[Ret 5.3 MoP] Retribution Concordance (RetCon) - Who ordered the Mogu'gai Panda? - Page 13

Balhale · 11/11/12, 5:50 PM

Originally Posted by Skyhammer

Hi guys, got a question about trinkets in this tier. Is there any clear way of determining accurate DPS values for trinkets once in raid gear? Even relative differences would suffice, but I'm struggling at the moment with finding accurate modelling. Most resources i've seen seem to focus on trying to isolate the average stat contribution from each trinket and valuing them accordingly. It seems to me, however, that their value would be more significantly changed by your choice of HA / SW / DP, your other trinkets and the specific fight.

As an example, I'm currently running Relic of Xuen and Wok. If i understand Execution Sentence correctly, the on-use haste doesn't benefit it at all, so I try to wait for Relic to proc and ES with the Str buff up. By that measure, taking Lei-Shin's Final Orders and praying for a proc overlap (possibly delaying ES by a few seconds, no more) and trying to pre-empt the ICD of both when popping class CD's would seem optimal.
However, on the fights where i find HA to be preferable to SW (Elegon being a good example - need HA for spark waves in 10 man and everything stacks for initial pull and final burn phase) , it seems like bursts of haste during HA are incredibly beneficial due to the sheer number of TV (or DS) that can be pumped out as a result. This would lead me to think that a combination of Wok and Darkmist Vortex (once it comes out, obviously) would pull ahead significantly while using HA.

Am i overvaluing haste in CD phases? Is there a simpler way of comparing trinkets? At the moment, the number of variables is really screwing with me and i can't come up with effective comparisons. Any input here would be greatly appreciated.
Cheers,

P.S. First post, hope i'm not being simple

Easy enough to just change which trinket you sim your character with. Lei Shin >> Wok though.

Nythan · 11/12/12, 11:48 PM

so, from what i've been reading, entering raid tier gear, 2 haste~>1 str. Does this mean that instead of using the 300 str food, we should be using a 250/275 haste food? because that would compare to ~500/550 str, right? or would taking away the 300 str buff lower the value of haste? i dont have much time to run stuff anymore, and was just wondering if someone has already thought about this.

Fierss · 11/13/12, 12:33 AM

2 haste is roughly equal to 1 strength. That means 250 haste food would be 125 strength, not 500.

aylen86 · 11/14/12, 6:41 PM

Originally Posted by Exemplar

2 and 4 piece tier bonuses could also tie in heavily. I notice both Tourism and aylen86 (presently on Armoury) do not have 2 or 4 piece bonuses. Bonus TV damage will encourage faster HP generation, thus Haste is good. More AWs will encourage more abilities within an AW window, thus Haste is good.

Got my hands on a heroic starshatter/2pc bonus and I can say (using simcraft results and 50k iterations) that the value of haste compared to strength wasn't changing much.
I've tried to balance the dps contribution of haste and strength with my gemming and somewhere around 9k haste I've reached a wall over and over again. Without the 2pc bonus this was around 8,500 haste. I have to admit that the difference between a full haste gem setup and a balanced one is very small (~200 dps) and lays well in the margin of error.
I could achieve a haste rating of 11,5k easily but I don't think it's worth at this time. Don't stack haste blindly and sim your gear, couldn't agree more.

Tourism · 11/20/12, 3:39 PM

So I was curious about the multi target rotation. You say switch to DS at two or more targets, does this take the two piece into account? If it doesn't could the two piece push DS to three or more targets?

Elidra · 11/21/12, 3:25 AM

Originally Posted by Tourism

So I was curious about the multi target rotation. You say switch to DS at two or more targets, does this take the two piece into account? If it doesn't could the two piece push DS to three or more targets?

Even with 2t14 DS is greater than TV on two targets provided Inquisition is up.

Savius · 11/21/12, 4:23 AM

I was thinking about the right number of targets to switch to HotR. My conclusion is that two targets are enough to switch from CS to HotR.

The formulas are (from the OP of this thread):
- CS: 125% wpn dmg + 791 (physical, normalized at 3.3s)
- HotR: 20% wpn dmg (physical) + targets*35% wpn dmg (holy)
For HotR both the physical and the holy portion proc Hand of Light so for the comparison of HotR vs CS we can ignore HoL to simplify calculations.

