Monday, May 9, 2016

TrainingGrounds update and Brewmaster calculators


TrainingGrounds update

First things first: we've updated TrainingGrounds to reflect build 21655.  Among the changes made, Strike of the Windlord once again deals full damage to all targets, and we've rearranged the keybinds to emphasize its priority.  If you haven't had the chance to play 7.0 Windwalker yet, give this addon a try.  As always, let us know if anything seems missing or inaccurate.



Stepping away from WW for a bit, we're going to take a quick look at Brewmasters.

Mastery

First, a calculator which determines your overall avoidance from your current Mastery%. We haven't been able to find such a calculator online, so we made our own:

To better track diminishing returns, we've also assembled three charts for you which track overall avoidance, avoidance per mastery, and relative damage reduction.

As you might expect, diminishing returns kick in not only as our mastery increases, but also as our base avoidance goes up. Additional mastery contributes less not only to our avoidance, but also to our relative damage reduction. Looking at the third chart specifically, there seem to be some awkward breakpoints where the value of mastery drops off sharply. I'm not sure whether to attribute that to statistical fluke, rounding error, a problem with the formula, or just misinterpretation of the results. You can mouseover any of the numbers in the charts to see how we arrived at them.

Self-mitigation during dangerous events

Next, a "dangerous event" calculator. This one was written to test the efficiency of Purifying Brew and the new Gift of the Mists talent: when you take a chain of very damaging attacks (such as Nightmare Visage, Felblaze Flurry, or Glaive Combo), how much healing do you need to stay alive, and how much damage can you self-mitigate? When you use the calculator, keep in mind the default values reflect a Warlords situation where you might easily take many times your maximum health in damage over a very short amount of time.
(You could also use this calculator to look at incoming autoattacks over a long period of time, but it won't include avoidance or extra damage.)

GotM does a good amount of extra healing, provided either the incoming damage hits for a large portion of your maximum health, or your health drops low during the event. For example, if you run the calculator with the default values, GotM drops an extra 4 orbs, but the damage per hit is very high and you escape the event within an inch of your life. Now compare that with an event where instead of taking 7 hits of 50% each, you take 14 hits of 25% each, and your health never drops below 50%: GotM drops only one additional orb compared to other talents.

Brew charges

Now that you have an idea of how Purifying Brew compares to Healing Spheres, take a look at this brew calculator. This calculator will tell you how many brews you'll have available and how often you can expect to use Purifying Brew if you want to maintain a given Ironskin Brew uptime. It's designed to compare Light Brewing and Gift of the Mists specifically, looking at average and worst-case Face Palm proc rates. Black Ox Brew is trickier to simulate as it won't recharge at the same time on every cycle, but you can get a decent approximation by adding 3+ brews to your Gift of the Mists results.

Posted by at No comments:

Thursday, April 21, 2016

TrainingGrounds 21531 update

TrainingGrounds has been updated for build 21531.
http://wow.curseforge.com/addons/traininggrounds/files/2-training-grounds-demo-2016-04-21a/

We've updated Strike of the Windlord's attack power and cost, and we've also changed the order SotW and FoF appear in the tutorial.

(SE&F was already using build 21531 mechanics.)

...

Additionally, we've gone back and updated the old Spinning Crane Kick calculator with the latest numbers and mechanics.
Posted by at No comments:

Monday, April 18, 2016

TrainingGrounds demo

We've created a playable demo of TrainingGrounds.
TrainingGrounds is a combat simulator addon which is played inside of World of Warcraft.

It is currently uploaded to http://wow.curseforge.com/addons/traininggrounds/files/2-training-grounds-demo-2016-04-21a/.
Place it in your addons folder, move to a safe place in-game, and type /tg to play.

This demo may be relevant to you if:
  • you do not currently have access to the Legion alpha, and you do not mind playing a crude approximation of 7.0 Windwalker Monk with mechanics only as accurate as Wowhead's datamined tooltips.
  • you do have access to the Legion alpha, but you would like to see how you play at 50% haste with the as-of-yet unreleased changes to SE&F.

The demo is designed for players who are already familiar with the Warlords Windwalker Monk rotation.  There is a "tutorial" which explains the basic controls, but no attempt is made to teach you how to play effectively.

