Demonology Warlock 7.3

[Guest JPax]

On 14/10/2016 at 1:27 PM, Realmouse said:

this has never been true for me. if i cast hand  every time i have dread stalkers out and 4 soul shards, i never get doom damage except during high haste procs (like bloodlust) my warlock is Gerbiler on Azjol-Nerub . he has low item level and poor stats still.

It's a dot and functions like all other dots, re-applies don't stop the damage timer, just increase the total duration for it to damage in.

 

Say you were an spriest and cast shadow word pain, it's got a bunch of ticks going on over it's duration. If you recast inbetween two ticks, it won't reset the interval there. It'll just extend the duration and carry on ticking as if it was the original shadow word pain. Same principle applies to Dooooooooooooom. Although the HoG'D doom talent is acting a bit hinky atm

 

[Guest JPax]

2 hours ago, Furty said:

I think you might want to check your calculations, or rather, how you do a calculation - you make many assumptions about absolute values of stats (for example, calculation the value of haste based off of ... imp damage?). Stat weights are (roughly) correct, and though they do fluctuate based on your current number of stats and the number of targets, the priority is the same as listed at all times (except Versatility is occasionally actually better than Mastery, depending on your ilvl).

I've just done a simple calc of stats, it comes out as Haste>Mastery>Crit>Versatility.

Crit is weaker by a wide margin than haste and mastery.

When I calculate a single wild imps dps, I'd use (Normal damage x normal chance + Critical damage x Critical chance) / (Cast time/(1+haste values)) x (1+Mastery) x (1+Versatility) to calculate the average dps a single wild imp will do.

[Furty 51]

2 minutes ago, Guest JPax said:

I've just done a simple calc of stats, it comes out as Haste>Mastery>Crit>Versatility.

Crit is weaker by a wide margin than haste and mastery.

When I calculate a single wild imps dps, I'd use (Normal damage x normal chance + Critical damage x Critical chance) / (Cast time/(1+haste values)) x (1+Mastery) x (1+Versatility) to calculate the average dps a single wild imp will do.

Can you show your work/sims? Not seeing this either.

[Guest JPax]

I didn't use sims, I used pen and paper and dummy samples. I should probably mention I'm using a different build than your suggestion as well. My talent build is 1, 2, 1, 2, 1, 2, 3 or:

Shadowy Inspiration

Improved Dreadstalkers

Demonic Circle

Power Trip

Demon Skin

Grimoire Of Service

Soul Conduit

 

With the Demonbolt build, it upped the damage of my casts but not my overall dps. This was when I was using a crit/haste build and your guide suggestions. I changed to the build detailed above and had better single target damage and aoe damage as well. When I found the new build worked better, I then decided to do manual calculations to see how the stats affected my dps samples. Overall:

For equal values ~9000 rating in each of the stats:

Crit showed a 28% increase

Haste showed a 49% increase

Mastery showed a 48% increase

Versatility showed a 29% increase

[Guest JPax]

Also from my findings:

Crit/Haste dps increase =  87.4%

Crit/Mastery dps increase =  85.3%

Mastery/Haste dps increase =  115.3%

[Orthios 271]

11 minutes ago, Guest JPax said:

I didn't use sims, I used pen and paper and dummy samples.

You just discredited everything you've said so far, unless you've managed to do 10,000+ dummy tests with you build.  Individual tests (or also, small numbers of tests, sub 1000) are unreliable as RNG has a heavy influence the outcome of each individual iteration.  And no, there is no reason to not trust Simcraft, as it gets updated very frequently around patches and hotfixes so that it is always up to date.

[Furty 51]

10 minutes ago, Orthios said:

You just discredited everything you've said so far, unless you've managed to do 10,000+ dummy tests with you build.  Individual tests (or also, small numbers of tests, sub 1000) are unreliable as RNG has a heavy influence the outcome of each individual iteration.  And no, there is no reason to not trust Simcraft, as it gets updated very frequently around patches and hotfixes so that it is always up to date.

I think it is fair to say SimCraft isn't quite demonstrative evidence unless you are providing your own APL, but I also agree it is the best place to start for quickly comparing a wide variety of talent builds and finding a determinate winner. Basically I think it is fair to look at someone else's SimC results with a skeptical eye, but not before first actually looking at their APL and parameters to see why the Sim is giving the results it does (and if they are logical results).

