus damage (573%)
 
calculations for 0 armor >15 m: 
Dragon bow:
non crit
2539.6888 
crit
4698.42428
Cruel
non crit
1896.92605
crit
5045.823293
 
 
I though that something was wrong. Damage looks clearly, based on gameplay, just too high. I've had few 4k+ hits on demons but normally damage wasn't even close to this. 
This in general made me think:
1. How this > 15 m really works, maybe it is longer distance I though and it probably is true.
2. Let calculate armor factor also and maybe this randomization of damage
 
 
So now let take armor of enemy as 60, from what I know that "common one"
Minimal damage
Dragon bow: 1546.90136   ////  2861.767516
Cruel 983.2627475 ////// 2615.478908
 
Maximal
dragon: 1800,87024 ///// 3331,609944

cruel: 1172,955353 /////// 3120,061238
 

Average:

Dragon bow 1673,8858  /////  3096,68873
cruel 1078,10905 ////// 2867,770073
 
 
With that calculations damage looks way more like this in game (and shows how cruel sucks on armored targets)
 
This makes armor huge factor for archer, especially for longshot, which makes every even little change in base damage (and this is what armor really do) quite huge for end result. I've also calculated how much armor penetration is worth (we have 2 sources of it)
1. Sunder armor effect from passive on crits, probably won't help with longshot because it applies status after damage, not before it.
2. We have passive giving 25% armor penetration.
 
With 25% armor pen, and average "damage" we have something like this:

Dragon bow 1890,33655 /////// 3497,122618
cruel 1282,8133 //////////// 3412,283378
 
So this one passive can pretty much add 400 damage so around 14% more damage on crit. 
 
 
I would be pleased with comments on that, maybe I've calculated something wrong here.