Actual DPS is damage over time. You can increase DPS by increasing damage, or decreasing time, while controlling for the other. With Stormstrike, the time is static (9 or 10 s), so you must increase damage.
Agreed.
And as long as that 3s cooldown is part of the denominator of the WF DPS calculation (2xDPS*Weapon Speed / 3s + Avg WF Lag), slow weapons will always be better than fast ones with the same DPS, because there is no way to increase the proc rate to overcome for lost damage.
This is the part I'm not following. If we look at procs per 1000 swings, for any number between 1.5 and 2.9, you end up with the same proc rate. Therefore, if WF damage is proportional to weapon damage per swing, and weapon damage per swing and per speed are inversely proportional, they should be interchangeable.
Now, in fact, I don't think that's quite true -- I think that WF damage is slightly more than proportional to weapon damage. So, while a 1.5 and 3.0 second weapon have the same theoretical DPS, I think the WF procs are probably a bit more than twice as much with the 3.0.
This has been proven multiple times in game and in simulation. I can therefore say that if your model of windfury predicts greater or equal damage from faster weapons, the model is wrong.
That may be, but I'd like to understand why it's wrong.
For one thing, it's a bad idea to use 2.6s in your computations, as no shaman has an average wspd of 2.6s. Most of us have close to 80% flurry uptime, mongoose and a ton of passive haste. That 1.5s weapon? Yeah, it's going to swing a minimum of 2 times inside of the WF CD. So's a 1.9s weapon if you have any haste at all.
Agreed. That said: My current analysis (assuming a flat 20% proc rate out of cooldown) is that on a single weapon, WF should proc 1 in 5 if your speed is 3 or higher, 1 in 6 between 1.5 and 2.9, and 1 in 7 between 1.1 and 1.4. If this is wrong -- and it may well be -- I'd like to know why.
I certainly don't accept the "3s + avg proc time" analysis, because that ignores the parts of weapon swings which could have occurred within the 3s cooldown. Look at just "time until first possible proc". At 1.5s, or 3.0s, it's 3s exactly. At 1.6, it's 3.2. At 2.9, it's 5.8.
This leads to the question of what the other tradeoffs are. Should I never use a 1.6s weapon? What if I have a 2.8s weapon that's 10 levels old, and I just got offered a 1.6s blue which has higher damage per swing?
For another, it's also a bad idea to assume perfect hit and expertise ratings, and to ignore the affects of crits.
Shouldn't all of these have the same effect on everything, though? 15% of attacks do double damage = +15% damage over time. Doesn't matter what damage is.
You don't seem to be considering haste effects, such as flurry/lust/drums/rating. You also don't seem to be calculating WF cooldown as linked across hands.
There have been thousands of man hours put into modeling enh dps. Asking people to explain it all to you when you could peruse the theorycrafting perma-threads on the subject of windfury is wasteful of everyone's time and needlessly clouds the thread.
Hmm, i think the big thread was deleted so it is no longer checkable. Not to mention there was a reason why it was called megathread :P
Seebs, WF has 3.0 sec cooldown shared by both weapons in DW. That is due to oversight or complications at WF coding which makes it check both enchantments when weapon attacks. So you never have 20% proc rate with DW both WF, either it is 0 or 36% depending if less than 3 seconds has passed from last WF.
Concerning actual dps and speed compromise, run the simulator to see exact cut off for specific speed and dps.
You don't seem to be considering haste effects, such as flurry/lust/drums/rating. You also don't seem to be calculating WF cooldown as linked across hands.
I am indeed asking about 1H. Not all of us are full-enh DPS.
And I'm not ignoring haste effects -- I'm granting that they change things, and I've discussed how they change things.
There have been thousands of man hours put into modeling enh dps. Asking people to explain it all to you when you could peruse the theorycrafting perma-threads on the subject of windfury is wasteful of everyone's time and needlessly clouds the thread.
Let me try again.