It's not exactly easy to compare those two because CS is physical, HotR is physical and holy. So here is how I looked at it: check the logs and see the how much larger are the physical numbers compared to the holy numbers of HotR and then get a formula for total dmg done of HotR as multiple of the physical part which is multiple of wpn dmg. For simplicity, I've ignored the weapon dmg normalization but it has more benefits for HotR than for CS.

I checked my logs, looking for more hits with the same time stamp like the ones below:
[19:26:32.981] Vibius Hammer of the Righteous Jasper Guardian 3888
[19:26:32.981] Vibius Hammer of the Righteous Jade Guardian 11555
[19:26:32.981] Vibius Hammer of the Righteous Cobalt Guardian 12287
[19:26:32.981] Vibius Hammer of the Righteous Jasper Guardian 13305

in this case the ratio is 11555/3888 = 2.97; 12287/3888 = 3.16; 13305/3888 = 3.42
I found the holy/physical ratio to range from about 4 on fully debuffed bosses to 3 on trash.

So, for the lower ratio of 3.0 the total dmg of HotR for one target expressed in terms of the physical dmg is:
(physical + holy dmg) = (1 + 3)*20% wpn dmg = 4*20% wpn dmg = 80% phy wpn dmg.
At two targets this becomes (1 + 2*3)*20% wpn dmg = 7 * 20% wpn dmg = 140 % wpn dmg and that beats the 125% of CS.

Other things to consider: CS is normalized at 3.3s, HotR is not. This favors HotR by almost 10% (3.6s vs 3.3s).
CS has an extra 791 flat dmg but at my current numbers this is less than 2% of weapon dmg so it doesn't change anything.

So, on two targets HotR wins over CS by a good margin.

Edit: I'd be curios to see these formulas by computing the holy/physical ratio through the armor dmg reduction values (which I don't know) instead of empirical values.
Edit2: text more clear and better log example

Exemplar · 11/21/12, 9:27 AM

Ask and ye shall receive. Napkin math CS vs HotR using presently understood values.

At how many targets does CS equal HotR?

CS: (633* + Weapon Damage + (AP/14 * 3.3)) * 125% * 110% (Sword of Light) * 104% (Judgements of the Bold) * 67.9% (fully debuffed Armour mitigation) * 1.x% Mastery
CS: (633 + Weapon + (AP/14 * 3.3)) * 97.1% * 1.x% Mastery

Let's pretend we're using a 3.3 weapon speed. Yes, this is a lie, but it simplifies the calculations. CS is normalized to 3.3, while HotR should use true speed (3.6 across the board at the moment).

HotR Base: (Weapon Damage + (AP/14 * 3.3)) * 20% * 110% (Sword of Light) * 104% (Judgements of the Bold) * 67.9% (filly debuffed Armour mitigation) * 1.x% Mastery
HotR Base: (Weapon + (AP/14 * 3.3)) * 15.5% * 1.x% Mastery

HotR Splash: (Weapon Damage + (AP/14 * 3.3)) * 35% * 105% (CoE) * 130% (Inquisition) * 1.x% Mastery * Targets
HotR Splash: (Weapon + (AP/14 * 3.3)) * 47.8% * 1.x% Mastery * Targets

We'll call (Weapon + (AP/14 * 3.3)) "Mod Weap" just to shorten it.

Balance point would be where CS = HotR base + (Splash * targets).

Balance: (633 + Mod Weap) * 97.1% * 1.x% Mastery = (Mod Weap * 15.5% * 1.x% Mastery) + (Mod Weap * 47.8% * 1.x% Mastery * Targets)

First thing to notice - both sides have identical 1.x% Mastery multiplier. The value of Mastery is inconsequential, the ratio between the abilities remains identical. Divide this out and it drops cleanly.

Balance: (633 + Mod Weap) * 97.1% = (Mod Weap * 15.5%) + (Mod Weap * 47.8% * Targets)
Balance: 614 + .971Mod Weap = .155Mod Weap + (.478Mod Weap * Targets)
Balance: 614 + .816Mod Weap = .478Mod Weap * Targets
Balance: (614/.478ModWeap) + 1.71 = Targets

That 614 damage divided by 1/2 your modified weapon DPS is an ugly value, but we can deal with this.

So we need somewhat more than 1.71 targets. X is modified weapon damage. 614/X needs to be 0.3 or greater to require more than 2 targets. Thus .3X = 614, or X <= 2047. We would need less than 2047 modified weapon damage to require more than 2 targets.

