Retribution DPS Theorycraft - Page 149

Ariasis · 04/29/08, 3:51 AM

The new 2.4.2 ptr patch has this following addition for Paladins:

"Seal of Crusader now increases the damage dealt by Crusader Strike by 40% while active."

-

If I remember correctly, CS does 110% weapon damage. Now with activating SotC you attack 40% faster, but weapon damage is also reduced by around 40%. This meant that your CS did much less damage then intended while SotC was active. This is fixed by Blizzard.

The fact remains that SotC is still not the seal of choice as a damage seal and only usefull as a Judgement.

I am just prot paladin, so correct me if I am wrong...

Avitus · 04/29/08, 4:00 AM

Originally Posted by antilog

I had a question on expertise and dps:

I currently have none because I am using an axe and do not have a Shard of Contempt, so would it be better to replace my Pendant of the Perilous / Shattered Sun neckpiece with the Brooch of Deftness from badges, or is the ap/crit from the previous neckpieces better no matter what?

Use Rawr?

Originally Posted by Sapp

Actually, it's more specific than that; it's an artifact of the old-style "Reckoning". Basically, Seal of Command can only occur once per combat round, but every single white attack made during that duration is allowed to roll for a Command attack until one actually succeeds.

Is this confirmed?

If so, does Rawr Ret take this into account? This pretty much turns everything we assume with SoC ppm on its head.

urukai · 04/29/08, 4:44 AM

I am wondering about Blackened Naaru Sliver - WoWWiki - Your guide to the World of Warcraft
Is there a trinket which has the same effect or a similar?I ask because i want to know if a SoB procc or a JoB would grant attackpower too in battletrance.

If it is so it shouldn't take very long to get the 10 stacks.

Prinsesa · 04/29/08, 4:55 AM

Originally Posted by urukai

I am wondering about Blackened Naaru Sliver - WoWWiki - Your guide to the World of Warcraft
Is there a trinket which has the same effect or a similar?I ask because i want to know if a SoB procc or a JoB would grant attackpower too in battletrance.

If it is so it shouldn't take very long to get the 10 stacks.

Until someone actually has this trinket, anything we say is simply speculation at this point.

The closest thing we have to compare is the [Darkmoon Card: Crusade]. Correct me if I'm wrong, but autoattacks, Crusader Strike, SoC procs and SoB procs all trigger the stacking AP bonus of the Crusade, so it would be a fair assumption that this trinket would follow the same behavior.

Anarkii · 04/29/08, 5:09 AM

@Avitus
Rawr doesn't take that into account. If I understand this correctly, the actual damage from SoC if this theory was true would be *less* than predicted?
Currently Rawr(and the spreadsheet) assume that WF can proc Command. If you're saying that when a white attack procs Command and windfury, the subsequent extra attack from WF has no chance of proccing Command, then the total number of SoC procs will be less than if SoC could proc off all WF procs.

Am I understanding this correctly?

Tpyo · 04/29/08, 5:22 AM

Originally Posted by burghy

Wouldn't a small (1s) internal cooldown do just that?

This was my understanding as well. I even recall the animation being slightly delayed sometimes as a result of the WF attack proccing SoC. I'll be testing it in the near future, as I'm sure some of you already have. Let us know your results.

Originally Posted by Anarkii

@Avitus
be *less* than predicted?
Currently Rawr(and the spreadsheet) assume that WF can proc Command. If you're saying that when a white attack procs Command and windfury, the subsequent extra attack from WF has no chance of proccing Command, then the total number of SoC procs will be less than if SoC could proc off all WF procs.

Am I understanding this correctly?

Yes. It would be another explanation to SoB's dominance if this is proven to be correct.

Mithar · 04/29/08, 8:21 AM

Originally Posted by Avitus

My only issue with ArP has always been a theoretical one, rather than pratical:

Basically, if ArP is that good while it only affects ~70% of our attacks, how much better is it for warriors who pretty much share our loot table?

As said this is purely theoretical, but I feel it's a pretty silly mechanic. For example SoC only does 70% weapon damage since it's balanced around being an attack that completely ignores armor.

However with the latest itemisation you can almost fully mitigate boss armor with all your attacks as a warrior, hmm.

Bolded for my emphasis

If you can stack enough ArP to almost fully mitigate boss armor, would there be a point where SotC haste + haste from other sources + WF + boss having no effective armor = a greater damage output than SoB/SoC?

Other posters have claimed that SotC will always result in less damage (even with the new changes) because white attacks are mitigated by armor. If you have an amorless boss, could SotC become viable?

• Zurm · 04/29/08, 8:31 AM

Originally Posted by Prinsesa

Correct me if I'm wrong, but autoattacks, Crusader Strike, SoC procs and SoB procs all trigger the stacking AP bonus of the Crusade,

You are wrong. SoB Procs do NOT trigger the darkmoon card. SoB also does not trigger JoW, or pretty much any other proc effect. Except maybe vindication (not sure, haven't tested so I'm not gonna assume either way).