I am looking at 1H DPS. WF appeals to me for a few reasons. I read the theorycrafting perma-thread on the subject, and it compared the damage from procs between two weapons, asserting that PPM for a 1.5 speed weapon would be at most 20, and that PPM for much slower weapon was "conservatively" 15.
But 15 is ludicrously high.
So, for a hypothetical 1H user... What is the actual impact of weapon speed? The theorycraft I see on this topic has, on several occasions, contained obvious errors. So far as I can tell, there is a repeated assumption that, since there's a 3s cooldown, the closer you are to 3s the better -- but that doesn't seem to make any sense if you don't have two weapons which are potentially out of sync.
So, let me start from the top.
I am fighting 1H+shield, for sinister reasons of my own. I don't have two weapons, I don't have stormstrike.
WF proc damage appears to be roughly proportional to damage per swing. WF proc rate appears to be very close to 1/6 if you have a weapon which swings once during the cooldown, and 1/7 if you have a weapon which swings twice during cooldown. (This is intuitively obvious, if you just assume that you get 1/5 of swings out of cooldown, and then add the swings during cooldown to the set of misses.)
What this suggests is:
1. If you are fighting 1H, there is very little difference between a weapon speed of 1.6 and 2.9 in total WF proc damage. Probably some difference, but not very much.
2. The cutoffs seem to be:
1.4s: You will get about 1/7 procs
1.5s-2.1s: You will get about 1/6 procs, but it'll drop to 1/7 during flurry.
2.2s-2.9s: You will get about 1/6 procs even with flurry.
Of course, other haste effects move these.
Now, since each proc adds substantially over 2x the damage of a basic swing, these would turn to ballpark figures of:
1.4: 7/7 white damage + 2.5/7 yellow
1.5-2.1: somewhere between that and
2.2-2.9: 6/6 white damage + 2.5/6 yellow
3.0+: 5/5 white damage + 2.5/5 yellow
So WF is adding about 36% to base damage on a 1.4 speed weapon, and 41% to base damage between 2.2 and 2.9, and 50% at 3 and higher.
("2.5" is pretty approximate; I'm guesstimating 1.25x normal swing damage for WF hits, that could be wrong.)
But, for instance, I don't show any obvious difference (if the only haste you have is flurry, which is true for us lowbies) between 2.2 and 2.9. If the WF proc damage isn't normalized, there's some.
But, if these numbers are even close, then the difference, while not completely trivial, is smaller than the difference I'm likely to see between a lower-level and higher-level weapon. So, for instance, if I get a chance to upgrade from a 2.9s weapon with 20dps to a 1.4s weapon with 30dps, I should do it, even though everyone says "shamans should use slow weapons".
No one you paraphrase saying "use slow weapons" is talking about a lvl 48 using a 1H+shield. Don't worry about weapon speed: you are correct that as long as your hasted weapon speed is >1.5 seconds and < 3 seconds, you will notice no meaningful difference in windfury damage output.
Next time you want to discuss leveling in Azeroth, I would make that very explicit. This thread is populated mainly by those asking about max level raiding and so answerers will naturally assume that is your focus as well.
No one you paraphrase saying "use slow weapons" is talking about a lvl 48 using a 1H+shield. Don't worry about weapon speed: you are correct that as long as your hasted weapon speed is >1.5 seconds and < 3 seconds, you will notice no meaningful difference in windfury damage output.
Next time you want to discuss leveling in Azeroth, I would make that very explicit. This thread is populated mainly by those asking about max level raiding and so answerers will naturally assume that is your focus as well.
I see. Thanks, that clarifies a lot. In raiding, there's really not much reason for a shaman to be meleeing and NOT have two weapons, so far as I can tell. (At which point, all the WF proc damage stuff is probably dwarfed by stormstrike, meaning slow weapons would win even if WF proc damage slightly favored faster weapons.)
