Talk:Armor penetration

Your math is bad, Ollj. No matter what someone's AL is, if they increase it by 40 (assuming the damage dealer is level 20 for spells and has 12 in the attribute for weapons), the damage received will be halved. --Fyren 04:51, 13 Aug 2005 (EST)

my math is from [here], whats your source, general rumors? death penalty is NOT simply subtracted fron the armor (though its a nice simplification that works fine within lower AL) !


 * Fine, let's use that site. With base damage of 100 and armor of 67, to use one of your examples, damage dealt is 88.58.  Your text said the "half life" for 67 armor would be 44 armor.  67 + 44 = 111.  Base damage of 100 and armor of 111 results in 41.32.  88.58 / 2 != 41.32.  With 100 base damage and 107 armor, resulting damage is 44.29.  This is half of 88.58.  Adding 40 armor halved the damage.  Your own source refutes your numbers.


 * The formula empirically derived by SonOfRah for weapons is actual_damage = damage * 2 ^ ((attribute * 5 - AL * (1 - penetration)) / 40) for an attribute of 0-12. For spells, actual_damage = damage * 2 ^ ((caster_level * 3 - AL * (1 - penetration))/ 40).  "Damage" is either the spell's listed damage or the random roll from the weapon's range.  There are other modifiers, of course, but this is the basic gist of it.  --Fyren 05:23, 13 Aug 2005 (EST)


 * Also, death penalty has nothing to do with armor. --Fyren 05:24, 13 Aug 2005 (EST)

hey fyren. you should read my text more closely and stop confirming it if you want to prove it wrong! your examples are all limited to 0% armor penetration, mine are not! 100 damage on 60/100/140 armor with 0% armor penetration, shows a half life evers 40 AL. 100 Damage on 75/125/175 armor with 20% armor penetration shows a half life every 50 AL. btw we are all using the same formular, I just understand it! in (y)our formular 1-"penetration" gets multiplied with "AL"/40. casterlevel doesnt have to be variable, however the half live depends on "penetration"

This page, as written is FUBAR
Ollg, it is apparent that English is not your first language. Please, please, refrain from wholesale edits requiring the competent and extensive use of it. You have almost completely trasmogrified this article into a nigh incomprehensible explanation of armor level, and invoked terms from chemistry to obfuscate an already complicated value.

IMO, my rewrite left this article clearly defined this term as RELATIVE to the AL of the situation in question, something your rewrite completely ignores. Armor penetration DOES effectively modify the AL of the attackee, although other factors go into the actual calculation of damage, I noted this.

Take a step back and see how astray this article has become. Define the TERM. --Jackel 07:39, 13 Aug 2005 (EST)


 * For the record, I am an engineer, I studied math and physics in college, and I am an experienced player who's been around since almost when the game was first started and I want to state for the record that I have no idea what this article is talking about. None. What is "Armor half life"? How did Gordon Freeman get in this website? :)
 * More importantly, if your math is from that site, then how about you refer readers to that site, instead of reporting it as if you actually came up with it? Isn't that the honest thing to do? --Karlos 09:03, 13 Aug 2005 (EST)
 * this is not about armor level. its about armor penetration. of course thats linked to armor.
 * half life is more than a game by valve, it has a physical (and matematical) meaning: http://en.wikipedia.org/wiki/Half_life . you must have missed that in college.
 * I think you're the one missing my point. The article is NOT readable. I promise you this: I will rewrite it, and I will rewrite it into English rather than Gibberish and I will move the Gibberish to a spearate math article and free the users from your oppressive reign of mathematical terror over their articles! :) You have to understand that the average user clicking on Armor Penetration wants to know what it means, that's it. --Karlos 09:22, 15 Aug 2005 (EST)
 * "Armor Penetration gives a damage boost." For someone determined to include incomprehensible formulae, this is a gross oversimplicifation.


 * "The higher a playrs AL the more armor gets weakened, because it ignores a percentage of armor and weaker armor is exponentially worse." Huh? I think I know what you mean, but such a poorly worded phrase will be misunderstood by most new players, I should think.


