Diablo® III

Skill Proc Coefficients [Complete List]

(Locked)

I thought Cleave: Broad Sweep proc coefficient was 50%?

Says 80% on the doc.
12/08/2012 06:31 PMPosted by PlonX
The angry chicken rune on hex do have proc when you explode them, it's 33% or 0.33 for each enemy hit
Thank you, not sure how I missed that.

I thought Cleave: Broad Sweep proc coefficient was 50%?

Says 80% on the doc.
I re-checked and 0.800 is correct.

------------------------------------
Updated: 12/8/2012
- Witch Doctor, Hex -> Angry Chicken. Fixed proc coefficient (to 0.333 from 0.000)
Edited by ZzEzZ#1377 on 12/8/2012 10:16 PM PST
Probably a stupid question here but I can only see the active skill chances, are there any for passives? The one I'm looking for is Juggernaut ie what % it has of healing you 15% of HP. I'm getting tired of facerolling everything then having very rare occurrences of near death after being chain-frozen and jailed.
Thank you so, so much for this.
I have a question and a couple comments though:

First, and sorry if this has been answered already somewhere, but I'm a bit confused by the difference between the LoH and item/skill proc rates. I always assumed those were the same, I could even have sworn having read at one point a blue post saying exactly this. Could you please elbaorate on why this isn't the case?

Second, like the monk's signature skills, many other skills have hidden attack speed modifiers. I happen to know first hand, from a lot of testing, about 2 of them: Dashing Strike has a hidden 1.1x ias coefficient, and Cyclone Strike has a hidden 1.2x ias coefficient. Now I assume not many people know this, since to the best of my knowledge I am the only one crazy enough to have built a full Cyclone Strike reduced casting cost set + spirit gen to be able to spam it 24/7 as my main skill :p

Now the issue is that your testing cleary shows a 0.125 loh/skill proc coefficient for it, as do the patch notes from when it was upgraded from 0 to 0.125 a while ago. But on the other hand my testing shows that CS hits 125 times per 100s, with a 1.0 a/s Skorn and no other ias gear, every time I run that test. Same for DS and it's 110 hits per 100s.
So it seems that when Blizz stated a 0.125 coeff in the patch notes, they actually set a base 0.100 per hit coeff for the skill and multiplied it by the hidden attack speed coeff.
And of course now I'm assuming the same could be true for many other skills, which could affect your results and make things interesting :)
Edited by Santosvella#2387 on 1/6/2013 8:58 AM PST
So I ran some tests with a 1.00 a/s 100 LoH weapon, obviously getting rid of any other ias gear or life regen items I could have.

CS gives me alternatively 12 and 13 LoH per ennemy per hit.
DS gives me exactly 100 LoH per ennemy per hit.

Both of those values are in line with the respectively stated 0.125 and 1 so your per hit charts are correct.
However, as I said in my previous post, my tsting has shown that CS has a 1.2x hidden a/s modifier and DS a 1.1x one, so the correct values for your (PP)-per second charts for those skills should be:

CS: 1.2*0.125 = 0.15
DS: 1.1*1 = 1.1

Of course I don't assume that many people would particularly care about those 2 skills in particular, but I doubt these are the only ones working with hidden attack speed modifiers, so more testing would be required to discover the true per second coeffe of every skill in the game.

Actually I just remembered that I know of a 3rd one already. Unfortunately I don't have my notes on hand, but Wave of Light has a negative a/s modifier too: if my memory srves me right, its around 0.67x for all runes except Explosive Light, which is 0.75x.

I hope this helps.
Edited by Santosvella#2387 on 1/6/2013 8:57 AM PST
12/30/2012 11:14 PMPosted by Lazybugger
Probably a stupid question here but I can only see the active skill chances, are there any for passives? The one I'm looking for is Juggernaut ie what % it has of healing you 15% of HP. I'm getting tired of facerolling everything then having very rare occurrences of near death after being chain-frozen and jailed.
Passives either have a flat percent chance to proc (Like Juggernaut) or, if they rely on "hit effects" like Crit Mass, then they have percent*proc coefficient chance to proc.

So http://us.battle.net/d3/en/class/barbarian/passive/juggernaut Juggernaut has a 15% chance to proc always. Keeps HC interesting eh? ;D

01/05/2013 03:15 PMPosted by Santosvella
the difference between the LoH and item/skill proc rates. I always assumed those were the same, I could even have sworn having read at one point a blue post saying exactly this. Could you please elbaorate on why this isn't the case?

First, np ^-^. Then, the proc coefficient for LoH and item/skill effects are the same, but the total proc they give per second is different. That's because it's not a chance to trigger Life on Hit, you get it every time. On the other hand skill/effects only have a chance to proc, so this means the math in calculating Life on Hit per second (an additive thing) and skill/effect proc chance per second (a percent thing) are different. The formulas are:

(Life on Hit)*PC*(# of hits in 1 second), ie Life on Hit + Life on Hit + Life on Hit, etc. for however many hits in 1 second
(Skill/Effect Chance)*(1-(1-PC)^(# of hits in 1 second)). This is the way you calculate the chance of something happening at least one time given x # of tries (where x# would be number of hits in 1 second)

-----------------
@ Your interesting hidden IAS finds:
I never knew about these (also grats on the unique CS build ;D). How did you stumble upon these? Just random testing?

