Log in

View Full Version : Why are weapons with higher delay better?


ZerkerDwarf
04-27-2010, 11:35 AM
<p>I never understood why single wield weapons with a higher delay are better than a comparable weapon with a shorter delay.</p><p>E.g. Enervated berserker weapon vs. the 6s delay weapon from Madoril & Co (seems to be named "Hairsplitter" in English according to the /weapons command, whereas the German translation is "Patron"... weird).</p><p>/weapons shows these results: Dragon's Temper - Hairsplitter (ungrouped with same equipment)</p><p>Base damage: 96-543 vs. 200-1133</p><p>Actual damage: 340-1926 vs. 709-4020</p><p>Actual delay: 3.0 vs. 6.0 (maximum haste mod: 1.3 vs. 2.7)</p><p>Proc chance: 9% vs. 18%</p><p>Procs per minute: 1.8 vs. 1.8</p><p>The 6s weapon strikes twice as hard as the 3s weapon but therefor only half as often. When the 6s weapon strikes, there is a 18% chance to cause a proc and the 3s weapon has 9% chance.</p><p><strong>So why then is the 6s weapon (said to be) better?</strong> Double amount of damage and proc chance, but striking only half as often as the 3s weapon. That seems to be equal. Those hidden combat mechanics nobody has an insight into are quite confusing.</p>

Banditman
04-27-2010, 11:46 AM
<p>Higher delay, in general, means that the weapon has a higher number at the top end of it's damage spread.  This is what is important these days.  Because crit is so easy to get, you're almost never concerned with what the low end of the weapon is because every single swing is a crit, meaning that the minimum number you get back is the top end of the spread.</p><p>Higher delay also makes it easier to cast CA's / Spells while the weapon is in it's delay phase.  Let's say you have a weapon with an effective delay of 1.5 seconds, and another with an effective delay of 2.5 seconds.  Further, you want to cast an ability with a cast time of 2.0 seconds.</p><p>With the first weapon at 1.5 seconds delay, if you cast the ability immediately after the auto attack hit, your next auto attack swing comes due while the ability is still casting, delaying the auto attack swing even further.  So, the 1.5 delay weapon doesn't actually fire until 2.5 seconds have passed - 2.0 for the cast and .5 for the recovery.</p><p>With the second weapon, at 2.5 seconds delay, casting the same ability immediately after the auto attack hit does not delay your next auto attack swing at all.</p><p>There are a few instances where faster weapons, in theory, could be better.  Specifically, they are related to those abilities which yield a proc on every single auto attack swing . . . Stampede from a Mystic for example.</p>

Anastasie
04-27-2010, 11:48 AM
<p>One of the main benefits of using a slower weapon is being able to time auto attacks. With the slower weapon you are much less likely to interrupt your auto attack (which reduces your dps a lot) and can squeeze more ca's between each hit.</p>

Grabs
04-27-2010, 11:56 AM
<p>Dont forget,  if you are capped on haste, a fast weapon may be way too fast as result.</p><p>I think 1.5 Delay is the cap right?  cant go below that even with haste right?  so in the end,  you get more benefit from haste on a long delay weapon. If my number above is correct,  a 3.0 delay weapon,  is as low as you can go and get full benefit from max haste.</p>

Dareena
04-27-2010, 11:59 AM
<p>Also Attack Speed / Haste lowers a weapon's delay speed by a %.  Therefore a specific attack speed modifier would lower a 4 second delay by an amount, but the same attack speed would lower a 6 second weapon by an even greater amount.  Since weapon damage ratings are based on the weapon's starting delay, it very advantageous to use weapons with high delays so you can get the greatest possible benefit out of your Attack Speed score.</p>

Cusashorn
04-27-2010, 12:09 PM
<p>I asked this question in the Monk forums when I found out that our weapons were being changed from 2.5 to 4.0 delay.</p><p>They told me that a slower delay weapon has a lower chance of being reposted by the enemy, and that a weapon procs more times a minute.</p><p>If a weapon has a % chance to proc, then a faster weapon is better. If a weapon procs 1.7 times a minute or whatever, then a longer delay weapon means it procs FAR more often.</p>