 * ''"The higher a playrs AL is the lower is the foes boost by armor penetration in Percent-points, but high armor lowers percent points more than armor penetration raises them.
 * "25% penetration doubles the damage on 160 armor by just adding 10% penetration damage to 10% damage while 25% Armor penetration on 60 armor adds ~35% penetration damage tn 100% damage." Do you proof read anything you edit? This has almost 100% incomprehensiblilty.
 * I can't make any sense of that graph either, can anyone else, besides Ollj that is?--Jackel 02:33, 18 Aug 2005 (EST)
 * Yeah, this article is pretty incomprehensible. Ollj's typos don't make it any easier to understand either. Also i dislike the use of word half life, i understand what it means as a physics term, but some people probably don't and people should not be forced to read this wiki with a dictionary (or read another article) just to understand the words. Also I think it is not quite the right word here, the meaning is slightly different than Ollj seems to think. I am voting for complete reword/reformat by someone not Ollj. The formula (yeah it's formula not formular) could also be scrapped and just link [here] --Geeman 02:46, 18 Aug 2005 (EST)
 * the examples below the graph should help reading the graph. Its not much different from fast casting. just rewrite whatever you want to--Ollj 02:50, 18 Aug 2005 (EST)
 * There, how about that? I think this is an equitable comprimise. I don't want to throw out all the potentially interesting data Ollj has added, but it is too far beyond what the average reader is going to want to know, or likely to understand.--Jackel 05:46, 18 Aug 2005 (EST)

Research Questions section
This section seems to be entirely speculative, not to mention dubious. I am deleting it. The author is free to argue for its reinclusion here, but I will oppose a simple revert. F G 15:08, 18 April 2006 (CDT)


 * Can you be more specific about what you thought was deleteworthy about that passage? Your comment on the talk page was very cryptic. Feel free to respond on the article's talk page -- just trying to get your attention. &mdash; Stabber (talk) 15:37, 18 April 2006 (CDT)

Well, to start with, I simply dislike how big a deal this wiki makes with Judge's Insight. By my count, the mention of its anomalous behaviour can be found on: And possibly others that I've missed. It is horribly overdone.
 * 1) Judge's Insight
 * 2) Holy damage
 * 3) Light damage
 * 4) List of skill anomalies
 * 5) Ignore armor

Secondly, the passage was all about a theoretical interaction of armor ignoring damage with armor penetration. To date there is no skill or attack in the game that has both conditions. The ignore armor article states it pretty nicely, I think: "it is generally believed that when armor is ignored, armor penetration isn't relevant (for both technical and semiotic reasons)."

Thirdly, it is completely a mystery to me what a matter of mostly specialist interest needs to be featured so prominently on such an article. This sort of arcana belongs in damage, or perhaps in armor -- not in armor penetration, which is mostly a list of skills that cause armor penetration.

Fourthly, there is no fourth reason. That's it. Just three reasons. F G 16:00, 18 April 2006 (CDT)


 * You are quoting my own text to me from the ignore armor article. Also, JI is not really made a big deal of in holy damage and light damage as you claim -- just one line notes. Now, although I'm peeved that you are dismissing this passage simply because you don't like the "big deal" we make of JI, you do have a point that the interaction is theoretical, and therefore would interest only crazies. I'll leave your deletion unopposed for now unless someone else joins my side. &mdash; Stabber (talk) 16:10, 18 April 2006 (CDT)

Re-open damage calculation discussion.
An anon user (which is actually Kuntz on www.guild-hall.net) edited the article with the following:

''The two types of armor you can wear are base armor and bonus armor. There is an easy trick to tell them apart. Base armor could be called semi-colon armor as there is a semi-colon in its description. For example; "Armor: 60". Bonus armor could be called plus-sign armor as there is a plus-sign in its description. For example; "Armor +10". Base armor can be penetrated while bonus armor cannot be penetrated. This is important to remember when calculating damage. For example consider a Monk has 60 AL and +10 AL while enchanted, a shield with 16 AL and +10 AL against Lightning, and a +5 AL + 7 AL against Elemental sword. This Monk would have a total of 100 AL (60+10+8+10+5+7) against Lightning, however the 25% penetration from Lightning Orb will only effect 68 of that AL. The Monk's final AL would then be 83 ((68 * 0.75) + 32), and not 75 AL (100 * 0.75).''

