Diese Seite macht ausgiebigen Gebrauch von JavaScript.
Bitte aktiviert JavaScript in Eurem Browser.
Live
PTR
10.2.7
PTR
10.2.6
Beta
Lvl 80 Ele shammy
Antwort verfassen
Zur Forenübersicht zurückkehren
Beitrag von
Alitruae
I was always told it was Fs>Lvb>Lbx4-5>lvb repeat for boss and single target dps. For pack pulls it was fs>lvb>CL>Lbx2>CL>Lvb repeat
Beitrag von
lucius
And that's fine. Even on single target CL will do slightly more damage, at the cost of much mana. It's much simpler to have new shaman not use CL on single mobs because you don't need to give them more things to think about (ie, asking them to judge if they'll run out of mana before this 7 min fight finishes).
Beitrag von
Haxzor
And that's fine. Even on single target CL will do slightly more damage, at the cost of much mana. It's much simpler to have new shaman not use CL on single mobs because you don't need to give them more things to think about (ie, asking them to judge if they'll run out of mana before this 7 min fight finishes).
ffs, stop screwing with the newbs.
STICK TO THE
fs>LvB>LBx4(or 5)>LvB>LBx3(or4)>repeat
Beitrag von
lonephoenix
I'm depressed at the amount of fail going on here.
So lets do some real math?
Let's first calculate the Damage per second of both Chain Lightning and Lightning Bolt the at the magic 1900 spellpower number. Pulling the formula from WoWWiki:
Damage = (Base Cost + (Spell power Coefficient) * Spell Power) * x
(Where x is any number of spell modifications.)
For all these calculations, We'll be ignoring trinket procs, As they tend to vary highly in how they factor into any given spell cast.
All spell damage is calculated using 1900 Spell Damage, and enough haste to generate a 1.7 Second Lightning bolt, and a 1.3 second Chain Lightning. (This numbers will change over the course of this post.)
There are around four "time frames" we want to consider when determine the best spell rotation:
1. Per second.
2. Per 8 seconds. (Lava burst Cool-down focused)
As I write this post, I'm doing the mathematics, thus I have no true bearing on the best rotation that this moment.
So lets consider the first case:
1. Per second (cast).
This number is attained through simply the casting of a single spell alone.Typically, we use this to determine the "value" of any given spell. Thus, we can calculae the value of any spell for it's time casted, in order to detemine what spell we want to use to fill our rotations. Also, with this value, we can determine the "value" of any given rotation.
Lightning Bolt: using the damage formula listed above, we see a single spell can create the following average damage:
(1)Average base damage per cast: (715 + 815)/ 2 = 765 Average damage per cast.
(2)Average damage (Untalented): (765 + (1900*0.794) = 2273.6 Damage per cast.
(3)Average damage (Talented, No Procs, No Criticals, Glyphed): (765 + (1900* (0.794 + 0.10 ) ) * 1.05 * 1.04 = 1519.1 Damage per Time Spent Casting
(8) Average "Value" of a Lightning bolt Critical (Talented): (4) / 1.7 = 3797.6 Damage per Time Spent Casting.
Chain Lightning: using the damage formula listed above, we will repeat the same calculations with Chain Lightning:
(1a.) Average base damage per cast: ( 973 + 1111) / 2 = 1041.5 Average damage per cast.
(2a.) Average Damage (No Bounces) (Untalented.): (1041.5 + (1900*0.641) = 2259.4 Damage per cast.
(3a.) Average damage (Talented, No Procs, No Criticals.): (2a.)* 1.05 = 2372.3 Damage per cast.
(4a.) Average damage (Talented, Critical): (3a) * 2.50 = 5930.9 Damage per critical.
(5a.) Average Lightning Overload damage (Talented.): (3a.) * 0.5 = 1186.2 Damage per overload.
(6a.) Average Double Lightning Overload damage (Talented.): (5a.) * 0.50 = 593.1 Damage per Double Overload.
(7a.) Average "Value" of a Chain Lightning (Talented.): (3a.) / 1.3 = 1824.9 Damage per Time Spent Casting
(8a.) Average "Value" of a Chain Lightning Critical (Talented.): (4a.) / 1.3 = 4562.2 Damage per Time Spent Casting.
Therefore, we can conclude from this first examination, that for a simple cast "value" of a Chain Lightning is greater than a Lightning Bolt, even at 1900 spellpower. But we notice that Lightning Overload Procs are generating higher damage than the Chain Lightning Overloads, and the Critical strikes from Lightning Bolt leap ahead of the Chain Lightning Criticals by nearly 600 damage.
