I did a quick optimization with putting FF back into the BM rotations now that the bug is fixed.
It appears though that it is best to use FF when you have nothing else to do but CoS to regain focus. What I mean by this is that everything else is on CD (including ET) and you do not meet the minimum focus requirement desired to cast AS (around 63). Since I am not really familar with BM, I am not 100% sure concerning about how that affects its number of uses since FD does not indicate when its available for use relative to when it is used as far as I can see.
Looking at the times between FF, I see a decent amount of variability between 34s and 49s, which indicates to me that the cast is indeed being delayed in many cases. Here is the breakdown with some additional information:
| Time | Time Since Last | Focus | Note |
| 42.3 | - | 26.3 | A few seconds after RF pair and then BL has completed |
| 80.3 | 38 | 29.0 | |
| 123.1 | 42.8 | 54.9 | |
| 165.7 | 42.6 | 32.3 | |
| 209.3 | 43.6 | 21.9 | Occurred during RF |
| 245.0 | 35.7 | 51.9 | |
| 288.6 | 43.6 | 54.6 | |
| 337.9 | 48.3 | 57.6 | Had Readiness use with double DB for focus and a RF |
| 377.4 | 39.5 | 21.2 | |
| 411.5 | 34.1 | 14.5 | |
From a binomial probability, I would expect for between 11 and 12 pet basic attacks to fully stack Frenzy. That would mean on average that fully stacked Frenzies should occur roughly every 36 to 40s, assuming a basic attack speed of 3.3. Although in the sim, the pet did attack in most cases at around 3.3 to 3.4s, there were a few cases of almost 5s between basic attacks, which will skewthings a bit.
I also do not know how Frenzy stacks are modeled in FD - whether each basic attack actually has a probability to gain a stack or whether some average number of attacks is used. I assume the former, which also results in more variability in the number of attacks between FFs.
But I would assume that the FF cast is being delayed a little in some of the situations in which it occurred. I am wondering whether it would be hard to update FD to instead just indicating whether Frenzy is up to also indicate the number of stacks by appending the number or to at least indicate when Frenzy is fully stacked so we can evaluate what the delays in FF casts are and whether it is actually beneficial to delay it in certain situations.
While it may be good to delay casting FF is some situations, it may not be good in other situations. It is obviously good to delay casting FF when you are under procs that increase your stats that are inherited by your pet. It is also good to delay casting FF when your pet has plenty of focus and is attacking at its maximum rate with a lot of WH uptime, this includes cases when the hunter is hasted and/or is high on focus and is casting more shots that can result in more GftT procs.
09/18/12, 1:01 AM
Similar Threads
