Steady Focus Duration Change from 10s to 20s on PTR
Concerning the change to SF proposed for 5.1, I will not complain about having the buff duration extended since it does provide MMs a little more flexibility in shot choices since we are not required to do as many SS pairs as previously; however, to be honest, I really do not see this change as much of a benefit, at least not currently.
Here are my thoughts on this change:
1) First, on average to balance focus, MMs still need to do at least 2 SSs per CS cycle. In the best case scenario during the Standard phase of an AS focus-dump cycle during a RF with doing only 2 SSs at 12.93% haste from gear without the T14 4P and 9.09% haste from gear with the T14 4P (which very few hunters if any can achieve currently), here is the focus cost in different scenarios:
| Scenario | Focus Cost w/o T14 4P | Focus Cost w/ T14 4P |
| CS-SSx2-ASx6 | -38 | -40 |
| CS-SSx2-ASx5-GT | -33 | -35 |
| CS-SSx2-ASx5-MMM AI | -18 | -20 |
| CS-SSx2-ASx5-DB* | -8 | -10 |
| CS-SSx2-ASx4-MMM AI-GT | -13 | -15 |
| CS-SSx2-ASx4-GT-DB | -3 | -5 |
| CS-SSx2-ASx4-MMM AI-DB* | +12 | +10 |
| CS-SSx2-ASx3-MMM AI-GT-DB* | +17 | +15 |
*Note that at this haste level that DB should perform 10 attacks. Hence, it provides 50 focus over 15s. You can expect for about 6 of the DB attacks at most on a CS cycle if the DB was cast during the cycle. This is a benefit of 30 focus.
So even in this best case scenario (with not assuming that other haste effects are stacked on RF), most of the cases are still focus negative and require 2 SSs. The rare cases that do generate a positive amount of focus do not generate enough focus where an SS could be replaced with another AS, which requires 23.32 (amount of focus gained over a SS cast) + 20 (AS focus cost) - 43.32 focus. The best case of 17 focus gained when a MMM AI proc occurs and both GT and DB are available is less than 40% of the focus needed to do that substitition.
Furthermore, the focus gained on these cycles are needed to help balance the focus loss of other cycles, which it does not fully do requiring an extra SS about every third cycle.
Even with considering the case with additional haste on top of RF, 2 SSs are still required. Since the SS casts are already at the 1s GCD with RF, all BL does on top of RF is increase focus regen by 1.9 FPS. Thus, over a 9s CS cycle, another 17.1 focus is generated making the best case generate 34.1 focus, which is still too small to allow an AS for SS replacement since it is only 79% of the cost. Furthermore, this situation is moot anyway, since if you are stacking haste buffs on top of RF, you will want to be using AI as the focus dump instead of AS anyway, which requires a lot more than 2 SSs per CS cycle ( 4 to 5 actually) to counteract the larger AI focus cost.
So from the average CS cycle focus balancing point of view, the change has little impact for most of the fight. This applies to the CA phase too in which we use AI and need extra SSs to balance focus.
2) The is one on average situation though where these extended duration can possibly help currently - during the KS phase. Below is an updated version of the table with including cycles with 2 KSs (since you would normally shoot a pair). Note that at beast 2 KSs every 11s that some CS cycles would have no KSs and revert to those in 1).
| Scenario | Focus Cost w/o T14 4P | Focus Cost w/ T14 4P |
| CS-SSx2-ASx4-KSx2 | +2 | 0 |
| CS-SSx2-ASx3-GT-KSx2 | +7 | +5 |
| CS-SSx2-ASx3-MMM AI-KSx2 | +22 | +20 |
| CS-SSx2-ASx3-DB*-KSx2 | +32 | +30 |
| CS-SSx2-ASx2-MMM AI-GT-KSx2 | +27 | +25 |
| CS-SSx2-ASx2-GT-DB-KSx2 | +37 | +35 |
| CS-SSx2-ASx2-MMM AI-DB*-KSx2 | +52 | +50 |
| CS-SSx2-AS-MMM AI-GT-DB*-KSx2 | +57 | +55 |
As can be seen, every one of these CS cycles is at least focus neutral, with most being way focus positive. In fact, the last 2 cycles are sufficiently focus positive where an SS can be replaced by another AS. Thus, the extension of the SF buff could possible allow for one to perform only 1 SS on some CS cycles during the KS phase when under the effects of RF and make up the focus loss on the cycles with the 2 SSs to maintain SF. In fact, it is even possible to perform no SSs on 1 of the cycles during a RF pair and make up for the focus loss on other cycles.
There are a few rubs to this though:
- You do not always have RF available during the KS phase.
- If RF is available during the KS phase, it only covers about 1.5 to 3 CS cycle depending on whether you have a single or RF pair available.
- As gear scales, AI should become increasingly better as a focus dump choice during RF (as opposed to the option choice with the T14 4P), making this situation moot.
3) The next consideration is whether this flexibility has any real benefits to shot order in the CS cycle.
