WotLK castsequences for shaman/druid (theorycrafting/long)

[Fuzzyboy]

won't have one piece of haste gear or the totem / druid imp moonkin talents?

Ok, so I read up on haste and did some thinking. What we want to do is this:

1) Get flame shock off as soon as possible to burst it when its about to expire
2) Cast CL as soon as it cools down
3) Cast Lava burst when flame shock is about to expire

Point 3 is debatable - you might wanna cast it when it fits your rotation better, since shock doesn't tick for that much damage, but that would depend on whether you have a focus on mana save (long fights) or high dps (shorter fights).

According to blue posts, threat generation will be high in relative comparison to now, meaning that if your dps and tank is equally well geared your dps can pretty much go nuts and won't grab aggro (less focus on aggro management - again according to blue posts). So we have two scenarios - scenario A (dpm efficiency) and scenario B (dps efficiency).

Assuming 10% haste, that would be 5% from wrath of air and 5% from other sources (1% spell haste = 15.7 haste rating - 10% is equivalent to 157 spellhaste):

- Lightning Bolt (LB) will have a cast time of 2,0 / 1,1 = 1,818181818, lets assume 1.85s to cover for lag etc
- Lava burst will have a cast time of 1,5 / 1,1 = 1,363636364, lets assume 1.4s to cover for lag
- CL will have a cast time of 1,5 / 1,1 = 1,363636364, lets assume 1.4s to cover for lag
- Wrath will have a cast time of 1,5 / 1,1 = 1,363636364, lets assume 1.4s to cover for lag
- GCD will be at 1.4 on 10% haste also.

Assuming that EM is applied at 1.4, we have our two scenarios:

Scenario A:

0.0 Flame shock
0.0 GCD start
1.4 GCD end
1.4 CL start cast
2.8 CL end cast
2.8 LB start cast
3.0 FS tick
4.65 LB end cast
4.65 LB start cast
6.0 FS tick
6.3 CL CD
6.5 CL start cast
7.9 CL end cast
7.9 LB star cast
9.0 FS tick
9.75 LB end cast

Because CL hasn't cooled down yet, we can't cast it here. If we cast LB here, we will finish at 11.6 which is just

0.2 after CL CD which is ok.

9.75 LB start cast
11.4 CL CD
11.6 LB end cast
11.6 CL start cast
12.0 FS tick
13.0 CL end cast
13.0 Lava burst start cast
With only 2 secs until FS expiration, this could be a good time to chuck in lava burst
14.85 Lava burst end cast (consumes flame shock)

<restart>

14.85 Flame shock
14.85 GCD start
15.0 Flame shock would expire if not bursted
16.25 GCD end
16.5 CL CD

this looks fine and would allow us to restart our cycle after, giving a cast-sequence of:

/castsequence Flame shock, chain lightning, lightning bolt, lightning bolt, chain lightning, lightning bolt, lightning bolt, chain lightning, lava burst

or in other words exactly the same sequence as before, only it takes less time to complete (14.85 secs instead of 15.5), but keep in mind that we end up waiting 0.25 secs for CL to cool down at the end and "waste" 0.2 earlier in the sequence. Because we want flame shock to tick, we can't burst it without losing dpm efficiency, but that is accounted for in the DPS scenario:

Scenario B:

0.0 Flame shock
0.0 GCD start
1.4 GCD end
1.4 Lava burst start cast
2.8 Lava burst end cast
2.8 CL start cast
4.2 CL end cast

Since there is still 1.8 sec to shock cooldown and 3.5 sec to CL cooldown we can choose between waiting or casting a

LB (which would only delay the flame shock cast by 0.05 since we have a 1.85 cast time on LB.

4.2 LB start cast
6.0 FS CD
6.05 LB end cast

<restart> Since we can see from the continuing pattern below that we won't run into big trouble with CL CD (only a 0.05 delay)

6.05 Flame shock
6.05 GCD start
6.45 GCD end
6.45 Lava burst start cast
7.85 Lava burst end cast
7.9 CL CD


The DPS sequence will be:

/castsequence flame shock, lava burst, chain lightning, lightning bolt

or in other words exactly the same sequence as before, only it takes less time to complete (6.05 secs as opposed to 6.5) so the 10% spellhaste will increase dps but not alter the castsequence. Interestingly, increasing spell haste beyond 10% will be wasted on this sequence once the haste pushes the LB after the first CL below the FS CD. A way to fix this would be casting the LB before CL (FS, Lava, LB, FS and continue into a longer rotation) but that would seem to require a lot more haste and regardless be less efficient on multiple targets.