Zombies + Sacrifice: unbeatable burst AOE?


The DiabloWikiwitch doctor has the ability to summon Zombie Dogs and Sacrifice them, obliterating everything nearby in a similar fashion to corpse explosion in Diablo 2. At first glance, Sacrifice seems strong, but not overwhelmingly so; however, its greatest feature is the high level of synergy it has with a number of other abilities.

With the appropriate skill selection, a flurry of zombie-explosions can net 25,000%+ weapon damage in a matter of moments. Skeptical? We were too – read on for the gory details.

As with all pre-launch content, keep in mind that much of this is pure speculation and theorycraft; much can change between now and the 15th.


TL;DR: Conservatively, 10,000-25,000% weapon damage burst is achievable by using:
1. Soul Harvest – 5 stacks
2. Sacrifice – Next of Kin
3. Zombie Handler (passive)
4. Pierce the Veil (passive)

Round up a bunch of mobs, spam the sacrifice button – that’s all there is to it. Dogs will keep passively resurrecting, so keep smashing that sacrifice button. Our assumptions to get these numbers are very conservative on the low end, and more realistic on the high end – read below for details.


 

Skill Selection and Build Efficiency

The idea behind the Zombie Dog/Sacrifice build is pretty basic – you summon Zombie Dogs and blow them up, damaging everything around them in an AOE (12 yards). Various skills allow the dogs to resurrect themselves, allowing you to blow them up again. You’ll notice below that there are two categories of damage: “1 cast” and “2 cast”. This refers to the idea that you can summon your dogs and wait for the summon spell to come off cooldown before blowing them up, effectively allowing you to re-summon and sacrifice them again. “1 cast” assumes the summon is on cooldown and you can only use this attack combination once, “2 cast” assumes you can re-cast Zombie Dogs immediately. All results will be expressed in terms of percent of weapon damage.

The core build:
1. Summon Zombie Dogs
2. Sacrifice – Next of Kin
RESULT: 1,265% (1 cast), 2,530% (2 cast)

There are several abilities that provide additional direct sources of potential damage for this build, though they are not equal in terms of the amount of damage they provide. The following abilities are listed in order of the amount of  their damage potential:

1 cast 2 cast Dmg Boost
Core Build 1265% 2530%
Soul Harvest (5 stack) 3163% 6326% 150%
Zombie Handler 1678% 3356% 33%
Soul Harvest (1 stack) 1645% 3290% 30%
Mass Confusion 1574% 3148% 24%
Big Bad Voodoo 1574% 3148% 24%
Pierce the Veil 1518% 3036% 20%
Circle of Life 1306% 2612% 3%

Key assumptions here are that Big Bad Voodoo and Mass Confusion will be affecting 75% of the enemies in the pack. If this number is more or less, these abilities are more or less valuable. At 100%, they are even with Zombie Handler and at 50% they are worse than Pierce the Veil. We also (conservatively) assume that each sacrificed dog is hitting 1 monster per explosion and killing 1 monster per explosion.

  • Soul Harvest is the clear winner in terms of potential damage, adding roughly the same or better than any other option and has the potential to provide much more.
  • Zombie Handler is the clear second choice, as it provides more damage than other options and has a guaranteed effect

Making those changes leaves us with the following build:

The efficient build
1. Summon Zombie Dogs
2. Sacrifice – Next of Kin
3. Soul Harvest (5 stacks assumed)
4. Zombie Handler
RESULT: 4,194% (1 cast), 8,388% (2 cast)

And leaves the following options:

1 cast 2 cast Dmg Boost
Efficient Build 4194% 8388%
Mass Confusion 5225% 10450% 25%
Big Bad Voodoo 5225% 10450% 25%
Pierce the Veil 5033% 10066% 20%
Circle of Life 4331% 8662% 3%

Here things get a bit trickier – the only obvious choice is to remove Circle of Life. Pierce the Veil provides a 20% damage boost to every spell you have, but comes at the cost of some mana. Mass Confusion and Big Bad Voodoo can provide more damage than Pierce the Veil, but requires that you kill monsters affected by their respective debuffs.

