Quote Originally Posted by Argyle_Darkheart View Post
Your calculations are wrong.

You're assuming you start each rotation from a UI3 F3, but that isn't the case. Every time you Transpose, you're re-entering Astral Fire with an Umbral Ice 2 Fire 3 (which is a 3.5s cast), but you're never calculating for this.

That is, each rotation should be starting from Astral Fire 3 Fire 1, and ending with the appropriate Umbral Ice 2/3 Fire 3.

For example, you have the Priority 4 rotation calculated as a slight DPS increase, but it's actually a DPS loss. The correct rotation would be:

[Astral Fire 3] F1 (2.5s) > F1 (2.5s) > F1 (2.5s) > F1 (2.5s) [Proc] > F1 (2.5s) > Hardcast Flare (4.0s) > Firestarter (2.5s) > Transpose > B1 (2.5s) > T2 (3.0s) > Umbral Ice 2 Fire 3 (3.5s) - 3087 potency / 28s = 110.25 potency/s

Versus:

[Astral Fire3] F1 (2.5s) > F1 (2.5s) > F1 (2.5s) > F1 (2.5s) [Proc] > F1 (2.5s) > Firestarter (2.5s) > Astral Fire 3 Blizzard 3 (2.5s) > T2 (3.0s) > Umbral Ice 3 Fire 3 (2.5s) - 2593 potency / 23s = 112.74 potency/s

All that said, I don't know why you parsed a DPS increase; I certainly haven't experienced any DPS increase (usually the opposite) while testing your rotations.
I did make a math mistake on the fire 3 casting time. I'll adjust math later. To be totally honest, I'm not sure why my experience is so radically different from anyone else testing the rotation. I'm currently checking to see if I actually am just under performing using the standard rotation. If that is not the case I will then review all of my math and revise it. I plan to try and simulate the effect of waiting for mana ticks in future dps comparisons.

The only reasons I can think of for why my personal experience using this rotation differs so widely from the rest of the community is that I wrote my guide wrong and it is misleading people in some way or I am doing something wrong myself when using other more conventional rotations. I certainly don't have a small sample size for comparison though, I've been using this rotation since 2.2 and found it preferable again and again.

Thanks for the feedback.