54 replies to this topic

[Pandara_RA!]

Let's talk about damage. By damage I mean...

(P + M * 35) * (W/ 1000) * SC / (T + A - (V * 30) ) = Total Damage

P = Your power
w = Your weapon's skill coefficient
M = Might stacks
V = Vulnerability stacks
T = Toughness
A = Target's Armor
W = Weapon power

Thanks reddit for the fixed formula

(Bear in mind this isn't exactly perfect math, it's just something that seems super close to what is there in game after testing hits about a billion times and recording the findings. No one truely knows the real formula as of yet, though in the case of using vuln/might this one works pretty close.)

THIS! whew, now that we got that out of the way, don't worry if you aren't a LvL 40 Math Mage yet because I'll walk this thing out partly.

The weapon coefficant:
Every skill on your bar has a coefficant, you can very quickly get a rough idea what yours is by looking at your damage, then add 100 power from your gear and compare the two numbers, if you want a more in depth look you'll need to dig deeper than just that and test power levels of 100-200-300 etc to find those pesky .153 ratios like on the engi!

Example:
Before adding armor: 400
After: 500

400/500 =.8 meaning for every 100 power you add you'll get 80 damage added to that skill!

Why Might stacking in general tends to be a poor choice for sPvP:
In short the reason why might is bad is because:

• Very few skills have coefficants above a .3 ratio, with only a few exceptions hitting ~.5 meaning at 10 stacks of might (350 power) only might add 175 raw damage at most to only the strongest skills, for most ranged attackers the coefficants never break .3, or even .2!
• Might can be hard countered by boon removals
• In order to might stack for most classes you will need to give up many powerful traits/armor/sigils to get it, it's rarely something "free", besides the cases where you get might without any investment (built into a weapon set you were going to use anyways)
• Might stacking for many Professions require skill cooldowns or taking stills that offer no utility.
• Duration of might is on the short side, meaning stacking up to the higher numbers of 20-25 stacks isn't possible w/o min-maxing on runes.
• Extending might duration is hard to do through runes as the most possible is +60% from runes by using 2 runes from 3 different sets to grab all 3 of the +20% duration extentions
• Choosing Might on hit/might on weapon swap on sigils is a trap, compare them to the sigil of Air/fire (1,400/1200 damage crits respectivly) 3 stacks of might = 105 power or about 50 extra damage per hit for the next 6-8 hits or 300 damage and about 650 only if all 6 crit.

Why Vuln Stacking is even worse choice for sPvP:

• Vuln removes only 30 armor per vuln stack vs a lvl 80 target (everyone in sPvP) (-0.3125 * Level + 5 Defense per stack says the wiki) some say it's 30, though even at 30...
• 1 stack of vuln raises damage by only about 1.5% per stack, but only for the next attack and there after.
• There isn't really any way to extend vuln duration by too much, meaning Vuln will tend to never be in a high stack amount, but rather around 10-14 stacks though it may touch into the higher stacks 20-25 for a very short duration wit the aid of the "condition duration" runes.
• This is extremely important because at 25 stacks you increase damage to the target by only 33%
• at 15 stacks you increase your damage by about 15%
• 15% of 500 is peanuts at 75 damage
• 15% of 5,000 (3-8 autos, or a chain of power skills) is still peanuts. (750 damage additional)
• 15% of 15,000 is 2250. remember this number for later.
• One air sigil hits 1,500 damage ( about every 5 seconds, compare that to 2,250 damage over 6-8 seconds single target...with somehow dealing 15,000 damage during that time it's not even a close choice, it's ludicrous.
• Gets removed by condition removal. Which is everywhere and spamable.
• Doesn't increase condition damage at all. Meaning running vuln extentions through condition duration on runes doesn't synergize.
• Most Weapon sets that use Vuln have poor coefficants and are ranged.

Why do you keep saying for sPvP?
SPvP isn't a 20vs20 , it's a 3v3 if your lucky and a 4v4 rarely. Where this matters most is vulnerability, as vuln not only effects your damage but everyones elses damage as well, in sPvP you can't expect everyone to be able to focus on your target just for a bonus 15% extra damage, though in world vs world a random zerg will more than likely use it if you threw it on a WvW boss.

Finally why toughness isn't very good.

• Toughness doesn't have many amulets with stats you want, where zekers amulet or knight do
• Toughness isn't directly % based like how armor works in many games, it's closer to flat scaling.
• Vitality has a crazy good 10:1 flat scaling ratio. Making toughness only efficient with really high hp pools. Though currently you can't really get vitality high enough to take advantage of toughness without killing your healing or damage output.
• Toughness only blocks normal damage.
• Per 100 toughness will block about 34 damage for every 1,000 applied and only if it isn't from conditions.
• Per 100 Vitality will absorb 1,000 damage. Striaght up. conditions/damage Vitality don't care.
• Toughness does get better as you use heals though you can't always heal multiple times in fights
• In sPvP you have no clue who you are fighting, if it's a full spike/burst or condition team your toughness isn't very effective at all compared to living.
• Surviving a full burst by having a high pool of hp allows for you to heal at least once, meaning peopel with toughness want high vitality, toughness, and healing, though currently that combo is easy to obtain due to the way the amulets give stats.

