I'm confused (quite possibly) in your calc you have the Rank 7 CoA (level 67) but compare it against the level 60 CoD? At 70 you would have CoD rank 2 which out DPS's CoA
btw, I'm still not sure about correct numbers for CoD at lvl 70. At thottbot I could find only lvl 60 CoD, although one of them (not available at trainer) has more damage (4200 dmg for 380 mana), but it is not clear if that is player castable spell.
So, anyone have info what is actual CoD base damage/mana for lvl 70 ?
I'm confused (quite possibly) in your calc you have the Rank 7 CoA (level 67) but compare it against the level 60 CoD? At 70 you would have CoD rank 2 which out DPS's CoA
btw, I'm still not sure about correct numbers for CoD at lvl 70. At thottbot I could find only lvl 60 CoD, although one of them (not available at trainer) has more damage (4200 dmg for 380 mana), but it is not clear if that is player castable spell.
So, anyone have info what is actual CoD base damage/mana for lvl 70 ?
btw, i recalculated numbers with changed 5% on SL for demonology and with inclreased bonus from DK from 165 to 200 - it now made gap in damage between demonology and affliction, resulting in demonology as best DPS option . ANd those mixed builds like 21/40/0 benefits same from those changes.
With my original numbers, before I entered suggested corrections, afflic was better than demonology related to DPS. Now, demo>afflic , and that also does not look as something I expected, just like I did not expect destruction to be lower DPS.
http://www.thottbot.com/beta?c=Warlock
Yeah, it's 4200 damage for 380 mana.
well, that is significant increase in damage compared to lvl 60, and while lvl 60 CoD was barely better than lvl 70 CoA, this lvl 70 CoD is clearly better than lvl 70 CoA.
Is there any way of calculating the expected DPS values taking into account "casting lag"? With the death of fastcast, I assume this would be interesting for high latency players, as there should be an expected damage lost, moreso evident on builds that are more "nuke" reliant vs DoT reliant.
I love you warlocks, and I love how you all do math, but if we ARE talking about raid bosses you need to be practical. Your CoA is not going to match what CoR or CoE or likely even CoS brings to the raid. So in most situations what you're able to curse with is actually up to the number of other warlocks in the raid.
Furthermore, the affliction warlocks bring shadow embrace, and malediction.
Malediction alone should be enough damage to compensate and improve over damage-total than most likely any other build. You have to stop thinking about your own DPS in these situations, and look at what you're contributing on the whole. Sometimes imps and CoE/malediction are more worth it than doing another 100 DPS... not to mention max DPS is likely not going to be max-personal DPS... and often times you're going to be aggro capped well beforehand.
Your corpse does 0 DPS, and when your MT dies - or you pull aggro, you're probably next as well.
Every class in the game (with Rogues being pretty low relative to the others), have very significant synergistic dps increasing abilities. From totems to blessings to misery, curses, shouts, etc. When people give feral druid spreadsheets they almost always ignore what LoTP brings to your raid, or what innervate provides in DPS when applied to a mage. Those both count to the feral druid's DPS. Just like TSA for hunters, and so on and so forth. Raid builds and raid environments are not the sum of 40 individuals but the network and merging of it all.
1st: for the Destruction-ruin setting you have a full listed as .5 and a none as 0, however you put 1 in the setting column. The multiplier at the top is also taking .5 and adding it to this value. Currently a 1 will set the crit multiplier to behave normally, a 0 will make you crit for half damage and a .5 will make you crit for hit damage... unless i'm reading this wrong.
Also 2nd: you are accounting for crit by multiplying by a % and doing another multiply to account for miss%. The game does a single roll to determine whether or not you have a hit crit or miss. The roll for partials is another roll however.
If you had 5 hit and 10 crit and you were up against a +3 level mob (17% base miss) then the roll table would be something like this:
1-10 crit
11-88 hit
89-100 miss
The following formula from my own spreadsheet works for computing the change in average damage from the effects of hit and crit:
This was for calculating the average damage modifier against a +3 level mob (thus the .83) and capped the bonus from +hit to 16 and the bonus from +crit to 99 which would allow the maximum modifier to be 1.98 which is a 99% crit rate and 1% miss rate.
Furthermore, the affliction warlocks bring shadow embrace, and malediction.
