Procs + CB & 1 more Q

[Nikalia 1]

From my reads, Chaos Bolt should be cast only with procs (or to not reach 4 full embers). My question for the experts is: are there any procs that AREN'T worth casting a CB when they are the only things up?

 

Here are my available procs:

 

Jade Spirit

Ember Master (4pc)

Toxic Power (KTT)

Wrath of the Darkspear? (BBoY)

Tempus Rumpet (sp?) - Meta

 

Is CB worthwhile if ANY of those are up? Or only the bigger ones like trinkets?

 

I can't locate a log as of now but a lock 10 ilvls lower than me is creaming me on CB dmg. My immolate uptime is solid, I hardly ever sit on 2 stacks of Conflag, and I don't overwrite backdrafts (unless I'm @ 4 embers I'll CB). So CB & Procs is my self-diagnosis. 

-----------------------

Additional Q: Purified Bindings has yet to drop for me this entire xpac - we have the Heroic on farm of course but nothin! Is this alone going to be a huge increase to my dps, or should BBoY & KTT be keeping me up there rockin' on the charts? I wonder if I need to try for an LFR or Flex binding and use that instead of my ktt for better single targeting.

 

Thoughts? 

 

As always, thanks guys <3

 

edit: p.s. I cannot seem to find the right thread on this other Q but... if BBoY hits 10 stacks but I started my cast before it falls off, will it still hit harder or must I finish the cast before the proc falls.

Edited March 3, 2014 by Nikalia

[Lockybalboa 618]

PBi is number one thing to get. Flex, LFR, normal, heroic, doesn't matter. Just get one. The DPS gain from going from LFR to heroic WF isn't that huge, but getting one at all is major.

 

BBoY is a pain in the ass to min/max in heroics, I would lose it for the PBI ASAP.

 

Tempus Rumpet is a haste proc, it does nothing for CB.

 

Jade and Ember I only cast a CB with if I'm close to capping embers. The 4pc proc has a 10 second ICD with it so if you know it's been about 10 seconds since you got a proc and are at 28 ember bits, you can normally start a CB cast, hit 30 with ticks of Immo, and get the CB with the small proc up. It's not a major DPS gain at all, it's very micro manage and not going to really make an impact, but it's something.

 

Jade also I only use if I'm close to capping, I rather cast a CB when I'm at 29-34 ember bits with a JS proc then at 37 ember bits and no proc. Both serve the same goal, one is slightly stronger.

 

Only DUMP embers with a real proc like trinkets. 

[Nikalia 1]

 If you're very competent at your Warlock, start trying to time your Chaos Bolts with trinkets so that they overlap with Ember Master.  This is one of the last things you should concern yourself with as Destruction.

-Zagam, Fixing your DPS FAQ

 

 

 

  Chaos Bolt's damage is calculated when the cast is executed, not when it hits the target. Travel time is a non-point for  Chaos Bolt, so keep that in mind when preparing your big nuke launches.

-Destro guide

 

I'm glad I found these, and I'm so thankful for the advice on purified bindings!

TY Balboa for the excellent response!

Edited March 3, 2014 by Nikalia

[Lockybalboa 618]

I didnt see the edit when I posted, the buffs on YOU are counted when the cast leaves your hand.

[Lothrandir 12]

This is a little late, but worth noting I think:

 

PBI works on ICD, while KTT (I think these two are the optimum setup for destro) works on RPPM. There is a difference: it is possible to sort of say when PBI will proc, but it is neigh-on impossible with KTT; therefore PBI is a must in any shape or form (I'm sadly still sporting the Flex one myself )!

 

[Lothrandir 12]

Just found this from a blue post in the era of 5.2:

 

 

 

For 5.2, we made a change to trinkets to make them less predictable and less boring with the RPPM system. Sometimes you can get lucky and get several back to back procs. This feels cool. Sometimes you can also be unlucky, and get long streaks when the trinket just refuses to proc. A trinket that is intended to proc every 1 minute could in theory go 8 minutes or more without activating.