[Max B]

I love math
Thank you

My only question is whether or not there is a direct point you are trying to make? Because it seems like you are saying the buffs and debuffs suck. Go with condition damage if you want to hurt people.

[Orz]

A good math read and gives really good food for thought. In the end, the numbers don't lie.

[Hova]

Hum this is interesting. So in your opinion what's the difference between having something like 1000 toughness and 23k hp, or 1500 toughness and 19k hp? This was something that was doing my head in in BWE2. I couldn't decide to build for more toughness and sacrifice a bit of hp, or the other way around.

[Bridger]

This really makes me wonder why everything is so complicated. I feel like simpler would be better for these combat systems. I feel like they are trying to make it harder to find the "optimal" build by making the math hard, instead of relying on opportunity costs which cannot be quantified (though this is done as well).

As far as I'm concerned all the math should be straightforward and visible in the tooltips. Look at LoL, for example. You don't have to be Archimedes to do the math on Attack Damage vs. Armor Penetration. The system is very simple and has a simple mathematical answer. If your enemy has more than X armor, go for more AD, if they have less, go for AP (for any given situation). It feels like GW2 throws a lot of smoke and mirrors to make it harder for players to figure out, as if that means they'll never figure it out. In the end I think that really just makes it harder for them to balance properly. Look at this post for instance. If the math had been simpler, it would be stupidly obvious how much more useful vitality is than toughness. I've wanted the damage formula since I heard about the game, but the one posted in this thread seems stupidly complex. Why?

P.S. I stole your Karma, sorry. Your post was too awesome to not share.

Hum this is interesting. So in your opinion what's the difference between having something like 1000 toughness and 23k hp, or 1500 toughness and 19k hp? This was something that was doing my head in in BWE2. I couldn't decide to build for more toughness and sacrifice a bit of hp, or the other way around.

Well one thing to consider (however the math works out) is that since healing is based upon a specific number and not a percentage, the higher HP and lower toughness would probably be good for quick fights where you don't get to use a lot of healing, but the high toughness and lower HP would be better for protracted fighting with a lot of HP use. This concludes (to me) that high HP is the way to go in sPvP but high toughness might be better for something like WvW where you have constant harassment and use of the heal skill.

[Aggelos]

The math seems solid and I love this kind of stuff! I remember reading for hours on Elitist Jerks reading all the math that Redcape and Exempler did for Ret paladins and I loved it!

Ok sorry for the rabbit trail up there but the thing I noticed is if I didnt gear for more toughness I felt really squishy but I will do some more testing next BWE to see how this math pans out.

Ill try my toughness build 30/20/20/0/0 with a Soilder amulet and beserker gem

then ill try a HP build 30/20/0/20/0 with a knight amulet and beserker gem

roughly the toughness build sits with 16.9k health 1685 toughness

and the hp build will look like this : 21k health 916 toughness

[Hova]

Well one thing to consider (however the math works out) is that since healing is based upon a specific number and not a percentage, the higher HP and lower toughness would probably be good for quick fights where you don't get to use a lot of healing, but the high toughness and lower HP would be better for protracted fighting with a lot of HP use. This concludes (to me) that high HP is the way to go in sPvP but high toughness might be better for something like WvW where you have constant harassment and use of the heal skill.

I see what you're saying. But you mention quick fights where you don't get to use a lot of healing. I had multiple fights (even 1v1) on Spvp where i used my healing skill 2 and sometimes 3 times. So is vit still > than Toughness in that scenario?
This is why i had such a hard time during the BWE trying to decide to go for more vit, or more toughness. For example, should i go with a condition+precision+toughness amulet? Or a condition+precision+vitality ammy? This was just doing my head in during the beta...

[SolidSteve]

One thing to take into account about the Toughness Vs. Vitality, is also that vitality reduces the strength of heals applied to your self.

If you have 10K health, and stack toughness, and a 5K heal, your heal is 50%. If you stack 15K health and no toughness, now your same heal is only 33%.

I still think vitality wins out, but its another effect that should be taken into account.

EDIT, another example to explain the effect.

Player 1 = 10K health + 1000 Vitality = 20K health. The player also has a heal 5K on a 10s timer, or 500health/s.