Mostly, though, at this point I'm just curious about the numbers in that theorycraft thread. Even with dual-wield and dual-WF, 15ppm seems pretty high to me for slowish weapons. Assuming a pair of 2.9s... You have about 21 swings per, so 42 swings total; add in six stormstrikes for another 12 swings. Each WF proc gives you at least one swing on each weapon which cannot proc, so we're back to the 1/6 proc rate, or a little lower (in case stormstrike comes up during the WF cooldown), so 54/6 = 9, and that's on the high end; it would probably be a tad lower. Add in haste and flurry and so on, and it gets better, but I still don't think 15ppm is likely.
The thing is, with the numbers presented, if the slow weapon were proccing 11 times instead of 15, it wouldn't be much of an advantage. It's still better overall (because of the higher damage from stormstrikes, and because higher damage/fewer swings penetrates block better)... But it's no longer nearly as overwhelmingly better as suggested.
So where did the 15ppm come from, and why is it tagged "conservative" when it's about 35% better than casual number-crunching suggests?
Is there an easy way to get LaTeX to work in posts?
The model I posted above assumes that the mob falls over and dies after you attack it n times, no matter how many WF procs and crits you get. My interest is much more academic than practical, although it might be of more useful in a raid setting where the Shaman's damage is only a fraction of the damage a mob is taking. In that case a few more procs or crits is not going to effect the time it takes to kill the boss too much.
Assuming that the effective attack speed is one of {.9, 1.4, 2.6}, then in 5 min a shaman will attack {333.333, 214.286, 115.385} times. The missed chances per proc would be {3,2,1} respectively. So using the density:
Where n is the number of attacks, m is the missed chances per proc and k is the number of procs. C is some constant to normalize, it depends on n and m but not k.
So for a 2.6 weapon:
expected value:30.3166
SD:3.24633
1.4 weapon:
expected value:44.6349
SD:3.11655
.9 weapon:
Expected value:57.4749
SD:2.92496
So it looks like my model is wrong/missing something, it predicts that a faster weapon will get more procs in the same amount of time. The increased number of attacks more than make up for the lost chances for WF to proc in this model. The only thing that could save it is if a WF proc for a slow weapon was a lot more valuable than that of a fast weapon, which I think is true, but not quite to the extend needed.
Oh, certainly a faster weapon will get some more procs.
If WF damage scales precisely with base damage, then as long as you're in the same bracket, the extra damage and extra procs should precisely balance out. When you cross a bracket (say, going from 1.5 to 1.4), you should lose some noticable amount of efficiency. When you're in a range where flurry crosses the bracket, you'll lose some efficiency but not much.
The main reason I'm curious about this is that I've seen a lot of forum posters (not so much here, but on the WoW forums) who have turned "slower weapons better" into a kind of pure dogma, where a 2.9 with no benefits is clearly superior to a 2.4 with the same DPS, but +20 each strength, int, stamina, and attack power. I'm mostly convinced that slower is better, but I'd like a better grasp on how MUCH better, so I can make more reasoned choices. Is .1 speed equivalent to 5% DPS? 1%?
Okay -- this is difficult to predict with an equation, because shaman dps has so many feedback loops. My own testing, both in game and simulation, has never resulted in a graph that was stable enough to post -- though it always scaled upward as weapon speeds decreased, and in all cases both WF and total DPS has gone up.
But above 2.4s, testing in simulation seems to predict an increase of about .5-.7% of TOTAL DPS simply by switching to a MH weapon with the same DPS but a .1s slower speed. The OH effect is actually pretty close to that as well. (Below 2.4 things are pretty chaotic...I've seen increases of 2%+)
I underscore total DPS, because you need to understand -- this doesn't mean that a 100 dps 2.6s weapon is the same as a 100.5 dps 2.5s weapon. The slower weapon is actually MUCH MUCH better than that. E.g. if you do 1000 dps with a 2.5s weapon in your mainhand, you'd do at least 1005 dps with a slower weapon in your MH. This means with WDPS = 9 AP and .25 TDPS = 1 AP, you'd need a 102.3 dps 2.5s weapon to match it.