Overall, Pierce the Veil seems to be the winner, but at this point it’s down to preference.

The all-out build:
1. Summon Zombie Dogs
2. Sacrifice – Next of Kin
3. Soul Harvest (5 stacks assumed)
4. Zombie Handler
5. Pierce the Veil
6. Mass Confusion – Devolution
7. Big Bad Voodoo
8. Circle of Life
RESULT: 6,600% (1 cast), 13,200% (2 cast)

 

Playing with Assumptions

Note again our conservative assumptions:

  • Each dog hits 1 monster per explosion
  • Each dog kills 1 monster per explosion
  • 75% of enemies are affected by confusion and/or ritual zone
  • 5 Stacks of Soul Harvest
  • 0% Bonus weapon damage from other sources


If we change these to something more realistic for an AOE bomb on a monster pack:

  • Each dog hits 2 monsters per explosion
  • Each dog kills 1.5 monsters per explosion
  • 75% of enemies are affected by confusion and/or ritual zone
  • 5 Stacks of Soul Harvest
  • 50% Bonus weapon damage from other sources


Or take a look at ideal conditions:

  • Each dog hits 3 monsters per explosion
  • Each dog kills 2 monsters per explosion
  • 100% of enemies are affected by confusion and/or ritual zone
  • 5 Stacks of Soul Harvest
  • 50% Bonus weapon damage from other sources
Assumptions Core Build Efficient Build All-Out Build
Conservative 2530% 10065% 13200%
Realistic 7590% 24156% 31680%
Aggressive 11385% 36234% 47520%

As you can see, the damage scales upward rapidly as conditions improve. Click here to download a spreadsheet that allows you to tinker on your own (File, Save As).

 

Important factors for calculation:

  1. Pierce the Veil damage is multiplicitive
  2. Soul Harvest damage is additive
  3. No more than 4 dogs can be alive at any given time
  4. Each “cast cycle” assumes all monsters are dead after two cast cycles – Big Bad Voodoo and Mass Confusion will allow this combination to be chained for a longer period of time.

Page 2: Show me the math!

Tagged As: | Categories: Diablo 3

Comments

You're not logged in but can still post comments. Register or login to remember your details.
  1. Wow, this seems to be an amazing strategy against bosses/rough elite packs. Now I’m wondering about viability as a general build. To be it’s most efficient it depends on some decent cooldown abilities to function properly, meaning that going from one pack to the next and blowing up the dogs remaining after the last pack, without the debuffs of BBV and MC cooled down, probably won’t resurrect any dogs. HOWEVER, awesome looking strategy for boss/elites 🙂

    • Assuming, you actually kill each champ with the burst. On level special packs took a ton of dmg relative to your character to kill even in early act 1. 10k% weapon dmg may seem like tons but if other more spammable skills end up around 2.5k (just a random number), I doubt they plan on tuning the spammable to kill in 4 shots. I think the 2500% weapon dmg pick is reasonable if you look at the way the damage is calculated. The stat based damage boost can get pretty large and reduce the relative impact of soul harvest.

      Also in the elite context you need to think about the risks of standing there blowing up your meatshields over and over. You run out of burst and their hp isnt zero, youre probly dead. Also the fact that youre relying on luck to even be able to chain it more than once without a resummon. A 19% chance of fizzling is still a 19% of fizzling.

      TLDR: even weapon % dmg numbers mean nothing outside of context.  

    • This isn’t really any amazing strategy. It might be good for elite packs because it’s a kind of one shot big burst with a long cooldown, but if it doesn’t kill them, then you’re probably in a pickle with not much left in your repertoire.

