Talk:Damage calculation

just wannt check something

What happens if I type up an article from scratch, and 95% of the content, format, and structure just happens to coincide with the version on GuildWiki? Things doesn't seem to have changed much since I last took a stab at majorly restructuring the damage calculation sections on GuildWiki last time... -PanSola 21:13, 22 February 2007 (EST)

Speculations on the official wiki?

On GuildWiki, a great part of the Damage calculation article contains speculations about how damage is calculated in Guild Wars, with many math formulas which were derived from observation. These formulas seem correct as far as any player can tell, but are not necessarily exact from a game developer's point of view. Maybe things are just coded another way, but appear to give a rather similar result. Given that, what policy should we adopt about content which is subject to speculation on this official wiki? Is there any chance that "official formulas" will be released, to be included in this article? Or should we follow the same policy as in GuildWiki, indicating that some parts of the article are subject to speculation? I would not personnaly be totally satisfied with the second solution, as I feel that this wiki should contain mainly exact information. --Funky Jah 07:56, 6 April 2007 (EDT)

Very good point. I don't believe the powers that be monitor recent changes (like I do :)), so maybe it would be an idea to put a note to this effect on Izzy's or Gaile's talk pages. --Snograt 08:17, 6 April 2007 (EDT)

Indicating that some parts of the article are subject to speculation would be usefull to prevent other users from making the same mistake I did. I mistaked the article informations for "official" ones and linked to it then realized my error and had to revert back an edit. Serge Yseron 18:43, 17 April 2007 (EDT)

Hmm, Mike O'Brien uploaded this as "a base revision for this article." I can't see this being added to by anyone without a developer's POV. Is Mike saying this is gospel? :D --Snograt 18:52, 17 April 2007 (EDT)

Disagree. The source is a guru post by Ensign, and "It has since been enhanced with additional findings and clarifications from the Guild Wars community and development team." Thus what Mike uploaded is everybit as uncertain as what is on GuildWiki. -PanSola 19:15, 17 April 2007 (EDT)

(ec)Well yes - that's the way that I see it. But if it's that uncertain, why would someone of Mike's standing upload it? It's almost official recognition... --Snograt 19:28, 17 April 2007 (EDT)

(edit conflict) You are kidding aren't you? The fact that Mike O'Brian is coming through this page and editing it shows that it is probably the *most accurate* information anywhere on damage calculation. I find it funny that you suggest that one of the top arenanet people doesn't know his own game. --Aspectacle 19:30, 17 April 2007 (EDT)

Ensign's article is close enough to the actual implementation that it's far easier to start with his article and correct any inaccuracies than it is to write a new article from scratch. I added the text about "enhanced with additional findings and clarifications" in hopes of the community and ArenaNet working together to perfect this article. To get us started, I just provided some corrected information about hit locations.--Mike O'Brien 19:39, 17 April 2007 (EDT)

Thanks for the hit locations. That and the critical hit probability are the only two parts of damage calculation that I can't type up from scratch with my eyes closed (due to their statistical nature). (-: -PanSola 23:08, 17 April 2007 (EDT)

removing the "level 20 / AR60 / Attribute 12" base

If no one strongly disagrees, I'm gonna rewrite the information without using level 20, 60 armor, and 12 rank of attribute as basis. It greatly simplifies the mathematics while being completely equivalent.

60 armor cancels out 12 martial attribute and level 20 spell/wands/staff. There's no reason to throw all of them into the formula just to watch them eventually cancel out after you do all the extra work. -PanSola 19:22, 17 April 2007 (EDT)

I think you're right that this article would be easier to understand if everything weren't normalized around AL 60. --Mike O'Brien 20:39, 17 April 2007 (EDT)

Moving away from the "Armor level" terminology

This was a major movement I pushed for back on GuildWiki, and I would like to reuse all my arguments from there. Basically, the in-game terminology is simply "armor". Additionally, there is a general correspondence between the value of "armor" versus character level, which is profession-dependent.

Given several characters (Heros/Henchies/NPCs/Pets/Monsters) all at the same level, we expect them to have different armor depending on their profession.

The most concise and simple way to represent this idea, is to set up a one-to-one correspondence between the "level" of an armor and the expected character level of the Hero/Henchman/NPC/Pets/Monster; then use a different terminology for the stats that the game simply calls "armor".

The proposal I made for GuildWiki is to refer to the stat as "armor (rating)", or "armor rating". I'm fine with the use of other terms, as long as they don't contain the word "level" in it.

Correspondingly, the "strike level" used in this article should also be renamed to avoid the usage of the word "level". For example, if we decide to call the stat "armor" as "armor rating", then we can rename "strike level" as "strike rating".