Long and short of it is, slow weapons ARE better. The better your stats are, the more true this becomes. This has been proven in game, and in simulation. If your model does not predict it: fix your model, it is wrong.
PS: One thing I haven't seen discussed is Stormstrike. Stormstrike is actually the source of a large number of Windfury procs -- at least 20%, perhaps as much as a third. It's higher than you think, because for most Shamans SS can't miss. Stormstrike actually selects FOR slow weapons, because, again, you can't speed SS up but WF procs with slow weapons are bigger.
Long and short of it is, slow weapons ARE better. The better your stats are, the more true this becomes. This has been proven in game, and in simulation. If your model does not predict it: fix your model, it is wrong.
Well, once again, I'm modelling a single 1H, and no SS, and all the calculations I've seen have been for two weapons, which do seem to be different.
The thing is, I'm fine with revising my model. What I'd like to know, though, is what is wrong with my model. Are WF procs hitting for more-than-proportional damage? Are faster weapons proccing less often than I predict? Are slower weapons proccing more often?
PS: One thing I haven't seen discussed is Stormstrike. Stormstrike is actually the source of a large number of Windfury procs -- at least 20%, perhaps as much as a third. It's higher than you think, because for most Shamans SS can't miss. Stormstrike actually selects FOR slow weapons, because, again, you can't speed SS up but WF procs with slow weapons are bigger.
Even without WF, SS selects strongly for slow weapons.
If your weapon speed were 2.5, SS would give you 25% extra damage; if it were 2.0, only 20%. (This is overgeneralized, and ignores different hit chances, but that just makes it even more noticable.)
My current suspicion is that the analysis for single weapons will be noticably different from the analysis for dual weapons. And yes, I know that dual weapons are better for full enhancement, and this is the kind of thing which probably won't affect raiders. However, it does seem like it'd matter some to people who aren't raiding at a given moment -- say, who want to go solo stuff for a while, or who are levelling.
Look, I have no interest in helping you with your single weapon testing. It's ridiculous and has nothing to do with Enhancement shamans, who should at all levels be either using a 2H or dual wielding. This is an Enhancement shaman thread, for Enhancement specced shamans who dual wield, have UR, have Flurry, have Enhanced Weapon Imbue talents, aren't being parried, can Stormstrike, etc.
You can probably remove some number of those variables and come up with a model that predicts no difference between slow and fast weapons above 1.5s. But that prediction has as much bearing on anything we care about here as a post in the shadow priest forum.
Originally Posted by seebs
Are faster weapons proccing less often than I predict? Are slower weapons proccing more often?
Yes, and yes. If you add misses and haste into the mix, fast weapons will have a much lower number of actual hits. Granted, slower weapons will to...but when you add SS into the mix, you have 2 (okay, in your case 1) extra chances per 10 sthat can't be missed and occur regardless of weapon speed. In the real world for real enhancement shamans, this helps keep slow weapons great.
Let's consider a 1.6s and a 2.6s weapon (don't use a 1.5, because that swings twice in a single CD). And what the hey, we'll consider your 1 hand and no misses. In 60s, the 1.6 will have 37.5 + 6 chances to proc WF. The 2.6 has 23 + 6 chances. Ignoring the cooldown, that's 50% more chances; however, the 2.6s weapon is doing 62% more damage.
A 10% miss chance brings this to 48% more chances from a fast weapon vs. the same 62% more damage.
Long and short is, that there are fewer chances to proc than you predict for both speeds of weapons due to misses and that the number of actual procs is much closer than you predict from weapon to weapon due to 20%+ of these procs coming from SS.