And Mithar, while it is POSSIBLE that in some crazy scenario with silly amounts of ArP SotC might be better, think of the stats you would sacrifice to get there, and then re-evaluate what your dps might be at that point. ArP only became good (even though it was only introduced at that point) once people started breaking the ~3k AP mark raid buffed.

As a final side note, we killed Eredar Twins last night (finally!), and I frapsed the fight. If any of you are bored and want to watch the video of that kill, I should be posting it sometime this evening (or for you EU people, next morning).

Avitus · 04/29/08, 8:55 AM

Originally Posted by Anarkii

@Avitus
Rawr doesn't take that into account. If I understand this correctly, the actual damage from SoC if this theory was true would be *less* than predicted?
Currently Rawr(and the spreadsheet) assume that WF can proc Command. If you're saying that when a white attack procs Command and windfury, the subsequent extra attack from WF has no chance of proccing Command, then the total number of SoC procs will be less than if SoC could proc off all WF procs.

Am I understanding this correctly?

You're understanding it correctly, I'm pretty sure I've never seen White + SoC + WF + SoC. Can anyone confirm or deny this? A counter example would definitely disprove it if someone has one.

I'm not sure how you would model this however.

Originally Posted by Mithar

Bolded for my emphasis

If you can stack enough ArP to almost fully mitigate boss armor, would there be a point where SotC haste + haste from other sources + WF + boss having no effective armor = a greater damage output than SoB/SoC?

Other posters have claimed that SotC will always result in less damage (even with the new changes) because white attacks are mitigated by armor. If you have an amorless boss, could SotC become viable?

Blast from the past, digging up a post from page 103?

(A minor correction that this would be possible for 6600 armor bosses if you have the highest level arm pen in every slot, but not for more).

Anyway I think if you stack this much Arm Pen you will gimp your other stats. SotC definitely scales best with Arm Pen and Haste (better than SoC for this one), but I don't think it's going to be worth using still.

The reason is SotC effectively gives no real damage increase, even the extra WF procs will hit for less as people have pointed out in the last few pages.

Avitus · 04/29/08, 9:08 AM

Shameless plug: I've written a small mod to help with drum rotations.

For all of those posting about acquiring leather working recently and using drum rotations, you can check it out here: SimpleDrumAnnounce

Not sure if I should post this in the UI thread too since it's just a "mini mod" so to say

Nex_moongladeEU · 04/29/08, 9:51 AM

Not exactly on the topic of retribution paladin DPS theorycraft, but since it might be of interest. I noticed on yesterdays brutallus that my threat was through the roof. From one day to the next i was completely screwed and just kept pulling aggro over and over again.

The end of the raid, with brutallus down, despite two threat steals on my behalf gave pretty good proof of the threat generation being completely off.

Fight: Brutallus
Scenario: I pull threat and die after ~45 seconds, get combat resed and resume fighting one minute into the fight. Recieve an almost instant blessing of salvation. Maintain a meager 1600 DPS for the rest of the fight and STILL manage to pull aggro off the tanks before the enrage. So with tanks having a 1 minute head start, and me dealing 1600 DPS (with salvation and maxed out fanaticism) i manage to steal aggro off the tanks. It is not a tank issue as they've been keeping threat just fine with me dealing over 2k DPS on several tries other days, just yesterday i was royally screwed as soon as i even tried to DPS.

Has anyone else gotten any clear evidence of their threat reduction failing? All i had before yesterday was anecdotal evidence and a strong gut feeling that sometimes my threat just wasn't working as it should.

More details here: WoW-Europe.com Forums -> [Bug] Fanatisism

Flip · 04/29/08, 9:57 AM

Did you fat finger your taunt button?

Not that I've ever done that...

Lansky · 04/29/08, 10:15 AM

Nex, you were not by chance using Righteous Fury or perhaps a mod that would auto buff it when it was not active? When I initially went from mainly prot to mainly ret this auto buffing mod caused some funny situations. You must be buffing your threat somehow though as otherwise a raid doing 1.6k dps and still pulling off tanks would obviously not be able to kill Brutallus.

Have not had any personal evidence of Fanaticism failing. Only seriously been Retribution since the start of Sunwell however.

Nex_moongladeEU · 04/29/08, 10:19 AM

Originally Posted by Flip

Did you fat finger your taunt button?

Not that I've ever done that...