This might come off as somewhat GuildWiki-centric, as I am pushing for a system currently used by GuildWiki. However, back in the days GuildWiki also had this exact problem, when the usage of the term "armor level" was way more established over there than it is here right now. -PanSola 23:28, 17 April 2007 (EDT)

Your argument seems to hinge on a supposed confusion of the word "level". Due to how armor becomes available to the player at specific stages of the game with precise values (unlike typical RPGs which feature piecemeal armor values), it's not hard to understand where the concept of a level for armor comes from. When used together, the term "armor level" is using armor as an adjective to describe what kind of level, like "character level" or "dungeon level", exactly like your proposition of "rating" coupled with "armor" or "strike". AL is also a common acronym to express one's armor rating. It's simply part of the game's community vernacular and isn't likely to change at this point. -- Hercanic 12:30, 8 January 2008 (UTC)

"heritage" of the research results

Does anyone know if Ensign's article was an extension of SonOfRah's research results, or if SonOfRah's research was based off an older article of Ensign's, or if they were completely independent of each other (or if they are actually the same individual)? My knowledge of damage calculations all came from the extensive discussions on GuildWiki talk pages, whose foundation was in turn based on SonOfRah's research results. If SonOfRah's research results were not based on Ensign's, then I will feel very uncomfortable contributing to this article if Ensign gets credit but SonOfRah doesn't.

I can actually type up a damage calculation article from scratch that is actually more detailed and informative than the current article here (with exception of hit location probabilities that Mike added), with my eyes closed (or with one hand tied behind my back plus monitor off, but not with eyes closed and one hand tied behind my back, because that will produce too much typos). The main omission in Ensign's article is how skills modifying damage factor in the equations, especiallly with respect to effect stacking and order of operations, which I all know by heart (assuming they haven't changed in the past 5 months).

If the "heritage" of my knowledge originated from Ensign's work, gone through additional research by SonOfRah, added by extensive findings of the GuildWiki community, and eventually trickled to me, then I'm fine with SonOfRah remaining uncredited. Alternately, if I am allowed to produce an article (from scratch) strictly more informative than Ensign's original document, and add in Mike's info on hit locations, then I'm fine if only a general credit towards the Guild Wars fan community is acknowledged (but not to any individual in particular).

I don't personally know who SonOfRah is, but if there is only one single thing about Guild Wars that I know very well, it's because of the prior work SonOfRah has done before me. Of course, what I know is also publicly known to many other guild wars fans, and I can't stop them from adding the information without crediting SonOfRah. Not to mention I already released all my contributions to GuildWiki under GDFL as well (and I also already typed up from scratch a damage calculation article in Traditional Chinese released under GDFL). Still, to me the heritage matters. -PanSola 23:51, 17 April 2007 (EDT)

I really don't know the heritage of Ensign's work, and I guess other people must not know either since there hasn't been a lot of discussion about this. Anyway, since SonOfRah clearly made a big contribution to the Guild Wars community, why not just credit his work too in the "References" section? --Mike O'Brien 14:04, 26 April 2007 (EDT)

Cool, will do. (-: -User:PanSola (talk to the File:Follower of Lyssa.png) 03:33, 27 April 2007 (EDT)

Question About % Bonus Damage

Would, say, +15%+20%+20%, equating to slightly more than 55%, also provide bonus damage for the armor-ignoring damage of skills? Say, I had +5% bonus damage on my weapon, and I used a skill that dealt +100 damage, and NO weapon damage. Would my final damage then be 150 damage, or would it remain at 100, as if % bonus damage only affected actual weapon damage? -Omigawa

It'd remain at 100. -User:PanSola (talk to the File:Follower of Lyssa.png) 17:11, 23 April 2007 (EDT)

Note about weapon requirements

It's been this way since day 1, but I don't know if it's a bug but it should be noted that unless you have a level 12 in the appropriate requirement, even a weapon with no requirement (candy cane weapons) will not do it's max damage vs AL60.

And if you have rank 12 in the appropriate attribute, the weapon with no attribute requirement will deal twice its damage vs AL20. It's noted in the Weapon Damage section of the article already. -User:PanSola (talk to the File:Follower of Lyssa.png) 00:08, 15 May 2007 (EDT)

Hit Locations

Hmm so all this time the 3/8 stuff for the chest was kind of right but not really...--Life Infusion «T» 20:13, 21 May 2007 (EDT)

Clarify weapon damage calculation

To assist me in understanding the damage calculation for weapon damage, could someone clarify the following for me --IAmAI 19:04, 15 June 2007 (UTC)

From the article, Effective Damage calculation for weapons is defined as:

[Effective Damage] = [Weapon Damage] × 2^((5 × [Attribute Levels Under Threshold] + 2 × [Attribute Levels Over Threshold] - 60) / 40)

From my interpretation of the above, I believe the variables Attribute Levels Under Threshold and Attribute Levels Over Threshold can be expressed, for the purpose of calculation, in terms of the maths functions min and max:

[Attribute Levels Under Threshold] = min([Attribute Threshold], [Attribute Level])

[Attribute Levels Over Threshold] = max(0, [Attribute Level] - [Attribute Threshold])

Expressed as a whole equation:

[Effective Damage] = [Weapon Damage] × 2^((5 × min([Attribute Threshold], [Attribute Level]) + 2 × max(0, [Attribute Level] - [Attribute Threshold]) - 60) / 40)

Physical Skill Damage

I was recently told that armour DOES NOT affect the + damage from weapon skills. I don't beleive this to be the case and I may very well be wrong. I'm trying to find somewhere on the wiki which clarifies this as I can't seem to come to the conclusion myself. I was pretty sure, almost certain, you can reduce physical skill damage to 0 with + armour as this is how I used to farm trolls with my warrior before the armour stacking nerf. I belive this is also how Kinetic Armour now works, + damage from skills does not ignore armour unless it otherwise states it does. This is why Illusionary Weaponry is a fun skill. Can someone clarify this for me as I am confused from the layout of this page and what I have been told. Dancing Gnome 06:48, 24 July 2007 (UTC)

+damage has always ignored armor. In increasing your armor, you reduced their base damage (which is what they're using most of the time) and thereby kept most of your damage to a minimum. --Kale Ironfist 07:13, 24 July 2007 (UTC)

Is there somewhere on the wiki where this is mentioned? I did some more testing with blessed insignia armour and kinetic armour skill and to my suprise the trolls outside droknars were able to hurt me without skills, but with skills they did +40 dmg to me through my armour. I was unaware (or forgot) +dmg skills ignored armour and I don't beleive the skill description reflects this. Perhaps we should mention this specifically somewhere in the wiki? Dancing Gnome 09:17, 24 July 2007 (UTC)

Sorry I keep forgetting to completely rewrite the article on this wiki. The GuildWiki damage calculation article should have a much more detailed explanation. -User:PanSola (talk to the File:Follower of Lyssa.png) 12:36, 24 July 2007 (UTC)

Random variable added?

I was looking at the various formulas on the stuff here, and i came across an integer that just seems to randomly appear in places. For example, using the formulas at the end of the article.

When talking about Level 15 Elementalist fireballs against the Charr.

69 × 2^((3 × 15 - 21 ) / 40) = 104.6 damage

When talking of the level 2 Warrior critting a Flash gargoyle.

10 × 1.20 × 2^((5 × 2 + 20 - 6 ) / 40) = 18.2 damage

I've bolded this seemingly random number in the formula above.

What is this number for? I've looked at the entirety of the article, and have seen nothing about this random number. It seems to only be there and in effect when talking about weapon attacks. Are weapon attacks under the effect of this random integer? If so, is it changeable, mutable, or otherwise affectable in such a way to be in your advantage? MysticSword 07:45, 28 January 2008 (UTC)MysticSword

Errr, nevermind. Forgot about Customized weapons. MysticSword 07:49, 28 January 2008 (UTC)MysticSword

Discrepencies

It is important to note that this equation, though close, is not completely accurate. I've tested every weapon's critical damage in game, both with +35% damage and +0% damage, and many of the results don't match what the equation would give. For example, a vanilla max damage axe criticals for 40 at rank 12, yet the equation comes out to 39. As a more extreme example, +35% damage scythes in-game crit for 92 at rank 16, and the equations only come out to 89. I've considered how the game may or may not round, and some of the numbers still don't match. I've tried getting my calculator as close as possible by multiplying the character's attribute rank by a constant, and though it gets closer results, there are still problems. ~Shard (talk / Nerf List) 00:43, 4 March 2008 (UTC)

How does the game round damage?

If I deal 49.5 damage, will the target take 49 or 50 damage? Anyone knows how the game rounds damage? --SkyHiRider 21:19, 23 August 2008 (UTC)

I would assume up because of the rules of rounding. — ク Eloc 貢 19:51, 25 August 2008 (UTC)

Computers always truncate decimals to get integers. It wastes less time processing. ~Shard 20:30, 30 August 2008 (UTC)

Unless the programmers used something other than a basic cast. It takes the same amount of procession power to always round up as it does to trunicate, and while true rounding might take an additional step, theres no reason why the programmers couldn't use it. I've seen games that round rather than trunicating for damage. 71.84.240.55 02:47, 28 October 2008 (UTC)

It's inefficient. It takes almost twice the time to round to the closest integer than it does to drop decimals. It's certainly possible, but unlikely. Maybe we can ask someone who wrote it. ~Shard 06:53, 23 January 2009 (UTC)

Now shard whats a big noob like you using a big word like truncate? Oops, NPA, i meant noob in a good way. Like armond, he's a noob but thats good... right? I have a feeling this comment is going to be deleted --adrin 07:49, 23 January 2009 (UTC)