I suggest confirming this edit, which seems correct (and Kuntz is a respected member of www.guild-hall.net) but I cannot verify at this moment. The revert seems to be in violation of GW:YOU and GW:AGF --Draygo Korvan 12:07, 16 June 2006 (CDT)


 * For reference, when did Kuntz did the test? Also did he test against the non-Factions Virtuoso's armor (which uses completely different wording)?  My prior test results are in Talk:Damage/Archive4, done in December 2005. If his tests are newer, then I have no issue switching to his, as long as numbers and methods of testing are posted here (no re-test needed, just want the data from his original test).   The revert was done because I had actual test results (and I linked to it when I performed the revert), whereas the edit being reverted simply stated stuff.  I wasn't suggesting the person was vandalizing the article, but between the anon being misinformed/misguided, relying on a test prior to December 2005, relying on a test after December, two out of the three "good faith" scenarios involve the anon being wrong (but with good intentions).  In the end, I simply chose the version of the article that is backed by a talk page posting.- 12:18, 16 June 2006 (CDT)

Here is his test, which is reletively simple: ''Testing is simple, roll a PvP Monk with Boon and the enchanters armor, have a friend bring some Pen damage skill, and away you go. I've known this for awhile, re-tested with Kriegar multiple times in the last week, nothing has changed. Pen works out to (60 * .75) + 10, not 70 * .75.''

It's exactly why they list AL two different ways, just like they list damage two different ways.

''Armor: 60  Armor +10''

''Two different ways. Just like Evis has a +42 which is never ever effected by armor. The base damage of Evis will be effected by armor, then +42 is added onto that result. The lowest you'll ever hit with Evis is 43~ because of this.'' Copied from http://www.guild-hall.net/forum/showpost.php?p=492290&postcount=15 --Draygo Korvan 12:22, 16 June 2006 (CDT)


 * BTW, you might want to carefully re-read the actual content of the policies you cited. Since I'm an involved party, I cannot avoid biase, but to the maximal ability of my attempt at objective judgement, I believe I was in perfect compliance with both policies (take into consideration my edit summary of the article also).  - 12:18, 16 June 2006 (CDT)
 * I was specifically talking about the So, when you can reasonably assume that something is a well-intentioned error, correct it without just reverting it or labeling it as vandalism. clause in GW:AGF. I dont mean to pick on ya but your action caused a negative post to show up in www.guild-hall.net about guildwiki, something I think the GW:YOU and GW:AGF are ment to help avoid. For something that can be easily tested as this, i think the proper action should have been to test it ingame first, then edit any mistakes and post about it in the discussion page. --Draygo Korvan 12:31, 16 June 2006 (CDT)


 * I read the thread on guild-hall, and I thought the negative feeling is mostly caused by the "Die spammer die" from Rainith. Of course, there were a few posts in the thread that were deleted by the time I got there, adn I could not read what they said.  - 13:34, 16 June 2006 (CDT)


 * On a side note, can the "Die Spammer Die" message possibly be changed to something a litle more... tactful? Especially considering the bug... I get this message fairly frequently. To a new user it is very likely to be taken as discouraging or insulting, as the guild-hall forum posts demonstrate. Something like "Automatic spam filter block. Please retry in a few moments if this is in error." would be much more appropriate. -- 66.92.33.187 14:31, 16 June 2006 (CDT)
 * You misunderstand the bug, the bug's ban message cannot be changed. It is basically a bug that happens when a ban is incorrectly applied. It takes the message from a DIFFERENT ban and applies it instead. So what happens is people are seeing the "Die Spammer Die" message ranith used to ban a different user. There is no "Spam filter" that automatically blocks users from the site. --Draygo Korvan 14:40, 16 June 2006 (CDT)
 * That message is taken from an actual ban that exists, and I have no clue how the software operates. I'll try something from the MediaWiki side, but it'll depend on luck. - 14:35, 16 June 2006 (CDT)