What the hell... it's bound to a grid/clique binding only... that should just not be possible. Thanks anyway, time to take a look at what addons i updated before that raid. Problem solved now, nothing to see. Move along.

crásh · 04/29/08, 10:24 AM

Deleted. sorry for the crap QQ
Have done my own research. Great site

• Zurm · 04/29/08, 10:44 AM

Rather than manually reply, I will ask you to look for main post of this thread on page 1 for rotation questions. For ALL gear questions, please refer to:

Rawr

It allows you to pull your character off the armory, and from there you can see upgrades, see which enchant is best, and all of that is shown on graphs so you can see how much of an upgrade it is exactly. There is also an option to do a stat comparison, which shows a graph that compares each stat point for point.

Avitus · 04/29/08, 10:47 AM

Originally Posted by Nex_moongladeEU

Originally Posted by Flip

Did you fat finger your taunt button?

Not that I've ever done that...

What the hell... it's bound to a grid/clique binding only... that should just not be possible. Thanks anyway, time to take a look at what addons i updated before that raid. Problem solved now, nothing to see. Move along.

Weird how that WWS lists Brut swinging at you 328 ms before RD went off, probably lag, but

Anyway shit happens

• Zurm · 04/29/08, 10:54 AM

That's why I don't have my taunt key bound... lessons from BT learned harshly

I rarely use it, so clicking is fine.

Nex_moongladeEU · 04/29/08, 11:30 AM

Originally Posted by Avitus

Weird how that WWS lists Brut swinging at you 328 ms before RD went off, probably lag, but

Anyway shit happens

Yeah, apparently. But i still have no idea how it could have happened. Got clique bound to grid only, grid is in my upper left side corner, i pretty much never release the right key other than to drink elixirs. To get the mouse pointer to G1, the tank group, i need to pass the entire grid frame, slowly edging my mounse pointer (while holding the right key down mind) from the centre of the screen to the very top left, and while hoovering anywhere over grid my seal would be a FoL and judgement would having me start tossing a HL. As well as any CS hit would have tossed righteous defence on a random raid member rather than the 7 pinpoint ones on the actual tank.

I still can't see how it happened. But that's it. Checked clique setting now when i got back to the game-puter. Settings looks okay. I know what happened, i just can't grasp "how" it happened.

Astrylian · 04/29/08, 11:45 AM

Since 2.4, RD works by casting it on your hostile target, and the server infers that you want to cast it on your target's friendly target. ie, WWS shows you cast it on the tank, but you probably had the boss targeted and cast it.

• Zurm · 04/29/08, 11:51 AM

Has anyone done any thorough testing post-2.4 on whether or not Vindication still procs JoW and darkmoon buffs even when the mob is immune? I haven't been able to tell from WWS parses, so before I went along to test it I was going to ask to make sure no one else did.

Fiola · 04/29/08, 12:15 PM

Originally Posted by flyingtoastr

There is a very simple solution to the larger numbers of Windfury procs.

Random numbers are random.

This is just another case of wanting to see a pattern where there is none. Windfury can not proc off yellow attacks. It was patched that way a long time ago, it is a given fact. If by some odd reason it is broken you can rest assured that it will be fixed rapidly. There is a reason no one uses Hamstring in PvE anymore.

On the other hand, it is entirely possible that WF increases the PPM of SoC, which should be looked into.

Alright, I dug out my statistics book and refreshed my rusty stats skills. Let's try this again.

WF procs will have a binomial distribution, as each swing has a chance p of success.
Binomial distribution - Wikipedia, the free encyclopedia

As such, we can use the following:
E[X] = np (This is expected value of X. ie: 100 swings with 20% chance to proc, expect 20 WF procs)
Var[X] = np(1-p)
StdDev[X] = sqrt(Var[X]) (For 100 swings, SD = 4)

Where n is the number of swings, p = 20%, and 1-p = 80%.

Once we know the standard deviation, we can use the three sigma rule.
68% of values should be within +- 1 SD of E[x]
95% of values should be within +- 2 SD of E[x]
99% of values should be within +- 3 SD of E[x]

68-95-99.7 rule - Wikipedia, the free encyclopedia

Originally Posted by Fiola

Horde SoB: (as a baseline)
Ashoram - WWS - 2000 DPS
112 swings, 19 WF procs (20.4%)

Fiorina - WWS - 1799 DPS
127 swings, 22 WF procs (20.9%)

Celina - WWS - 1307 DPS
122 swings, 23 WF procs (23.2%)

Kózu - WWS - 1699 DPS
119 swings, 19 WF procs (19%)

Fiorina - WWS - 1680 DPS
115 swings, 22 WF procs (23.7%)

For the first WWS report:

n = 112 - 19 = 93
p = 0.2 

E[X] = 18.6
Var[X] = 0.2 * 0.8 * 93 = 14.9 
StdDev[X] = 3.9    

E[x] + SD = 22.5 procs

# of SD diff = (WF procs - E[X]) / SD = (22.5 - 22) / 3.9 = 0.1

So SoB WWS 1 is 0.1 SD from expected value.