Truncating and rounding both take a good nanosecond to perform on a high-quality server. I'm fairly certain that, considering everything else that is being performed on Guild Wars servers, the speed of rounding a floating point value is not at the top of anyone's list. As for the answer, I can only make an educated guess; I think it would round instead truncating. --24.113.82.206 03:35, 3 April 2009 (UTC)

But when your server is catering to nearly a million players at a time, each of them needing hundreds of calculations per millisecond, that processing time adds up. I've seen frame rates drop from simple things like rounding. ~Shard 03:40, 3 April 2009 (UTC)

Actual ingame damage

Something I don't understand. If these results are true, and the damage calculater link is true, with an axe hitting a caster(in pvp), even using a skill such as evicerate or excutioner strike, it should be impossible to hit beyond 100 damage. Yet, I often hit well above 100 damage while using attack skills, even going near (and sometimes above) 100 without using any skills. Therefore, ether this page is wrong or theres more to the damage calculations then what is writtin here. --The preceding unsigned comment was added by User:125.237.14.144 (talk).

tl;dr: hitting moving targets, bonus damage, and armor penetration from strength quickly add up.

It's difficult to account for what you're describing if you don't provide data or screenshots we can check against; there's a variety of reasons that could happen. I know for a fact autoattacking a moving caster with a non-vamp customized 15^50 max axe at 14 axe mastery hits for 59 damage, and the calculator confirms that (it gives 58.6 for a critical hit). Try the same example, but using Bull's Strike at 13 Strength with a vampiric axe instead of autoattacking. Now you have not only bonus damage from Bull's, but armor penetration from Strength (since you're using an attack skill) and 5 unconditional damage from vamp, and your damage should be 72(calculated)+27(bonus)= 99 damage. The same situation with a hammer gives 89+27 = 116 damage. Eviscerate and Executioner's should do 42+29=71 and 42+38=80 damage to a stationary target, on average, and 72+29=101 and 72+38=110 damage to a moving target.

There's also the possibility that the people you're fighting don't have max armor, though. -- Mafaraxas 21:17, 6 December 2008 (UTC)

Weapon bonus damage gets added on top of all this, so a rank 15 power attack from an axe critical will do 69 (axe crit vs 60 armor) + 40 (power attack damage ignores armor) = 109. Sundering also works into this. The highest nonskill damage you can normally get is 92, a rank 16 scythe critical. Oops, that's nonsundering. It would probably be closer to 97 or 98 then. ~Shard 06:50, 23 January 2009 (UTC)

Percentile damage reduction and damage multipliers

How do percentile damage reduction and damage multipliers fit into the order of operation? I've brought it up at Guild Wars Wiki:Requests for research. -- Gordon Ecker 00:45, 25 August 2008 (UTC)

A formula for calculating damage increase or decrease

If I understand the normalized math correctly in this page you can easily calculate (with a calculator) how much of an increase or decrease in damage is done when the attackers strike level or the target's armor level change. If you know the level change in strike level - armor level then:

new damage = original damage * 2^( level change /40)

or

The change in damage (as a multiplier) = 2^( level change /40)

For example:

If you add 20 armor points to your defense you take 2^( -20/40) or 70.7% of what you would've taken before. So an attack that once did 100 points of damage would now do 71.

Similarly if you have a spell that reduces your opponents armor by 50 the multiplier is 2^( 50/40) or 238% of the damage previously done with skills or weapons.

If you increase your sword attribute from 8 to 10 the level change in strike level is 10 (5 * 2 skill levels), so the multiplier would be 2^( 10/40) or 118% more damage with a sword swing than you would have.

If you gain 1 level, any level, all your skills do 5% more damage. The level change is 3 (3 * 1 level) so the damage increase is 2^(3/40) or 105% of the old damage when attacking the same target with the same skill.

Maybe this should get a by-line in the page.

76.216.23.160 06:58, 18 December 2008 (UTC)Mercus Maximus

Hit Locations?

What about spells (especially AoE spells)? Do those count as ranged attacks when determining armor location hit? Do spells like Lightning Strike and Earthquake have greater chances of hitting the head or feet?151.201.236.169 15:55, 12 February 2009 (UTC)

Weapon Requirements

I'm finding that some bonus/collector staves are behaving differently. How does the requirement affect the energy these staves provide? The staves I'm referring to are the Soul Shrieker and any of the BMP collector staves. I receive the full energy bonus from them. Alt-tabbed back just then, equipped two staves I didn't meet the requirement for - both provided full energy. One was in-class (Necro), one was off-class (Ele). Adding a note now. RandomJF 10:07, 1 July 2009 (UTC)

Actually, forget it. Just checked the staff page. Also, forgot to sign. RandomJF 10:07, 1 July 2009 (UTC)

IIRC the damage is based on attribute, not the energy. Same with shields that add hp. The "Req att XX" only affects the stat it is on. StatMan 14:42, 1 July 2009 (UTC)

First chart

Someone needs to clarify that first chart. What the heck does "to RANK 0 / to RANK 12" mean? What, in fact, does this chart apply to and why is it there? --The preceding unsigned comment was added by 70.50.100.207 (talk).

