Windwalker (DPS): a flurry of tender fisting. - Page 34

Crevan · 12/11/12, 11:51 AM

I've played a little with possible heroic BiS tier piece setups in SimCraft to get a very rough idea of what we should be aiming for. I've used the default action list for a Patch-style boss over 25k iterations and my own char as a baseline (a human with alchemy/engineering). The items I've kept constant in all sims were:

neck - from Blade Lord
cloak - from Garalon
wrists - from Blade Lord
weapons - double Shek'Zeer fists
belt - from Lei Shi/Tsulong
boots - from Zor'lok
rings - from elite Protectors and Wind Lord
trinkets - from Elegon and Lei Shi

Each gear set was reforged/gemmed/enchanted to be close(ish) to hit/exp caps, minimum mastery and about equal amounts of crit/haste. I didn't do any in-depth optimisation, since I was only looking for rough estimates here. The numbers are:

2p_shoulders_head*	118943
2p_chest_head		118698
1p_head			118633
1p_shoulders		118619
2p_head_legs		118559
1p_legs			118528
2p_chest_shoulders	118493
4p+_off_legs		118492
2p_gloves_head		118478
2p_gloves_shoulders	118393
1p_gloves		118370
2p_shoulders_legs	118360
1p_chest		118291
2p_gloves_chest		118244
5p			118210
no tier			118114
2p_gloves_legs		118107
4p+_off_shoulders	117961
4p+off_gloves		117947
2p_chest_legs		117922
4p+off_head		117846
4p+off_chest		117634

The error on these numbers was around 40-45 dps points. The delta between lowest and highest in the list is 1309 dps, about 1.1% of the top result. My main conclusion from this list is a further confirmation of how bad our set bonuses are this tier. We could use any combination of tier and non-tier armour and still be close enough to the theoretical optimum for it not to matter. What is best for you personally will depend on your play style (how often, if at all, you want to use FoF; how comfortable are you with your energy management) and on your guild's progress (i.e. what gear you have access to).

*Keep in mind that the actual theoretical highest dps set might not be tier head and shoulders with 3 offpieces; finding out what is will require more in-depth testing and optimisation.

Xiaoryn · 12/11/12, 12:13 PM

I wonder if our time isn't better spent figuring out relative values of haste and crit, and at what point (hastecap or otherwise) we should be stacking the other.

I mean, It seems to me that there are still several unanswered core questions about this spec:
-Assuming Ascension, how much haste rating or energy/sec should we be aiming for? How much w/o Ascension?
-How does the above change if we're using FoF on near cooldown (and this is a fight-specific question)?
-How does 2pc and 4pc factor in, exactly?
-How do we account for Mastery procs in our haste cap math? Is there any way to make it relate-able via numbers?

I think that once we get these figured out, we can really start to generate 'gold-standard' BiS lists, rotations, and gemming.

I wish I was better at creating sims or pure numbercrunching, but I've tried many different things on my character from week to week...but that data isn't reflective of 'best case scenario' because I am a human who raids with other humans.
So far, though, I've found that my numbers are generally better when I cap haste around 5.5k, then roll crit. I'm gemming all agility or agi hybrids, and using FoF at every viable opportunity.

Crevan · 12/11/12, 1:44 PM

I don't have much time to do the actual theorycrafting right now, but here are some tips about how to do it.

The amount of haste you want can be calculated in terms of your total energy gain per minute, the amount of jabs that energy will allow you to perform, and the amount of GCDs you will need to perform those jabs and spend the generated chi.

Your total energy gain per minute is energyPerSecond * 60 + energyFromEB = totalEnergy, which will result in totalEnergy / 40 jabs and (totalEnergy / 20) + a chi (a = 0 with Ascendance and 3 with Power Strikes).

As far as I know, TP and BK mastery procs are independent of each other and both have the same chance to occur. So, to use every mastery proc you get you will need to spend jabsPerMinute * 2 * ( masteryPercent + masteryPercent ^ 2) GCDs per minute. (I've corrected this part to account for a double proc occuring)

The above formulae already account for the 4 set bonus (you simply replace energyFromEB with the correct value), and it will alter the number of jabs, GCDs used and chi generation.

Two-set bonus is more tricky to account for. Using FoF close to or on cooldown might result in wasted resources or other abilities delayed. Simcraft reports the time between Fists uses as about 24 seconds with set bonus and about 29 seconds without. It's probably better to use these values when calculating the GCDs and chi you spend on the ability.

khel · 12/12/12, 7:51 AM

Originally Posted by Eyedore

Conclusion: when I am below 90 energy I can safely do a FoF without loosing dps

This looks ok to me. I recommend that you modify a simulationcraft priority order for dps though and actually sim this to be more confident in your findings. You can set different "fof energy caps" at 40-100 energy for example in increments of 10, and then come back with a more convincing evidence to support your argument.