Malediction alone should be enough damage to compensate and improve over damage-total than most likely any other build. You have to stop thinking about your own DPS in these situations, and look at what you're contributing on the whole. Sometimes imps and CoE/malediction are more worth it than doing another 100 DPS... not to mention max DPS is likely not going to be max-personal DPS... and often times you're going to be aggro capped well beforehand.
This is extensively and quantitatively addressed in the "Warlocks in the Expansion - Theorycraft" thread over a few full pages - it's neither a new issue, no even a resolved on. There's many situations where a DoT curse outperforms a Maledicted CoS/E in a smaller group.
DOT and rot.
Travian: Phased Weasel, -144 | 61, Damascus.
I love you warlocks, and I love how you all do math, but if we ARE talking about raid bosses you need to be practical. Your CoA is not going to match what CoR or CoE or likely even CoS brings to the raid. So in most situations what you're able to curse with is actually up to the number of other warlocks in the raid.
Previous to 2.0 this was always true. With DoT scaling its questionable. If you read the full post and looked at all the data points you will notice provisions were left in for not casting CoA. ON TBC Dooms do 10k in 60 seconds. That is 166DPS. For casters their collective DPS must be >1660 DPS to hit the break even point where casting it will make a difference. In a 10 men that will never happen. In a 25 man, ya a warlock will probably be casting Raid curses instead of DPS curses.
I love you warlocks, and I love how you all do math, but if we ARE talking about raid bosses you need to be practical. Your CoA is not going to match what CoR or CoE or likely even CoS brings to the raid. So in most situations what you're able to curse with is actually up to the number of other warlocks in the raid.
Furthermore, the affliction warlocks bring shadow embrace, and malediction.
Malediction alone should be enough damage to compensate and improve over damage-total than most likely any other build. You have to stop thinking about your own DPS in these situations, and look at what you're contributing on the whole. Sometimes imps and CoE/malediction are more worth it than doing another 100 DPS... not to mention max DPS is likely not going to be max-personal DPS... and often times you're going to be aggro capped well beforehand.
This is all true, and is why in today's 2.00 40-man raiding environment, every warlock that you bring over 3 is going to do about 20% more dps than his fellows. I am by all rights a pretty decently geared warlock (6/9 T3, Polarity, Wraith Blade, C'thun gear, etc.) but I've taken time to show my guild that it doesn't matter which warlock we bring for class spots 4-6, each extra warlock you bring is going to be in the top 10 DPS for pretty much every encounter (excluding Gluth ... and mostly because aggro is a bitch, our DPS is sick anyways [we usually kill him right at the 2nd decimate] and I don't want to take ~any~ chances). It's not just "oh, bring Nfar, he does great damage"... any warlock not constrained by the holy trinity (CoE/CoR/CoS) is going to do nice top 10 or top 5 DPS if he has a modicum of skill. Anything more than 6 should really be Demo or Destruction spec to get the most out of 40 debuff slots, but even then our DPS is very good.
But in a 25 man scenario, we will rarely see CoDoom or CoAgony on a standard Tank and Spank encounter. HOWEVER, I think in 10 mans it will be just the opposite. Is 10% more damage from one fire or frost mage worth more than 20% more damage from my CoAgony? Particularly when that Agony is returning something like 300 mana to me over 24 seconds from JoW? Is 10% more damage from a rogue and a hunter (discounting a Prot warrior) worth more than 20% more damage from me? Especially when I'm getting 35% reduced aggro (6/9 Plagueheart + 10% from talents) for it?
Affliction is definitely the way to go for me. It offers true raid usefullness (although I still don't think I'll be speccing the points in Malediction in it's current state), it offers synergy with other warlocks via Soul Siphon, and it offers self-sustainability through Dark Pact and Siphon Life. Blizzard finally put in 40 debuffs... I would consider it a shame if warlocks ignored it and played purple mage all night.
1st: for the Destruction-ruin setting you have a full listed as .5 and a none as 0, however you put 1 in the setting column. The multiplier at the top is also taking .5 and adding it to this value. Currently a 1 will set the crit multiplier to behave normally, a 0 will make you crit for half damage and a .5 will make you crit for hit damage... unless i'm reading this wrong.
Cell for ruin talent value should have 0 if you dont have ruin talent, or 0.5 if you have. That will add in 'crit multi' cell to 0.5 as you noticed, which will result in value showing how much bonus damage you get from crit. If I remember correctly, we have 50% bonus crit damage without any talent, and ruin increase crit for '100%', basically making bonus crit damage 2*50%=100%.
