Community

 
Jump Menu:
Post Reply
Page 1 of 2  •  1 2 Next
Switch to Forum Live View The Sun Warlock:143+ at-will dpr
4 years ago  ::  Oct 30, 2009 - 1:52PM #1
furious_kender
Date Joined: Oct 25, 2007
Posts: 2,129

This is one of the sample builds I'm thinking of putting in my guide


This build was sort of requested by Ytterbium_Dragon. 


Goals of the build: do as much damage as possible using vulnerabilities while putting enemies in a lose-lose situation via divine challenge and arcane ripostes. 


I didn't pick encounters or dailies, because they do little to increase (and can even decrease) warlock DPR.


I picked level appropriate gear.


====== Created Using Wizards of the Coast D&D Character Builder ======
sun warlock, level 30
Dragonborn, Warlock, Morninglord, Demigod
Eldritch Blast: Eldritch Blast Constitution
Eldritch Pact: Infernal Pact
Arcane Admixture Damage Type: Arcane Admixture Thunder
Arcane Implement Proficiency: Arcane Implement Proficiency (Staff)
Divine Spark: Divine Spark Constitution
Divine Spark: Divine Spark Charisma
Dragon Breath Key Ability: Dragon Breath Constitution
Dragon Breath Damage Type: Dragon Breath Cold
Arcane Admixture Power: Hellish Rebuke


FINAL ABILITY SCORES
Str 15, Con 28, Dex 15, Int 10, Wis 12, Cha 24.


STARTING ABILITY SCORES
Str 11, Con 18, Dex 13, Int 8, Wis 10, Cha 12.



AC: 43 Fort: 41 Reflex: 37 Will: 40
HP: 185 Surges: 15 Surge Value: 55


TRAINED SKILLS
Arcana +20, Religion +20, Thievery +22, Intimidate +29, Endurance +29


UNTRAINED SKILLS
Acrobatics +17, Bluff +22, Diplomacy +22, Dungeoneering +16, Heal +16, History +17, Insight +16, Nature +16, Perception +16, Stealth +17, Streetwise +22, Athletics +17


 


FEATS
Level 1: Armor Proficiency (Chainmail)
Level 2: Arcane Implement Proficiency (staff)
Level 4: Implement Expertise (rod)
Level 6: Draconic Spellcaster
Level 8: Soldier of the Faith
Level 10:Dual Implement Spellcaster
Level 11: Radiant Breath
Level 12: Lasting Frost
Level 14: Called Shot
Level 16: White Lotus Riposte
Level 18: White Lotus Master Riposte
Level 20: Wintertouched 
Level 21: Warlock implement expertise
Level 22: Warding Curse
Level 24: Astral Fire
Level 26: Arcane Admixture
Level 28: Echoes of Thunder
Level 30:Weakening Challenge


 POWERS
Warlock utility 2: Fast Hands
Warlock utility 6: Red Leeches of Nihal
Warlock utility 10: Etheral Sidestep
Warlock utility 16: Warp Space
Warlock utility 22: Wings of the Fiend


ITEMS
Rod of the Dragonborn +6, Eladrin Spiritmail +6, Boots of Caiphon (epic tier), Staff of Ruin +6, Symbol of Divine Light +4, Hands of Hadar (epic tier), Razordark Bracers (paragon tier), Cloak+5, Ring of the Radiant Storm (paragon tier), War ring, ankmon's bracers, Siberys Shard of the Mage
====== Copy to Clipboard and Press the Import Button on the Summary Tab ======


At will teleport of 4 spaces, which is enough to trigger shadowwalk consistently


Hellish rebuke does fire, thunder, cold and radiant damage. It causes cold vulnerability 5 and radiant vulnerability 15. 


If needed, boots of caiphon are used to always trigger hellish rebuke secondary damage via a minor action each turn.


