Monk Windwalker Rotation

I’m here again to talk about Windwalker Rotation Table, just a few things I noticed.

The Opening sequence should look like this :
Fist of the White Tiger > Tiger Palm > Touch of Death > Storm Earth and Fire > Trinket > Fist of Fury > Blackout Kick > Whirling Dragon Punch > …

  1. Fist of the White Tiger should be a top priority
  2. Chi Burst should be cast one sec prepull (with travel time it will hit the boss on 0), otherwise use it when everything else is on CD, or if you are not in melee.
  3. You should try to maximize Touch of Death usage (in single target situation) :
  • Try to have 5 Chi before to cast it (most of the time you should be able to reproduce Opener sequence every 2 minutes)
  • You need to hold Fist of Fury and Whirling Dragon Punch for ToD window.
  • Always hold one Storm, Earth, and Fire to sync it with ToD. Here SEF is just used on cooldown.
  • Second potion should also be use during ToD

I will update this thread if I see something else :slight_smile:

I’ve tested most of these things before because I play WW a lot, but I’ll definitely post up the tests for other people to see.

Using a lot of logic around saving up for Touch of Death never really ends up being a DPS increase overall. It seems like it should, but really it has a very small effect. Just using abilities whenever they are ready ends up being optimal in the simulator. And as you are learning, I have a strong preference for factoring rotations down to the simplest set of rules possible :wink:

The reason for this is because we do our simulations with a variance on the fight length. Sometimes you’ll squeeze in an extra use of an ability, sometimes not. If we wanted this logic to result in a DPS increase every time, we’d have to try to include some math to determine if the delayed time on ToD and/or SEF, FoF, WDP results in one less use or not based on the exact fight length (which isn’t really realistic). That’s about the gain you could get from doing these things, so even missing out on one use (or partial use) of an ability nullifies the gains.

In practice, I find that the timing of fight phases more often dictates whether or not I hold some abilities a few seconds to get a bigger burst at an important moment. It doesn’t matter much if my overall DPS goes down a bit.

I understand that, and I would like to know what is your % threshold to determine if a change is a real gain. Because most of the time any change (except if it’s huge) will not go over 0.3% I guess. So relatively speaking, may be 0.1% is also a valuable boost ? But we also have to think about the margin of error … how can we be certain that those changes are not valuable ?

Of course on a fight length of 6mn everything will be smooth out, but what about a dungeon boss ?

That’s why I like to do different kind of sim :

  1. 1mn length, to crystallize the burst.
  2. 6mn fight without any big DPS cooldown, to understand the ‘sustain’ DPS. I guess it is a bit what you are referring to ‘factoring rotation’.

For a lot of players, the ‘burst sequence’ (or open) is the most important part. During a fight, I have to play mechanics and focus on other stuffs than DPS, then I like to find a ‘window’ without too much mechanics to focus on my damage.

Now I really would like to see your test, specially on a 1mn length fight for instance.

Edit: I just noticed that we can’t go under 100sec with AMR simulator, do you think is something that could be added ?

You could make your own boss script and allow a minimum time of 60 seconds if you’d like. The Training Dummy script allows fights down to 30 seconds if you want to use that one.

If you want to change one of the other scripts, go to the wiki: Not Available - you can expand the Bosses and pick the boss type you want to use. Click on that boss and then use the COPY button in the top center of the page to make your own copy of that script. Edit it and then change the minimum fight length.

As for what I consider a “real gain” in rotation development - I have a few rules of thumb that I go by. First, I see if what I am testing always results in a DPS gain. If it does, then it doesn’t matter if it is small, I’d consider it worth doing unless it is really complicated. If it is difficult/complicated, then I check how much of a gain it is. If the gain is very small - less than 1% - then I might not consider it worthwhile. Anything smaller than 1% is really questionable considering that simulation models are not perfect models of the game.

Another thing to keep in mind: Lets say you test 5 different rotation tweaks individually and find that each one gives a potential 0.75% DPS gain. If you then put all 5 of those things into one rotation and test that - you rarely would see a 3.75% DPS increase with all 5. The only way that would be true is if those rotation tweaks all had zero interaction with each other, which is almost never the case. This is why we use the method we do for testing rotation tweaks, and then we usually include an “overall” test as well to see what the sum total potential DPS increase is compared to just doing a bare bones rotation.

In the case of Windwalker and Touch of Death - during that 8 second window you essentially get 10% increased damage on any ability you use. If, for example, you hold FoF for a few gcds to let it happen during ToD - and then end up using FoF one less time in the fight - you lose way more than 10% of the damage of a FoF. Even missing one tick of a FoF would not be worth it. If you have to use a BoK while you are holding FoF, you lose out on the CDR from that BoK - so the opportunity cost of holding abilities as a WW is generally too high to make it worth it in the long run. You’d be looking to instead delay ToD for FoF, instead of the other way around. I’ll test out some logic for trying to sync a SEF with ToD as part of the tests I do.

Thanks for this long explanation, all this is new to me, and it helps a lot !
I did myself some tests, doing a custom rotation for WW, and I can clearly see what you’re talking about, it seems to be a really small gain overall.
I’m pretty disappointed to see those kind of min/max doesn’t change too much our DPS. (in every spec apparently …)
But in the other hand, I have a lot of fun playing around with your simulator !

+~0.7%

It could be nice to have an option to compare two reports (side by side or showing up/down with color on each spell damage). Or may be an option to export result into xml/csv/plaintext whatever.

If you do a batch of simulations with the Test Combinations simulation type, there is an option on the output to export the results to a csv file so that you can apply custom analysis to it.