UPDATE: So as you stop bothering us, I just tested on Yo's simulator: without Flurry, Dual Wield, Stormstrike, Mongoose or haste procs, all weapon speeds between 1.6s and 2.6s do roughly the same DPS. 1.5s weapons do slightly less. Congratulations, your model is correct; for non-enhancement shamans weapon speed doesn't matter. But if you add in any of the above criteria, slow weapons immediately start to offer increased DPS. Add them all, and it's a big, big deal.
Last edited by Toots Hepcat : 05/27/08 at 5:28 PM.
Look, I have no interest in helping you with your single weapon testing. It's ridiculous and has nothing to do with Enhancement shamans, who should at all levels be either using a 2H or dual wielding. This is an Enhancement shaman thread, for Enhancement specced shamans who dual wield, have UR, have Flurry, have Enhanced Weapon Imbue talents, aren't being parried, can Stormstrike, etc.
Well, hang on. 2H = single weapon.
So it's relevant to anyone who, for some reason, has no dual-wield. Admittedly, not raiding.
But you know what? It's fine for you to say "we don't care, it's irrelevant". Maybe this belongs in a different thread.
But what doesn't make sense to me is, if all of your math is for dual-weapon, insisting that you've got a general proof for a case you're saying you don't care about and haven't looked at.
You can probably remove some number of those variables and come up with a model that predicts no difference between slow and fast weapons above 1.5s. But that prediction has as much bearing on anything we care about here as a post in the shadow priest forum.
Okay. Well, if you'd like to suggest a more appropriate thread for asking about windfury mechanics with a single weapon, I'd be happy to discuss it there.
Let's consider a 1.6s and a 2.6s weapon (don't use a 1.5, because that swings twice in a single CD).
Does it? I would have assumed that 1.5s would swing once in cooldown and once immediately after cooldown.
And what the hey, we'll consider your 1 hand and no misses. In 60s, the 1.6 will have 37.5 + 6 chances to proc WF. The 2.6 has 23 + 6 chances. Ignoring the cooldown, that's 50% more chances; however, the 2.6s weapon is doing 62% more damage.
Right -- but that's still with stormstrike, and there's not much reason to have stormstrike without dual weapons.
So I'm comparing 37.5 to 23, and that's 63% more chances, against 62% more damage.
And that's my whole point.
A 10% miss chance brings this to 48% more chances from a fast weapon vs. the same 62% more damage.
This is assuming that SS never misses, yes?
UPDATE: So as you stop bothering us, I just tested on Yo's simulator: without Flurry, Dual Wield, Stormstrike, Mongoose or haste procs, all weapon speeds between 1.6s and 2.6s do roughly the same DPS. 1.5s weapons do slightly less. Congratulations, your model is correct; for non-enhancement shamans weapon speed doesn't matter. But if you add in any of the above criteria, slow weapons immediately start to offer increased DPS. Add them all, and it's a big, big deal.
Yeah. My prediction would be that flurry matters (maxed out) between speeds of 1.6 and 2.2, and that stormstrike hugely favors slow weapons.
Now, maybe none of that matters to people whose only use of a shaman ever will be endgame DPS raiding as pure enhancement, or at least mostly enhancement. But this discrepancy matters a lot to pretty much every shaman who is level 39 or lower, or who has any kind of mix of talents, or who isn't enhancement at all. Many shamans will end up in melee; even resto shamans have to hit something sometimes when soloing.
I understand that there's a particular focus on pure endgame raiding in theorycraft, but I think it might not hurt to occasionally consider the tiny little chunks of time some of us might spend outside the raid. Sometimes that matters too, and I'd think that theorycraft ought to be a more general discipline.
This thread is for Enhancement Shamans as the title says, what you are describing is... well dunno what. Any non enhancement shaman wouldnt really care about the specifics of your scenario so I dont see the value of analysing it. Any non enh shaman would have a shield and weapon oriented to their spec and their combat style would be in that aspect as well.