 

'less predictable and less boring' is really subjective Blizzard marketing talk, and to get several procs in a row and then go 8 minutes without is as Destro-uncool as it gets ...

[Zagam 1,982]

The difference between Flex and HWF PBI is very small.  Just having PBI is great. 

 

Also, PBI has trolled many of us.  Locky and I last night varied on horrible RNG when trying to burn Skeer on Heroic Paragons.  I had PBI not proc until Skeer was dead which is mathematically a chance of a snowball surviving in hell for 3 minutes.  He had a similar pull a few pulls before that.  PBI works as a 15% chance on spell hit to proc when not on the internal cooldown.  It SHOULD proc within 5-10 casts reliably, but every now and then, you'll get some shit RNG.

 

KTT, however, almost ALWAYS procs within the first few casts.  This is because RPPM trinkets have a built in proc protection in that the coefficient proc multiplier is always maximum before you pull.  With the coefficient being a number between 6 and 10, you're looking at an RPPM value of 0.92 * 6 or 10 meaning you have a RPPM rate between 6 and 10, or an average proc rate between every 6 and 10 seconds.  The first proc of KTT is almost guaranteed to happen within 10 seconds.  Any procs after that are purely RNG related.  If you ever get another KTT proc as soon as the first one wears off, you've struck gold - this has a low chance of happening, but it does happen.  The maximum duration you can go without getting an RPPM trinket proc is in the neighborhood of 3 minutes and 15 seconds.  At that point, the coefficient grows to where it's simply impossible to avoid a proc.  You will see an average proc rate of KTT with about every 65 seconds.  However, that can vary to 10 seconds and go up to 190 seconds.  This is why players must be able to adapt to changes in your rotation instead of building macros to just perform a standard rotation. 

[Lothrandir 12]

Your PBI woes are unusual tho. So far I've seen it reliably proc around when expected. I dread for the situation you found yourselves in

 

Doesn't change the fact imo, that this is what makes PBI so much better than KTT - the reliability. I had 4 KTT procs in a row (felt like in a row - real close is what I mean) on a dummy last night. This would be horrible in a raid setting. Do you know how 4 such procs affect the future proc rate? Is it then a certainty that you wont see a proc for 4 mins, or should you count your blessing and expect a proc in 1/0.92 minutes on average?

 

Also, I'm thinking if the reliability factor is so big that Frenzied Crystal is worth considering (in favor of the two RPPM based trinkets)? Any thoughts?

 

[Lockybalboa 618]

Also, I'm thinking if the reliability factor is so big that Frenzied Crystal is worth considering (in favor of the two RPPM based trinkets)? Any thoughts?

 

[Lothrandir 12]

Seem to reply to my own posts a lot ... but found this formula in the same blue post referenced earlier, which computes the proc chance factor applied to the base chance for an RPPM trinket:

 

MAX(1, 1+((TimeSinceLastSuccessfulProc/AverageProcInterval)-1.5)*3)

 

This seems to indicate that the answer to the question regarding the multiple procs in a row is indeed count your blessings and expect one in x seconds on average. It also means that the full base proc chance of the trinket is only increased after 1½ x average interval is reached, in the case of SoO RPPM this is around 1m38s. Still, as stated, it's very RNG after the first proc (and I find no mention of the base proc chance nowhere).

 

Edited for correctness.

Edited March 12, 2014 by Lothrandir

[Zagam 1,982]

To echo Locky's loud sentiments, FCoR is garbage.  Here's a small breakdown...

 

KTT passive - Multistrike which is typically between 4% and 6% of your damage.  This is significant.

KTT proc - Intellect proc for 10 seconds with 0.92 RPPM.  This will average to about a proc per minute.

 

PBI passive - Amplification which is just awesome.  End of story.  There are lots of small effects (flat Chaos Bolt booster with Crit bonus, flat bonus across all spells, etc.)

PBI proc - Intellect proc for 20 seconds with 115 sec ICD - should line up with Dark Soul in most cases.  Bad proc rate is very rare.