Hellish Rebuke


To hit: 15 (base)+9 (con)+1 (prime shot)+ 6 (enhancement) +2 (CA)+3 (rod exp)+3(draconic spellcaster)+1 (bloodied)=40 versus 41 reflex= 95% hit ratio


Roll all radiant damage twice and use the better result (ring of the radiant storm)


Primary Damage: 7 (base)+9(con)+12 (enhancement)+6(item)+ 10.5 (curse) +5 (called shot)+5 (cold vulnerability)+15 (radiant vulnerability)+3(astral fire)+3 (echoes of thunder)+7(hands of hadar)+5.5 (Ankmon's Bracers)+5(dragonshard)+4(rolling 7d6+1d10 twice approximation given by Molecule)= 97 average damage


Critical hit damage= 12 (base)+9(con)+12 (enhancement)+6(item)+ 18(curse) +5 (called shot)+5 (cold vulnerability)+15 (radiant vulnerability)+3(astral fire)+3 (echoes of thunder)+12(hands of hadar)+10 (Ankmon's Bracers)+ 21(critical bonus dice)+3.5(war ring)+5(dragonshard)+2.55(rolling 7D6 twice approximation given by Molecule)= 142.05 


Secondary Damage: 7 (base)+9(con)+12 (enhancement)+6(item)+ 5 (cold vulnerability)+15 (radiant vulnerability)+3(astral fire)+3 (echoes of thunder)+5(dragonshard)+1.4(rolling 2D6 twice approximation given by Molecule)= 66.4 average damage.


Hellish rebuke DPR= 159.735


.10*(142.05+66.4)=20.845


.85*(97+66.4) =138.89


95% of the time, the enemy must either take Divine challenge or arcane riposte (9 (con)+15 (radiant vuln)+5 (cold vuln)=29 damage) & master riposte damage (159.33 damage).  Assuming they always take the smaller damage, which is divine challenge, they will take 9(base)+6(cha)+15 (radiant vulnerability)=30 damage, they also will be weakened.


Average DPR will be somewhere around 188.235


 


Defenses


Defenses are reasonably good once warding curse and shadowwalk are added in (+4 to all defenses).  This warlock is also extremely tough, with a good number of hp, a high surge value, and 15 surges. 


Mobility is decent, with an at-will teleport of 4 and a speed of 6. 


Please let me know if I made mistakes anywhere or could up my dpr somehow.


 


 


 

Quick Reply
Cancel
4 years ago  ::  Oct 30, 2009 - 5:09PM #2
Melos
Date Joined: Nov 7, 2005
Posts: 1,444

How are you able to apply Divine Challenge each round?  The damage from it only works for the first attack the enemy makes so you need to keep applying it.


You are attacking reflex 39 but the standard is Level + 12, so 42.  Not a big deal since this does not change the 95% chance to hit.


I don't think you would include the secondary Hellish Rebuke damage in critical hit calculations.


Item set weapons count for the bonuses even when sheathed.  Since staff implements can function as magic quarterstaffs you would be able to use the staff from the set like that.  You could use it to get an extra +1 or use Ankhmon's Bracers instead of the item set ones.


You are relying a lot on damage types adding keywords to the attack, which is not something that everyone agrees on.

Quick Reply
Cancel
4 years ago  ::  Oct 30, 2009 - 7:53PM #3
furious_kender
Date Joined: Oct 25, 2007
Posts: 2,129

Oct 30, 2009 -- 5:09PM, Melos wrote:


How are you able to apply Divine Challenge each round?  The damage from it only works for the first attack the enemy makes so you need to keep applying it.



You don't need to reapply DC.  I was however assuming I didn't kill the enemy when I hit it.


 


Oct 30, 2009 -- 5:09PM, Melos wrote:

You are attacking reflex 39 but the standard is Level + 12, so 42.  Not a big deal since this does not change the 95% chance to hit.




You're right.  The 39 instead of 41 was a typo, which doesn't change the dpr outcome.  For some reason the DPR builds on Borg's top striker list typically measure against NAD values between 40 and 41 instead of 42.  Until they start measuring against 42, I'll measure my builds against thier baseline of 41.   


Oct 30, 2009 -- 5:09PM, Melos wrote:


I don't think you would include the secondary Hellish Rebuke damage in critical hit calculations.




You don't max secondary damage on a crit, but you still get it.  Otherwise, you'd be better off not criticing at all with multi-damage roll powers like Hellish Rebuke.


Oct 30, 2009 -- 5:09PM, Melos wrote:


You could use it to get an extra +1 or use Ankhmon's Bracers instead of the item set ones.




 Wow, I forgot ankmon's bracers.  Good catch. 


Oct 30, 2009 -- 5:09PM, Melos wrote:


You are relying a lot on damage types adding keywords to the attack, which is not something that everyone agrees on.