According to that, I consider that when you crit. total crit damage is calculated as NoCritDamage+ NoCritDamage* 1 if you have ruin, or as NoCritDamage+ NoCritDamage* 0.5 if you do not have bonus, where 0.5 or 1 are my cell 'crit multi'. Based on that, I calculate average spell damage as
AvgSpellDmg= NoCritDamage+ NoCritDamage * CritMulti * Crit%
Also 2nd: you are accounting for crit by multiplying by a % and doing another multiply to account for miss%. The game does a single roll to determine whether or not you have a hit crit or miss. The roll for partials is another roll however.
If you had 5 hit and 10 crit and you were up against a +3 level mob (17% base miss) then the roll table would be something like this:
1-10 crit
11-88 hit
89-100 miss
yes, I know that, but basically difference is small for normal miss/crit values, and it always apply across all builds - and considering I initially did not treat misses at all, I ignored this case ;p
But I changed anyway in XLS, and i used simpler method , for which I hope i was correct. Basically I divided my 'Crit Multi' by ( 2- MissRate)
It is based on :
RealCrit% = Crit% / (1- Miss%)
If Miss%=1%, then RealCrit%= Crit%/(1-0.01)= 1.01*Crit%
In extreme case, where Crit%=40% and Miss%=60%, RealCrit%= 0.4/(1-0.6)= 1 = 100%, as it should be
Since MissRate in my XLS is shown as 1+Miss% (1.14), (1-Miss%)= (2-MissRate). And since 'Crit Multi' is multiplied to every Crit% in calculations, I just divided Crit Multi with that value.
Particularly when that Agony is returning something like 300 mana to me over 24 seconds from JoW?
I thought JoW gave a chance of restoring mana per spell/melee hit, which means you have 1 chance when you cast CoA to get 59 mana not 1 chance per dot tick.
A very valid point... my understanding was that 0/21/40 pumped out more damage in all situations where you couldn't use a pet. And that it was based around an incimolate cycle. Or is the increased efficiency from Imp LT too important?
I checked that other thread, and it is also very interesting, but with few different things:
- most scenarios considered there are not quite typical raid scenarios ... they include all posible raid buffs includig group only buffs, all possible debuffs on mob, very high +dmg gear of about 1000 and, most unrealistic, large nmber of potions, flasks and consumables. There is more realistic option, which still use potions and more buffs ,but is more comparable
-i could not find how to change scenarios there and recalculte results, or how to check used formulas. it is possible that some program or XLS was posted on later pages, but i couldnt find it
- there is only "max dps" consideration there, without "no heals" or "sustained"
But it is informative thread even if some results are different from what i get here... for example, when i include buffs/potions used there, i also get DPS for affliction around 1800,but demonology still gets higher at around 1900-2000
I included relevant suggestions that i found both here and in other threads in XLS file, and while all builds got fairly close in max dps, demonology still leads, which is different from that other thread. I would appreciate any specific comment pointng where my XLS is still wrong, as opposed to "lot of misinformation" type of comment which, even if we do not analyze how accurate it is,is certainly not helping in finding what is wrong ;)
I checked that other thread, and it is also very interesting, but with few different things:
- most scenarios considered there are not quite typical raid scenarios ... they include all posible raid buffs includig group only buffs, all possible debuffs on mob, very high +dmg gear of about 1000 and, most unrealistic, large nmber of potions, flasks and consumables. There is more realistic option, which still use potions and more buffs ,but is more comparable
-i could not find how to change scenarios there and recalculte results, or how to check used formulas. it is possible that some program or XLS was posted on later pages, but i couldnt find it
- there is only "max dps" consideration there, without "no heals" or "sustained"
But it is informative thread even if some results are different from what i get here... for example, when i include buffs/potions used there, i also get DPS for affliction around 1800,but demonology still gets higher at around 1900-2000
I included relevant suggestions that i found both here and in other threads in XLS file, and while all builds got fairly close in max dps, demonology still leads, which is different from that other thread. I would appreciate any specific comment pointng where my XLS is still wrong, as opposed to "lot of misinformation" type of comment which, even if we do not analyze how accurate it is,is certainly not helping in finding what is wrong ;)
Unfortunately when there is another thread that is mostly consistent it's probably not worth the effort to fix all the minor inconsistencies here. Given the game is still in flux and there are many areas your knowledge of the mechanics/class seems to be sub optimal then there is even less.