ZerkerDwarf
04-27-2010, 01:52 PM
<p>I'm afraid I am not fully into it so far.</p><p>Why does a longer delay cause MORE on-successfull-attack-procs?</p><p>When it reads something like "on successful attack this has a chance to inflict 1500 damage on target - triggers about 1.8 times per minute". Why should a higher delay cause this to trigger more often within the same time?</p><p>Let's say a fight takes 60 seconds. With a weapon of the double delay I strike half as often as with the half delay. When I strike ~30 times during the fight instead of ~50 times, why then do the 30 times cause more "on successful attack" procs?</p><p>I really also don't understand why they a long time ago changed e.g. "4%" into "1.8 times per minute". This is a thing I will never understand. Supposingly I am too stupid to recognize the advantages of "per minute" instead of "4% chance on attack".</p><p>Isn't the "1.8 times per minute" just a VERBAL CIRCUMSCRIPTION for "4% (or whatever) on successfull attack" ?</p>

Ambrin
04-27-2010, 02:08 PM
<p><cite>Cusashorn wrote:</cite></p><blockquote><p>I asked this question in the Monk forums when I found out that our weapons were being changed from 2.5 to 4.0 delay.</p><p>They told me that a slower delay weapon has a lower chance of being reposted by the enemy, and that a weapon procs more times a minute.</p><p>If a weapon has a % chance to proc, then a faster weapon is better. If a weapon procs 1.7 times a minute or whatever, then a longer delay weapon means it procs FAR more often.</p></blockquote><p>It's not that you have a lower chance of being riposted, it's that you you are hitting the mob less often and therefore have fewer attacks to actually riposte.</p><p>Longer delay weapons also proc more because the standardized proc system doesn't take into account your haste (AFAIK).</p><p>Slower weapons also work better with things like flurry and AE auto attack which only go off the mainhand, producing a higher amount of damage when they hit. This basically works out to slower weapons producing better burst DPS.</p>

SageGaspar
04-27-2010, 04:14 PM
<p>On procs that are normalized (per minute), the delay of the weapon literally does not matter at all. That's the point of procs per minute. On 100% procs like chimes the faster weapon actually has an advantage. 100% haste will a little more than double the procs per minute a weapon will get but this does not give an advantage to faster or slower weapons.</p><p>As far as high end damage goes, the way it works is when you crit it multiplies the base damage you rolled times your crit multiplier and rounds up if you're under the max. Thus a weapon that has a high spread can be advantageous. If you had a weapon that did 1-1000 damage for instance and you crit on a damage roll of 1, the crit multiplier would bring that up to like 3 damage, and then it'd round up to 1001 damage. It's not really the high end damage that matters, it's the damage spread of the weapon combined with its average non-crit DPS.</p><p>Otherwise delay is mostly important for timing attacks. The goal is to slip CAs in between autoattacks. If you had for instance an obscenely fast weapon like .25 every time you used a CA you'd lose an autoattack, which is no good.</p><p>Slower delay does not improve double attack or flurry or AE autoattack or anything like that. It does mean you will do more burst damage. In PvP this is very advantageous, in PvE most of the time you'd prefer a steady damage output versus random bursts.</p>

Cusashorn
04-27-2010, 04:45 PM
<p><cite>Ambrin@Nagafen wrote:</cite></p><blockquote><p><cite>Cusashorn wrote:</cite></p><blockquote><p>I asked this question in the Monk forums when I found out that our weapons were being changed from 2.5 to 4.0 delay.</p><p>They told me that a slower delay weapon has a lower chance of being reposted by the enemy, and that a weapon procs more times a minute.</p><p>If a weapon has a % chance to proc, then a faster weapon is better. If a weapon procs 1.7 times a minute or whatever, then a longer delay weapon means it procs FAR more often.</p></blockquote><p>It's not that you have a lower chance of being riposted, it's that you you are hitting the mob less often and therefore have fewer attacks to actually riposte.</p></blockquote><p>Right, that's what I ment to say.</p>