Rank and value are used synonymously. The chart illustrates the relative effect of the various attribute values (ranks) on weapon damage: the first row is relative to a rank of zero; the second relative to a rank of twelve. --DryHumour 15:25, 22 September 2009 (UTC)

Rank is how many attribute levels you have in your weapon skill. So if you have 16 Axe Mastery and are using an axe, you will do 115% of the damage you would do if you only had 12 Axe Mastery. Is that clearer? 141.14.133.231 17:07, 22 September 2009 (UTC)

Where is the information about Level disparity?

There used to be info on level disparity between creatures and the damage mitigation due to being a higher level.~>Sins WDB 00:35, 1 October 2009 (UTC)

Late reply, but I was looking for it too, and eventually found it at Level. Manifold 20:14, 3 February 2010 (UTC)

Based on the artical its not technically a dmg addition/reduction based on lvl difference. Why it appears this way is because your spells gain an "armor shift" of 3 per lvl. Foes generaly gain 3ar per lvl. In pve a spell against a caster profession tends to deal the listed dmg provided they are the same lvl as you because of this. Characters never get above lvl 20 so there ar shift ends at 60. Monsters keep gaining 3ar per lvl AND their spells keep gaining that same 3ar shift per lvl. To make this simple, a naked lvl 1 player will take the same dmg as a naked lvl 20 player. This is easy to see using a 20dmg flare. Lvl 20 60ar barrels take 20 ofcoarse. The closest barrel in the istan training ground is a lvl 5 with 55ar. 20 flare does 22dmg. So as can be seen, its an "armor shift" applied per each character lvl. Lvl 30 monsters get 3x30 "ar shift" which is a bonus 30ar to their spell dmg. Easy to see why reg lvl 30 monsters using air spells can hit double dmg and bosses can hit x4. If it were the case of bonus armor/dmg based on lvl difference things like cracked armor/armor penetration would be laughably ineffective. Justice 20:54, 24 September 2010 (UTC)

Damage calculator

This web link shown at the bottom of the page contains a virus and should be removed!!!! --The preceding unsigned comment was added by 92.17.58.126 (talk).

I will check it out right away. Hangon...it's an online calculator, how can it contain a virus? Oh well. I checked it out and it seemed to be fine anyway. -- Lacky 20:05, 10 January 2010 (UTC)

The web page hosting it could have been compromised and could contain a malicious JavaScript or it might have an XSS problem, etc.. --DryHumour 20:35, 10 January 2010 (UTC)

Yeah I tried it out and the loading bar will not load, and Norton pops up saying it has Malicious content :\76.185.20.44 19:37, 17 January 2010 (UTC)

Armor Penetration Table

The table that's currently displayed on the page displays values for 0, 5, 10, 25, and 50 percent armor penetration which is not particularly helpful as it includes a value no one will ever be using (5%) and ignores a much more common value (20%). Just trying to make sure I'm not missing something huge before I go ahead and change it. Norwegian Thunder 03:35, 15 April 2010 (UTC)

As I read the article, I noticed the same. The typical 20% of weapon mods are missing. 5%AP are only possible for a warrior with 5 strength. 10% come from weapon mods (minimal sundering), ranger and ritualist skills. 20% from weapon mods (maximal sundering), dwg and warrior skills. 25% from air magic and spear of lightning. 30% and 40% percent are additive for weapon mods and skills, so the may be useful.

i think also, that the formula for the AR is wrong. it should be AL * (100 - AP[0<=x%<=100%])/100 or AL * (1 - AP[0<=x<=1]). greets. --91.42.177.180 12:57, 24 February 2011 (UTC)

I agree on both points. The table should include at least 10%, 20%, and 25% as those are the values most easily generated through single effects; and the equation does seem to be incorrect. --Irgendwer 15:44, 24 February 2011 (UTC)

Critical Chance

I'd like some more eyes on whether 12 weapon mastery against a lvl 20 target results in 12% crit chance rate, or 20% or what. --JonTheMon 14:38, 12 May 2010 (UTC)

According to the formula, I end up with 12.55% - WeaponCritChance. I have yet to figure out how this last term is calculated though. Everything I have seen just generalizes it to "it depends on your weapon attribute and the level of the attacker and defender". That is NOT a proper mathematical description. 155.97.193.46 00:33, 7 December 2010 (UTC)

The formula implies that having a higher [weaponcritchance] results in a lower chance to actually get a critical hit, so shouldnt it be a plus instead of a minus, or should the term be called a penalty?217.122.170.106 21:37, 20 February 2011 (UTC)