There are many other features missing from the game.  One such feature is actionbar customization.  You cannot rearrange spells on your actionbar.  Sorry about that.  We wanted to get this demo into a playable state as soon as possible, and this wasn't a priority.

If you decide to play this demo, let us know if any of the existing abilties or game mechanics feel wrong or inaccurate.  (For example, we're not sure if the recent change to Roll is just a tooltip error.)  Our top priority at the moment is fixing any broken class mechanics.


Known issues

Automatic shutdown on Lua Error
The current build will automatically shut down if it detects a Lua Error... even if that error didn't originate from TrainingGrounds.  We recommend using the "Display Lua Errors" option in the Interface > Help menu.

Misaligned text
Occasionally, a fontstring will become misaligned.  In the example shown, the text is incorrectly centered instead of left-aligned.  There is some evidence that this is a bug in World of Warcraft, so we may have to find a workaround instead of fixing it.

Aspect ratio distortion
The game was designed for windowed mode.  Depending on the aspect ratio of your monitor, the game window may appear distorted when World of Warcraft is played in fullscreen mode.




Where do we go from here?


There are a lot of things that need doing before we can pretend this addon is even remotely feature complete.  At the very least, we want to implement the remaining Windwalker abilities (defensive cooldowns), allow actionbar customization, and rewrite the program to be more class-agnostic.  Our long-term goal for this addon is to support different classes and specializations, including specializations that do not, technically, exist.
Posted by at No comments:

Friday, April 8, 2016

Spinning Crane Calculator

If You're Reading This, The Page Hasn't Broken (Yet)

Big change to WW rotation recently. We made this calculator. We were bored.

- Disclaimer -
This calculator is designed to tell you whether you could press Spinning Crane Kick in lieu of another button at some hypothetical point in time when you already have an arbitrary number of stacks.  We make no statement on whether it is a good idea to acquire those stacks if you do not already have them.  Because that's starting to sound like Actual Math.

Fill out this form according to your current situation, or just leave in the default values if they sound close enough.  Then click Calculate.
Haste:%
Mastery:%
SCK cost (chi):
SCK base damage:
FoF damage:
SotW damage:
WDP damage:
RJW damage:
RSK damage:
BoK damage:
CB damage:



---


Join us next time, when we'll work on a "when should you press SEF" calculator.  Or not.  Maybe we have something else planned for next time.  Ooh, the anticipation is kil[Lua Error: attempt to index field '?' (a nil value)]
Posted by at No comments:

Monday, April 4, 2016

Progress report

It's been a few weeks since our last video and we figured this was a good time to post an update regarding addon development.



That's all for now.
Posted by at No comments:

Friday, March 18, 2016

Posthaste

Haste is now assuredly more consequential

We were never a fan of Haste in build 21215.  You might have noticed already, but all the videos in the previous post were recorded with 0 haste.  As we'll discuss further down the page, haste was the stat to avoid, just as mastery is currently the stat to avoid in Warlords.  There wasn't much point in showing you our rotation at 30% or 50% haste if doing so would compromise our throughput.  (After all, who* wants to push more buttons and deal less damage?)
* ...okay, but besides me, who wants to do that?

It turns out the developers didn't think haste was all that great either, and they overhauled it for quite a few classes.

For Monks, Haste now decreases the cooldown of Fists of Fury, Whirling Dragon Punch?, Rising Sun Kick, and Rushing Jade Wind*.
* It already decreases the cooldown of Rushing Jade Wind.

To explain how much of a difference this makes, we compiled a table of every haste benefit we could think of, spanning from Warlords to Legion ... to Legion again.

We should mention that we are bad at designing tables, so we hid this one away inside this spoiler tag that may or may not work when you click on it.




Don't want to look at all that?  That's okay, we probably got the details wrong.  We'll summarize the important stuff:

In Warlords, we get more Blackout Kicks/Chi Explosions (moderate attack) and more Rushing Jade Winds (moderate talent).  That alone increases our damage by a fair amount, but we also generate Tigereye Brew more often (as a side effect from BoK/CHEX).  TEB stacks are consumed to temporarily boost all damage by 60% -- that includes the damage from our strongest attacks, which are otherwise unaffected by haste.