Bloodvalor · 12/12/12, 5:14 PM

Originally Posted by Xiaoryn

I wonder if our time isn't better spent figuring out relative values of haste and crit, and at what point (hastecap or otherwise) we should be stacking the other.

I mean, It seems to me that there are still several unanswered core questions about this spec:
-Assuming Ascension, how much haste rating or energy/sec should we be aiming for? How much w/o Ascension?
-How does the above change if we're using FoF on near cooldown (and this is a fight-specific question)?
-How does 2pc and 4pc factor in, exactly?
-How do we account for Mastery procs in our haste cap math? Is there any way to make it relate-able via numbers?

I think that once we get these figured out, we can really start to generate 'gold-standard' BiS lists, rotations, and gemming.

I wish I was better at creating sims or pure numbercrunching, but I've tried many different things on my character from week to week...but that data isn't reflective of 'best case scenario' because I am a human who raids with other humans.
So far, though, I've found that my numbers are generally better when I cap haste around 5.5k, then roll crit. I'm gemming all agility or agi hybrids, and using FoF at every viable opportunity.

I agree with this completely. Being that Xiaoryn and I are not great and sims ect... If anyone who knows how to do that stuff needs help with providing logs or anything I'd be more than willing to help.

Eeinx · 12/12/12, 5:47 PM

My apologies for not getting to work on that, I've been meaning to, but I just haven't had time lately with the holidays and influx of last minute work and IRL commitments.

The module is sound now, so we should be able to change around the profile and gearsets to test for this. we've had some issues, but i think the best solution is if someone wants to run a fairly simple sim and do a reforge and stat weight plot around 6k haste then stack crit (which seems to be around the point in simulation)

My recommendation is probably as said above @ 5.5k haste then the rest into crit.

I really think gearing options will have alot to do with how things change. For instance, more mastery will devalue haste and value crit.

Since haste directly affects mastery through jab, they have an interaction which can mess up weights. Its going to depend solely on the players current gear whether a certain stat is better. This is the unfortunate fact.

Unless they make mastery more beneficial than just limiting our haste potential (through extra damage, etc), we wont have a realistic stat weight priority set up because it can fluxuate with mastery.

tastysnack · 12/12/12, 7:35 PM

Originally Posted by Eeinx

My apologies for not getting to work on that, I've been meaning to, but I just haven't had time lately with the holidays and influx of last minute work and IRL commitments.

BADDIE. Just kidding. Your work is noticed and appreciated.

Originally Posted by Eeinx

My recommendation is probably as said above @ 5.5k haste then the rest into crit.

I really think gearing options will have alot to do with how things change. For instance, more mastery will devalue haste and value crit.

Since haste directly affects mastery through jab, they have an interaction which can mess up weights. Its going to depend solely on the players current gear whether a certain stat is better. This is the unfortunate fact.

Unless they make mastery more beneficial than just limiting our haste potential (through extra damage, etc), we wont have a realistic stat weight priority set up because it can fluxuate with mastery.

In my own experience, I noticed that mastery provided much more damage for me pre-Ascension buff and when I was hanging around 5K haste (keeping in mind, too, that my gear options are rather limited at the moment to reputation rewards and whatever our feral druid/rogues/brewmaster don't want). I'd have to run the numbers again, but I believe that right at about 6.5K haste and 20% crit, Mastery was stronger - but following that, Mastery flatlined and Crit took off (according to DPS scaling). Keeping in mind that this utilized a 2H weapon. Link to where I posted that up. Again, I unfortunately am stuck with a 2H, but I know that the main effect DW would have is haste will end up more valued (if my TC is correct). Keep in mind, too, that most top-end monks (including those that are DW) are only packing around 4-4.5k haste (referencing Method's Justmonk) - I myself went with a similar build (using 2H) and have found it effective. But, there's also the fact that 2H receives increased weapon speed. I'm currently running a plot at home weighing Crit against Haste as DW to get a better sense for it. I'll edit once I'm able to provide results.