Blu
04-27-2010, 05:06 PM
<p><cite>ZerkerDwarf wrote:</cite></p><blockquote><p>I'm afraid I am not fully into it so far.</p><p>Why does a longer delay cause MORE on-successfull-attack-procs?</p><p>When it reads something like "on successful attack this has a chance to inflict 1500 damage on target - triggers about 1.8 times per minute". Why should a higher delay cause this to trigger more often within the same time?</p><p>Let's say a fight takes 60 seconds. With a weapon of the double delay I strike half as often as with the half delay. When I strike ~30 times during the fight instead of ~50 times, why then do the 30 times cause more "on successful attack" procs?</p><p>I really also don't understand why they a long time ago changed e.g. "4%" into "1.8 times per minute". This is a thing I will never understand. Supposingly I am too stupid to recognize the advantages of "per minute" instead of "4% chance on attack".</p><p>Isn't the "1.8 times per minute" just a VERBAL CIRCUMSCRIPTION for "4% (or whatever) on successfull attack" ?</p></blockquote><p><strong>Proc Chance</strong></p><p>Almost all offensive procs are phrased as x times per minute so it can be normalized for weapon delay. I'm told by a GM that even my monk's AA procs that say percentages in the descriptions are really normalized to a weapon delay.</p><p>Your average proc is normalized to a 3.0s weapon day IIRC. (Even if this assumption is incorrect, the comparison that follows still stands to reason.) In 60 seconds, there are 20 swings of a 3.0s delay weapon. So, "2.0 times per minute" would equate to around 2 swings in 20 swings or 10% of all your swings in a minute.</p><p>So now let's say that your weapon is actually 6.0s delay. So in 60 seconds, you can have 10 swings. So, 2.0 times per minute would mean that 2 swings in 10 need to proc to maintain the average. That equates to 20% of all swings, so the proc chance is 20%.</p><p>Now let's factor in haste (which I caps at 125% (can't recall the exact +haste cap)). I believe the proc percentage is applied before haste is applied.</p><p>So your 3.0s base-delay weapon has a 10% proc. With 100% haste, your actual weapon delay is 1.5, so you will actually get 40 swings in per minute, and your weapon will proc an average of 4 times per minute. (40 swings * 10% = 4 swings proc on avg.)</p><p>So let's assume 6.0s base-delay weapon with it's 20% proc chance. With 100% haste, your actual weapon delay is 3.0, so you will get 20 swings per minute, and your weapon will proc the same 4 times per minute as the 3s weapon. (20 swings * 20% = 4 swings proc on avg.)</p><p>So in general, delay makes NO difference in frequency of (normalized) procs. (I actually started writing this expecting a different outcome, but the numbers said otherwise.) Haste will increase the number of (normalized) procs you get per minute, but the increase is the same regardless of weapon delay.</p><p>In specific cases, there may be more or less procs depending on what the RNG feels like that minute.</p><p><strong>Fast vs Slow</strong></p><p>Slower weapons have advantages in that they allow you to cast more CAs between swings. This can provide an initial spike of dps before you start running out of things to cast. With a lot of zones and in a decent group, you are going to parse higher just because you are unloading on mobs up front and the travel time to the next mob might be enough time for your spells to come back up. But on long fights, faster weapons should average to about the same as the slower ones.</p><p><strong>2h vs DW</strong></p><p>2-handers have advantages in very specific situations. One of the main ones is AE auto attack. AE auto attack only uses mainhand swings for the aoe hits. So, an AE-auto hit is going to be much higher damage with a 150 DR 2-hander than it will be dual-wielding (an equivalent) pair of 100 DR 1-handers. That's 50 more damage rating on each of the 3 extra targets == significant. I believe also that only the main hand has a chance to Flurry, but not positive.</p><p>For those classes that do not have AE attack (or Flurry), Dual Wielding will probably be better, because 2x 1-handers tends to give more stats and more procs (as in each weapon having it's own proc) than a single 2-hander.</p><p><strong>Offhand Swing Speed</strong></p><p>Also note that the offhand swings at half the rated speed. That's why a pair of 100 DR 1-handers is about the same as a 150 DR 2-hander, theoretically. Obviously other factors play into that like stats and procs (and ae attack and flurry like mentioned above).</p><p><strong>Duplicate Procs</strong></p><p>Also note that having 2 of the same proc is not detrimental (unless the proc casts a buff, like the MC strength ring -- then each new proc'd buff overwrites the old one). One proc doesn't overwrite the other (though their effects may, but usually not). They don't magically combine into one proc with twice the proc chance, either. All procs that fire on attack have their chance to fire on any mainhand swing. So if you have Divine Restoration on 3 pieces of gear, then on each mainhand swing 0-3 Divine Restoration procs will fire, depending on what the RNG feels like that moment.</p>