So, omg Blizzard, what the hell. I just tested to confirm and I was able to pull off an average of 75 cyclone strikes in 60 seconds with exactly 1 attack per second. The http://us.battle.net/d3/en/class/monk/active/cyclone-strike tooltip of Cyclone Strike doesn't say anything about it being faster unlike http://us.battle.net/d3/en/class/monk/active/fists-of-thunder Fists of Thunder which does (but doesn't say by how much >.<).

So this won't help in knowing which skills to test. I guess everything skill/rune will need to be tested again for this hidden attack speed.

Just so everyone is clear, the "per hit/tick" page for every class should still be 100% accurate, but some of the values on the "per second" pages may be slightly off due to hidden attack speed modifiers.

Oh Blizzard, how you tease me...
**About the difference between LoH and S/E chance:

Thanks for clearing that up for me, and sorry, it should have been obvious if I had bothered to plug in my brain at the time, it's not as if I dealt with probabilities daily or anything -.-

** About the CS build and my findings:

Well first ty for the compliment, finding and making work builds that are way off the beaten path is my calling in arpgs. I'm that guy who rolled a dozen necros in D2 and not a single one of them ever summoned a single skeleton ^^

I didn't actually find it through random testing. When planning that build I built an Excel spreadsheet to input the total spirit gen and reduced casting cost on gear needed to at least be able to breakeven on CS, allowing me to to spam it 24/7. But when I tested it ingame, much to my surprise, I was slowly losing spirit. So after checking my spreadsheet calculations a billion times and starting to pull my hair out, I finally had the idea to test CS' animation speed and that was the Eureka! moment :)

** About other skills:

As I said, I can 100% certify that Dashing Strike has a 1.1 hidden ias coefficient.

Wave of Light is about 0.67 or 0.7 and the Explosive Light rune is 0.75. Unfortunately those are hard to test reliably (especially the latter) since it is tough to be able to spam then for a long duration. Also, I lost my paper notes from that time.

Lastly, I started to randomly test a few other skills earlier today and Poison Dart, for example, seems to reliably trigger 52.5 or 53 times per 50s period. So it would seem that many skills could be concerned by this effect, however subtle it may be.

I don't have a lvl60 char from each class at hand yet, but I'd definitely be willing to test every Monk and WD skill/rune for a start.
Edited by Santosvella#2387 on 1/6/2013 5:40 PM PST
Cool, thank you. I'll go through and check the other three classes when I get the time.
- Diablo III (Hardcore)
Posts: 1,582
View profile
Thanks for posting. Sticky requested.
I'll just start posting my results here as I'm getting them.

Methodology used was to test all of those in A3 town, with a 1.0 a/s weapon, getting rid of all followers and all non ressource-generating gear.
Each skill was spammed non-stop for a periode of 60s exactly, and the number of occurences of the skill animation manually counted during that timeframe.
For each skill I started with two 60s tests of the unruned version, and repeated the test two more times if there was any difference in the count.
After that each rune (even the ones obviously using the exact same animation) was tested at least once, and two more tests were run if any difference with the baseline count appeared.

For skills that aren't easily spammable because of high ressource cost, many tests over shorter periods of time have been ran, outliers excluded and the remaining results averaged.
For those and for cooldown skills that are non-spammable, a frame-by-frame count of video capture would give the most accurate results, but this is currently beyhond my capabilities.

Lastly, all skills were tested at varying mouse pointer range (e.g. dashing against a wall or towards a screen corner, lobbing a firebomb at the character's feet or as far away from him as possible, etc...) and so far this made absolutely no difference in the results.

---------- Witch Doctor ----------

Poison Darts - All runes : 64 occurences per 60s period, 1.067 ias factor

Corpse Spiders - All runes : 62 occurences per 60s period, 1.033 ias factor

Plague of Toads - All runes except RoT : 62 occurences per 60s period, 1.033 ias factor
Plague of Toads - Rain of Toads : 90 occurences per 60s period, 1.5 ias factor

Firebomb - All runes : 63 occurences per 60s period, 1.05 ias factor
This seems a bit weird since it looks like both "lob" spells are using the exact same animation, but Firebomb was consistently getting 1 more occurence per test period that Plague of Toads. Or maybe listening to jars breaking for 10mins straight had driven me insane already...