Player 1 is being attacked by another player doing 500dps, obviously the heal and the dps cancel out and he wins. Now another player comes doing 600dps. Player 1 is now taking effectively 100damage/s and will die in 200 seconds.

Player 2 = 10K health + 1000 Toughness. He also has the same 500health/s effective heal.

He starts being attacked by the player who does 600dps. After toughness he is doing 579.6dps, after the heal, he is doing effectively 79.6 dps and player 2 takes 125.628 seconds to die.

Now player 1s stat decision was obviously the smarter one, but its easy to see a scenario where toughness would win, such as if the attacker was doing 517.6dps. In that scenario he wouldn't kill the toughness player at all, but the vitality player would.

[Lockon]

more hp stat is the most popular choice in any game. I've played a couple of games with the ability to put attribute points for customization and in all cases, i've found the hp/vitality route stays constant in almost every games. Even in GW1 today, the survivor insignia that give more hp stat to your armor is more popular and the superior vigor rune is the most expensive rune in game.

[FuelTherage]

Does the damage bonus from might get affected by traits that manipulate damage?

[Anon06]

Have you taken into account traits such as empowered(warrior) or bountiful power(elementalist) before dismising might stacking? Empowered being most important as it doesn't specify the damage increase value, it's probably not enough to offset air/fire though. =/

[Deft]

Great topic Pandara, mega kudos! I think I'm going to try a high vitality bleed build now and see how that goes. I was planning on a toughness might stacking warrior, but that seems pretty useless now after reading this, lol.

Have you taken into account traits such as empowered(warrior) or bountiful power(elementalist) before dismising might stacking? Empowered being most important as it doesn't specify the damage increase value, it's probably not enough to offset air/fire though. =/

I tried testing Empowered during the stress test and it seemed fairly useless. Maybe they've changed it since then.

[Gogims]

Yeah really appreciate the info!! I was planning to get some traits that boost might in fire spell as elementalist. Now I know that there are better traits instead

[Pandara_RA!]

One thing to take into account about the Toughness Vs. Vitality, is also that vitality reduces the strength of heals applied to your self.

If you have 10K health, and stack toughness, and a 5K heal, your heal is 50%. If you stack 15K health and no toughness, now your same heal is only 33%.

I still think vitality wins out, but its another effect that should be taken into account.

EDIT, another example to explain the effect.

Player 1 = 10K health + 1000 Vitality = 20K health. The player also has a heal 5K on a 10s timer, or 500health/s.

Player 1 is being attacked by another player doing 500dps, obviously the heal and the dps cancel out and he wins. Now another player comes doing 600dps. Player 1 is now taking effectively 100damage/s and will die in 200 seconds.

Player 2 = 10K health + 1000 Toughness. He also has the same 500health/s effective heal.

He starts being attacked by the player who does 600dps. After toughness he is doing 579.6dps, after the heal, he is doing effectively 79.6 dps and player 2 takes 125.628 seconds to die.

Now player 1s stat decision was obviously the smarter one, but its easy to see a scenario where toughness would win, such as if the attacker was doing 517.6dps. In that scenario he wouldn't kill the toughness player at all, but the vitality player would.

Common misconception, but this isn't because your concept is wrong! it's becuase the options you have for toughness don't match your battle plan, for toughness to be effective you need as much vitality and healing as possible to increase it's value, but becuase toughness has such poor scaling you need that high vitality...but in order to get tughness and healing you need to forgo vitality on your amulet, both shamen and cleric provide 0 Vitality. which is just a big no no!

Have you taken into account traits such as empowered(warrior) or bountiful power(elementalist) before dismising might stacking? Empowered being most important as it doesn't specify the damage increase value, it's probably not enough to offset air/fire though. =/

Yes, if you get might/vuln from free scources it's icing on the cake, but if you trade a really good utility trait for it it's a bad trade, on warrior I rather be invulnerble for 5 seconds/never get knocked down in sPvP then do an extra 30 damage each hit, or get fury on immobilize

[Bridger]

I see what you're saying. But you mention quick fights where you don't get to use a lot of healing. I had multiple fights (even 1v1) on Spvp where i used my healing skill 2 and sometimes 3 times. So is vit still > than Toughness in that scenario?
This is why i had such a hard time during the BWE trying to decide to go for more vit, or more toughness. For example, should i go with a condition+precision+toughness amulet? Or a condition+precision+vitality ammy? This was just doing my head in during the beta...

This is true, though I imagine as players get better at focusing down specific enemies as a team fights will be shorter? Just a guess.

[kwlpp]

I see what you're saying. But you mention quick fights where you don't get to use a lot of healing. I had multiple fights (even 1v1) on Spvp where i used my healing skill 2 and sometimes 3 times. So is vit still > than Toughness in that scenario?
This is why i had such a hard time during the BWE trying to decide to go for more vit, or more toughness. For example, should i go with a condition+precision+toughness amulet? Or a condition+precision+vitality ammy? This was just doing my head in during the beta...