Particularly when that Agony is returning something like 300 mana to me over 24 seconds from JoW?
I thought JoW gave a chance of restoring mana per spell/melee hit, which means you have 1 chance when you cast CoA to get 59 mana not 1 chance per dot tick.
Coa used to give a chance for mana on each hit of the spell which was absolutely wonderful :D It however has not done this for a long time.
Particularly when that Agony is returning something like 300 mana to me over 24 seconds from JoW?
I thought JoW gave a chance of restoring mana per spell/melee hit, which means you have 1 chance when you cast CoA to get 59 mana not 1 chance per dot tick.
Coa used to give a chance for mana on each hit of the spell which was absolutely wonderful :D It however has not done this for a long time.
Ah, hadn't noticed. I remember it behaving the old way, back in MC when we figured out that you could judge multiple ranks of JoW at the same time for some truly ridiculous mana regen.
I've been tooling around with my warlock alt (aff spec). Never raided with him, so I'm not an expert on optimal dps cycles and all that. But I have some questions...
It seems to me that the affliction tree has a lot of really nice utility talents that raiding warlocks will be expected to take, such as shadow embrace, malediction, hell maybe even improved CoW (maybe...) and maybe improved drain soul for the threat reduction. And once you've committed to being a utility bot, why not be the guy who picks up improved imp in the demonology tree, and maybe even improved healthstone.
So what I'm getting at is that maybe raiding as an affliction warlock will mean you don't get enough points in destruction for Bane. And without Bane, shadowbolt becomes a lot less attractive to use. And in your pie chart, shadowbolt is still far and away the most used spell.
So ultimately my question is this: assuming affliction warlocks end up going over 41 points in afflliction and the rest in demonology, is heavy use of shadowbolt the right thing to do?
You can get into a pretty steady cycle of DoTs, tap/dp, lifedrain and sustain that indefinitely. And with the new soul siphon talent, that doesn't seem like such a bad thing to do in theory, if you have mulitple warlocks in the raid. And I'm wondering if that is not the optimal dps cycle.
So ultimately my question is this: assuming affliction warlocks end up going over 41 points in afflliction and the rest in demonology, is heavy use of shadowbolt the right thing to do?
Similar question was asked in some previous post, but it was more comparing DrainLife to SB, even with bane/imp.SB. In that scenario SB resulted clearly in better overall DPS.
It is fairly easy to simulate this scenario in XLS. I started from "no heals/CoA" scenario with afflic, and removed points in improved SB/bane/devastation (put tehm in demonology) ... resulting DPS (with only basic raid buffs) was 727.
Then I put 9 full DrainLife in 108sec cycle, and changed "real cast time" for SB to 1.5 sec - that simulate "insta" casted SBs due to nightfall. It assumed added 2 SBs due to Nightfall, which is about right for 9 DrainLifes with Corruption - chance for Nightfall to proc with both of those should be bit under 25% during 5sec cast time of DL ..so 9 DLs result in 2 insta SB.
Result was about 700 DPS for above combination, which is very close (about 3% lower) than using only SB. It shows that, while theoretically SB is still more DPS, in realistic raid situation without CoA, Drain Life for afflic lock result in about same damage, but much more healing. If it is raid boss fight where you get damage, then using DrainLife for afflic lock is clear choice. In fights where you do not take damage, it could still be used without noticeable loss of DPS.
Edit: above assumed that full bonus from Soul Siphon for Drain Life is in effect. If not enough locks to support enough debuffs on mob, then SB became again clear advantage for DPS.
Thank you nenad. I would imagine that for only 3% less dps, the drain life route offers a great many advantages in sustainability, survivability (and you already accounted for nightfall-proccing, but those are always fun)
Oh, did you also factor in that drain life will (depending on gear) have a lower resist rate thanks to the Supression talent? Because if you were to dump all 5 points into Suppresion, that would give you a 10% +hit advantage over shadowbolts that would have to be overcome by gear.
And finally: Giving up only 3% dps for so much beneficial raid utility! Truly warlocks have the best of all worlds.
12/21/06, 2:25 PM
Similar Threads