 * First of all, there is (or was) an actual spam filter that automatically blocks users from the site, and you misunderstood my understanding of the bug. Anyways, my hunch didn't work.  I knew that message came from one of Rainith's actual bans, and my hunch was it might be either his most recent or oldest ban.  Turns out that is not the case.  So, one solution would be to unban everyone currently banned, and whenever we ban anyone, we use a more generic reason keeping in mind some random person might see this message too.  I'm not going to support that plan (and it wouldn't work if the spam filter use an expired ban reason instead of a current one).  The alternative is to get Gravewit to look into the software.  Unless, of course, the ban is caused by a correctly placed ban on a spammer who used dynamic IP.  The solution to that really will be to unblock the currently blocked spammers, and use more toned down reasons for banning in the future. - 14:49, 16 June 2006 (CDT)


 * BTW, I did not "just reverting it or labeling it as vandalism", I reverted it with a edit summery explicitly stating the anon edit conflicts with a recorded test result on guildwiki, and I provided direct anchored link to the test result. The Assume Good Faith policy is meant to discourage unexplained reverts or calling edits vandalisms, I have committed neither (I assumed the user made a mistake with good intentions).  The You Are Valuable policy is meant to discourage self-deprecation (which in some cases is also used as a maytredom device) and by corallary can also mean "Just because you have been around longer doesn't mean you are better"; I also have not committed those.
 * What I am guilty of is "Not re-verifying previously reported test results before reverting an edit not backed by known test results". I have nothing against codifying that into a new GuildWiki policy, but I'm not going to take the initiative to do it.  The above is my inevitably biased attempt at trying to objectively analyse my actions and intentions.  It is most unfortunately that the other user involved took it badly, and I welcome furthur critism at the analysis of my part in the incident. - 14:12, 16 June 2006 (CDT)
 * I dont think it needs to be policy. I'm just saying because the article was edited in good faith, it should have been treated fairly, especially when it could be easily tested. Thanks for your hard work on this though, we now have a deeper understanding of how armor penetration works thanks to Kuntz and you for verifying it. I recommended to Kuntz that he register with the wiki so he can avoid that ban bug, which is more of a cause of the PR problem than your revert.--Draygo Korvan 14:21, 16 June 2006 (CDT)


 * By the way, the current way the article is written, suggests Kuntz's edit was partially wrong. However, this suggestion that he is partially wrong is based on assumptions not explicitly tested.  I will leave it up to other people to run the tests, if anyone care enough.  If you can't crunch the formulas, just collect the numbers and I'll crunch stuff out for you.  I simply don't want to run the additional tests.  See my test results for an example of what data I need (level, damage vs naked, etc) - 14:35, 16 June 2006 (CDT)


 * Not sure where to put this, so it's going here. I don't think PanSola has done anything wrong here. To the best of his knowledge PanSola believed that he was reverting a mistaken contribution. He explained that it was opposed to tests that he had carried out himself. I would say that it is Kuntz who has not assumed good faith here:


 * Ensign & that other guys oldschool guides are outdated and/or incorrect in some areas, those areas have been carried over to Wiki copy/paste style assuming they were correct to begin with. No amount of testing would have given you different results for how Pen works on armor because it's always funcitoned the same way. If you still have a copy of what I wrote on the Pen page, please place it back, as it is correct.