Vitality > Toughness still. If you're running into this issue in a 1v1, then you and your enemy aren't optimized for the current aggressive sPvP meta. I would know, because I ran a build that would do exactly what you're saying Hova. I would eventually win or just stall out, but the opportunity cost is terrible for it. You should look for ways to maximize your damage while keeping your vit up if you're in that scenario, which is what I'm doing now in preparation for BWE3.

[Pandara_RA!]

This is true, though I imagine as players get better at focusing down specific enemies as a team fights will be shorter? Just a guess.

You can't tell who you are fighting in sPvP and if you the guy on split/roaming you need to know you aren't goign to get 1 shot by some random. It's just not viable to risk it, this post is in no way saying toughness is worthless, just that

1. it's bad vs spike/burst damage
2. It's bad vs conditions as it blocks none of it (they are pure damage)
3. You can't risk not having 6k more HP
4. In the best case where you are fighting a target with consistant low damage the difference isn't really that strong.
5. you can't predict who you are fighting.

[Jacobin]

• Per 100 toughness will block about 34 damage for every 1,000 applied and only if it isn't from conditions.
• Per 100 Vitality will absorb 1,000 damage. Striaght up. conditions/damage Vitality don't care.

I would be interested if you could break that down further.

If I have 1800 toughness and I get hit for 3000 how much damage am I blocking?

And how does tougness affect the armor stat, and what does the armor stat do?

[Ivarr_Ironfist]

From my testing Power also had diminishing returns rather than a fixed co-efficient on abilities. On my Warrior at the end of BWE2, I tried calculating the co-efficient for the Rifle spam ability. For the first couple of hundred points of Power I noticed that the ability had a 1.1 co-efficient (If I remember correctly), however, when going up to over 2000 Power this didn't hold true. The co-efficient was slightly less than 1.1. The difference wasn't huge though (26 points I think), so I am thinking it could also be down to a round somewhere as well.

What I do know for definite though is that Malice and Expertise really need a re-think as well. At high levels of Malice (say 1500) you can get close to 2k DPS from a 25 stack of bleeds. However, you can only get a couple of hundred DPS out of Burning and Poison. My maths is in this thread. This makes Malice next to useless for some professions/builds that it really should be useful for, notably Elementalists (using Fire) and all Guardians. You don't even need to do maths to realize that Expertise is rubbish. Why would you want to elongate an already long duration Bleed effect when it is likely going to be removed? On the shorter duration effects the amount of Expertise you can pick up in the trait trees (which is probably all most builds will have) give next to no benefit. If you elongate a Cripple by 2 seconds or an Immobilize by 0.5 seconds, what significant benefit is that going to provide?

No interaction between Malice/Expertise and the non-bleed offensive stats (except Precision) is also a huge problem, basically gimping hybrid direct damage/condition damage builds. As I am sure most of you realize, as you increase one offensive stat the other offensive (with an interaction) become increasingly more beneficial to stack (as demonstrated by Vitality/Toughness interaction in OP). Therefore, you are losing some stat benefit if you choose to go hybrid direct damage/bleeds.

All in all, the OP + what I have outlined details a seriously broken attribute system, that I think really need iterating on (possibly multiple times) before launch. If this doesn't get fixed before launch I seriously feel like the game would be released in a half finished state.

[Hova]

Vitality > Toughness still. If you're running into this issue in a 1v1, then you and your enemy aren't optimized for the current aggressive sPvP meta. I would know, because I ran a build that would do exactly what you're saying Hova. I would eventually win or just stall out, but the opportunity cost is terrible for it. You should look for ways to maximize your damage while keeping your vit up if you're in that scenario, which is what I'm doing now in preparation for BWE3.

Yeah i know what you mean. I spent all afternoon looking at that new build creator (the one with the runes and amulets) to try and figure out my build for the next BWE. My goal is to increase my power while not losing that much on the condition dmg. I wanna try and see if my bleeds stay at around at least 100 dmg per tick while increasing my power so my direct dmg hits for more then it does right now. Maybe if i can achieve this i won't stall out like you said. Anyway, with just a few changes i managed to get my hp from 18k to 22k. Although my toughness went from 1.5k to 990.

edit: these are the stats with the the build i was testing in that build creator:


Power: 1345
Precision: 2139
Toughness: 916
Vitality: 1245
Attack: 2409
WeaponDamage: 1064
ConditionDamage: 1127
CritChance: 62.24
CritDamage: 30
Armor: 1836
Defense: 0
HP: 21662
Healing: 0


Now i just need to know exactly for how much my bleeds tick with that much condition dmg.