[Balance] Haste breakpoints and Legendary Metagem.

[chowchig]

I tried searching for answer to this question, but couldn't find anything about it.

Is there any new breakpoint I should be aiming for to get another tick on my dots for when the effect is active?

[manamonster]

I personally wouldn't worry about it. I'd just hit the normal 5273 breakpoint which accounts for Nature's Grace being active (Source). I consider the meta as being something that smooths out our rotation more than an increase to our dots. If you can manage to refresh dots while it is up, that's just an added bonus. I wouldn't break your rotation to reapply dots with the meta active (as the proc is semi-random and unpredictable). This is the same reason I opt not to use the Lei Shen trinket.

I use the meta as a way to insure I move between Solar and Lunar eclipses as fast as possible. It can be very handy during times where you need to multi dot and might lose out on Nature's Grace. Sacrificing crit for more haste (in order to reach another breakpoint) is very likely a loss of dps (but the only way to confirm is to sim your own gear using Simcraft).

[boomkinhero]

http://www.totemspot.com/vb/entry.php?b=44'

I prefer the 9692 breakpoint because of such high uptime on the meta gem proc. You should too. The extra dot tick would make the haste rating worth more than crit. You (on average) don't break rotation, as mentioned above.

[Deleted]

If you're geared for tier 15, you shouldnt have to worry about breakpoints at all and just stack crit on yourself.

In 5.3 and if you're nearly full BiS you'll be able to hit the 10289 cap without any haste gemming/crazy reforges so dont worry about any haste cap for now. But it would help alot if you actually gave us an armory link to your gear.

[manamonster]

http://www.totemspot.com/vb/entry.php?b=44'

I prefer the 9692 breakpoint because of such high uptime on the meta gem proc. You should too. The extra dot tick would make the haste rating worth more than crit. You (on average) don't break rotation, as mentioned above.

The article you linked has an assertion that intentionally going for haste breakpoints is a bit pointless. I tend to agree as even with 7500 haste I currently have seems like a lot (and my cast times are very very fast). Regemming / reforging out of crit to get an additional 2k haste seems like it would be a DPS loss (and this is considering my ilvl is 528). But at the end of the day the only real way to know is to sim your own gear I guess.

I would say if you're somehow close to 9692 then sure. But otherwise it's pointless to drop crit intentionally to get another breakpoint.

[land]

1) Can anyone pls math out what would be the amount of haste needed to reach the 11th tick (10289) WHILE the legendary meta gem is proc'd (30% haste) + NG?

2) On a similar note, can someone math out what is the haste to get GCD capped for Wrath while NG + leg meta + hero are all up?

Not sure how much haste is a 1%, also not sure how legendary meta stacks with other things like NG and hero...

[chowchig]

I personally wouldn't worry about it. I'd just hit the normal 5273 breakpoint which accounts for Nature's Grace being active (Source). I consider the meta as being something that smooths out our rotation more than an increase to our dots. If you can manage to refresh dots while it is up, that's just an added bonus. I wouldn't break your rotation to reapply dots with the meta active (as the proc is semi-random and unpredictable). This is the same reason I opt not to use the Lei Shen trinket.

It seems to proc too much for me to not take advantage of getting another tick with the meta + NG.

[manamonster]

1) Can anyone pls math out what would be the amount of haste needed to reach the 11th tick (10289) WHILE the legendary meta gem is proc'd (30% haste) + NG?

The link provided above by Boomkinhero (http://www.totemspot.com/vb/entry.php?b=44) shows haste breakpoints for varying combinations of haste buffs. 5% = raid buffed, 15% = Nature's Grace, 30% = Lust or Meta Gem. Using those numbers the chart shows different combinations of 5 ... 5 + 15 ... 5 + 15 + 30. It then shows the corresponding haste rating needed to get X ticks of Moonfire. Not using 4pc from the previous tier, you would need 9,692 haste rating with 5% + 15% (NG) + 30% (Meta) to get tick #14 on Moonfire.