Skipping the math for the remainder of trials:

SoB WWS 2
22 WF procs, E[X] = 21, SD = 4.1, 0.24 SD difference

SoB WWS 3
23 WF procs, E[X] = 19.8, SD = 4.0, 0.8 SD difference 

SoB WWS 4 
19 WF procs, E[X] = 20, SD = 4.0, -0.25 SD difference

SoB WWS 5 
22 WF procs, E[X] = 18.6, SD = 3.9, .88 SD difference

So far so good - all SoB WWS are within 1 SD of expected values.

Alliance SoC:
Calzone - WWS - 1597 DPS
114 swings, 24 SoC procs (21%/25%), 20 WF procs (21%/16.9%)

Tekzor - WWS - 1546 DPS
115 swings, 30 SoC procs (26%/31%), 19 WF procs (19.8%/15%)

Calzone - WWS -1780 DPS
122 swings, 33 SoC procs (27%/35%), 28 WF procs (29.8%/22%)

Saltycracker - WWS - 1995 DPS
134 swings, 43 SoC procs (31.4%/42.2%), 32 WF procs (31.4%/22%)

Const - WWS - 1663 DPS
114 swings, 40 SoC procs (35%/47%), 29 WF procs (34%/23%)

Using the same math as above:

SoC WWS 1
20 WF procs, E[X] = 18.8, SD = 3.9, 0.3 SD difference

SoC WWS 2
19 WF procs, E[X] = 19.2, SD = 3.9,  -.05 SD difference

SoC WWS 3
28 WF procs, E[X] = 18.8, SD = 3.9, 2.37 SD difference 

SoC WWS 4 
32 WF procs, E[X] = 20.4, SD = 4.0, 2.87 SD difference

SoC WWS 5 
29 WF procs, E[X] = 17, SD = 3.7, 3.25 SD difference

SoC WWS 6 - (the set of data from my first post; 151 swings, 57 SoC procs, 34 WF procs)
34 WF procs, E[X] = 23.4, SD = 4.3, 2.45 SD difference

I didn't cherry pick this set of data. I just went down the list of search results and picked out SoC parses that showed WF procs and 10 mill raid damage done (full fight parse). I didn't look for "high DPS SoC Ret paladins", since the WWS list was sorted by total raid DPS. You want me to believe that 4/6 SoC paladins just "randomly" got lucky over a combined 400 independent swings?

So what happens if we include SoC procs? (ie: Assume SoC can proc WF)

SoC WWS 1
20 WF procs, E[X] = 23.6, SD = 4.3, -0.8 SD difference

SoC WWS 2
19 WF procs, E[X] = 25.2, SD = 4.9, -1.48 SD difference

SoC WWS 3
28 WF procs, E[X] = 25.4, SD = 4.5, 0.58 SD difference 

SoC WWS 4 
32 WF procs, E[X] = 29, SD = 4.8, 0.62 SD difference

SoC WWS 5 
29 WF procs, E[X] = 25, SD = 4.5, 0.89 SD difference

SoC WWS 6 - (the set of data from my first post; 151 swings, 57 SoC procs, 34 WF procs)
34 WF procs, E[X] = 34.8, SD = 5.3, -.15 SD difference

Only one set of data shows substantial deviation with this type of calculation, and under-proccing can at least be explained by non 100% WF totem uptime. There's no such explanation for a proc rate 2~3 SDs above expected. Saying "it's random" is just handwaving.

So far, your only counter-argument is that we "know that SoC can't proc WF". How do we know that? Where are the patch notes specifically changing SoC? SoC is not a controlled spammable ability like hamstring, and we know that seals often play by their own rules. For instance, SoR procs flametongue weapon. SoR used to proc weapon procs, and anecdotally, I'm sure it could proc WF back before they specifically changed SoB/SoR to not trigger weapon procs.

This isn't just one WWS. Surfing through Alliance reports, I can find more that show a 2~3 SD difference (# WF procs > 25). More importantly, we're only seeing this deviation on Alliance side. What's so special about Alliance paladins that we can sometimes get 30+% WF proc rates?

flyingtoastr · 04/29/08, 1:19 PM

WoW-Europe.com Forums -> 10/08 Windfury totem hotfix

Sorry Fiola, if it is acting as odd as you insist it is bugged. SoC is not listed as an "on next melee" ability so it should not be proccing WF.

Saltycracker · 04/29/08, 1:26 PM

Originally Posted by Fiola

This isn't just one WWS. Surfing through Alliance reports, I can find more that show a 2~3 SD difference (# WF procs > 25). More importantly, we're only seeing this deviation on Alliance side. What's so special about Alliance paladins that we can sometimes get 30+% WF proc rates?

Could Blizzard have ninja patched this to shore up the parity between SoC and SoB?

• Zurm · 04/29/08, 1:28 PM

Now if only the same "bug" would apply to SoB!