Crismorn
04-27-2010, 05:13 PM
<p>More delay between auto attacks lessens the human error factor.</p>

Carthr
04-27-2010, 06:37 PM
<p>Just like almost everything in this game, it depends.  Take for instance, an inquisitor.  Fanatical Devotion procs on every single hit, once it has been proc'd by a critical hit. So once it has proc'd you want to get the max number of hits in with it.  Also they have a chance to proc 20% flurry, however with excessively long delays(8s from Hammer of Hard Knocks) the flurry's hit HARD, but you don't get many of them, and there's the potential to drop the flurry proc, in comparison to a shorter delay weapon like the Dogwood Staff.</p><p>Things like DPS mod, double attack, haste all come into play with this as well.  Most people agree that 4-6s weapon delay's generally offer the most dps.  However, the longer the delay, the more important it is to get accuracy and CPS(crushing/piercing/slashing) up, so that you don't miss!</p>

Aule
04-28-2010, 07:51 AM
<p><cite>Xerq@Crushbone wrote:</cite></p><blockquote><p><strong>Proc Chance</strong></p><p>Now let's factor in haste (which I caps at 125% (can't recall the exact +haste cap)). I believe the proc percentage is applied before haste is applied.</p></blockquote><p>It is indeed 125% speed increase at 200 haste modifier</p><p>3.0s = 1.336.0s = 2.67</p><blockquote><strong><cite>Fast vs Slow</cite></strong><p>Slower weapons have advantages in that they allow you to cast more CAs between swings. This can provide an initial spike of dps before you start running out of things to cast. With a lot of zones and in a decent group, you are going to parse higher just because you are unloading on mobs up front and the travel time to the next mob might be enough time for your spells to come back up. But on long fights, faster weapons should average to about the same as the slower ones.</p></blockquote><p>Slower weapons have the advantage in that usually you see a higher spread on longer delay weapons.  Once you factor in the effects of the crit, the longer delay weapons deal more normalized damage from the autoattacks.  This is generally made up a fair amount by 100% on swing proc's like peace of mind, cob, etc, where it doesn't make up on is aoe autoattacks.  Particularly for brawlers with crane flock, and berserkers, you want the highest max damage you can find on a weapon for aoe fights.k</p><blockquote><strong>Offhand Swing Speed</strong><p>Also note that the offhand swings at half the rated speed. That's why a pair of 100 DR 1-handers is about the same as a 150 DR 2-hander, theoretically. Obviously other factors play into that like stats and procs (and ae attack and flurry like mentioned above).</p></blockquote><p>The delay penalty for dual wielding is 30%.  So 4.0s delay weapons become 5.2s, with 200 haste mod that drops to 2.31.  Damage rating is a largely meaningless number now, you really want to look at the max damage divided by the delay.  Give some brief thought to the minimum damage but all other things equal, the one with the higher max damage will do more auto attack damage.</p>

Ocello
04-28-2010, 12:37 PM
<p><cite>Cusashorn wrote:</cite></p><blockquote><p>I asked this question in the Monk forums when I found out that our weapons were being changed from 2.5 to 4.0 delay.</p><p>They told me that a slower delay weapon has a lower chance of being reposted by the enemy, and that a weapon procs more times a minute.</p><p>If a weapon has a % chance to proc, then a faster weapon is better. If a weapon procs 1.7 times a minute or whatever, then a longer delay weapon means it procs FAR more often.</p></blockquote><p>This is ALL wrong...a slower weapon has less --CHANCE-- to be riposted simply becuase it hits less often.</p><p>Similarly, procs don't proc MORE, tbh they probably proc less with a slower delay.  You just have a higher CHANCE for it to proc with each swing.</p><p>Think about it--every time you hit something you have a chance to proc--sometimes procs will hit consecutively (if you add a trigger to ACT to listen for them you will know this).  If you hit something 10 times vs hitting something 5 times you could make that proc go off up to 10 times theoretically vs only 5.  Of course it doesnt always work that way, but you have more <em>chances</em> for that proc to hit.</p><p>And as far as the haste thing goes, as a Berserker I kind of preferred the 3.0 delay because you could go CA-auto-CA-auto and it was nearly flawless.  Now using a 4.0 delay, I have to wait a split second to use that next CA.  I'm not sure if it is worth it at this point, I just know that the 4.0 delay weapon I have is simply a better weapon than my E-Dragon's Temper.  TBH depending on my group setup I can get 2 CA's off when using the 4.0 but it will usually delay my auto by split-seconds...not sure how bad that hurts me but I'm sure it does.</p>