Thus, the next examination occurs over a span of 8 seconds. This is more technically a span of 7 seconds, as we're assuming the application of a single Flame Shock into each cycle. Ideally, this is time best spent casting a different spell if the debuff is already active on the current target.
Ideally, we're looking to maximize the average damage created over 8 seconds. Thus, we're going to examine multiple rotations, factoring no critical strikes, then factoring in critical strikes.
Assuming 1.7 second Lightning bolts, we can fit 4 into an 8 second cycle, for 6.8 seconds, or 5 casts in 8.5 seconds. Considering the clipping which would occur at 8.5 seconds, which slows our next Lava Burst, it would be ideal to fit a Chain Lightning into this rotation at the end, for a 8.1 clip. However, clipping is clipping, Thus this ideally NOT the best rotation, as it exceeded our time frame.
Another rotation to consider is Chain Lightning on Cooldown, filling with Lightning bolts.
This rotation would be: CL -> LB -> LB -> CL -> LB for a total time of 7.7 seconds casting. If we account for latency, we might find this rotation is too "tight" to effectively execute, which would generate clipping of the Lava Burst cool-down during the final Lightning Bolt. This can be considered a valid sacrifice for an 8 second rotation, but will likely faulter during a 45, or even 10 minute rotation.
Thus, these will be the two cases we will consider:
(A)LB -> LB -> LB -> LB
(B)CL -> LB -> LB -> CL
Using (7) and (7a)
(11) Average "value" of Rotation (A): (7)*4 = 6076.4 Damage per time casting.
(12) Average "value of Rotation (B): (7)*2 + (7a.)*2 = 9726.2 Damage per time casting.
Now, this is assuming no critical strikes land. It appears simple that Chain Lightning is overcoming the value of Lightning bolt, However, if we begin to factor critical strike chance into the value of the rotations, we should notice changes.
Assuming a base critical strike chance of 19%, and all procs, as well as talents considering, a Shaman should witness a critical strike chance of approximately the following:
(13) CSC (Critical Strike Chance) = 19% + 5% + 5% + 3% = 32% chance to critically strike with Lightning Bolt and Chain Lightning. This is entirely self-buffed, with no external asistance. A more "reasonable" result may be 45%, accounting for full Raid buffs and consumables.
So now, we will weight the value of each cast by it's Critical strike chance, to determine the net "value" in each rotation.
(13) Lightning Bolt's Weighted value: ( (.78*1519.1)+(.32*3797.6) ) = 2400.1
(14) Chain Lightning's Weighted value: ( (.78*1824.9)+(.32*4562.2) )= 2883.3
Rotation (A) CSC Weighted Value: 2400.1*4 = 9600.4
Rotation (B) CSC Weighted Value: 2400.1*2 + 2883.3*2 = 15367
Now, assuming 45% CSC, and recalculating (13) and (14)
(13a.) LB's New Weighted Value: ( (.55*1519.1)+(.45*3797.6) ) = 2544.4
(14a.) CL's New Weighted Value: ( (.55*1824.9)+(.45*4562.2) ) = 3056.7
Now factoring in Lightning Overload Procs by weighting each spell with the potential for an overload, and the overloaded value. and the double overload value, and the fully buffed values:
(15) Final LB Value: 2544.4 +(.33*(.55*1291.1 +.45*3227.8)) + (.11*(.55*645.6+.45*1614)) = 3617.6
(16) Final CL Value: 3056.7 + (.33*(.55*1186.2 +.45*2965.5)) + (.11*(.55*593.1 + .45*1482.8)) = 3821.7
Therefore, we see a final rotation value of:
(17.)Final Rotation (A) value: 3617.6*4 = 14470.4 Damage per time spent casting.
(18.)Final Rotation (B) value: 3821.7*2 + 3617.6*2 = 14878.6 Damage per time spent casting.
Therefore, it is safe to assume that at 1900 spellpower, Chain Lightning should be weaved into rotations on cooldown, if your mana pool can support the strain,
(Cont.)
Beitrag von
lonephoenix
(Cont. from Above.)
as the difference in mana spent is nearly triple fully talented (Assuming full Clearcasting uptime. and NO Convection):
(19.)Mana per Rotation (A): (424*.4)*4 = 678.4 Mana per rotation
(20.)Mana per Rotation (B): (424.4)*2 + 1142*.4)*2 = 1252.8 Mana per rotation.
Which means you can perform 1.8 Rotation (A)'s for each Rotation (B).
Assuming 400 MP5 while casting, and 20k Mana, and requiring 75 rotations for a 10 minute boss fight we see:
Rotation (B) needs 93960 Mana to support it's execution, and Rotation (A) only requires 50880 mana to execute. Rotation (B) will be nearly out of mana by Rotation 40, Which is nearly 5.5 minutes. Rotation (A) can function far longer.