Not sure how much haste is a 1%

425.19 haste rating = 1% haste at Level 90

, also not sure how legendary meta stacks with other things like NG and hero...

It stacks multiplicative.

5% (raid buffed) convert to decimal # = 0.05
15% (Nature's Grace) = 0.15
30% (Legendary Meta) = 0.30
30% (Heroism / Bloodlust) = 0.30

To combine these percentages multiplicatively you would need to first add 1 to each percentage (after it is converted to decimal).
5% (raid buffed) = 0.05 + 1 = 1.05
15% (Nature's Grace) = 0.15 + 1 = 1.15
30% (Legendary Meta) = 0.30 + 1 = 1.30
30% (Heroism / Bloodlust) = 0.30 + 1 = 1.30

Total Haste Bonus = 1.05 * 1.15 * 1.30 * 1.30 = 2.040675

So Total Haste Bonus = 104%

Total Haste = (1 + (haste rating)/425.19) * 1.05 * 1.15 * 1.30 * 1.30
Total Haste % = (Total Haste - 1) * 100

2) On a similar note, can someone math out what is the haste to get GCD capped for Wrath while NG + leg meta + hero are all up?

The equation to calculate the cast time of Wrath given some haste bonus is as follows:
Base Wrath Cast Time = 2s

Wrath Cast Time = 2 / (Total Haste Caculation)
So lets say you had 0 haste rating on your character and had all 3 buffs up. (See calculations above)
Total Haste Calculation = 1.05 * 1.15 * 1.30 * 1.30 = 2.040675
Wrath Cast Time = 2 / Total Haste Calculation = 0.98s

This means the cast time of wrath is below the GCD with all 3 haste buffs (Nature's Grace, Legendary Meta, and Bloodlust/Heroism) procced at the same time. Therefore you need 0 haste rating to achieve the GCD cap with all 3 buffs active.

To adjust and see the cast time otherwise just convert your haste to the proper decimal value like this:

Assume Nature's Grace (15%) & Legendary Meta (30%) and Raid Buffed (5%) are active
Haste = 5000
Haste Decimal = ( 5000 / 425.19 ) / 100 = 0.1176
Total Haste Calculation = 1.1176 * 1.05 * 1.15 * 1.3 = 1.7543526
Wrath Cast Time = 2 / 1.7543526 = 1.14s

Here is a more in depth reference on how haste works: http://www.wowpedia.org/Haste

[Tarm]

I was wondering about any possible NG + meta gem breakpoints too. So I ran a quick analysis in WrathCalcs with my gear and 20% Starsurge proc wastage.

I was sitting at 6812 haste and 11752 crit. I was curious how my dps would change as I replaced crit with haste. In real life of course I could only really do this in gem increments of 160 or 320, but for the heck of it, I checked how every additional exchange of a point of crit for a point of haste would affect my dps.

The following plot shows the percent dps increase I get over my initial 6812 haste build for higher amounts of crit:

i.imgur.com/4pWZ9we.png

Features of note:

• The breakpoint at 9692 represents the 14th tick of moonfire/sunfire buffed by NG and the meta.
• The breakpoint at 10289 is the 11th tick with just NG.
• Haste sims stronger than crit at all haste levels (up to 10289) even ignoring the breakpoints.

So, as people have said before, meeting the 10289 haste breakpoint definitely gives a higher dps in WrathCalcs.

Whether or not it is actually a dps increase in game is open to debate. The theoretical 1-2% difference in WrathCalcs is so small that you'd never be able to see it in a real fight where things like proper rotation, luck with procs, luck with movement, luck with fight mechanics, and CD stacking will have effects that are orders of magnitude higher than any potential difference from going haste or crit.

So ultimately it comes down to whichever style of play you like better.


1) Can anyone pls math out what would be the amount of haste needed to reach the 11th tick (10289) WHILE the legendary meta gem is proc'd (30% haste) + NG?

