[BETA] Destruction Warlock Discussion

[striderZA]

GoSac is great for dungeon trash. It's not as good as GoServ for single-target, but it's less of a loss than picking Cataclysm or F&B.

Yeah I've been going with Sac almost exclusively.

[Deleted]

Going to segway away from whatever's going on here.

On a lighter note, anyone find a good dungeon spec? It's feeling like you're gimped for AoE when speccing single target (eradication, service, roaring blaze), and vice versa.

i think that is intended, they want you to make the choice between aoe and ST.

[Baconeggcheese]

With the exception of you, ppl seem to have reached a consensus about this issue.

I seem to have read a different thread than you.

Here I'll put it this way, on an actual log of nythendra I spent 38 shards in 6 minutes with CDF talented. Assuming a proper average I'd expect to have gotten an additional 9 shards with a decent chance at a 10th, so I'm just going to assume 10 shards to favor SC. I had roaring blaze, reverse entropy, service, and eradication talented. I have no idea what my haste was scaled to during the testing so I'm just going to base all this on if I had 0 haste.

I did 22 CDF casts during the fight, for a total of 8.10m dmg. Assuming 0 haste this is 66 seconds of time spent. So that's the number we're looking to beat here by filling that time with the additional cb's generated through SC, the additional eradication up time, and filling the rest with incinerate.

My average CB was 430,269. Since we're assuming I gained 10 shards that'd be 5 CB casts at 2.5 seconds so 12.5 of my 66 seconds that I need to fill, and I'm going to assume optimal situation with perfect play and figure that we gain exactly 30 seconds of eradication and that every single one of these additional CB's are being cast perfectly to clip the end of the old eradication and even that there was already an eradication up from the non SC cb's for the first SC CB. This means we have 430,269 x 5 = 2,151,345 dmg from CB, x 1.12 (eradication) for a total of 2,409,506 dmg.

Now I have 53.5 seconds left, which is 26.75 incinerates. I am going to just round up to 27 to again favor SC again. There's 17.5 seconds left of eradication up time, which is 8.75 incinerates, I'm going to round up to 9 in favor of SC again even though the 9th would not have clipped the eradication buff. My average incinerate was 125,660 x 9 = 1,130,940 x 1.12 = 1,266,652. There's 18 more unbuffed incinerates so 125,660 x 18 = 2,261,880 + 1,266,652 = 3,528,532 total incinerate damage.

Now with 27 additional incinerates we should average 1 dimensional ripper proc, that should actually replace one of the incinerates. The damage of this can vary wildly depending on what portal spawns, but I'm going to just favor SC here and add 400k without removing the 27th incinerates damage as well as include that 400k in the eradication window to again favor SC so +448k damage for the dimensional ripper proc (again not removing the 27th incinerates dmg from the previous incinerate total).

Now to factor the eradication up time for the passive damage sources (I'm just looking at what actual events happened in 30 seconds for each of these which is including my haste which favors SC), I'm going to assume I precast service imp before the 30 second window perfectly, this is 9 x 93,922 average = 845,298 x .12 (we just want the eradication bonus) = 101,435. The regular imp does 8 casts in a 30 second window at 45,807 average x 8 = 366,456 x .12 = 43,974. And finally immolate damage, I'm just going to look at the dot damage and ignore that at least one of the incinerate casts would have been a immolate in that 30 second window and incinerates initial damage is 70k, this again favors SC a bit. Including my haste I get 12 immolate ticks in a 30 second window at 69,347 x 12 = 832,164 x .12 = 99,859. 101,435 + 43,974 + 99,859 = 245,268 dmg.

That leaves us with 2,409,506 + 3,528,532 + 448,000 + 245,268 = 6,631,306 damage I would have done in that 66 second window that I spent on CDF if I had instead taken SC. CDF did 8.1 mil dmg, which leaves me with a 1,468,694 deficit to make up just to break even, yet alone do more damage than CDF. This is assuming the stars would align in an impossible way, and even that some events would happen that actually wouldn't which favors SC.

So what am I forgetting here that would bring me up another 1.5m dmg, cause I did this really fast and lose. I'm sure I'm missing something.