      One problem is zombie handler. This doesn’t increase the number of dogs summoned by the skill, so you’re still only getting the 4 you start with + whatever revive. Where zombie handler becomes useful is with mass confusion and big bag voodoo (and circle of life to a lesser degree), which require killing things. Normally you’d go into battle with 3 or fewer dogs, and if you kill a boss or unique pack, you’ve probably already won so the extra dogs won’t matter.

      I think sacrifice is a viable strategy, but it’s like a long cooldown spell because of the dogs cooldown against anything tough. It’s far more useful for random monsters (especially in group play) when you can alternate summoning, big bad voodoo, and mass confusion to create dogs.

      I wouldn’t want to use this against a boss. 275% damage per dog exploding and ~6 dogs * 1.35 = 8 dogs. That’s 2200 damage before you count pierce the veil and soul harvest, which work for any other skill just the same. If you don’t happen to have the 3 dogs going in, then you only get an average of 4 dogs = 1100 damage. Zombie wall pileon is 765% with half the cooldown of the dogs, takes only 1 skill slot, and isn’t dependent on dog AI and keeping them alive.

      I’m not gunna play around with sacrifice much. Big bad voodoo and mass confusion cooldowns are far too long, and reliance on killing enemies for this to be effective means hard fights will just be even harder. If I’m already getting kills, I don’t need some awesome synergy to make me kill better; I need the help when things aren’t dying fast enough.

  2. me and a few others had been looking into the effectiveness of a sacrifice witch docter on the forums a  while back but its great to see the hard-coded evidence! its gonna be fun 😀

  3. You got a few mistakes 1: how the Hell does killing a mob increase your damage in any way shape or form for the 1st cast of Sacrifice? Also number of dogs rezzed should only effect sacrifice cast no 2 or later,
    Your next of kin numbers don’t work with info on Blizz Calc 35% chance of rezzing a killed dog 4*0.35=1.4 not 2.10 for 4 dogs and 1.05 for 3 dogs on average.
    The correct code for 1st cast damage is =(I2*K2)*B14 and for second cast damage is =(I7*K2)*B14 for a total of =I9+I10 over both.

    Any way the important number is its only 825% damage per dog per mob.

    BTW are you sure Sacifice is +30% damage per mob up to 5 stacks the Blizz Calc says 130 INT per mob up to 5 stacks.

    • I didn’t really understand all of what you said, but I’ll go ahead and respond to the bits that are clear:

      1. You’re missing a critical part of the calculation: killing a mob increases the damage by triggering abilities that spawn more dogs – the overall damage presented takes into the fact that you’ll be getting multiple waves of dogs to sacrifice.

      2. You clearly didn’t read the entire article – go check out page 2 for the math. I explain very clearly why it’s 2.1, not 1.6. It has to do with the dog spawns from sacrifice.

      3. As to the first cast/second cast: Again, please read the article. I specifically lay out what those numbers represent – each cast includes sacrificing until all dogs are dead. The “second cast” is NOT the “second wave of dogs after sacrificing”. It represents the ability to re-start the chain of sacrifice/respawn after all dogs have died the first time.

      4.  Blizzard’s calculator disagrees with what was in the Beta – as of the last patch, each stack of soul harvest provided 30INT, which is roughly 30% damage per stack (which is exactly what is in the spreadsheet).

      • Lets see 1, 2 & 3 ok I guess you confused me with how you said it. BTW I dont see the point of 2nd cast in this case as any one can see that it will double your damage.
        Anyway if I factor that in according to your conditions above you will get ~4.4 dogs back on first cast (1.4 from Kin and 1.5 from both Confusion and Voodoo) that is more than the 4 allow = possible infinite bomb chain as you always have 4 up until theres less then 4 mobs on screen.
        4: Number from the beta right OK but that was value at lvl 13 or so if I am right, Blizz Calc is the boost for a level 60. There will be a boost in effectiveness at higher levels for skills like this, since to get the build you listed you need to be level 58+ I think useing the Bliz Calc 130 INT boost for a Lvl 60 Char will make more sense then use a 30 INT boost of a Beta Character.