You get the 11th tick at 0 haste rating when NG and the meta proc are active. NG and meta gives 56.98% haste at 0 haste rating.
12th tick at 1976
13th tick at 5869
14th tick at 9692
15th tick at 13584

You can see all of this by playing around with the Haste tab in WrathCalcs. This Google spreadsheet is also helpful and easily modified to look at the meta gem proc too: docs.google.com/spreadsheet/ccc?key=0AtT3sAEBV2QfdEtWeThnVWx4X3NoT29ZRGJjbkxjS0E#gid=0

I couldn't find who put together the Google doc. Anyone know?

[earthwormjim]

I did not like the play feel with that much haste and the meta. I also missed the shooting stars procs I would regularly get by having 15k crit rating vs 10k I dropped down to to reach 10k haste.

[manamonster]

Another thing to note is that if you're using the Volatile Talisman of the Shado-Pan assault, it procs 8800 haste rating which is 20.7% haste. With 10k haste & the trinket proc you're looking at approx. 44% haste from STATS alone. This is before adding in 5% raid buff, Legendary Meta, Nature's Grace, or Lust/Hero.

[chowchig]

Another thing to note is that if you're using the Volatile Talisman of the Shado-Pan assault, it procs 8800 haste rating which is 20.7% haste. With 10k haste & the trinket proc you're looking at approx. 44% haste from STATS alone. This is before adding in 5% raid buff, Legendary Meta, Nature's Grace, or Lust/Hero.

I have the Shado-Pan trinket and the trinket Essence of Terror, along with having the herbalism haste on use spell. I just get some super dots when that line up at the start of a fight, even though now I know not to pop Lifeblood, since I'll probably be capped on GCD.

[Tarm]

The debate here about whether it is worth it to reach these breakpoints made me want to share a little analysis I did for myself. Everything below is done in WrathCalcs with my current gear setup, full raid buffs, and 20% shooting star wastage.

I was sitting at 6812 haste and 11752 crit and I was curious how my dps would change as I replaced the crit with haste. So I measured my dps in WrathCalcs as I exchanged each point of crit for a point of haste.

The following plot shows the percent dps increase I get compared to my initial 6812 haste build for higher amounts of haste (and correspondingly lower crit):

i.imgur.com/4pWZ9we.png

Features of note:

• The jump at 9692 represents the NG + meta breakpoint for the 14th moonfire tick.
• The jump at 10289 is the NG breakpoint for the 11th tick.
• Haste sims stronger than crit at all haste levels (up to 10289) even ignoring the breakpoints.

So, with my gear setup, meeting the 10289 haste breakpoint maximizes dps in WrathCalcs, giving a ~1.6% dps boost over a 6812 haste build.

Whether or not a haste build is actually a dps increase in-game is hard to know. The theoretical 1.6% difference in WrathCalcs is small enough that I'd never be able to see it in a real fight where things like rotation execution, CD stacking, and luck with procs, movement, and fight mechanics will have effects many times greater than any potential difference from choosing a haste build over a crit build. I do see the boost when I compare 30 minutes of single target rotation on a training dummy, though that might not be a long enough test to really be sure.

So ultimately it probably just comes down to whichever style of play you like better. You're not going to noticeably gimp yourself either way.

__________________________________________

As a bonus for the interested, here's a similar analysis I did with a few different shooting star wastages:

i.imgur.com/hz1FsLf.png

The values of the breakpoints themselves stay constant, but, as we'd expect, haste becomes more valuable the more shooting stars we are wasting. The potential issue with this type of analysis though is that as we get higher haste (i.e. lower crit) we would waste fewer shooting stars which would then increase the value of crit relative to haste. It's not super clear to me how we can model how shooting star wastage varies as a function of crit and haste levels (besides trying to do it empirically on a training dummy), but it could be an important factor. "Important" being a relative term - we're talking about fractions of a percent of dps...

[Deleted]

Just on a note, haste has always simmed stronger then crit. But on movement (wich is pretty much any fight) your haste will do nothing for you. Even counting in the SS waste, i still don't think haste is superior. And certainly not when you need to drop your crit to such low levels where SS procs really become scarce when you need them.