Blu
04-29-2010, 04:53 PM
<p><cite>Aule@Guk wrote:</cite></p><blockquote><p>Slower weapons have the advantage in that usually you see a higher spread on longer delay weapons.  Once you factor in the effects of the crit, the longer delay weapons deal more normalized damage from the autoattacks.  This is generally made up a fair amount by 100% on swing proc's like peace of mind, cob, etc, where it doesn't make up on is aoe autoattacks.  Particularly for brawlers with crane flock, and berserkers, you want the highest max damage you can find on a weapon for aoe fights.k</p></blockquote><p>Here's the explanation of why higher spread weapons give higher average damage with criticals factored in.</p><p>For example, let's say you have 100% crit chance (pretty easy to get with TSF gear). Your weapon damage is 100-400. Your base crit modifier is 1.5. Your hits should range from 150-600, before the min crit rule. The min crit rule states that your actual crit hit will be the larger of a) max base weapon damage (400 in this case) + 1 or b) the calculated critical damage (150-600). So real actual hits will be between 401-600, with about 55% of your hits being 401 damage, because the RNG * crit modifier was less than 401 and was rounded up to 401. (RNG generated base damages from 100 to 266 will all produce a 401 hit. 267-400 will all be the given value times 1.5.) Your average hit will be 401 55% of the time and 500ish ((600 + 401) / 2) the other 45% of the time. So the real average hit is around 445.55 due to the min crit rule whereas it would be 375 with just a straight 1.5 times base damage crit calculation.</p><p>On a closer delay weapon with the same DR, let's say 200-300, your actual crit hits will range from 301-450 (but it would be 300-450 without the minimum crit rule, which is nearly identical range). The average hit in this case is going to be around 376. So clearly, higher spread gives better damage because of the minimum hit rule.</p><div><p><cite>Aule@Guk wrote:</cite></p><blockquote><blockquote><strong>Offhand Swing Speed</strong><p>Also note that the offhand swings at half the rated speed. That's why a pair of 100 DR 1-handers is about the same as a 150 DR 2-hander, theoretically. Obviously other factors play into that like stats and procs (and ae attack and flurry like mentioned above).</p></blockquote><p>The delay penalty for dual wielding is 30%.  So 4.0s delay weapons become 5.2s, with 200 haste mod that drops to 2.31.  Damage rating is a largely meaningless number now, you really want to look at the max damage divided by the delay.  Give some brief thought to the minimum damage but all other things equal, the one with the higher max damage will do more auto attack damage.</p></blockquote></div><p><span style="color: #888888;"><span style="color: #ffffff;">I thought the offhand penalty was half... otherwise auto-attack timing would be difficult between 2 same-delay weapons. But you could be right... it just must be that the offhand auto-attack is delayed by CAs a lot because it doesn't fall on the same interval as the main-hand swing.</span></span></p><p><span style="color: #888888;"><span style="color: #ffffff;">DR can still be a meaningful number for statistical comparison if a few things are assumed to be equivalent among the weapons. Then it's just useful for calculating which effects are more beneficial. When actually comparing weapons in-game though, you are right. You need to look at spread vs delay. With 100% crit, a 100 DR weapon with a 1.5:1 spread is much worse than a 100 DR weapon with a 4.0:1 spread, as demonstrated above.</span></span></p>