Elementals would def. go for spells and so would Resto after the healing/dmg ratio was added. This specific case would be in a soloing environment where they are out of mana and well in that case theyll just go for what they have and I would be very surprised if they had an extra weapon in their bag to swap in for those cases.
Going around chasing the universal shaman theorycrafting theory is just a waste of time, there are so many variables and in the end all shamans fall into 3 categories and within those categories there are minimal differences so lets just focus on what the majority of the community is focused around.
If you want to theorycraft around random talent people at level 40 or under then plz do so but create a thread and just dont expect anyone to really care.
PS: SS is def. of use if you are using 2H weapons as an extra means to maybe proc WF and thats what anyone using a 2H is waiting for, the big WF proc that oneshots cloth.
Dunno if Yo still reads this but someone else might remember if this has been covered before (yes I have used the search function).
I was wondering if Yo's Simulator timed the on-use procs to trigger at roughly the same time or not, ie. triggering berserker's call, haste pots, blood fury, drums etc at the same time for optimal synergy? Assuming that these are triggered as soon as CD is up and ie from the start they should but then it all comes down to priority order (many has a CD of 2min) and proccing blood fury/berserkers call before drums is non optimal.
Anyway I was wondering if these issues where taken into consideration in Yo's simulator or not.
If not it would be nice if it could be added, for example by trying to get as many overlapping procs as possible, ie. Drums and Haste pot comes out at the same time, drums have longer uptime then haste so it is procced first etc.
This thread is for Enhancement Shamans as the title says, what you are describing is... well dunno what. Any non enhancement shaman wouldnt really care about the specifics of your scenario so I dont see the value of analysing it. Any non enh shaman would have a shield and weapon oriented to their spec and their combat style would be in that aspect as well.
What is "oriented to their spec" and "style in that aspect as well"? It's a bunch of vague, meaningless, verbiage which isn't even wrong, just free of meaning. What orients a weapon to resto? Fastness? Slowness? Everyone just says "always wield the slowest weapon you can get and use WF", but so far as I can tell, that's not true without DW...
What I care about, mostly, is that the theorycrafting page I was pointed to makes claims about WF proc rates which are, so far as I can tell, totally wrong -- as in off by a factor of two or so. The conclusion (slow weapons better for DW) happens to be true... But I'd think it'd matter to get the argumentation correct.
Anyway, if you think another thread would be better, that's cool. I went here because when I asked about windfury proc rates, I was directed here.
Searching the thread brought up some nice reads on DS in the offhand, but nothing conclusive. Any word from Yo if we will see DS offhand in the simulator soon? I'd like to play around with other MH weapons in the sim with DS in the offhand and see what I get, and I'm sure I'm not the only one.
Seebs -- you're pretty indignant for a guy who has little understanding of the theory he's disputing.
I think it's obvious what "orients" a weapon to a resto shaman: it has + Healing on it. + Healing also gives you spell damage; it doesn't give you AP, doesn't give you crit, etc. There are also several resto talents that grant bonuses to lightning bolts, while only one of them (NG) grants a bonus for melee -- and it's one few resto shaman take.
I mean, it just happens that when I go resto, I have a set of great enhancement gear. Yet when single wielding my Dragonstrike with 30% crit chance, 1550 AP, 15 Expertise and a ton of armor penetration, I only do about 300 dps. If I stay in my resto gear and spam lightning bolt, I'm doing about 350 dps -- and if I put on my elemental gear (which has more crit), I can top 400.
If you intend to be a resto or elemental you should be shooting lightning. This is the combat style that matches that spec's strengths, both in terms of gearing and talents. Only enhancement shaman will be best served by entering melee range, and at no level does using a 1H weapon allow for the best DPS.
Look: you kinda, sorta found something: if you don't have flurry, dual wield, stormstrike or any haste procs, it doesn't matter what your weapon speed is as long as it's greater than 1.6. Our point is: nobody should ever be doing that. Those are the key talents of the enhancement tree.