[Dastey]

Now I have 53.5 seconds left, which is 26.75 incinerates.

That would be 29.72 incinerates.
Also if we took backdraft instead as our first tier that would be a lot more incinerates as we would probably have something close to 41% uptime on the 30% reduced cast time ending up in average of 12,3% haste taking the cast time of incinerate without haste down to 1,58 seconds which would then be 33,86 (lets say 34) incinerates instead of the 27 you presumed.

Which would take numbers to 11 x 125,660 x 1.12 = 1.548.131
As for unbuffed incinerates that would be 2,890,180

A total of 4,438,311 compared to your 3,528,543 which is a difference of 909,768 simply by changing 1 talent and taking the incinerate artifact trait into consideration.

Let's conclude that these 2 talents are atleast quite close

- - - Updated - - -

Also on a different note I noticed GoSac is not affected by our mastery, so am I correct to assume that our mastery only works on spells we cast and not proccs?

[Gaidax]

First of all you miss tier which at least would give at lest another 2 shards from SC rounded down, then the fact that 53.5 seconds is 31 incinerates and not 27 due to Artifact (rounded down in your favor as a courtesy to your favor), finally you never included things like Summon Doomguard which is peanuts really, but more importantly Lord of Flames which is quite stronger and definitely would benefit from extra Eradication uptime a lot.

But your incinerate mistake is already half a million damage you casually missed. In short - that plus tier, plus cooldowns and you are extremely close to CDF even in your case. CDF would still pull ahead by maybe half a million and that's about it, but then if you would take Backdraft instead - it would actually lose damage because of many more incinerates and many more opportunities to get another rift.

So yes, CDF is better than SC in your case... by what... 1-2% in the end? Basically this gap will be instantly breached the moment you increase your shard generation thanks to better gear or god forbid get Legendary Item that gives 15% chance to gain shard with fire spells - which you sure as hell will be farming for as a raider. Or hell... just take Backdraft for that matter...

---

This is why I claim that CDF was overnerfed, it was okay-ish when it was 750% (although I'd have it at 850%), but at 630% it is simply a waste of time, which is what I am trying to get across. It's basically a dead talent that only just barely is better at the start of the tier and in single target only encounter and this gap will be eliminated fast with gear. Heck, even this "barely better" is questionable as it seems.

This is wrong, IMO.

[wholol]

Guys, if 2 talents are so close in terms of dps value that it causes several people to get into arguments over which talent is superior over several pages and still not have reached a conclusive agreement as to which one is superior simply means they are both viable options ? that is simply good design right there... give people a choice for once and they tear each other apart

On a separate note, does anyone know if chaos bolts mimicked by havoc provide an additional chance to proc soul conduit ? if so, the value of soul conduit when more than 1 target exists increases substantially.

[Gaidax]

Probably not, I would be very surprised it actually works that way.

And the design is not terribly good, simply because you have two talents that are extremely close in pure single target, but anything other than that and one of them grows much stronger and dusts the other.

[striderZA]

On a separate note, does anyone know if chaos bolts mimicked by havoc provide an additional chance to proc soul conduit ? if so, the value of soul conduit when more than 1 target exists increases substantially.

Don't think it does, since you only spend one shard, and that's the shard that has the %refund.

[wholol]

Don't think it does, since you only spend one shard, and that's the shard that has the %refund.

Oh right, I can't believe I forgot how soul conduit works.

When you think about certain details for too long you start forgetting other details , or I'm going senile... both perfectly viable scenarios

[Baconeggcheese]

That would be 29.72 incinerates.

That should be with the artifact trait yeah? Not sure how it factors it in tbh. Definitely something I forgot about tho.

Also if we took backdraft

CDF benefits from backdraft, and this entire window is time spent casting CDF so backdraft is moot.

Let's conclude that these 2 talents are atleast quite close

Considering I massively biased all of my rounding and assumptions towards SC in completely unrealistic ways that would never actually happen in practice all together, especially fitting things into the eradication window and rounding up for extra casts that technically would have just not happened its still that far behind I'd say they aren't all that close yet alone being competitive to better.

It's close enough if you're not on the cutting edge of progression, but you wouldn't catch me knowingly losing 1m+ dmg. And that's assuming the current tuning stays which is the lowest its been.