Therefore, at 1900 Spellpower, Chain Lightning is still required in Rotation to Maximize damage done over time.
Will we see it drift toward LB spam as the SP Coefficient begins to overtake the flat damage additions? Yes. I didn't calculate the exact figure.
Beitrag von
66520
Oh, I'm very much aware, but I was just hoping you'd proving the exact point of the statement. And you did, thanks. (lol U CANT START SENTANCES WITH AND)
It is not any more incorrect to start a sentence with a conjunction than end one with a preposition or use double-negative for emphasis.
They are part of a group of grammatical rules invented purely to make English grammar more like Latin, despite the fact that it's a completely different type of language and therefore shares very few of the same grammatical rules.
It is taught as incorrect in schools, and journalism studiously avoids them simply because they're still considered unprofessional, but they haven't been considered incorrect by Grammarians for years.
When someone says 'above 1900 spellpower you should remove CL from your rotation,' and I respond saying "you'll do more damage if you use CL," you take that to mean damage per cast. I'm sorry if that's not entirely obvious to you that you are wrong.
You're the very first person in this thread to mention the 1900 figure.
Beitrag von
Haxzor
If you're raiding with 1900sp fully buffed, I feel sorry for you
Beitrag von
142776
This post was from a user who has deleted their account.
Beitrag von
lonephoenix
If you're raiding with 1900sp fully buffed, I feel sorry for you
When people are looking to tell someone how they should play out a particular rotation, it's nearly impossible to account for everything, but when we're gearing up, it's possible to be under the theoretical "caps" since things like Naxx are so trivial now.
I just wanted to point out at 1.9k SP isn't the magic number, it's more than likely 2k Spell power. The LB rotation I list is only down by 400 Damage per time spent casting, and it's also significantly more simple to execute (Which is reason enough for me to do it, not to mention that the mana saved means I can be useful for emergency healing more often, which happens far more often than I'd like to say.)
Beitrag von
lucius
Oh, I'm very much aware, but I was just hoping you'd proving the exact point of the statement. And you did, thanks. (lol U CANT START SENTANCES WITH AND)
It is not any more incorrect to start a sentence with a conjunction than end one with a preposition or use double-negative for emphasis.
Sorry, I was actually referring to my sentence just before the parentheses. Taking a cheap shot at myself, not at you.
When someone says 'above 1900 spellpower you should remove CL from your rotation,' and I respond saying "you'll do more damage if you use CL," you take that to mean damage per cast. I'm sorry if that's not entirely obvious to you that you are wrong.
You're the very first person in this thread to mention the 1900 figure.
This is true, unless you read the post 2 spots above my first. But the number honestly pops up in every thread because 'someone heard someone say it'.
-------
There are two things about the CL v LB and 1900SP arguement that I find hilarious.
1. Ignores mana use and actually suggests it's better to use the vastly more mana consuming rotation for shaman just gearing up. These are the same shaman that will run into the MOST mana issues.
2. It's just wrong about the scaling.
Will we see it drift toward LB spam as the SP Coefficient begins to overtake the flat damage additions? Yes. I didn't calculate the exact figure.
They do scale very similarly through spell damage when you look at their cast times.
Gripes: 1.7 / 1.3 cast time choice favors LB. Just as choosing 1.75 LB 1.3 CL cast time would favor CL. Not huge, but haste will keep the CL / LB cast time ratio constant at 0.75. While it's only off by 2% and that may not seem like a lot, we do bother to include the glyph which is 'only' 4%.
(4) and (4a), these spells crit at 200%, the multiplier should not be 2.50x.
Random thought: Lightning overload proc rate. It is currently unknown how this works, as far as I've been able to find. More specifically, how does it work on CL?
Does CL: A) have the same 33% to proc on each jump, B) have 11% chance to proc on each jump, or C) have a total 33% chance to proc on each spellcast, split up between each target it hits.
Beitrag von
lonephoenix
Will we see it drift toward LB spam as the SP Coefficient begins to overtake the flat damage additions? Yes. I didn't calculate the exact figure.
They do scale very similarly through spell damage when you look at their cast times.
Gripes: 1.7 / 1.3 cast time choice favors LB. Just as choosing 1.75 LB 1.3 CL cast time would favor CL. Not huge, but haste will keep the CL / LB cast time ratio constant at 0.75. While it's only off by 2% and that may not seem like a lot, we do bother to include the glyph which is 'only' 4%.
(4) and (4a), these spells crit at 200%, the multiplier should not be 2.50x.
Random thought: Lightning overload proc rate. It is currently unknown how this works, as far as I've been able to find. More specifically, how does it work on CL?