 

25 minutes ago, Guest JPax said:

Also from my findings:

Crit/Haste dps increase =  87.4%

Crit/Mastery dps increase =  85.3%

Mastery/Haste dps increase =  115.3%

I'm curious what leads you to the conclusion that Mastery is so strong. Even if I do simple pen & paper comparisons as you suggest, Mastery takes 194 rating for 1% increase for pets during empowerment. Looking at a few log samples, I think it's a fair approximation that pets do about 50-55% of our damage currently, so 2% Mastery or 388 rating for about a 1% damage increase. It takes 400 Versatility for a 1% flat damage increase, and since this has no preconditionals (re; Mastery only affects empowered pets) and it affects cleave it would seem that Mastery is an inferior all round stat though marginally stronger on a single target only (Nythendra, basically).

[Guest JPax]

7 minutes ago, Orthios said:

You just discredited everything you've said so far, unless you've managed to do 10,000+ dummy tests with you build.  Individual tests (or also, small numbers of tests, sub 1000) are unreliable as RNG has a heavy influence the outcome of each individual iteration.  And no, there is no reason to not trust Simcraft, as it gets updated very frequently around patches and hotfixes so that it is always up to date.

That's not completely true about small sample builds. If you know the variables and mechanics you can average damage and interactions reliably and extrapolate for indefinite periods of time. Once you identify the pattern plateaus in a set of data you can divide it into subsets of peaks and highs and apply values over those periods.

 

What you're talking about is real world interactions for data like plant growth, not a controlled test environment like a video game which has far fewer variables to deal with. And almost no random variables that can't be averaged. I don't use simcraft purely because I like to do my own calculations and statistic farming, that and my computer is a potato when it comes to processing 10k+ runs. I'm not doubting simcraft in any way, and if this guide is a direct result of the findings in SC then fair enough. But I have found that the build posted here is not the optimum build through personal testing. The only reason I investigated in the first place is because I had logical doubts about the values given here.

 

Demonology warlocks get a high value of crit inherently, crit scales inversely with itself. So does haste, but we start out on a lower value. Although haste is weighted heavier for rating, it double dips for it's use. A minor damage buff for our pets, faster damage intervals and faster soul shard generation resulting in more pets. I don't doubt hastes worth.

Mastery has a lower starting point than crit, scales better with rating than crit, but buffs a portion (although the lion's share) of our damage.

9000 crit is 27.692307692307692307692307692308... rounding along with the base value to 32.69% crit.

9000 mastery is, along with the base value, 59.40%

59.4:32.69 --->1 : 0.5503..

Against each other, crit is only more valuable if your pets do less than 55% of your total dps. And that's being generous, because I'm not accounting for the artifact weapons critical chance buffs. And before you say that discredits what I've just said, mathematically it lowers it's value in the mastery : crit ratio.

[Furty 51]

3 minutes ago, Guest JPax said:

That's not completely true about small sample builds. If you know the variables and mechanics you can average damage and interactions reliably and extrapolate for indefinite periods of time. Once you identify the pattern plateaus in a set of data you can divide it into subsets of peaks and highs and apply values over those periods.

 

What you're talking about is real world interactions for data like plant growth, not a controlled test environment like a video game which has far fewer variables to deal with. And almost no random variables that can't be averaged. I don't use simcraft purely because I like to do my own calculations and statistic farming, that and my computer is a potato when it comes to processing 10k+ runs. I'm not doubting simcraft in any way, and if this guide is a direct result of the findings in SC then fair enough. But I have found that the build posted here is not the optimum build through personal testing. The only reason I investigated in the first place is because I had logical doubts about the values given here.

 

Demonology warlocks get a high value of crit inherently, crit scales inversely with itself. So does haste, but we start out on a lower value. Although haste is weighted heavier for rating, it double dips for it's use. A minor damage buff for our pets, faster damage intervals and faster soul shard generation resulting in more pets. I don't doubt hastes worth.

Mastery has a lower starting point than crit, scales better with rating than crit, but buffs a portion (although the lion's share) of our damage.

9000 crit is 27.692307692307692307692307692308... rounding along with the base value to 32.69% crit.

9000 mastery is, along with the base value, 59.40%

59.4:32.69 --->1 : 0.5503..

Against each other, crit is only more valuable if your pets do less than 55% of your total dps. And that's being generous, because I'm not accounting for the artifact weapons critical chance buffs. And before you say that discredits what I've just said, mathematically it lowers it's value in the mastery : crit ratio.

Also curious why you think Crit scales inversely with itself. It's an analog DPS increase; it will give you a linear gain for every additional point of Critical Strike until cap barring any scale factors that cause an exponential effect (e.g. Pyretic Incantation for Mage).