I go by the following:


"A creature that is vulnerable to a specified damage type usually takes a specific amount of extra damage when it takes damage of that type, or it suffers a specific effect. For example, a creature that has vulnerable 10 radiant takes 10 extra radiant damage when an attack deals radiant damage to it or when it takes ongoing radiant damage."


In short, if it does damage of a type, the vulnerability is triggered. 

Quick Reply
Cancel
4 years ago  ::  Oct 31, 2009 - 1:26PM #4
Molecule
Date Joined: Jan 15, 2009
Posts: 1,962

If you're curious, my math suggests that if all the damage dealt by the power is treated as radiant (which I believe is the case but my knowledge here is kind of hazy), the expected value from rolling 7d6 twice (curse + power + gloves) is 27.05 (as opposed to the expected value of 24.5 from only rolling it once).  Also, the expected value from 1d10 twice pick the highest one is 7.15 rather than the 5.5 expected from just one die.

<Ioun> they're apparently making a MolIsCool pp
Quick Reply
Cancel
4 years ago  ::  Oct 31, 2009 - 5:44PM #5
furious_kender
Date Joined: Oct 25, 2007
Posts: 2,129

Oct 31, 2009 -- 1:26PM, Molecule wrote:


If you're curious, my math suggests that if all the damage dealt by the power is treated as radiant (which I believe is the case but my knowledge here is kind of hazy), the expected value from rolling 7d6 twice (curse + power + gloves) is 27.05 (as opposed to the expected value of 24.5 from only rolling it once).  Also, the expected value from 1d10 twice pick the highest one is 7.15 rather than the 5.5 expected from just one die.





That is cool.  I was playing around with math trying to figure out how to do it.  My thought was to generate a population and assume one roll is, on average, one standard deviation high and the lower rolls is, on average, a standard deviation low, but I don't think that actually is how it works.


If it's not too much trouble, I'd be curious to see how you got your numbers so I could add them into the totals and accurately depict their value in my handbook.


Oh I also forgot to add in a dragonshard, so I redid some of my defensive items to get the money, and my at-will DPR is now up to 154.75.


 


 


 


 

Quick Reply
Cancel
4 years ago  ::  Nov 01, 2009 - 3:55AM #6
Molecule
Date Joined: Jan 15, 2009
Posts: 1,962

I actually wrote a small program for it because doing it by hand would be intensely long.  But the method goes more or less like this.


Calculating the probability of any given die roll is actually the major difficulty for large numbers of dice.  The algorithm for calculating this probability is not super hard to derive but I won't go into it here; if you want to look at it though it's here: en.wikipedia.org/wiki/Dice#Probability


In short, my method is a brute force approach which iterates through every possible combination of two 5d6 rolls and calculates the odds of that outcome happening and the number that results.  There is a distinct possibility that there is a much more elegant way to do it, but I don't know what it is.


 I don't know how math inclined you are, so I won't go into the gory details unless you want me to, but as an analogy here is how it would work for 1d6 instead of 5d6:


The chances of the first roll being a 1 are 1/6.
The chances of the second roll being less than or equal to a 1 are 1/6.
The outcome of this result is a 1, so 1/6*1/6*1 = 1/36 is the contribution from this result.


The chances of the second roll being greater than a one is 5/6; to be more precise there is a 1/6 probability each of a 2, 3, 4, 5, and 6.
The outcomes of these results are 1/6*1/6*2, 1/6*1/6*3, 1/6*1/6*4, etc.


Adding up all of these combinations (in this case there are 36 combinations; with 5d6 it would be 24^2 or 576) gets you the expected value of the die rolls.


Performing these calculations will get you the same result as you would using an excel spreadsheet (which is easy for the situation where all numbers come up equally often); the expected result of 2d6 pick highest is 4.472


And then I just cycle through all the possible combinations of die rolls and add up the expected returns.


For 5d6, the major difference is that the probability is not evenly distributed among all numbers; there is a much higher chance of rolling a 17 than there is of rolling a 30.  This doesn't change the problem conceptually, it just means that instead of 1/6*1/6 we are looking at probabilities of 1/7776*5/7776 and the like.


Edit:  If you're interested in the exact program I could post the code if you'd like.  It's a pretty short Java program; 80 lines including whitespace and comments.