        • The point of second cast is just to illustrate the maximum potential burst in a short period of time – you’re correct that it’s simply doubling the damage.

          The calculations take the maximum possible dogs (it won’t allow a number larger than 4), but you’re correct – with the given assumptions, you will infinitely chain dogs until there are less than 4 mobs in the explosion area.

          Lastly, I see what you were saying now with that calculation – 130 INT does make much more sense than 30. I’ll update the calculations shortly to reflect that, thanks! As you said, that INT adjustment is going to be dependent on a number of factors; I’m merely trying to illustrate that the combination of skills has the potential to be very effective.

  4. Didn’t try to figure out all of Rockman’s either, but one part should be clarified cause it’s useful math to not get wrong. The highest probability of anything happening is 1.0 (or, 100%). Nothing in the world has a 210% chance of happening. The 2.1, as Jom just said, is number of dogs.  Probability of a dog spawning, btw, can be thought of as four not failing. The odds of a failed spawn are 1.0 – .35 = .65 So the odds of at least one success is 1.0 – (.65^4) = 82%, and 73% for 3 dogs

  5. Yes, but how do you kill mobs in the dog downtime?

    • The “efficient” build doesn’t take up all of your active skill slots – you would still have room to fit in some other abilities. That said, this may not be the best all-around build, but it’s certainly one to keep an eye on.

  6. You don’t, you just kite like 20 mobs together until you can 😀

  7. If we have Mass Confusion – Devolution then if we use ‘Plague of Toads – Addling Toads’ do enemies killed when confused by toads also have a 50% chance to spawn a zombie dog?

  8. Although I never actually crunched the numbers, the incredible potential of Sacrifice has been clear in my mind since the skill calculator came around.

    Something I don’t believe you mentioned is that Sacrifice costs NO mana, a Sacrifice based build should be able to get along without getting any +mana or mana regen (if you can get by with all Sacrifice and possibly a cheap filler). Looking for +damage over +resources makes the item-hunt simpler.

    I was sorely dissapointed by Sacrifice in the beta though. It might just be that there weren’t enough synergies etc. But the radius seemed to small to me. Thoughts? 

    TL;DR: Sacrifice-based builds can be incredibly mana efficient, but is the radius sufficient?

    • Edit button is gone?

      Things you should mention in your article: This build is extremely mana-cheap (see calc below), gold-pickup radius is an important affix and radius of Sacrifice is the most important factor when it comes to the viability of this build.

      Mana costs:
      Sacrifice, Soul Harvest, Big Bad Voodoo, Mass Confusion all have NO mana cost.
      Summon Zombie Dogs costs 176 (228,8 PtV) mana

      Now if you get Gargantuan as your 6th skill you will need 528 (686,4 PtV) in your mana pool (for very infrequent casts). If you want to supply with a direct damage spell Poison Dart or Haunt will allow you not to think about Mana at all (freeing up affix slots for damage and gold-pickup). 

      Gold-pickup radius increases the effectiveness of Circle of Life and allows you to play slightly more safe.

      I wonder if there will be anything (item-affixes) which will improve Sacrifice radius? Also, has anyone actually measured? 

      P.S. They have confirmed 4 dogs max in the Best Buy Q&A.

      Here is a stab at a Sacrifice build that is in line with your calculations: http://eu.battle.net/d3/en/calculator/witch-doctor#YVdjQT!aWd!ZcYZcZ
      * SZD – Final Gift: They’ll be dying a lot so we don’t have to.
      * SH – Soul to Waste: To make it more reliable.
      * Gargantuan – Restless Giant: Tanking when we get in trouble, movement speed helps. (benefits from Handler as well)
      The rest should be self-explanatory.

      • Exchange Circle of Life for Grave Injustice. The latter is far better with this build.