As long as modified weapon damage is greater than 2047, you can swap to HotR at 2 targets. Hint: questing greens and a large club should be well over 2047 damage.

Also recall we pretended we had a 3.3 speed weapon. At 3.6 speed, HotR gains a bit more.

Bottom Line: HotR should be an increase on 2 targets at every level of MoP gear.

* - wait, what's this 633? 633 * 125% = 791. It's cleaner math if we move it in with weapon damage. Lets us combine our multipliers.

Podima · 11/21/12, 9:33 AM

Originally Posted by Smyte

HoW > J > HoW > TV is a self-sustaining rotation. Never use CS during SW, J will always be off cooldown and is a significantly harder hitting ability. Replace J with Exo when Exo procs or is off CD.

Going back a little bit for what may be ultimately a silly question - is this statement accurate, given that our priority is CS > J? I want to make sure I'm using wings properly.

Also thanks to those who double-checked the multi-target rotation - good to know that rets don't need to be in that strange in-between rotation (CS+DS) for 2-3 targets anymore.

Edit: Since we're reexamining the multi-target rotation, could someone confirm that Seal of Righteousness is still not worth using unless you have 6 targets up for 30+ seconds?

Theck · 11/21/12, 12:47 PM

Originally Posted by Exemplar

Let's pretend we're using a 3.3 weapon speed. Yes, this is a lie, but it simplifies the calculations. CS is normalized to 3.3, while HotR should use true speed (3.6 across the board at the moment).

HotR has used normalized weapon speed since it was re-worked in beta to scale off of weapon damage instead of AP/SP. So you don't have to pretend, your calculation is correct as-is.

I did the same calculation for Protection several months ago - for us it's 2 targets at low Vengeance, but 3 targets at high Vengeance (>140k or so) because we have slightly different scaling. In practice, I think nobody bothers to switch back to CS at high Vengeance against two targets, though.

Balhale · 11/21/12, 2:38 PM

Originally Posted by Podima

Going back a little bit for what may be ultimately a silly question - is this statement accurate, given that our priority is CS > J? I want to make sure I'm using wings properly.

Also thanks to those who double-checked the multi-target rotation - good to know that rets don't need to be in that strange in-between rotation (CS+DS) for 2-3 targets anymore.

Edit: Since we're reexamining the multi-target rotation, could someone confirm that Seal of Righteousness is still not worth using unless you have 6 targets up for 30+ seconds?

The reason you use J > CS during wings is that you are GCD capped anyway so instead of prioritizing CS to give yourself more buttons to press you can prioritize J because it hits harder.

Kwayver · 11/21/12, 6:40 PM

Originally Posted by Balhale

The reason you use J > CS during wings is that you are GCD capped anyway so instead of prioritizing CS to give yourself more buttons to press you can prioritize J because it hits harder.

by this logic, does that rule apply for <30% hp targets also?

Kwayver · 11/21/12, 7:13 PM

I'm interested in posting some ret (and some class-wide) paladin macros. should i post them here or make a sub-thread in the paladin sub-forum?

in the meantime while on the discussion of swapping from 1 target rotation to 2+ rotation, i came up with some very handy macros to assist reaching keys in these circumstances.

/cast [modifier:shift] Divine Storm
/stopmacro [modifier:shift]
/cast Templars Verdict

This macro allows you to hold the shift button to swap between DS and TV. it can also be applied to CS/HotR, and the selected target of Holy Prism (but would work a little differently, you would change the target of Holy Prism for this mod)

This also updated the tooltip of the button which is a nifty feature. as CS/HotR share a cooldown, there is no conflict when swapping and changing "shift stance" with this macro.

Montoyal · 11/21/12, 7:14 PM

Now I'm confused. If judgement hits harder compared to CS why should we prioritize CS over judgement on the regular rotation? Does the difference of their cooldowns end up generating more HP in the long run if we prioritize CS?

My understanding was to always prioritize the strongest hp-generating attack unless we had 5 HP or ES was off CD.

Kwayver · 11/21/12, 7:44 PM

Originally Posted by Montoyal

Now I'm confused. If judgement hits harder compared to CS why should we prioritize CS over judgement on the regular rotation? Does the difference of their cooldowns end up generating more HP in the long run if we prioritize CS?

My understanding was to always prioritize the strongest hp-generating attack unless we had 5 HP or ES was off CD.

this is to prevent idle spots in our rotation. see forum article World of Logs Analysis (Help me not suck thread) for an in-depth description