For an AS focus dump cycle, where ASs are instant and only cost 20 focus each, it is not difficult at all to cast an SS pair and avoid focus capping in most situations. An SS pair gains about 46.64 focus over the cast for most cycles. Even with starting at 100 focus, casting a CS and an AS clears enough focus for an SS pair without overcapping in most situations. During a RF, an SS pair could cause a little overcapping if starting at 100 focus, but its not a concern if you are not near 100 focus and can be avoided by casting CS and 2 ASs before the SS pair.
The thing that complicates the situation and where the flexibility helps is the impacts to the rotation due to MMM AI procs and when abilties like DB, GT, and Readiness (which allows an extra DB and GT) come off CD. In the rare situation where many of these are available during a CS cycle, its nice to have the flexibility to shift where the SS pair occurs in the CS cycle or to even skip the SS pair one cycle every so often if trying to burn down a low health target (like the energy cores on Elegon).
For an AI focus dump cycle, the situation is a little different since AI costs 50 focus. With factoring in the regens over the cast time, an SS pair regens a little more than the focus cost of an AI. However, during a RF, which is when we usually cast AIs, the extra focus from RR results in AIs cost being less than 8 focus more than the focus regened from a single SS. If a single DB is active too, then an SS cast actually regens more focus than the cost of the AI cast where casting a single SS for each AI is sustainable during the DB.
4) The next consideration is how the increased flexibility can potentially help us reduce or prevent focus overcapping in real game play and let us use that focus instead to cast more ASs instead of SSs. There is definitely a possible DPS gain here; however, it is not much if at all. First, if doing things properly, you should ideally be wasting little to no focus. But then again, no one is perfect.
But with considering that an AS for SS substitution requires 43.32 focus, it requres a significant amount of overcapping having to previously been done to see any DPS improvements. With assuming that the increased flexibility does allow enough focus overcapping prevention to power an extra AS, the damage difference gained using the Max MM T14H DPS case is about 27K. Over a 5 min fight, this is only an 89 DPS improvement, which is only about a 0.08% DPS improvement.
5) The final consideration is how this change affects your ability to maintain the SF buff. For most players in most situations, maintaing the SF buff should not have been a problem. I know it usually was not for me since once I found the "rhythm" of the MM CS cycles, it was not hard to maintain usually.
However, if you are not paying enough attention, moving a lot, stunned/CCed at the wrong time, or trying to burn a target, it is easy to have the buff fall off, resulting in a DPS loss.
This change makes it a little easier to maintain the SF buff in game play situations.
However, it is a two-edged sword. This flexibility also reduces the "consistency" and rhythm in the MM CS cycles (since you are not required to perform a SS pair every CS cycle) where it is possible to actually end up losing the buff more often now if one is not careful.
6) Small possiblity for more MMM AI procs. The extended SF duration allows us to more often insert a MMM AI proc between SSs instead of waiting until after the SS pair. This results in additional SSs that can apply to a new MMM AI proc instead of being cast when MMM AI is already fully stacked.
Similarly, abilities coming off of short CDs, like KS, GT, and DB, no longer have to wait for an SS pair to complete every cycle. If there is still plenty of time on SF, we can insert one of these shots between SSs if they have come off CD. In rare cases, this lack of delay could possibly result in an extra one of these shots, but it is doubtful.
Summary:
The extension of SF duration from 10s to 20s is a quality of life change and not an overall DPS change. Over the course of the fight, we still have to cast the same number of SSs to provide the focus gains to power our focus costing abilities. The main benefit of the extended duration is a little more flexibility in our shot selection. We can better adjust our shot order to accomodate boss strategies, such as having a short burn stage with limited SSs. Another quality of life benefit is that it makes maintaining the SF buff in real game play easier to do. The only potential DPS benefit of the change is the shifting of shots may prevent a little focus overcapping for potentially a very minor DPS improvement.
FD Analysis:
Since the FD "Enable Public Test Realm Changes" Setting incorporates the extended duration of the SF buff, we are able to simulate the effect to the MM rotation. I did a before and after simulation using the MM Max DPS case for the T14H gear.
FD actually simulated the extended SF buff case as a 171 DPS loss. This DPS loss is due to shot shifting in the case.
Looking more closely at the shot numbers, here are the differences:
- CS -1
- AS +3
- SS -3
- Auto -3
- WQ -1
The loss of the CS is due to shot shifting due to focus levels at the time of the CS cast requiring an SS to get enough focus. This resulted in more focus later for more ASs instead of SSs. If the CS was not lost due to shot shifting, 2 of the 3 AS for SS replacements would not have occurred, leaving only the one AS for SS substitution.
Looking at the extended SF case, of the 45 CSs, 27 (60%) of the CSs had an SS pair to refresh SF before it and 18 (40%) of the CSs did not. The situations in which the SS pair was skipped include:
- Inserting an instant that has come off CD.
- Performng the MMM AI proc immediately after proc on the first SS instead of waiting until after the pair.
- Better focus balancing
- Use of Readiness resulted in a lot of instant cast abilities to be available.
- A RF supporting a lot of ASs
These results basically verify my analysis that the change is a quality of life improvement but not really a DPS improvement.