With that in mind, I feel like a certain amount of mastery is beneficial (say, ~14%, which is what I see most top-end monks carrying (Method's Justmonk again for reference, who is sitting at 15.6%)), at least so that the amount of Jabs we're required to use is decreased (keeping in mind, again, BoK and TP typically do more damage than a single Jab). I know anecdotes aren't true evidence, but I'm personally packing around 14% and A) have no issues refreshing TP without the need to use Chi to do it, and B) have zero energy issues. So, I think that some people really are undervaluing our mastery for what it is. Not to say there aren't things that I'd change about it, but it's not as bad as we're all trying to make it out to be - we just seem to have a sort of "mastery cap" right around 14-15% that's been apparent in most of my charts.

Gondlem · 12/12/12, 9:20 PM

I've been curious about something for a while but I'm unsure if it's possible to really get reliable data from simcraft or any other source for it. How much does the presence of extra skull banners impact the value of crit? With current class balance a lot of guilds have an eye on warrior stacking, and certainly it wasn't uncommon to have 3 or 4 warriors in a 25 man raid that was trying to beat DPS checks in T14. The DPS increase for having say 3 banners over 1 is pretty noticable, but that damage only comes from crits.

This, along with actual in fight mechanics which cause a tendancy towards energy capping (movement etc), is a reason I've been leaning more towards crit.

saboya · 12/12/12, 9:35 PM

Originally Posted by tastysnack

With that in mind, I feel like a certain amount of mastery is beneficial (say, ~14%, which is what I see most top-end monks carrying (Method's Justmonk again for reference, who is sitting at 15.6%)), at least so that the amount of Jabs we're required to use is decreased (keeping in mind, again, BoK and TP typically do more damage than a single Jab). I know anecdotes aren't true evidence, but I'm personally packing around 14% and A) have no issues refreshing TP without the need to use Chi to do it, and B) have zero energy issues. So, I think that some people really are undervaluing our mastery for what it is. Not to say there aren't things that I'd change about it, but it's not as bad as we're all trying to make it out to be - we just seem to have a sort of "mastery cap" right around 14-15% that's been apparent in most of my charts.

I don't see how Mastery can be better than any stat at this point, regardless of how much Crit or Haste you have (assuming non-ridiculous values like 100% crit). In low Haste situations, Haste is the best stat. The more Haste you have, the better Crit is. Assuming there is a point where Crit is better than Haste, Crit should never fall behind Mastery, there's no reason for that to happen.

I doubt any Monk is "sitting" on ~14% Mastery on purpose, they just can't reforge out of it. Especially when we have a 1.2k Mastery trinket.

Littch · 12/13/12, 4:13 AM

Can anybody write out a more useful/up to date version of our single target priority in simcraft, i have no idea what I am doing with it.

It currently is.

actions.st=rising_sun_kick
actions.st+=/fists_of_fury,if=!buff.energizing_brew.up&energy.time_to_max>(cast_time)&buff.tiger_p ower.remains>(cast_time)
actions.st+=/blackout_kick,if=buff.combo_breaker_bok.react
actions.st+=/blackout_kick,if=(chi>=3&energy.time_to_max<=2&!talent.ascension.enabled)|(chi>=4&ene rgy.time_to_max<=2&talent.ascension.enabled)
actions.st+=/tiger_palm,if=(buff.combo_breaker_tp.react&energy.time_to_max>=2)|(buff.combo_breaker _tp.remains=0&buff.combo_breaker_tp.react)
actions.st+=/jab,if=talent.ascension.enabled&chi<=3
actions.st+=/jab,if=talent.chi_brew.enabled&chi<=2
actions.st+=/jab,if=talent.power_strikes.enabled&((chi<=1&!cooldown.power_strikes.remains)|(chi<=2 &cooldown.power_strikes.remains))
actions.st+=/blackout_kick,if=((energy+(energy.regen*(cooldown.rising_sun_kick.remains)))>=40)|(ch i=4&!talent.ascension.enabled)|(chi=5&talent.ascension.enabled)

More than 1 change could be useful, so for example FoF > RSK > BoK, or the current but just updated so I have heard people saying as long as you have less than 90 energy before using FoF its fine ect. any help will be massively appreciated

Kagura · 12/13/12, 6:43 AM

To give everyone a rough idea of what they can expect about the globals they have available, I made a very small and dirty tool : WW global analysis

A couple assumptions :
- You use RSK on cooldown
- The amount of Tiger palm needed is calculated without taking into account the mastery procs, so in reality this will be a bit lower.
- Assumes 4 piece bonus (Energy from EB per sec = 120/6)

The tool will let you choose a fight length, if you want to use FoF or not, your haste rating and your mastery percentage. A sample output