Zombie Charger - All runes : 64 occurences per 60s period, 1.067 ias factor

Spirit Barrage - All runes : 62 occurences per 60s period, 1.033 ias factor

Acid Cloud - All runes except KoD : 61 occurences per 60s period, 1.017 ias factor
Acid Could - Kiss of Death : 60 occurences per 60s period, 1.0 ias factor

Wall of Zombies - Pile On : 64 occurences per 60s period, 1.067 ias factor
This is obviously the only rune I tested since it is the only spammable one with reduced cooldown gear

Firebats - All runes except Dire Bats : need more testing, but it looks like 60 occurences per 60s period, 1.0 ias factor
Firebats Dire Bats : 60 occurences per 60s period, 1.0 ias factor

Haunt - All runes : 62 occurences per 60s period, 1.033 ias factor

Locust Swarm - All runes except RoT : 64 occurences per 60s period, 1.067 ias factor

And it looks like that will almost be it for the WD. I still need to test Summon Zombie Dogs and Sacrifice, but I don't have the reduced cd gear for it. If someone could either test it and give me the results or lend me a cheap set of ZD reduced cd for 15mins that would be much appreciated.

I'll start working on the Monk next.
Edited by Santosvella#2387 on 1/8/2013 5:19 PM PST
Passives either have a flat percent chance to proc (Like Juggernaut) or, if they rely on "hit effects" like Crit Mass, then they have percent*proc coefficient chance to proc.

So http://us.battle.net/d3/en/class/barbarian/passive/juggernaut Juggernaut has a 15% chance to proc always. Keeps HC interesting eh? ;D


Thanks for responding but I can't see where it says the 15% is the chance to occur, seems to me that it is the amount of HP recovered if it does occur. I read some posts earlier that said it had a good chance just wondering what that is. 15% chance for me doesn't justify using it (I'd probably be better off getting a couple of CC reduction gears and continuing to use TaN for the armour instead), while 50% might.
Passives either have a flat percent chance to proc (Like Juggernaut) or, if they rely on "hit effects" like Crit Mass, then they have percent*proc coefficient chance to proc.

So http://us.battle.net/d3/en/class/barbarian/passive/juggernaut Juggernaut has a 15% chance to proc always. Keeps HC interesting eh? ;D


Thanks for responding but I can't see where it says the 15% is the chance to occur, seems to me that it is the amount of HP recovered if it does occur. I read some posts earlier that said it had a good chance just wondering what that is. 15% chance for me doesn't justify using it (I'd probably be better off getting a couple of CC reduction gears and continuing to use TaN for the armour instead), while 50% might.
Oh sorry, I misread the tooltip. Hmm interesting. I've never noticed passives that work from chance that aren't given specifics of the proc chance. I don't know ): sorry

Are you aware of any other skills like this? If there are a few more, I can create a page for them all.

--------------
@Santsovella. Yikes...from what you have so far, it looks like there's going to be a lot of revisions needed to the PP charts T.T
Edited by ZzEzZ#1377 on 1/7/2013 4:54 PM PST
@ZzEzZ

Yeah, sorry about that but on the other hand, this will be a massive step forward in understanding the finer points of D3's mechanics... provided it doesn't all change in 1.07 :p

We'll also need to add a third chart per class, with those hidden ias coeffs and the "true" dps of each skill but at least the good news is you already have the perfect spreadsheet template for that.
Oh sorry, I misread the tooltip. Hmm interesting. I've never noticed passives that work from chance that aren't given specifics of the proc chance. I don't know ): sorry

Are you aware of any other skills like this? If there are a few more, I can create a page for them all.


Yea I thought I might have trouble getting an answer on this one (I really think Blizzard should document stuff like this, they're not making it easy). I don't know of any other passives with "chance" in them but I suppose there could be, this is definitely an underused skill IMO. Thanks anyway for the help.
Edited by Lazybugger#1375 on 1/7/2013 6:43 PM PST
if life steal is not affected by the coefficient, how is it affected at all? is it a straight 1 to 1 dmg.
01/08/2013 02:11 PMPosted by Borderline
if life steal is not affected by the coefficient, how is it affected at all? is it a straight 1 to 1 dmg.


Yes, you get LS for all your damage, from all sources including those that normally don't generate LoH. However your total LS is reduced by 30/60/80% if you're in Nightmare/Hell/Inferno.
Updated: 2/12/2012
- Updated Wizard (per hit/tick) proc coefficients to reflect patch changes
- Updated per second proc rates. It turns out some skills are literally faster/slower than others. Ie: (47/60) on Seismic Slam would denote 47 Seismic Slams in 60 seconds.
-- Barb: Seismic Slam, Weapon throw
-- Monk: Fists of Thunder, Deadly Reach, Crippling Wave, Way of 100 Fists, Cyclone Strike, Dashing Strike
-- Witch Doctor: Plague of Toads (Rain of Toads), Firebats (Hungry Bats)

Note: All proc coefficients are exactly the same in PvE and PvP
Thanks for the update!

Great work!
Thanks for the update!

Great work!
Np :D

I got excited when I saw Blizzard say they added proc coefficients to the game guide.

http://us.battle.net//d3/en/blog/8649259

Turns out, they just defined them and gave a few examples. Ah well, spreadsheet will still be updated every patch until Blizz releases the procs themselves.
01/08/2013 05:12 PMPosted by Santosvella
Yes, you get LS for all your damage, from all sources including those that normally don't generate LoH.

Nope. Pets/hydra/turrets/reflect damages do not proc LS
This topic is locked.

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)

Reported!

[Close]