Those numbers you cite and question appear to be theoretical MAXIMUMS, not real world numbers; 20 being the max procs possible with a 3s CD and 15 being the max possible with a flurried 2.6s weapon with no other haste buffs (1 proc every 4s). So yeah, they are lot higher that a computed average for an unhasted weapon.
Last edited by Toots Hepcat : 05/28/08 at 4:45 PM.
Seebs -- you're pretty indignant for a guy who has little understanding of the theory he's disputing.
Well, I seem to have the central point right -- if you don't have SS/DW, the speed difference evaporates (or becomes small enough not to be easily measured). So apparently my theory is spot on with regards to the actual question I was asking, which was "does speed matter if you don't have SS/DW".
I think it's obvious what "orients" a weapon to a resto shaman: it has + Healing on it. + Healing also gives you spell damage; it doesn't give you AP, doesn't give you crit, etc. There are also several resto talents that grant bonuses to lightning bolts, while only one of them (NG) grants a bonus for melee -- and it's one few resto shaman take.
Yes. So, given a choice between two otherwise equally good restoish weapons, which one should I use if I sometimes need to hit things? If this is not the right thread to discuss questions of how talents work when you aren't using them in the single most precisely calibrated raiding environment, what is?
If you intend to be a resto or elemental you should be shooting lightning. This is the combat style that matches that spec's strengths, both in terms of gearing and talents. Only enhancement shaman will be best served by entering melee range, and at no level does using a 1H weapon allow for the best DPS.
Well, of course not. But, if for some reason you are entering melee with a 1H... does it matter what speed it is? Apparently some, but not nearly as much as it does with stormstrike.
Look: you kinda, sorta found something: if you don't have flurry, dual wield, stormstrike or any haste procs, it doesn't matter what your weapon speed is as long as it's greater than 1.6. Our point is: nobody should ever be doing that. Those are the key talents of the enhancement tree.
Sure.
But I don't spend all my time in raids, or even always in groups, and I found that enhancement isn't totally worthless without 35 points of talents; in fact, even 20ish makes a huge difference in solo viability. I really like it.
Those numbers you cite and question appear to be theoretical MAXIMUMS, not real world numbers; 20 being the max procs possible with a 3s CD and 15 being the max possible with a flurried 2.6s weapon with no other haste buffs (1 proc every 4s). So yeah, they are lot higher that a computed average for an unhasted weapon.
And the key issue is:
According to that page, if you are fighting 1H, you should go with a slower weapon, because the MAX procs for 1.5s is still 20, but a "conservative" estimate of real procs for 2.6s will be 15, and 15 heavy procs beats 20 light procs.
But, in the real world, where you do NOT get theoretical maximums, it makes no difference at all, and there are even some cases where the faster weapon will win.
I guess... Yeah, this isn't the case that makes the most different to raiding. But it seems to me like the theorycraft ought to cover, not just the theoretical best specs for raiding 70s, but questions like "how will I get to 70", or "what should I do after my guild wipes on a drama llama and I have to mostly solo and pug for a couple of weeks"? And it seems to me that, even though the error turns out not to affect the raiding case, where your goal is obviously pure DPS with no worries about other contributions, it is likely to adversely effect lots of other readers who get directed there and told incorrectly that, no matter their spec or anything else, slower weapons add a huge amount to DPS.
I'm sorry, 'cuz reading back I realize I've come across pretty badly. I got a lot of snide remarks on the WoW forums from people who were clearly incapable of even the most basic statistical work, so I sorta came in hostile here. I apologize. You guys are doing good theorycraft, and you are absolutely right that, in endgame raiding, where specialization is the name of the game, there is simply no reason ever not to dual-wield if you're buying enhancement points, because the points it would take to get partway into enhancement would be better spent elsewhere.
But down here at 48, I switched from dual-wield to 1H+shield, and went and cleared a bunch of quests which had been pretty rough going dual-wielding, because I DO need the damage mitigation from the shield, and the faster heals, to handle some of these groups.