 

FCoR passive - Cleave which is terrible.  Even on a fight like Protectors, if they're stacked, Multistrike out-performs Cleave.  You have to have 5-6 targets up consistently for Cleave to be worthwhile.

FCoR proc - Intellect proc for 15 seconds on 85 sec ICD - this doesn't line up with much. 

 

You have to consider the passive and the proc.  Using FCoR is like ignoring your passive.  The proc also doesn't line up with much, and lining up CDs is always far superior.

[Lockybalboa 618]

The base proc is listed on the trinkets and is affected by ilvl.

[Zagam 1,982]

full base proc chance of the trinket is only increased after 1½ x average interval is reached, in the case of SoO RPPM this is around 1m38s.

This is correct.  At 98 seconds, the bonus coefficient kicks in.  Your proc rate chance is the same from T=0 until T=98.  From there on out, the proc rate goes up exponentially until you proc. 

[Lothrandir 12]

To echo Locky's loud sentiments, FCoR is garbage.  Here's a small breakdown...

 

...

 

You have to consider the passive and the proc.  Using FCoR is like ignoring your passive.  The proc also doesn't line up with much, and lining up CDs is always far superior.

 

No, I'm not ignoring the passive - I know crystal is garbage on the passive, but if the proc lined up perfectly with ember generation or whatever, there could've been something in it. PBI lining up with DS:I is a good argument, but I was never suggesting replacing PBI, only the RPPMs :) (And I wasn't really suggesting; only speculating).

 

The base proc is listed on the trinkets and is affected by ilvl.

 

It's only listed on the ICD based trinkets, not the RPPMs. It must be pretty low, since it scales up pretty fast over time.

[Lockybalboa 618]

It's only listed on the ICD based trinkets, not the RPPMs. It must be pretty low, since it scales up pretty fast over time.

 

Kardris' Toxic Totem

 

(Approximately 0.92 procs per  minute)

 

 

That's the "base" proc. It can be found at the bottom of every RPPM trinket.

[Lothrandir 12]

Yes, I'm talking about the proc chance, which is what the aforementioned formula addressed.

[Zagam 1,982]

Yeah, the 0.92 RPPM is for the 553 version.  This will be modified by the trinket's item level and then modified again with the proc rate protection formula.

 

For example, the HWF version (572) has an RPPM of 1.1, so your proc rate goes down to (60/1.1) seconds which adjusts the time interval when the coefficient multiplier is added to around 85 seconds instead of 95.

[Lothrandir 12]

Still there must be a built-in proc chance, which has nothing do with the RPPM. A proc chance like the 15% in PBI. The result of the formula posted earlier would be multiplied by this chance to get a final proc chance. I just still cant see anywhere what this base chance is (i.e. the chance to proc until 98/85/whatever seconds).

 

It's like when PBI states e.g. 15% proc chance with 115s ICD, I'm just missing the first part for e.g. KTT, while the ICD part translates into RPPM.

[Zagam 1,982]

No, the RPPM IS THE PROC RATE.  RPPM = Randomized Procs Per Minute.  PPM = Procs Per Minute.  RPPM is random in the proc rate itself.  The only reason PBI has an internal proc rate is because it could be depended on too easily to always proc every 115 seconds.  The proc rate is the limitation and provides SOME randomness.

 

KTT's proc rate is it's RPPM rate.  Period.  There is no other factor.

 

When you start a fight, the proc rate is set at almost 100% because the game calculates your trinkets on pull as if you were out of combat for 120 seconds.  This is 30 seconds past the 'limit' where the coefficient kicks in which is why the proc rate at the start is increased, but it's only increased by the coefficient multiplied by the RPPM rate.  The proc rate of KTT is the exact same from T=0 until T=breakpoint where coefficient kicks in.  Then, the proc rate is a factor of the coefficient times the RPPM value until the point a proc occurs.

[Lothrandir 12]

Dazed and confused ....

 

Would you say this article is out of date now?