Fight length in seconds : 300

Mastery in % : 25

Haste rating in charsheet : 6000

Amount of FoF you want to use per minute : 2

-

Haste percentage : 14.117647058824

Energy regen per sec : 13.123529411765

Peak energy regen per sec : 16.573529411765

Energy gained from lust : 138

Fists of Fury cast time : 3.5051546391753

Energy from EB per sec : 2

Energy gained : 4675.0588235294

Amount of Jabs : 116.87647058824

Possible mastery procs : 58.438235294118

Total chi gained : 233.75294117647

Amount of TP needed : 15

Possible RSK : 37.5

Amount of FoF : 10

Amount of BoK : 56.876470588235

Total globals : 300

Jab globals : 116.87647058824

TP globals : 15

RSK globals : 37.5

FoF globals : 35.051546391753

Mastery globals : 58.438235294118

Xuen globals : 2.6666666666667

Globals left : -22.409389529007

Update : forgot to account for jab possibly proccing both Combo breakers

Eeinx · 12/13/12, 12:14 PM

Thanks all for the contribution.

The key thing i'm noting which are this:
1. Mastery quantity scales with haste. (more jabs, more rng for mastery procs)
2. Haste scales with haste.
3. Crit quantity scales with haste.
4. Energy puts a hard limitation on maximum benefit of haste.
5. Crit has no limitation but RNG.

But foremost I believe that because of the contribution by haste, mastery will never be as good as haste.

I think because of these interaction, we'll have to dig a bit deeper to get the sim to give us actual weights.

What i'm on about is:

How much would it benefit us to turn off the mastery buff provided by the raid?
At what point should mastery and haste be at to allow crit to be more beneficial?

These are all things we should think about, because as noted above, mastery is directly benefited from haste outside of chi generation.

Theres a ton of work to do, but right now, asking for simple weights is not faring well with the sim. We need to sort out this interaction before delving into BiS lists etc.

Kagura · 12/13/12, 1:07 PM

Adjusted a couple values for my tool (my calculated energy regen during lust was off, haste from lust is multiplicative and not additive).

I will leave everyone to conclude for himself, but I will go with the next stats for a week to provide some log data:

- Haste rating : 4848 (So I have 1 global free every minute)
- Will leave mastery buff on, so I will be on around 24.00% mastery
- Will use 4 piece meaning 120 energry from EB
- Will not use Fists of Fury

Eeinx · 12/13/12, 1:16 PM

Originally Posted by tastysnack

With that in mind, I feel like a certain amount of mastery is beneficial (say, ~14%, which is what I see most top-end monks carrying (Method's Justmonk again for reference, who is sitting at 15.6%)), at least so that the amount of Jabs we're required to use is decreased (keeping in mind, again, BoK and TP typically do more damage than a single Jab). I know anecdotes aren't true evidence, but I'm personally packing around 14% and A) have no issues refreshing TP without the need to use Chi to do it, and B) have zero energy issues. So, I think that some people really are undervaluing our mastery for what it is. Not to say there aren't things that I'd change about it, but it's not as bad as we're all trying to make it out to be - we just seem to have a sort of "mastery cap" right around 14-15% that's been apparent in most of my charts.

Well, since the effect on haste helps mastery scale, it would also cause mastery to diminish in value up to a certain point.

I'm not a super-mather or I could probably show this in graphs. I have it in my head, I just dont know the language to make a visual of what i'm saying. I'll give it a shot when I can.

---

@Kagura, something seems fishy with your tool.

If i set mastery to 25% it looks like its assuming half of your jabs will be mastery procs.

you will always have half mastery globals compared to jab globals. on the default setup. I think your tool is good for formulation, but I can't use it to test the interaction between stats because it seems like rng is potentially missing.

EDIT: Nevermind, I see your intention with the tool. Its formulative. My apologies.

Kagura · 12/13/12, 1:34 PM

Originally Posted by Eeinx

@Kagura, something seems fishy with your tool.

If i set mastery to 25% it looks like its assuming half of your jabs will be mastery procs.

you will always have half mastery globals compared to jab globals. on the default setup. I think your tool is good for formulation, but I can't use it to test the interaction between stats because it seems like rng is potentially missing.

EDIT: Nevermind, I see your intention with the tool. Its formulative. My apologies.

I was very very concerned about this as well, but I checked most of my logs for the last 4 weeks and my mastery procs (TP and BoK) are a bit blow half of my jabs (I have around 25% mastery in raids). The missing ones can be contributed to me sucking at playing and possibly refreshing some of my mastery procs.

My intention with the tool is finding a sweet spot where I don't feel I can't miss any globals.

I have also experimented with the mastery buff before, but I concluded on having CB:TP procs as the lowest priority.