[manamonster]

Just on a note, haste has always simmed stronger then crit. But on movement (wich is pretty much any fight) your haste will do nothing for you. Even counting in the SS waste, i still don't think haste is superior. And certainly not when you need to drop your crit to such low levels where SS procs really become scarce when you need them.

When you say "always simmed higher" are you talking with Patchwerk style sim? I don't use that Sim setting for myself anymore since there are literally 0 fights where you can stand still (and moonkins suffer so much from movement). When I was still using it, it seemed like crit still simmed higher. End of the day though, it's all relative to the person and their gear I guess.

[land]

Hey Manamonster, thanks for ur detailed answer. I'm going to try tonight going for the 9,692 cap and see how it feels. By the way, ask mr robot got updated to include different haste breakpoints included the meta gem one for balance druids.

[xtramuscle]

Hey Manamonster, thanks for ur detailed answer. I'm going to try tonight going for the 9,692 cap and see how it feels. By the way, ask mr robot got updated to include different haste breakpoints included the meta gem one for balance druids.

Firstly 9,692 is not a cap, its a break point, the only real cap is GCD (and even that only applies in currect gear with meta+ lust active).
As for Mr Robot, avoid that site at all costs, failure to do so may result in IRL Death.

[manamonster]

Firstly 9,692 is not a cap, its a break point, the only real cap is GCD (and even that only applies in currect gear with meta+ lust active).
As for Mr Robot, avoid that site at all costs, failure to do so may result in IRL Death.

Askmrrobot is fine as long as you know what you're doing. Blindly following suggestions without understanding stat weights is a recipe for disaster. Several times (using my own custom stat weights) it has given me a superior reforge / regem. Sometimes it has screwed stuff up. Just like anything, you take it with a grain of salt. Simcraft can be completely useless if you don't understand how to configure or interpret results properly. All these tools give you a rope -- it's up to you to not hang yourself with it.

[land]

Firstly 9,692 is not a cap, its a break point, the only real cap is GCD (and even that only applies in currect gear with meta+ lust active).
As for Mr Robot, avoid that site at all costs, failure to do so may result in IRL Death.

Thanks for correcting me, yes its a breakpoint. And failing to learn how to use a tool for sure will lead to bad results. Learning it, customizing the weighs, locking down gems you might not want to change etc makes that tool pretty decent imo.

[manamonster]

So I measured my dps in WrathCalcs as I exchanged each point of crit for a point of haste.

The theoretical 1.6% difference in WrathCalcs is small enough that I'd never be able to see it in a real fight where things like rotation execution, CD stacking, and luck with procs, movement, and fight mechanics will have effects many times greater than any potential difference from choosing a haste build over a crit build. I do see the boost when I compare 30 minutes of single target rotation on a training dummy, though that might not be a long enough test to really be sure.

So ultimately it probably just comes down to whichever style of play you like better. You're not going to noticeably gimp yourself either way.

Two things I notice in your analysis as being iffy. First of all, WrathCalcs is great for a rough estimation, but it will never hold up against Simcraft. Since Simcraft can literally run hundreds of thousands of sims and give you an average, it will always be more accurate (in addition to giving you multiple scenarios like modeling Movement in a fight).

You also admit that the 1.6% difference is too close to count. End of the day, WrathCalcs doesn't account for trinket procs (outside of averages), movement, etc. The reason Berserking is better as a racial is because it stacks multiplicatively with other buffs --- not because it is a "average haste" throughout the fight (as some websites try to compare when comparing different racials). Since Simcraft tries to model as if you're in the actual game, casting every spell, while account for randomness and RNG, it is much more accurate. It will run 250,000+ simulations of fights with / without movement and actually account for multiplicative bonuses with buffs. Especially since the introduction of RPPM trinkets (and weapon buffs), tools like WrathCalcs become even less accurate. There's no way (that I know) to accurately model something that can be as horribly random as a RPPM trinket with an Excel spreadsheet.