I'm trying to do multiple runs of the sim with slightly different gear.
I visited a website about a month ago where i could enter the gear I wanted to test, and it would provide str/agi hit attack power etc.. much like the wow screen. Poor description I know, but for the life of me I can't find the site.
Anyone know of any other sites where you can enter gear and get base stats to enter into yo's?
I am not sure if the EJ Class Mechanics forum is the correct place to discuss such matters. You are asking of things that do not feel relevant at all and frankly smell of trolling. Go ahead, equip any weapon that has high dps and level yourself up. It's not important what speed it is in such scenario.
There's lots of reasons we don't theorycraft for leveling:
1) You don't need to...anybody can level with no metagame knowledge needed.
2) Leveling doesn't take that long and most of that time is spent running. Thus, the effect of optimizing a spec on leveling time is going to be hard to measure.
3) There just isn't enough data. I leveled pre-BC, and have no interest in going back to diaper school to learn the effects of the 3s cooldown on my rattle.
4) EJ is a RAIDING guild, these are RAIDING forums, we are all people who RAID and want to be best in a RAID situation.
5) Most of the time, raiding theory works pretty well for farming and grinding, too. See, at your level, dual wield and storm strike would up your damage by at least 50%. Instead, you've chosen a shield, which is decreasing your damage intake by about 30% at best and does nothing against magic. So...fights last 50% longer, but you take 30% less damage from each blow. You're still taking more damage overall! In essence, you've decreased your group efficacy and made your leveling time longer. Some hybrid! And you're worried about weapon speeds?
I'm trying to do multiple runs of the sim with slightly different gear.
I visited a website about a month ago where i could enter the gear I wanted to test, and it would provide str/agi hit attack power etc.. much like the wow screen. Poor description I know, but for the life of me I can't find the site.
Anyone know of any other sites where you can enter gear and get base stats to enter into yo's?
warcrafter.net is also a great site for choosing equipment and will give you all the specific stats you need to input into the simulator.
There's lots of reasons we don't theorycraft for leveling:
1) You don't need to...anybody can level with no metagame knowledge needed.
2) Leveling doesn't take that long and most of that time is spent running. Thus, the effect of optimizing a spec on leveling time is going to be hard to measure.
3) There just isn't enough data. I leveled pre-BC, and have no interest in going back to diaper school to learn the effects of the 3s cooldown on my rattle.
Fair enough.
4) EJ is a RAIDING guild, these are RAIDING forums, we are all people who RAID and want to be best in a RAID situation.
I see the issue. I had been given the mistaken impression that the class mechanics material was for general discussion of theorycraft, not specialized discussion of exactly one boundary case.
5) Most of the time, raiding theory works pretty well for farming and grinding, too. See, at your level, dual wield and storm strike would up your damage by at least 50%. Instead, you've chosen a shield, which is decreasing your damage intake by about 30% at best and does nothing against magic. So...fights last 50% longer, but you take 30% less damage from each blow. You're still taking more damage overall! In essence, you've decreased your group efficacy and made your leveling time longer. Some hybrid! And you're worried about weapon speeds?
No, I'm curious about whether a generalization someone made is actually generally true. And it's not. It turns out, in fact, that I am *not* worried about weapon speed.
With DW+SS, a bad pull could easily kill me long before I could do anything about it. Now, I pretty much don't die. Fine by me.
In closing: quit typing nonsense and go level.
And yet, for all that you call it nonsense, after two pages of saying obviously weapon speed mattered for 1H just as much as for 2H, the one actual sim done confirmed my speculation: That between 1.6 and 2.9, it doesn't.
I guess I'll stop bugging you, though, sorry to waste your time with discussions of parts of the game other than the one you play. If anyone has any suggestions for a forum with a broader interest in theorycraft, I'd love to hear of one.
05/27/08, 1:04 PM
Similar Threads