[Guest JPax]

11 minutes ago, Furty said:

I think it is fair to say SimCraft isn't quite demonstrative evidence unless you are providing your own APL, but I also agree it is the best place to start for quickly comparing a wide variety of talent builds and finding a determinate winner. Basically I think it is fair to look at someone else's SimC results with a skeptical eye, but not before first actually looking at their APL and parameters to see why the Sim is giving the results it does (and if they are logical results).

 

I'm curious what leads you to the conclusion that Mastery is so strong. Even if I do simple pen & paper comparisons as you suggest, Mastery takes 194 rating for 1% increase for pets during empowerment. Looking at a few log samples, I think it's a fair approximation that pets do about 50-55% of our damage currently, so 2% Mastery or 388 rating for about a 1% damage increase. It takes 400 Versatility for a 1% flat damage increase, and since this has no preconditionals (re; Mastery only affects empowered pets) and it affects cleave it would seem that Mastery is an inferior all round stat though marginally stronger on a single target only (Nythendra, basically).

Because it's not fair to say that our pets do 50-55% of our damage. For me they do 70-75% of my dps.

 

Even with your Demonbolt build my pets did 60-65%. That may just be me however, although I can't see how much can change in my build/rotation to facilitate a more even distribution except not to cast my pet spells :D

[Furty 51]

4 minutes ago, Guest JPax said:

Because it's not fair to say that our pets do 50-55% of our damage. For me they do 70-75% of my dps.

 

Even with your Demonbolt build my pets did 60-65%. That may just be me however, although I can't see how much can change in my build/rotation to facilitate a more even distribution except not to cast my pet spells :D

I think you're doing something quite wrong if pets are 75% of your damage. I invite you to look at the 99th percentile parses on WCL as that is what I used as a reference for those numbers.

[Guest Jpax]

7 minutes ago, Furty said:

Also curious why you think Crit scales inversely with itself. It's an analog DPS increase; it will give you a linear gain for every additional point of Critical Strike until cap barring any scale factors that cause an exponential effect (e.g. Pyretic Incantation for Mage).

Because each percentile point increase is competing with the base that preceded it, which means it's not linear, it's inversely proportional. If you had 0% critical and then improved to 1% critical chance, at double damage for critical strikes, you'd average a 1% damage increase.

If you have 10% critical, then improved it to 11% critical you'd have 110% damage increased to 111% damage.

1/110 isn't 1%, so it's not linear. It's less than 1:1, and it's value is proportional to what

[Furty 51]

14 minutes ago, Guest Jpax said:

Because each percentile point increase is competing with the base that preceded it, which means it's not linear, it's inversely proportional. If you had 0% critical and then improved to 1% critical chance, at double damage for critical strikes, you'd average a 1% damage increase.

If you have 10% critical, then improved it to 11% critical you'd have 110% damage increased to 111% damage.

1/110 isn't 1%, so it's not linear. It's less than 1:1, and it's value is proportional to what

Too tired to explain how stats work but in conclusion I think I've pretty clearly illustrated why you're wrong about the stat priority and sims, logs, and logical all support that deduction. Would advise sticking to the outlined stat weights.

[Guest Jpax]

8 minutes ago, Furty said:

Too tired to explain how stats work but in conclusion I think I've pretty clearly illustrated why you're wrong about the stat priority and sims, logs, and logical all support that deduction. Would advise sticking to the outlined stat weights.

Get some rest then, because you can't explain away how statistics affect themselves. Crit may improve your damage by 1% per 1% equivalent rating you have, but the more of it you amass the more you need to maintain a straight up 1% overall improvement to your damage as you progress. If you want to believe that 1% increase in crit means 1% additional total damage done with no regard for how much you have in the first place, then you're the one doing poor maths.

 

I encourage you to prove me wrong, because I'm now curious as to how you could possibly think otherwise with regards to that logic.

[Furty 51]

16 minutes ago, Guest Jpax said:

Get some rest then, because you can't explain away how statistics affect themselves. Crit may improve your damage by 1% per 1% equivalent rating you have, but the more of it you amass the more you need to maintain a straight up 1% overall improvement to your damage as you progress. If you want to believe that 1% increase in crit means 1% additional total damage done with no regard for how much you have in the first place, then you're the one doing poor maths.

 

I encourage you to prove me wrong, because I'm now curious as to how you could possibly think otherwise with regards to that logic.

I don't think I said anywhere crit increases your "total" damage by 1% per point. 

[Guest jacksonville]

Random question, and sorry if its mentioned somewhere and I missed it, but do you have any guidelines for eventual softcaps etc? Right now I'm sitting at 30% haste. Is there a threshhold where the 1.15 stat weight goes down considerably, and I should start valuing haste below/equal to the other secondary stats or something?