      • Soul to Waste is a…waste. Cooldown is only 15 seconds, so you can just recast it every 30 seconds. Siphon would probably be best for survivability later on.

  9. What happens in Inferno when the mobs are so strong, they don’t die in 1 hit? Or on an act boss that survives the initial 2 casts? You run for 60 seconds between cooldowns?

    • Obviously it’d be crazy to dedicate all your skills to this one trick – the “all-out” build is going overboard. This build uses very little mana, so you gotta find other ways to spend it. I’d leave at least two slots open – maybe Locust Swarm to mop up groups, and a direct damage skill for bosses.

      Upon closer inspection, I can’t really find any good DD skills. Maybe pick Locust Swarm (Pestilence) and Haunt (Draining Spirit). For boss fights, cast both DoTs on him and dance around for a few seconds. Could work.

  10. The funny thing is we totally had this in D2X and no one used it.  I know, I know, different game.  The concerning part about this is the Soul Harvest (stacks assumed).  Assumptions are the mothers of all F-ups.  Now I did not end up working with the WD as much as would have wanted to in the beta as I fell in love with the DH and wiz.  However, I found the use of full stacking Soul Harvest slightly tricky.  I know over time you develop the twitch and the timing but in inferno you may not have those luxuries.  I too will watch this build though.  All I can say is- thank god for D3 infinite respec.  😀

  11. Very nice 🙂
    I have another build using parts of this here, but I am sure my build can be improved alot

  12. Couple of things.
     
    How did you calculate Soul Harvest’s damage bonus? Since at level 60 it gives you 650 INT with a full stack it’s damage bonus is only 150% if you have 1300 INT and It’s power is of course inversely proportional to your INT, relatively speaking.
     
    The power of Circle of Life for a build like this is grossly misrepresented in my opinion. It gets much better as you increase your pickup radius. It was very useful even in the beta when I extended it’s range to 20 yards. If you constantly use only sacrifice to kill enemies (which is an unrealistic scenario in my opinion), then the extra damage it gives you isn’t too good. However, the survivability you gain with it will be invaluable, as you’ll be much less reliant on cooldowns to summon dogs.
     
    I would seriously consider the Gruesome feast passive for this build. You will be picking up health globes anyway, why not use them to boost your damage even further? Against normal enemies, where globes were plentiful in the beta it should beat Pierce the Veil. Just something to think about.
     
    Also, would you consider posting your build ideas on the forums as well? We can discuss them more comfortably there. The comment section here is a mess with all the threaded conversations. 🙂

    • Thanks for the great response! Lets see if I can hit all of your points…

       1) On Soul Harvest: Since we have no idea what our stats will look like at level 60, I operated under the (possibly very wrong) assumption that the damage bonus will be roughly equal throughout the game. If that isn’t the case, you’re absolutely right that it’s damage bonus will be more or less than what we assumed here.

      2) On Circle of Life: It just seems so lackluster for a passive slot. Even if we assume that the range is the entire screen, Circle of Life will resurrect one dog every 20 mobs. When you consider the other options (e.g. 20% passive damage reduction), I have a hard time justifying using it. That isn’t to say that you’re wrong – outside of the “burst combo”, Circle of Life IS much more useful; I’m just not sure it’s EVER good enough to take the place of one of the other passives.

      3)  On Gruesome Feast: Yep, that would be a bit more “free” damage.

      4) On Forums: Absolutely. In the future, I’ll make sure all of the build ideas are posted on the forums as well. 

  13. I like AOE spells that I can place in my own choice of Area. Dogs have a mind of their own, there isnt any “alt + click” to order dogs to a certain place, is there?
    Now, please tell me. Soul Harvest clearly being so amazing. Why would blizzard make such a skill that just seems to give you so much increase in dmg? Any WD not using Soul Harvest would be under powered? Any WD using Soul Harvest will do tons of dmg? Please try to explain the thought process of creating this skill and talk about the fact that shouldnt every single WD use this skill @ 5 stacks 24/7.? :roll:
     

    • Soul Harvest has a huge drawback, namely that you have to be close to 5 enemies to get the maximum bonus, which may not be something that you want or even possible at times.