I'd like to ressurect that issue. I did other research the last days and now came across critical chances (at the german wiki though, since I mostly contribute there). I wanted to recheck them and started by comparing them to the values listed here to get some impression, but they totally do not fit. Especially Izzys statement about 1% per mastery rank seems pretty wrong to me. The german wiki states 15% and 17% atm (one tested, one calculated) for mastery 12, I don't trust in both. So I started myself and now have 1728 hits at rank 12 (there is a reason for that number, but don't ask) with a percentage of ≈17.809. I'll probably have more values in a few hours, atm I did rank 12, 0, 4 (again: don't ask now). Rank 0: 0.9995%, rank 4: 5.220%. – 16:04, 6 September 2011 (UTC)

As a quick update: I am still collecting data by afk autoattacking the master of damage with a peppermint spear (so I get constant normal and crit values). Since every cycle takes 3 minutes and I'm aiming for 10 for each rank, this takes time, but I'm on it. Results when it's done :) – 17:18, 7 September 2011 (UTC)

Ok, almost 36,000 hits are gathered, so I can "confirm" this correction suggestion with data, at least for a lvl 20 vs. 20 scenario. Each value in the "rank 0–12" lines is a number of crits out of 141 hits within 180 seconds, ranks 13–16 are out of 120 hits within 180 seconds. The amount of those testing packages differ (11, 15, 20), because I wanted to increase accuracy at some ranks. Picture, table file – 15:50, 9 September 2011 (UTC)

And just for the record: Rank 11 was a little troublesome, since I got a few quite high values. Thus I finally took 40 instead of 20 and "normalized" them down to 20 again by cutting off the lowest/highest ones. Thats why rank 11 shows so narrow results. – 18:32, 10 September 2011 (UTC)

Consolidation

A few of the examples in the Consolidation section are wrong. Penetrating Attack doesn't have 50% armor penetration, a customized and +15% modified weapon doesn't do 135% base damage, Strength's armor penetration doesn't stack with Penetrating Blow's armor penetration, and armor doesn't affect bonus damage from attack skills. If there are no objections I'm going to fix these. --Irgendwer 18:42, 28 June 2010 (UTC)

Could you please give here (or in the article) more detals about delusions you mentioned, esp. about modifiers, like 100% base weapon damage + 15% from inscription + 20% from customization != 135%. I think that lot of people suppose exactly 135% because it's near to intuitive. If the reality is different, need to explain the common mistake. --85.141.163.13 19:03, 29 June 2010 (UTC)

That's because the percentages don't add up like that. 15% of 100 is 15. 20% of 115 is 13. 115+13 = 138. 20% of 100 is 20. 15% of 120 is 18. 120 + 18 = 138. Percentages add up multiplicatively. Same apply for 40/40 sets, you don't get exactly 40% probability of effect, but actually 32% for single effect and 4% for double instead. The math for that is pretty different because it's a percent of probability and not an percent of damage. Before accusing someone of being delusional, please review your understanding of math. --Lania 19:22, 29 June 2010 (UTC)

More detail: damage modifiers stack multiplicatively, which makes mathematical sense because they are themselves applied multiplicatively to the base damage and multiplication is commutative (i.e. applying multipliers to each other and then to a base value has the same effect as applying them to the base value one after the other). Thus, a +20% modifier and a +15% modifier result in 1.2*1.15=1.38 times the base damage. Lania, I'm pretty sure 85.141.163.13 was calling the article delusional, not me. --Irgendwer 19:32, 29 June 2010 (UTC)

Ah yeah, you're right. After re-reading, he does say delusions you mentioned, which points to something in the article and not you. --Lania 19:38, 29 June 2010 (UTC)

Some mistakes

Hello there, I checked this page a few hours ago. I recognized some mistakes on the page:

• The damage of "Conjure Flame" is not affected by armor.
• The armor penetration from the stringth attrivute and AP from a sundering upgrade is not multiplied but added. --The preceding unsigned comment was added by 80.171.15.76 (talk).

Good catch. I guess I'll go fix the math. Again. --Irgendwer 20:23, 2 July 2010 (UTC)

Starter weapon damage

I'm thinking to myself, "I wonder if I should use a longbow I don't meet the requirements on or just use the (non-upgradeable) Candy Cane Bow for those situations where I want to be out of range." I look at the Requirement page:

"Weapon damage is reduced to that of a starter weapon with a slight bonus to damage depending on the related attribute, as described in damage calculation"

Hrmn. Nothing about actual bow damage reduction numbers. I look here as suggested:

"if you do not meet the requirements on a given weapon it will deal damage like a starter weapon of the same type"

Again, nothing about the numbers for starter bow damage. There is no 'Starter Weapon' page, so I look at the Starter Bow page:

"Newly-created Rangers begin with one equipped"