[Valks 2,373]

On 10/29/2016 at 0:49 AM, Guest jacksonville said:

Random question, and sorry if its mentioned somewhere and I missed it, but do you have any guidelines for eventual softcaps etc? Right now I'm sitting at 30% haste. Is there a threshhold where the 1.15 stat weight goes down considerably, and I should start valuing haste below/equal to the other secondary stats or something?

Weights are constantly changing depending on the amount of the stat you have. It's very rare for you to get an upgrade and see none of your stat weights change. The ones listed in the guides are basically general weights that you can use across item levels to assess upgrades, if it is that you can't find weights yourself.

I'm not aware of any caps, @Furty might know otherwise though?

[Furty 51]

23 hours ago, Blainie said:

Weights are constantly changing depending on the amount of the stat you have. It's very rare for you to get an upgrade and see none of your stat weights change. The ones listed in the guides are basically general weights that you can use across item levels to assess upgrades, if it is that you can't find weights yourself.

I'm not aware of any caps, @Furty might know otherwise though?

No caps or breakpoints worth mentioning. 

[Nytemare 2]

When using Implosion as pure AOE (Mythic+, etc), are you suppose to use Demonic Empowerment after summon them, or just play the HoG and build 4 shards again fast (Demonwrath) throw another HoG and blow up old imps, repeat, while just making sure Felguard and Darkglare (if talented and used) is always under DE?

[Bazdouf 0]

On 01/11/2016 at 4:55 AM, Furty said:

No caps or breakpoints worth mentioning. 

I'd say that past a certain ilvl it's good not to go under 33% haste to have doom tick faster than demonic synergy, but it wouldn't really be a cap as haste is too good for demo right now.

[wolf5659 0]

Hey guys, 

I am a new lock, been playing the game for 12 years but decided to go lock this expac. I heard from everyone that 7.1 is demonology, so i went demonology and I tried my best to follow up with the guide, I am at 861 and I still can't do more than 150k damage in heroic raid... watched videos, other guides, etc. Yet, lower ilvl people are doing better than I am doing, I am at 30% haste 51% masterty can someone help me please?

Regarding my rotation:
I open with stalkers > demonic empowement > dmon bolt x 4 > hand of guldan

Edited November 24, 2016 by wolf5659

[Valks 2,373]

On 16.11.2016 at 3:05 AM, Nytemare said:

When using Implosion as pure AOE (Mythic+, etc), are you suppose to use Demonic Empowerment after summon them, or just play the HoG and build 4 shards again fast (Demonwrath) throw another HoG and blow up old imps, repeat, while just making sure Felguard and Darkglare (if talented and used) is always under DE?

Yes, you should Empower after Imp.

[Valks 2,373]

On 18.11.2016 at 3:33 AM, Bazdouf said:

I'd say that past a certain ilvl it's good not to go under 33% haste to have doom tick faster than demonic synergy, but it wouldn't really be a cap as haste is too good for demo right now.

I mean, for most classes there is just a huge graph of fluctuation depending on where your other stats lie. It kinda sucks when gearing since you need to keep checking your own stat weights.

[Valks 2,373]

1 hour ago, wolf5659 said:

can someone help me please?

It's really difficult to provide help with just the information you've given. Can you post a log of you playing your Demo on the Warlock forums? You'll get far more accurate help, and faster, there.

[Pomkin 0]

I'm just on this because I main Demonology Warlock and I'm pretty sure that a lot of Demonology users looked forward into the guide videos. They said to use Grimoire of Synergy than Grimoire of Service. At first I thought it was to improve single targets and when fighting mobs. However it has huge disadvantage on the artifact ability, Thal'kiel's Consumption. It might boost the damage, but not by a lot, since what it absorbs to deal damage is the demon that is activated out there. Sure, it can do a lot of damage, but Demonology's first priority of secondary stats are Haste. If you have more Critical Strike than Haste, then you should just go with Synergy than Service, but for those who lack with DPS, they can use Service to have that huge boost that can actually be put up to 200~520K DPS (depending on what your gear, intellect number, and crit number is), not from 100K to instant 500K. It's just gonna give you a huge boost at the start.  They do lack a lot of DPS when it comes to mid to last bossing because now that their higher Demon, Summon Doomguard or Summon Infernal, and also lost Service's Demon.  Maybe that's why people would want Synergy so they don't lack DPS. I don't know, just throwing it out there to discuss about using Synergy or Service.