Talk:Flurry

Worth making a note that 0.75 * 1.33 = 1? Meaning that all else being equal, engaging this skill should have no net effect on damage/second.
 * Easy way to test it: get 2 ele/war with identical +15 always/-5 en swords, 12 swordsmanship, +health armor, and just slug it out 1 with flurry 1 without.  If health at the end is significantly different, then it approximates the effect of flurry - either no change, reduced overall damage, or increased overall damage with flurry up.  Ninja Quail 23 July 2006

33% faster attack speed means 4 attacks instead of 3, 33% more attacks in the same time; 4 seconds with axe/sword and 5,25 seconds with a hammer. The damage is the same, the adrenaline gain is not. You basically convert energy into adrenaline, intelligent warriors only triger it when they know they do not miss, giving them an adrenaline boost.
 * without Flurry: 3 attacks in 4 seconds deal 3*40=120 damage, you gain 3 strikes of adrenaline.
 * with Flurry: 4 attacks in 4 seconds deal 4*30=120 damage (25% less damage per hit), you gain 4 strikes of adrenaline.


 * A technical point. Consider an axe: An attack with it happens every 1.33 seconds or roughly 0.75 attacks per second. Under Flurry, the time between attacks should be reduced by 33%, changing the attacks per second to 0.75 / (2/3) or 1.125 attacks per second. Over four seconds, without Flurry there are three axe attacks, with Flurry 4.5 attacks, a 50% increase in number of attacks and adrenaline gain. Some testing with the actual skill supports this increase.


 * Your calculations are wrong. Following the description, the attack rate increases by a third, this means: .75 * 1.33 = 1. What you calculated is an increase by 50% ( / (2/3) == * 3/2 == * 1.5). So, on a 4 second period, with a rate of 1, it is quite possible that you'd get a count of 4.5 attacks, accounting for lag on start and end of the 4 sec period.--theeth 22:15, 18 January 2006 (UTC)


 * The same testing indicates the average damage per attack is actually roughly 33% less, not 25% as in the skill's description, resulting in the average damage per second being the same as without Flurry. So the skill notes are still basically correct. rhess 19:22, 18 January 2006 (UTC)

Personally Flurry has never left my skill bar since I last changed my build which was some time ago. It's not good on its own but its great when using it to charge other skills. Using a Zealous mod though is pretty essential when spamming it. Remember this skill is a stance so that you can gain the bonus's from "while in stance" equipment. --SK 01:52, 24 May 2006 (CDT)

With my recent research with this skill, I discovered that spells like Conjure Lightning that add extra damage seem to not affected by the 25% reduction, just like how Illusionary Weaponry is unaffected. The increase damage could offset the reduction. --SavageX 12:12, 29 May 2006 (CDT)
 * Attack skills are also unaffected. An easy test is Balthazar's Island with Dragon Slash, Flurry, a candy cane weapon, and 0 ranks in swords. 4 damage to the dummy without Flurry, 3 with Flurry (~ 25% reduction, as expected). Dragon Slash, which adds 10 extra damage at rank 0, brings that up to 14 and 13, respectively. (Any skill works. I just picked one already sitting on my skill bar.) &mdash; 130.58 (talk) ( 04:09, 17 June 2006 (CDT) )

Revert
The original content is probably correct. At the very least, the edit was incorrect. To present my own data, in the past I had two characters using candy cane weapons whack each other without any armor on. One used flurry all the time, the other just whacked. I don't remember how much weapon mastery we used, but it was chosen to be the lowest value that did not result in rounding due to flurry's 25% damage decrease. It was probably around 3 mastery to deal 12 base damage (versus 0 AL). Criticals skewed things a little, but with low mastery they were uncommon; I think 1-3 usually over the course of a single trial. The guy using flurry always killed the one not using flurry, so the simple "4/3 * 3/4 = 1" theory is incorrect. The guy using flurry ended up with around 40 health left every time, so in the time he dealt 480 damage he took around 440 damage. The current writeup claims a 112% damage output when using flurry which is about what my experiment produced. --68.142.14.19 06:48, 29 July 2006 (CDT)

Recent change
Anon changed a load of figures, is this correct? &mdash; Skuld 08:32, 2 August 2006 (CDT)
 * No, see above. He made essentially the same change on July 29.  --68.142.14.106 09:44, 2 August 2006 (CDT)
 * Why is it assumed that 33% of time is taken away from 1.33 rather than 33% added to 0.75?
 * Also, 0.75*1.33 DOES equal to 1.00 when sigfigs are considered.
 * While I'm glad you moved your comments to the bottom, I don't think you read the section above this one where I explained my test method. I did not assume anything.  --68.142.14.106 10:55, 2 August 2006 (CDT)

Misconceptions
The greatest misconception concerning flurry is the effect the skill has on hitting speed. For unknown reasons, some people prefer to believe that the 33% increase in the rate of attack manifests itself in a decrease of hitting period by 33% rather than increase in hitting frequency by 33%. Because there is no offical data on attack times, the community has collected its own data. The main way data was collected was in-game with the use of stopwatches. This has resulted in the publication of data in the form of periods (time per x) causing 1.33 s/h to be a more popular number than 0.75 h/s. Logically, why would a skill which states that it increases the rate of attack by 33%, actually it by ~50% as stated by the current edit? "it actually nets you an approximate 50% increase in the number of attacks." If one were to calculate what the effect on frequency would be if one were to decrease the period by 33%, one would find a similar number. *<"flawed" flurry effect>=<"flawed" affected period> (4/3)*(2/3)=8/9 -> resulting frequency = 9/8 = 1.125 -> 50% increase in hits per second. Through this demonstration, we can see that the current edit is changing the wrong attack values. Rates, especially ones which include time, are with respect to time. This means that flurry affects the attack frequency which is in units of hits per second rather than the period.
 * Please note that 0.75*1.33 DOES equal to 1.00 when significant figures are considered.
 * Also note that the period of attack would never equal to 1.33 considering that data types can hold up to 6 decimal points. It would probably be closer to 1.333333333, which makes it safe to use proper fractions (since there are numerous references to them in the skill descriptions anyways).


 * Have you tested this? As stated above, the numbers in the article are based on actual tests of this skill (and Frenzy which has the same wording with respect to increasing attack speed). If you have tested this, I'll gladly go retest it and present the data here, but if you haven't, I think you'll need to design a test that indicates you are correct before the community will agree with you. --12:10, 2 August 2006 (CDT) Forgot to log in. --Kryshnysh 12:12, 2 August 2006 (CDT)


 * Ditto. Everything else being equal, I would've gone for strict interpretation of the official wording.  However, if there has been tests conducted in support of the official wording being wrong, and there are no tests reported of the other side of the argument, I am more inclined to go with the test results.  This would not the first time Anet has messed up the skill description. - 13:52, 2 August 2006 (CDT)