Well, that's very helpful [/sarcasm], but there are no numbers at all. The Ranger page is no help, so I look on the Bow page:

nothing

ARGH! This is getting frustrating. Can someone please list somewhere the actual numbers for starter weapons? (Or point me to where they're listed?) -- Tometheus (talk) 23:14, 6 November 2010 (UTC)

GuildWiki has them. I don't know why they aren't listed here. At any rate, I'm quite sure the information you've cited is wrong (as I've noted here) and I would have replaced it except no one seems willing to comment on whether it is or not. In the absence of findings other than my own, I'm going to predict that the Candy Cane Bow will do more damage. I should also point out that there is no tactical reason to use a Candy Cane weapon, ever. --Irgendwer 23:48, 6 November 2010 (UTC)

Critical damage, skill level and threshold?

The formula for critical damage doesn't take threshold into account, is that correct? -Naryoril 9:22, 28 December 2010 (UTC)

There is no threshold at least for attribute 12 and lower criticals(12 gives a multiplier of sqrt(2)), however I do wonder what the multiplier becomes if say the attribute was 13. It should be 2^[((5x17)-60)/40] without threshold but maybe the critical multiplier threshold is just 4 levels higher, meaning the gain for the 17th point is 40% of normal. This is not mentioned. 217.122.170.106 14:31, 20 February 2011 (UTC)

help me understand?

so do any of the formulas take into account the requirement on the weapon? i dont know if im reading the wrong but i dont see it. can some one tell me the damage difference on Req 9 15-22 sword 15^50 20/20, and a Req 8 15-21 sword 15^50 20/20. both un-customized. AurenXneruA 17:06, 25 January 2011 (UTC)

Weapon requirements do not influence damage unless you do not meet those requirements. Koda Kumi 17:09, 25 January 2011 (UTC)

....then why is it in all the calculations -_-AurenXneruA 00:17, 28 January 2011 (UTC)

None of the calculations take requirements into account. What do you mean? --Irgendwer 00:24, 28 January 2011 (UTC)

In your example, if you have sword mastery rank 9, the first weapon is going to be slightly better. If you only have rank 8, the second weapon will be noticeably better (depending somewhat on how it was acquired).

The reason you can't see "requirement" in the calculations is that it's irrelevant to everything else; the calcs start with base damage. And any min-max strategy assumes that you have already maximized the base damage by ensuring that your toon meets min Q for your weapon.

Be careful about spending too much time investigating marginal differences. As with all gaming, there's only so much that theory can do to help you plan. Your best bet is to take the actual weapons out to Isle of the Nameless and attack the dummies.  — Tennessee Ernie Ford (TEF) 01:02, 28 January 2011 (UTC)

doubling of armour ignoring damage

Hi

My understanding is that armour ignoring damage is fixed, as per the page, but i've seen both Ancestors' Rageand Death Blossom do double damage. Have i just have overlooked some core game mechanic? 211.27.254.149 02:16, 12 May 2011 (UTC)

The target was probably using Frenzy, or Primal Rage. Manifold 02:18, 12 May 2011 (UTC)

ah yes i realized this just after posting, it was against foes using frenzy. nevermind 211.27.254.149 02:22, 12 May 2011 (UTC)

what is throwing this off?

i was bored and decided to flail (metaphoric, not the skill) on the master of damage with +15%+20% pvp daggers @ 15 dagger mastery. using spreadsheets and calculations in this article damage should be 10-26 and crits of 43 but that is not what i get in game. i get 11-26 with 37 crits. most simple answer is the rounding goes up always instead of what i read in this discussion above about "converting to integers [from decimals]", but that doesnt' answer the low crit damage. i read on the big dervish update a few months back that crit damage was reduced for scythes... was it reduced for everything then? is that a bug due to what was mentioned from dev notes? or what factors are missing in this extremely vanilla test run? 98.134.220.75 01:46, 10 June 2011 (UTC)

Your calculations probably didn't account for the attribute threshold, which applies to both regular and critical hits. Where the critical hit calculation says 5*AttributeLevel, that's shorthand for 5*AttributeLevelsUnderThreshold+2*AttributeLevelsOverThreshold, i.e. 60+2*(AttributeLevel-12) for a level 20 character with at least 12 in the related attribute. --Irgendwer 02:25, 10 June 2011 (UTC)

oh i read it verbatim instead of applying crit formula the same as normal. should someone state that somewhere? 98.134.220.75 02:41, 10 June 2011 (UTC)

I think it's implied by the earlier subsection on basic weapon damage, but I moved a bit of information around to hopefully make the calculation easier. --Irgendwer 03:04, 10 June 2011 (UTC)