Does CL: A) have the same 33% to proc on each jump, B) have 11% chance to proc on each jump, or C) have a total 33% chance to proc on each spellcast, split up between each target it hits.
My assumption, based on everything I've read up to this point, Lightning Overload functions upon every spell cast, including another overload. I'm not assuming that CL will proc off another bounce, because off wording of the talent alone. Not that empirically testing isn't worth doing, I'm just not interested, nor do I find the effects dramatic in their altering of the damage, as each overload suffers a massive diminishing return for damage, no matter how high it scales.
I calculated the critical strike damage bonus based on Elemental Fury adding an additional 100% critical strike damage. I can rework this to be 200% without much issue. It simply throttles back every critical by 50%, which merely makes the LB and CL "Break point" more difficult to reach.
I agree completely, and I mentioned that the mana issue becomes incredibly relevant for a gearing shaman, but even if we remove 3-4k from the mana pool that which I calculate my number of rotations, you'll find that so long as the shaman is gearing properly, 250 MP5 is a reasonable expectation, for simply craftables and quest items, and would still allow for the ability to perform in nearly every heroic boss, and perhaps most raid bosses, depending on the composition of the raid. That's mostly a result of Clearcast procs, which I'm demanding full uptime. More accurately, we would likely see a LB not being effected by a Clearcast Proc, opposed to the fact that the promised Lava burst crit will promise that two spells (One of which is a CL) will be at 60% their mana cost,
However, at significantly higher Spell power and haste, this all becomes moot as the Spell power coefficient will enable LB to run away, and adding an additional LB will almost always be more important than filling the slot with a CL, even if we assume small cooldown clipping. (Less than .3 seconds, would make for a single additional Lava burst every 4 rotations or so, which will be fighting against 4 Lightning bolts, or 4 CLs + 1 Lava burst, which would need to be modeled to determine it's value.
I just wanted to show people the math, Because I read 1900 in several places in the past also, and I didn't appreciate the misinformation.
Beitrag von
Mesoforte
This is true, unless you read the post 2 spots above my first. But the number honestly pops up in every thread because 'someone heard someone say it'.
Reason for removing the gylph of lava, is that its not very good if you have less then 1900 sp unbuffed, or the T9 4pc set bonus
Its talking about glyph of lava burst.
Sorry guys, but this thread really needs to die already.
Beitrag von
lucius
Doesn't matter, it's someone repeating the figure of 1900 SP (incorrectly) making something better than something else, because he 'read it somewhere'. It keeps popping up in threads, and people keep repeating it because they see it in threads.
------
Checking more of the math, I see errors in the base coefficient, step (2) and (2a). LB's base is 0.7143 (2.5/3.5), and CL's base is 0.5714 (2/3.5). Not sure where you got your figures. Also, the 'base crit bonus' is 50%. Yes it's worded oddly, but the crit bonus multiplier is increased by 100% (of 50%) to a total of 100%. So the new crit bonus with the talents is 100%. It is hard to tell with their wording when they're talking about a multiplicative or additive increase.
Beitrag von
binkenstein
There's no such thing as a "double overload" either (overloads cannot proc additional overload casts).
I'm depressed at the amount of fail going on here.
So lets do some real math?
I sense a large amount of irony in this statement, because it's pretty much how I feel every time I see something as monumentally bad as that post.
Adding CL into a typical rotation will result in ~20 dps gain at most, due to the extra damage LB now generates, and the fact that CL overloads are an 11% chance per hit (which gives a 33% chance per cast assuming three targets are hit).
As it stands, any Spellpower figure that is the "break even" point for changing spells or doing something different will be very fluid based on buffs/bonuses.
Beitrag von
Mesoforte
Its both awkward and strangely gratifying seeing Binkenstein on the Shaman Forums.
Seeing him regularly will definitely add some spice. :S
Beitrag von
66520
You're the very first person in this thread to mention the 1900 figure.
This is true, unless you read the post 2 spots above my first. But the number honestly pops up in every thread because 'someone heard someone say it'.
That post is this:
Reason for removing the gylph of lava, is that its not very good if you have less then 1900 sp unbuffed, or the T9 4pc set bonus
And refers not only to an entirely different spell, but is talking about glyphs
not
rotations.
Though, LvB glyph, is also one of those mythical numbers that pops up in every thread. People like 1900sp a lot.
Beitrag von
132826
This post was from a user who has deleted their account.
Antwort verfassen
Ihr seid nicht angemeldet. Bitte
meldet Euch an
, um eine Antwort einzusenden, oder
registriert Euch
falls Ihr noch nicht über ein Konto verfügt.