      Realistically you won’t be able to keep the maximum bonus up at all times and in many cases getting close to enemies will be a loss in damage, since you’ll have many abilities that you could use from a range to kill enemies without a need to boost your damage.

      Soul Harvest is amazing, but in general not as good as many people theorize.

      It’s also worth noting, that it get worse and worse as your INT increases, since it’s a flat bonus.

      • ^ This is the part about Soul Harvest that people don’t talk about.  I am not the most technically proficient gamer and I’m sure I will improve with time once we log some miles with D3, but getting the 5 stack with harvest is not easy.  Add in Inferno difficulty and I think it will not be so easy to achieve what is theorized here.  I too will watch the scientists with great curiosity.  WD is one of my favorites, but sacrifice is NOT corpse explosion.  

    • From our limited beta exposure to Soul Harvest, I’m of the opinion that you’re absolutely correct. Soul Harvest’s potential damage is simply too much to NOT take. What remains to be seen is if that boost tapers off at later levels. If it does not, I think Soul Harvest will be one of the more popular spells in the Witch Doctor arsenal.

       

  14. How do you get both Soul Harvest and Sacrifice?  They’re both under the “Terror” skill choice…

    • By enabling Elective mode in the gameplay options you can circumvent that restriction. It’s disabled by default so that new players have less ways to mess up.

  15. In general, I think the build would be fun, but I have some serious reservations about using this build in anything but a funzie/I’ll probably die sort of way.  Monsters have 1 million hit points in inferno.  At 1K base dps, that’s a 1000 fold to kill.  Your “conservative” assumption that each dog kills 1 monster is way too high.  Each subsequent assumption is again, way too high.

    I’d have to agree with a couple of other posters that Circle of Life is not as good as Gruesome Feast.  Hell, Grave Injustice is better.  Also would have to agree that Final Gift on SZD with Gruesome Feast would likely do more damage.

    You are assuming that each dog will hit and/or kill a DIFFERENT monster per explosion.  Dog explosions overlap.  I honestly have no idea if the overlapping damage stacks, but let’s assume so.  That being said, the number of monsters hit is probably the same where before it was too aggressive, the overlap makes up for the difference.  Additionally, monsters killed is probably likely to be 4 dogs kills 1,2,3 monsters total (see 1M+ hp above).
    [The google doc says mobs killer per sacrifice where above it says per dog.  I’m too tired to dig into the formulas, but maybe it’s just a text thing.]

    Dogs run out and attack whatever, wherever.  That’s not a problem, we want them to be close so we can sac them on the monsters.  The real problem comes with the Confusion and BBV assumptions.  The formulas can be adjusted for different assumptions, but something closer to 50 to 75 is realistic for mass confusion and something closer 0 to 25% for BBV.  How do I pull those numbers out of the air?  Think about dog pathing and how aggressively they run across the map to get into the fight.  In order to pull off the Confusion + BBV, that would mean a Soul Harvest up close (not a big deal), a Confusion where they start moving around, not necessarily next to each other or toward you, plus a BBV somehow capturing all the confused monsters in it’s area of effect.  Not feeling it.

    Yes, I will play a sac build.  Probably something close to this (“Exploding Dogs 24/7 build)  http://us.battle.net/d3/en/calculator/witch-doctor#fZYVQj!aYU!ZaZZcZ  ).  But even that build has some big holes.  Sac builds for the fun!
     

  16. What about “provoke the pack” instead of “next of kin”?

  17. Sacrifice states that it only works with summoned zombie dogs. Therefore Circle of Life doesn’t work.

Comments are closed.