 * What he's basically saying is that the contributors of this article have copied someone elses work, and that they haven't tested anything. That's not assuming good faith. It's unfortunate that he received the ban message, whether it was meant for him or not, but I certainly don't think PanSola has done anything wrong.  &lt;LordBiro&gt;/&lt;Talk&gt; 14:45, 16 June 2006 (CDT)


 * To be fair, GuildWiki policy shouldn't be used to govern (or even judge) people's behavior on other fansites d-: Different places, different cultures/expectations.  That's the very reason we have the We are not Wikipedia policy (-: - 14:52, 16 June 2006 (CDT)
 * Assumption: you mean "he" as in Kuntz. Yes Kuntz was saying the article was largly copied from previous research. But that doesnt mean Kuntz edit wasn't in good faith. I also do not think Pansola did anything majorly wrong here. Note i used the word Seem, and I was using it in reference to the result of the action, which is what I think the GW:AGF and GW:YOU policies are ment in some ways to mitigate. AGF says you should edit any incorrect data instead of reverting it if the edit was well-intentioned. A revert, no matter how justified with a summery is a bit harsh in most cases (except vandalism etc). I only cited those policies because I felt the revert was against the spirit of those policies, what those policies are ment to do and cover. I really dont want to get deeper in this arguement here, I dont think pansola's revert was really against any rules or policies - but I thought it could have been done differently, with a much better result. I would rather not carry on this discussion here, as it is completely off topic now. In addition, I dont want this to be precieved as an attack on PanSola's good charater, which I respect. --Draygo Korvan 15:01, 16 June 2006 (CDT)
 * I want to make it clear that, while I have learned my actions in this incident led to a rather unhappy user experience for Kurtnz, I do not see this experience altering my future behavior. In other words, in the future, if I see an edit made that says something in direct contradiction to test results posted on guildwiki, and I do not see any indication of tests in support of a newer edit (an edit summery of saying something like "GuildWiki is wrong, tested this last weekend" is bare minimum), then I will still revert the edit with a edit summary pointing to the test result available on GuildWiki, and an invitation (also in the edit summary) for the contributor who got reverted to discuss the conflicted content on the talk page.  One exception I will make is when I see something turning into a common misconception (or a common confusion, such as globalness of Knight's/Virtuoso's bonus), in which case I will edit the article to say what the contributor just wrote is actually a common misconception.  I am willing to change my behavior if the overall GuildWiki community consider this bad a practice, but until then, I will go on as I have done.  GW:FAITH's instruction on "do not simply revert", according to my interpretation, means "do not revert without explaination", because RV without explaination means the reason for revert is so trivial it's not even worth mentioning.  I believe a detailed edit summary demonstrates assumption of good faith. - 15:27, 16 June 2006 (CDT)

I don't think PanSola should have done anything different. If Kuntz had not recieved the ban message then perhaps he would have come to this talk page and discussed his findings in more detail. Unfortunately that didn't happen and we'll never know how things would have proceeded. What I can say is that Kuntz' justification for the edit is patronising and presumes that we will just allow anything to remain on the site untested. That is an insult to the GuildWiki and it's an insult to PanSola and the others who have carried out tests on armor penetration. I'm not judging anything he's done on other sites by 'GuildWiki policy'; Kuntz has every right to be upset with the banning incident, and I'm annoyed that it ever happened, but that doesn't change the fact that he has stated that "those areas have been carried over to Wiki copy/paste style" and as far as I'm aware he is incorrect.  &lt;LordBiro&gt;/&lt;Talk&gt; 16:07, 16 June 2006 (CDT)


 * Just want to state that, with respect to Kuntz's statement that implied we didn't do any testing on armor penetration and just copied it, I do not feel offended or insulted. He's simply misguided.  BTW, our previous version of the armor penetration formula should disagree with all the older research that exist out there.  I'm not aware of any work that uses two bonus terms for AP. - 17:09, 16 June 2006 (CDT)

Test results

 * Date: 13:34, 16 June 2006 (CDT) (minus 30 minutes for typing up this report)
 * Build: Level 20 Me/Mo, with Healing Breeze
 * Equipment:
 * Performer's Gloves from Prophecies Campaign with 60 Armor and gives 15 armor while casting spells.
 * Peppermint Shield
 * Location: Old Ascalon
 * Test subject: Flash Gargoyle (level 3 = Damage Level 9)