Percent to rank 0 is misleading

Either this needs to be removed from the first chart or it needs to be stated very clearly that the rank 12 figures are the ones that reference listed weapon damage. Using Rank 0 as 100% is a completely arbitrary frame of reference that has no relation to in-game stats. Similarly, with respect to threshold, it is not so much that bonuses diminish after the threshold, but rather that you receive small bonuses instead of large penalties. Hold Me Closer, Necromancer 15:29, 27 August 2011 (UTC)

yeah the article is far from perfect/right/accurate.188.67.153.216 00:17, 4 January 2012 (UTC)

Very informative article but the maths on your formaulae is just plain wrong! Armour effect calc. example 1: 30 x2((60-(80+20+16))/40)). 80+20+16=116. 60-116= -56. -56 divided by 40 = -1.4. -1.4 x2 = -2.8. 30 x -2.8 = -84!. The formula itself makes no sense, but also how on Earth do you get 11.4? Dividing into 30 instead of multiplying makes more sense but the result is still different: 10.7. I would be grateful for an explanation

94.7.125.225 16:00, 6 September 2012 (UTC)

Very low armor ratings

I have a level 16 warrior with 14 hammer mastery. One hit with Rhino's Charge deals 393 damage on Agnar the Foot, and 313 on Hulking Stone Elemental. The elementals are supposed to have 3AR against blunt damage (according to wiki) and take 313 damage in comparison. Most attack are criticals. Armor penetration and Strength 4 have no effect (AR too low ?). If I calculate things correctly, i should not be above 180ish damage. I get this type of value on a lot of regular foes in pre, including stone elementals, which are supposed to have 18 against blunt damage. Did I miss something or could they have a very low or even negative armor rating? --Ruine Eternelle 17:42, 9 November 2019 (UTC)

Full Damage Equation Makes ZERO Sense

The above formula referenced in the article makes no sense at all. If you are level 20 and have 12 in a weapon skill, then you meet the threshold and therefore wouldn't have any attribute levels under or over the threshold. The value would be zero. If you have 15 in a weapon skill, then your attribute level isn't under the threshold at all! So what do you put for the 'AttributeLevelsUnderThreshold' variable in the equation? There is almost no context or instructions for using the equation, so it ends up being useless. Any help is greatly appreciated. God Of Fissures 22:54, 4 April 2020 (UTC)

If you're level 20 with a weapon level of 15, then you have capped out the 'AttributeLevelsUnderThreshold' at 12, and have gained 3 'AttributeLevelsOverThreshold' as additional damage.

So [Weapon Damage] x 2^[( 5x12 + 2x3 - 60)/40] Which reduces to [Weapon Damage] x 2^[(2x3)/40]

Hope that helps. Greener (talk) 23:29, 4 April 2020 (UTC)

Thanks for the quick reply. That clears some things up, but that isn't made clear in the article unfortunately. Also, how does this work when your character level is less than 20? Do you substitute the appropriate attribute threshold for that level in the equation?

For example for a level 19 with 15 weapon level and an attribute threshold of 11.5 (per the equation in the article) [Weapon Damage] x 2^[( 5x11.5 + 2x3.5 - 60)/40] Is that correct? God Of Fissures 00:34, 5 April 2020 (UTC)

With the understanding that there will be differences between our model and Anet's actual algorithm (e.g. the handling of decimals), that would be correct. Greener (talk) 02:15, 5 April 2020 (UTC)

Hit locations

Why was this information removed from this article? It's relevant and can influence optimal locations for certain insignia combinations. 76.127.242.82 06:47, 7 May 2020 (UTC)

No idea. There's been a freakish number of edits since March 2017. Compare with a a previous revision. Whether the content is correct or not however seems kinda questionable- high/low attacks? -Chieftain Alex 07:03, 7 May 2020 (UTC)

Unless something has changed in the last 5-7 years, the hit locations were in fact accurate. In all fairness the information isn't very important in most cases, but that doesn't mean it shouldn't be here. 76.127.242.82 08:17, 7 May 2020 (UTC)

Hits location is part of the armor calculation article, iirc.--Ruine Eternelle 08:38, 7 May 2020 (UTC)

Except it's not. I'M for re-adding the info. Steve1 (talk) 14:17, 7 May 2020 (UTC)

Here I have recovered it. I put it on this page, but your call to put it on the damage or armor page. - - Ruine Eternelle 19:10, 7 May 2020 (UTC)

It probably bears mentioning somewhere on the armor page as well. I don't know where the chart itself should be, but whichever page doesn't have it should have a brief summary of it and a link to the other page.

Again, it's not critically important information, but a lot of people are completely unaware of the fact that your chest piece is more likely to be hit, and that information makes the variable bonuses on radiant and survivor insignias make much more sense. (Lots of people assume either that armor is cumulative or that the hit rates are identical so they assume that survivor on chest is providing a bigger effective hp benefit than it is). Almost nobody is aware of the influence of mob size on hit chance. 76.127.242.82 20:58, 7 May 2020 (UTC)