Unread postby Densor
Fri Dec 30, 2016 8:10 am
Couple things I've been looking in to:
1) As Barrage becomes a frequent spell in the standard ST DPS rotation, I was wondering if it was a DPS increase to case an Arcane Explosion at 0 charges, rather than an AB. Both do complete shit for damage with 0 charges, but AE gets done 50% faster. However, AE also has a reduced chance to proc AM's.
First, I needed to take proc munching into account. Using Barrage, there is a possibility of munching AM procs with your normal rotation, since you cast 5 spells (Barrage, 4xAB for example) before you get back to 4 charges and optimal time to cast AM. By casting 5 spells, you have a chance to proc AM up to 5 times, but only 3 would stick around. As a result, the average number of AM procs you finish those 5 spells with is less than the sum of their proc chances. For the 4xAB rotation, at 18% AM proc rate, you lose about 4% of an AM proc by taking munching into account, or you'll munch one AM every 25 rotations, on average. At 26% proc rate, you lose 15% of an AM, which is 1 AM munch per 6.66 rotations. Using the AE, 3xAB rotation, at 18% proc rate, you lose 2.5% of an AM, and at 26% proc rate, you lose 10% of an AM.
Now with more accurate proc rates, I looked at the number of AM's you lose going from the 4xAB rotation to the AE, 3xAB rotation. With an 18% chance to proc AM, using AE instead of AB at 0 charges results in losing about 1 AM proc every 10 rotations. At 26% AM proc chance, you lose 1 AM proc every 6.66 rotations.
So, finally with those numbers, I was able to compare the lost DPS from fewer AM's to the DPS gained by a faster 0 charge spell. The end result was that using AE was a slight DPS increase; at the numbers I ran, it increased DPS by less than 0.5%. However, I didn't take TotM into account, but I don't know how it would affect the numbers.
On the one hand, if TotM procs on the 0 charge AB, it will definitely increase the DPS of the 4xAB rotation to higher than the AE, 3xAB rotation. On the other hand, it has seemed to me that TotM can't refresh itself, so if it procs on the 0 charge AB, it can't proc again until it expires. As a result, by proccing on the 0 charge AB, you are preventing it from proccing on the 1, 2, and 3 charge AB's, which have a significantly higher return on the proc.
2) If you take Overpowered, I wondered if it was a good idea to sync MoA with AP. I'm assuming you always cast MoA with RoP, so I'm going to ignore it for this math. Using OP, if you cast MoA on cooldown, you get a 1.7x MoA when it lines up with AP, and 2 more MoA casts, for a total of 3.7 effective MoA's. If you delay MoA to AP, you get 2 MoA casts at 1.7x power, for a total of 3.4 effective MoA's. The question now is whether 30% of an MoA is worth the cast time it costs.
So, the cast time of MoA is 2 seconds, and the damage done is 720% of SP plus 15% of mana. 30% of that is 216% of SP plus 4.5% of mana, or 108% of SP and 2.25% of mana per second. Our lowest DPS rotation (4xAB, Barrage) does 272% of SP per second on average. Thus, if 2.25% of mana is greater than 164% of SP, we should use MoA on cooldown, otherwise we should use it only during AP. At 1.5 million mana (24% mastery and AF), 2.25% would be 33750. Thus, for that to work, our SP would need to be less than 21k.
Conclusion: it may be a slight DPS increase to delay MoA for AP if you take Overpowered. On the other hand, my analysis doesn't address the difference in mana usage. If you are trying to do a Barrage-less rotation, you may be planning to cast MoA as a non-mana using spell to help make your mana last to the next Evocation. There's also an issue of the first tick of MoA not being affected by AP (and RoP), but don't think that would change the results. I'm also using the average rotation DPS, which may not yield accurate results.
3) But what about without Overpowered? If that is the case, during AP, our MoA is worth 1.3x, plus 2 more if casting on cooldown for 3.3x MoA's. By using with AP, we cast two MoA's at 1.3x power for 2.6x MoA's. As a result, we lose 70% of an MoA.
70% of an MoA translates to 504% of SP and 10.5% of mana over 2 seconds, or 252% of AP and 5.25% of mana per second. Without taking into account the talent we chose other than OP, the rotation does 272% of SP per second. To make this work, 20% of SP would need to be greater than 5.25% of our mana, which is probably not going to happen. Even taking the other talents into account and the other factors I am skipping, it is probably impossible to come out on top in this situation.
Conclusion: without OP, cast MoA on cooldown as expected.