<Ioun> they're apparently making a MolIsCool pp
Quick Reply
Cancel
4 years ago  ::  Nov 01, 2009 - 8:28AM #7
Reinhart
Date Joined: Mar 4, 2003
Posts: 599

Nov 1, 2009 -- 3:55AM, Molecule wrote:


There is a distinct possibility that there is a much more elegant way to do it, but I don't know what it is.




There is an elegant solution but it involves deriving maximum likelihood estimators for each system.  Your brute force approach is probably less prone to human error.


en.wikipedia.org/wiki/Maximum_likelihood

Quick Reply
Cancel
4 years ago  ::  Nov 01, 2009 - 8:34AM #8
furious_kender
Date Joined: Oct 25, 2007
Posts: 2,129

Nov 1, 2009 -- 3:55AM, Molecule wrote:


I actually wrote a small program for it because doing it by hand would be intensely long.  But the method goes more or less like this.


Calculating the probability of any given die roll is actually the major difficulty for large numbers of dice.  The algorithm for calculating this probability is not super hard to derive but I won't go into it here; if you want to look at it though it's here: en.wikipedia.org/wiki/Dice#Probability


In short, my method is a brute force approach which iterates through every possible combination of two 5d6 rolls and calculates the odds of that outcome happening and the number that results.  There is a distinct possibility that there is a much more elegant way to do it, but I don't know what it is.


 I don't know how math inclined you are, so I won't go into the gory details unless you want me to, but as an analogy here is how it would work for 1d6 instead of 5d6:


The chances of the first roll being a 1 are 1/6.
The chances of the second roll being less than or equal to a 1 are 1/6.
The outcome of this result is a 1, so 1/6*1/6*1 = 1/36 is the contribution from this result.


The chances of the second roll being greater than a one is 5/6; to be more precise there is a 1/6 probability each of a 2, 3, 4, 5, and 6.
The outcomes of these results are 1/6*1/6*2, 1/6*1/6*3, 1/6*1/6*4, etc.


Adding up all of these combinations (in this case there are 36 combinations; with 5d6 it would be 24^2 or 576) gets you the expected value of the die rolls.


Performing these calculations will get you the same result as you would using an excel spreadsheet (which is easy for the situation where all numbers come up equally often); the expected result of 2d6 pick highest is 4.472


And then I just cycle through all the possible combinations of die rolls and add up the expected returns.


For 5d6, the major difference is that the probability is not evenly distributed among all numbers; there is a much higher chance of rolling a 17 than there is of rolling a 30.  This doesn't change the problem conceptually, it just means that instead of 1/6*1/6 we are looking at probabilities of 1/7776*5/7776 and the like.


Edit:  If you're interested in the exact program I could post the code if you'd like.  It's a pretty short Java program; 80 lines including whitespace and comments.




Well I'm convinced.  It gives the results I expected, a significant dpr (1-2) gain at small dice rolls with a bigger gain the bigger the dice, and a little bit larger when larger numbers of nice are rolled (2-3).  This isn't a bad dpr increase for the item level and the slot. 


Thanks so much for your help!


I will add these estimates into the numbers.  I'm also thinking Radiant One may be a good way to increase DPR some more.....if only it was still in the builder Tongue out

Quick Reply
Cancel
4 years ago  ::  Nov 02, 2009 - 4:01PM #9
furious_kender
Date Joined: Oct 25, 2007
Posts: 2,129

Hiya guys,


I just want to give a little bump to see if there were any more suggestions or corrections before I added this to my guide.


I also  awaiting confirmation from Molecule that rolling damage twice boosts Hellish Rebuke damage to DPR= 159.33


 


 


 

Quick Reply
Cancel
4 years ago  ::  Nov 02, 2009 - 11:54PM #10
Molecule
Date Joined: Jan 15, 2009
Posts: 1,962

I believe that the War Ring damage would be rerolled as well would it not?  The crit calc should be for 7d6 rather than 6d6 if that's the case.  Also, 2d6 rerolled twice averages a gain of 1.4, not 1 (I think my last post was a little confusing in that regard, when I said "2d6 pick highest" I meant essentially "1d6 roll twice"), so your secondary damage goes up by .4.

<Ioun> they're apparently making a MolIsCool pp
Quick Reply
Cancel
Page 1 of 2  •  1 2 Next
Jump Menu:
 
    Viewing this thread :: 0 registered and 1 guest
    No registered users viewing