 * Lightning Orb expected damage to 0 AL (naked):
 * Assume Air Magic 0: 11.69 (round to 12)
 * Assume Air Magic 1: 18.70 (round to 19)
 * Actual damage taken when naked: 12

Level 3 Flash Gargoyle is extrapolated to have 0 rank in Air Magic


 * Lightning Orb expected damage to 15 AL:
 * Assuming 25% armor penetration: 9.61 (round to 10)
 * Assuming no armor penetration: 9.01 (round to 9)
 * Actual damage taken while casting spells: 9

Armor penetration does not affect the "Gives 15 armor while casting spells." bonus from Virtuoso's Armor


 * Lightning Orb expected damage to 10 AL:
 * Assuming 25% armor penetration: 11.53 (round to 12)
 * Assuming no armor penetration: 9.83 (round to 10)
 * Actual damage taken while wearing shield: 10

Armor penetration does not affect the "Armor: 10" from Peppermint Shield

Conclusion: Armor penetration only affects BaseAL.

''Note: This is in direct contradiction with the test reported in December of 2005 at Talk:Damage/Archive4

- 13:34, 16 June 2006 (CDT)

The following are assumed to also not be affected by armor penetration, but were not verified in the above test.
 * Armor piece bonus of the syntax "+x armor"
 * Primary weapon bonus of the syntax "+x armor"
 * Primary weapon bonus of the syntax "-x armor"
 * Secondary weapon bonus of the syntax "+x armor"
 * Secondary weapon bonus of the syntax "-x armor" (doesn't seem to exist, listed for completeness sake)
 * Shield armor value of the syntax "Armor: x (requires y Attribute)" when requirements are met
 * Shield armor value of the syntax "Armor: x (requires y Attribute)" from collectors when requirements are not met
 * Shield armor value of the syntax "Armor: x (requires y Attribute)" from dropped loot when requirements are not met <- This is stated by Kuntz to be counted for armor penetration, but the current article assumes it is not affected by armor penetration.
 * Skill bonus of the syntax "+x armor" (outdated teste result from Descember suggest this is not affected by armor penetration)
 * Skill bonus of the syntax "-x armor" (outdated teste result from Descember suggest this is not affected by armor penetration)

- 13:48, 16 June 2006 (CDT)


 * I also just performed some tests, though they are by no means exhaustive. I took a warrior with 12 axe, 16 strength, a PvP axe without a damage mod (besides customization), and wild blow and whacked a target with two different sources of +armor while enchanted (monk armor for +10 and a focus for +5).  The purpose was to see if the AP (from strength while using wild blow) modified either +armor.

! Armor !! Observed damage !! EAL if AP affects total !! "Total" AP damage !! EAL if AP affects base !! "Base" AP damage
 * - align="center"
 * 60 || 57 || 60 * .84 || ~56.11 || same || same
 * - align="center"
 * 60+10 || 49 || 70 * .84 || ~48.52 || 60 * .84 + 10 || ~47.19
 * - align="center"
 * 60+15 || 45 || 75 * .84 || ~45.11 || 60 * .84 + 15 || ~43.27
 * }
 * To compute the damages from the EALs, I did 28 * 1.2 * 2^((60 + 20 - EAL) / 40).
 * Kuntz's argument seems to be incorrect for the case of monk armor with +10 armor while enchanted and focuses with +5 armor while enchanted. This of course says nothing for any of the other sources of +armor Pan listed above.
 * Of interest is the mismatch in the "control" test where there wasn't any +armor going on. Likely, I did something wrong, but I don't see my mistake.  Otherwise, perhaps there's some sort of mid-calculation rounding.  --68.142.13.99 01:08, 17 June 2006 (CDT)
 * Your theoretical values are correct, barring rounding effects. Instead of getting PvP monks, use PvE monks and have them buy cheap armor from Lion's Arch or earlier.  This way the damage will be bigger, and the difference will be more appreant.  Actually, with the PvP monks, you can test the focus bonus by having the monks strip.  The difference should (theoretically) be really clear. - 02:18, 17 June 2006 (CDT)