So, Haste didn't let us use our strongest attacks more frequently, but in return we got to amplify their damage more often, and so it stayed useful to us.  We reluctantly admit that doesn't sound much like "haste".  In fact that sounds like a more active version of our existing Mastery*. 
* which is kind of amusing because our original Mastery back in Mists was a more active version of our Haste

In Legion build 21215, we get many more Blackout Kicks (which is now a weak attack) and slightly more Rushing Jade Winds (moderate talent)... and that's it.  Tigereye Brew is no longer linked to how many BoKs we use, and haste doesn't affect any of our other abilities.

There was some talk recently about how Feral Druids (in Warlords) avoid haste because it doesn't affect their bleed damage at all, and their direct damage isn't strong enough in comparison to take haste over another stat which does affect bleed.  Replace Shred+Ferocious Bite with Blackout Kick and you have an idea of how this works out for monks.

In Legion build 21249, we get many more Blackout Kicks/Spinning Crane Kicks (weak attacks) and slightly more Fists of Fury (very strong attack), Rising Sun Kicks (moderate attack), Whirling Dragon Punch?es (strong talent), and Rushing Jade Winds (moderate talent).

So, Haste finally lets us use our strongest attacks more frequently.  And if that sounds familiar, there was going to be a stat in Warlords that did the same thing -- Readiness -- but they ultimately shelved it because it "doesn't compare well* against [other] secondary stats".  Combining it with haste sounds like a nice compromise.
* In its place, we got Versatility.  Do I sound bitter?  Tryin' to sound bitter here.


What's our maximum APM?

Pop quiz!  Let's say I mistype a number in the monk simulator and give myself 50% Haste.  Also, the cost of Tiger Palm mysteriously drops to 30 while nobody's looking.  What's our maximum APM?

It's not 60.  I've heard some of you say it's 60, but it's not 60.  Each individual tick of Fists of Fury does not count as its own action.  Off-GCD abilities do count as their own actions.

(Standard Disclaimer: everything written in Dagnerous Padna's bolg is based on napkin math and may contain glaring inaccuracies.)

(Special Disclaimer: the following message is based on napkiner math and may contain glaringer inaccuracies.)

At 50% Haste, cooldowns and channel times are reduced to (1/1.5) = ⅔ of their normal amount.
Fists of Fury lasts for 2⅔ seconds at 50% Haste
Each Fists of Fury is 1⅔ APM "lost" (because we're not taking any actions while it channels)

How many Fists of Fury do we use per minute?
Fists of Fury has a 16 second cooldown at 50% Haste

If we don't take Serenity, we simply use Fists of Fury 3.75 times per minute
for a total of 6.25 APM lost
If we're hitting a button every second we're not channeling FoF, that brings us down from 60 APM to 53.75 APM (on GCD)

But if we do take Serenity (and if our goal is maximum APM, we really should), we use 7 Fists of Fury every 96 seconds (delaying each Serenity 6 seconds so we can fit two FoFs inside the 10-second buff)
7 FoFs per 96 seconds is the same as 4.375 FoFs per minute.
4.375 times per minute, we lose 1⅔ APM
which is a total of 7.2917 APM lost
If we're hitting a button every second we're not channeling FoF, that brings us down from 60 APM to 52.7083 APM (on GCD)

Now we add in our off-GCD abilities:
If we take Serenity, we'll use it every 96 seconds: +0.625 APM
If we don't take Serenity, we'll use Tigereye Brew instead: +⅔ APM
If we take Xuen: +⅓ APM
If we take Energizing Elixir: up to 1 APM
Touch of Karma: up to ½ APM

There are plenty of other ways to gain additional actions: Diffuse Magic, canceling SE&F or Flying Serpent Kick, Rolling in between GCDs, etc.  (Then there's interrupts, potions, extra action buttons, and so on -- but those actions aren't exclusive to Monks, so you might not consider them part of your rotation.)

So, depending on our talent choices and encounter details, we can expect our top speed to be somewhere in the neighborhood of 53-55 APM.

Of course, we can also lose APM, either through mistakes in rotation, or in situations where the enemy is unattackable.  Diffuse Magic's nerf and Nimble Brew's removal may finally get us to back off when the boss does something nasty.

Finally, this model assumes we have unlimited resources.  In practice, we won't have enough resources to fill every single GCD, even at 50% haste.



Next time...

That was a lot of words just now.  We were going to talk about "interesting" vs "boring" artifact traits this week, but we'll save that for later.
Posted by at No comments:
Subscribe to: Posts (Atom)