 

http://us.battle.net/wow/en/forum/topic/8197741003

 

It explains the real procs per minute mechanics introduced in 5.2, and indeed talks about proc chances.

 

Sorry to keep pounding on this, but it is a fundamental understanding I'd like to have crystal clear ...

[Eddie 16]

That article is not out of date. RPPM trinkets do not have a set proc rate. Their chance to proc increases the longer it goes without proccing. There is no internal cooldown for those trinkets, they can proc back to back to back. Bindings is a different type of trinket. its proc has a 115 second cooldown, before it can proc. Once that cooldown is over, every spell has a 15% chance to proc when it hits. Once it procs again, the cooldown prevents it from going off until those 115 seconds have passed.

[Lothrandir 12]

Yes I know, but let me rephrase: what is the chance of a proc of Toxic Power on the spell cast immediately after a wear-off of Toxic Power?

 

[Zagam 1,982]

Crystal Clear Explanation

 

KTT or any RPPM trinket

Proc chance = not set, factored based on average proc rate assigned and factors in proc protection rate

Internal cooldown = non-existant

 

PBI or any ICD trinket

Proc chance = set, 15% on spell hit.  This means ANY spell hit that does damage has a 15% chance to proc the trinket as long as the trinket is not in the middle of an internal cooldown period

Internal cooldown = 115 seconds

 

RPPM trinkets have a built in proc rate called RPPM - this value is calculated as the average procs per minute.  In the case of an RPPM value being 0.92, it's proc rate isn't inherently known, but the average number of random procs it will receive is approximately 1 minute / 0.92 procs per minute which comes out to an average proc interval of 65.22 seconds.  This means, on average, your proc interval can be anywhere from 0 seconds to 999 seconds, with the later being very rare (think a bell curve in statistics).  RPPM trinkets used to give HUGE lag in proc intervals, but statistical studies will show that the longer time passes, the higher chance you have of occuring a random event with a set probability of happening.  They don't give a standard deviation or variance in the procs - I study those with logs.  The formula generated by them shows the point in time where proc chance is magnified in order to reduce the probability of going an extended period of time without a proc.  Therefore, if you exceed the average proc interval (RPPM value) by a factor of about 1.5, then a proc protection coefficient goes in place and grows based on time past this value.  This factor grows until a proc is impossible to not get, much like it is in the beginning of the fight. 

 

Do you understand RPPM now?  It's complicated and complex, but RPPM IS the proc rate.

[Zagam 1,982]

To add clarification...

 

The percentage chance to proc KTT from T=0 until T=1.5 X average proc internval is the same.  In the calculation in previous post, we calculated average proc interval to be about 65 seconds with the RPPM value of 0.92.  This means from T=0 until T=65, the proc chance is EXACTLY the same in that you could expect 1 proc within 65 seconds with some success.  From T=65 until T=98, the percentage would REMAIN THE EXACT SAME as from T=0 until T=65.  At T=98, a coefficient kicks in and begins INCREASING the percentage chance of proccing until the proc rate reaches 100% which is around T=190 (value not known to be exact).

 

RPPM has the word RANDOM in it for a reason - you can't calculate the proc rate of a random occurence on any one spell hit without figuring out exactly how many spell hits you land consecutively and knowing other values.  They give you the anticipated mean value which is 0.92, or 65 seconds as an expected interval.  Statistically speaking, that means the interval will NEVER be exactly 65 seconds as it can't ever be exactly any value within a random probability.  It can, however, be calculated in a confidence interval where you can safely calculate a 95% probability of a proc happening within a time range.  Unfortunately, with such a large mean value of 65 in a total range of approximately 190, a 95% confidence interval would include most of the potential values anyways, thus negating any effort in finding the interval.

[Belloc 6]

It's essentially pointless to try to figure out what the proc rate for RPPM trinkets is before the protection kicks in. It really, truly is random. I've seen situations where the proc will fade and then immediately proc again. There may have even been situations where the proc refreshes itself (though I cannot state that as a fact). Either way, it'll proc when it procs and that'll be once or several times between your previous proc and the protection mechanic.