Blu
04-29-2010, 05:29 PM
<p><cite>Ocello wrote:</cite></p><blockquote><p><cite>Cusashorn wrote:</cite></p><blockquote><p>I asked this question in the Monk forums when I found out that our weapons were being changed from 2.5 to 4.0 delay.</p><p>They told me that a slower delay weapon has a lower chance of being reposted by the enemy, and that a weapon procs more times a minute.</p><p>If a weapon has a % chance to proc, then a faster weapon is better. If a weapon procs 1.7 times a minute or whatever, then a longer delay weapon means it procs FAR more often.</p></blockquote><p>This is ALL wrong...a slower weapon has less --CHANCE-- to be riposted simply becuase it hits less often.</p><p>Similarly, procs don't proc MORE, tbh they probably proc less with a slower delay.  You just have a higher CHANCE for it to proc with each swing.</p><p>Think about it--every time you hit something you have a chance to proc--sometimes procs will hit consecutively (if you add a trigger to ACT to listen for them you will know this).  If you hit something 10 times vs hitting something 5 times you could make that proc go off up to 10 times theoretically vs only 5.  Of course it doesnt always work that way, but you have more <em>chances</em> for that proc to hit.</p><p>And as far as the haste thing goes, as a Berserker I kind of preferred the 3.0 delay because you could go CA-auto-CA-auto and it was nearly flawless.  Now using a 4.0 delay, I have to wait a split second to use that next CA.  I'm not sure if it is worth it at this point, I just know that the 4.0 delay weapon I have is simply a better weapon than my E-Dragon's Temper.  TBH depending on my group setup I can get 2 CA's off when using the 4.0 but it will usually delay my auto by split-seconds...not sure how bad that hurts me but I'm sure it does.</p></blockquote><p>If you go to a higher delay weapon and can not get more CAs in without delaying auto-attacks, then you really haven't gained much unless the weapon is that much better.</p><p>With faster swings you are not getting more average procs. The fact that you have more chances to proc is counter-acted by the fact that you have a lower chance to proc on each swing. You might get more than average procs on a specific fight, but on average across all fights it will balance itself to the listed proc amount (+ haste increases), regardless of weapon delay.</p>

Aule
04-30-2010, 04:26 AM
<p><cite>Xerq@Crushbone wrote:</cite></p><blockquote><p><strong><em></em></strong></p><p><span style="color: #888888;"><span style="color: #ffffff;">I thought the offhand penalty was half... otherwise auto-attack timing would be difficult between 2 same-delay weapons. But you could be right... it just must be that the offhand auto-attack is delayed by CAs a lot because it doesn't fall on the same interval as the main-hand swing.</span></span></p><p><span style="color: #888888;"><span style="color: #ffffff;">DR can still be a meaningful number for statistical comparison if a few things are assumed to be equivalent among the weapons. Then it's just useful for calculating which effects are more beneficial. When actually comparing weapons in-game though, you are right. You need to look at spread vs delay. With 100% crit, a 100 DR weapon with a 1.5:1 spread is much worse than a 100 DR weapon with a 4.0:1 spread, as demonstrated above.</span></span></p></blockquote><p>It's not actually an off-hand penalty.  It's a penalty applied to the delay of both hands because you have a weapon in your off-hand.  As long as the delays were matched up to start with they stay matched up.</p>

TheSpin
04-30-2010, 07:09 AM
<p><cite>Aule@Guk wrote:</cite></p><blockquote><p><cite>Xerq@Crushbone wrote:</cite></p><blockquote><p><strong><em></em></strong></p><p><span style="color: #888888;"><span style="color: #ffffff;">I thought the offhand penalty was half... otherwise auto-attack timing would be difficult between 2 same-delay weapons. But you could be right... it just must be that the offhand auto-attack is delayed by CAs a lot because it doesn't fall on the same interval as the main-hand swing.</span></span></p><p><span style="color: #888888;"><span style="color: #ffffff;">DR can still be a meaningful number for statistical comparison if a few things are assumed to be equivalent among the weapons. Then it's just useful for calculating which effects are more beneficial. When actually comparing weapons in-game though, you are right. You need to look at spread vs delay. With 100% crit, a 100 DR weapon with a 1.5:1 spread is much worse than a 100 DR weapon with a 4.0:1 spread, as demonstrated above.</span></span></p></blockquote><p>It's not actually an off-hand penalty.  It's a penalty applied to the delay of both hands because you have a weapon in your off-hand.  As long as the delays were matched up to start with they stay matched up.</p></blockquote><p>Yuppers... if you dual wield both weapons suffer the same 30% penalty, not just the offhand.</p><p>This was changed way back when one handed and dual wield weapons were combined... like pre RoK</p>