[JV Chequer]

I really just wish Blizzard buff CDF to 850% to avoid all this math discussion around something exclusively for Pure-ST fights.

[Zoopercat]

Hey, I wanted to jump in real quick. I'm from Ask Mr. Robot and wanted to address the statement that @almara2512 made about the simulator. We don't know of any bugs affecting destruction (but that doesn't mean there aren't any). Everything is verifiable on the wiki, so if you think something is wrong, you can easily see it and report it. We're constantly reviewing the specifics ourselves too.

The rotation for destruction likely needs some love. If anyone is interested in playing with it, it's pretty easy. I'm here to help, and we'll be modifying it as well. We've been working with existing TC apls and even making some of them better. For example, we compared our APL with Binkensteins, and both did basically the same damage. But we took the best parts from both and made an even better one. (Of course, everyone is still developing them, but that's why it's good to collaborate now). We still have to do that for destruction, so hang in there.

So if you think the AMR sim is 'notoriously wrong,' mind posting why? If you know of a bug that we don't, would love to hear about it. Otherwise, all I can do is assume it's just spreading gossip for some reason And then I picture parts from the movie, Mean Girls lol.

[Baconeggcheese]

So if you think the AMR sim is 'notoriously wrong,' mind posting why? If you know of a bug that we don't, would love to hear about it. Otherwise, all I can do is assume it's just spreading gossip for some reason And then I picture parts from the movie, Mean Girls lol.

Mean girls is pretty par the course for the wow community. Doing anything like this will require thick skin and not taking comments personally.

A huge red flag that stands out is that the imp is topping damage on the basic sim. My imps consistently done around 9% of my dmg during testing, yet for some reason the sims imp is doing 17%. That's probably the most egregious thing I see off the bat.

Will have to look into other things and play with it.

[Wenoxar]

So if you think the AMR sim is 'notoriously wrong,' mind posting why?

I simulated myself few times on Zakuun Heroic, I'm Destruction warlock.

Simulator kept taking those talents:
- Shadowburn
- Fire and Brimstone
- Wreak Havoc

All of those 3 talents aren't supposed to be used on Single Target, at all.

[Dastey]

CDF benefits from backdraft, and this entire window is time spent casting CDF so backdraft is moot.

I gotta say I'm surprised that CDF benefits from backdraft as the talent clearly states Chaos Bolt and Incinerate.

Also on another note. GoSac doesn't benefit from mastery, presumable because it says that your spells deal up to extra damage. So I assume this is intended? Quite a shame.

[Turturin the Warlock]

Going to segway away from whatever's going on here.

On a lighter note, anyone find a good dungeon spec? It's feeling like you're gimped for AoE when speccing single target (eradication, service, roaring blaze), and vice versa.

I've had decent results in dungeons with
Shadowburn
Cataclysm
Fire and brimstone
Sacrifice
Channel Demonfire

[Uzkin]

I seem to have read a different thread than you.

Here I'll put it this way, on an actual log...

That was a far better "pen-and-paper" analysis in support of CDF than the previous attempts in this thread. But as you can see these are quite complex problems due to all the interacting abilities, talents and traits and it is very easy to miss something. This is why I chose to use an AMR simulation; it is less likely to forget about things than I am

Still, I didn't use AMR without criticism; while the destruction simulator seems quite reliable to me what I really needed from it was the information of what is different in the two simulations (CDF vs. SC), so that I don't miss anything. All that remained, then, was to check whether these damage differences were correct/reasonable or if there was perhaps some problem with the simulator. I found that all the differences were of the same order of magnitude as given by a quick back-of-the-envelope estimate, so that the simulator seemed to be working fine. For example, the increase in CB damage in the SC profile was as expected, the increase in Incin damage in the SC profile was as expected, and so on. The simulation, even if you don't trust its numerical results, can be used to identify all the sources of damage differences so that you don't miss any.

I made some more simulations with both Roaring Blaze and Backdraft; here are the results for comparison:

Backdraft/SC - 155267 DPS
Backdraft/CDF - 154243 DPS
Roaring Blaze/CDF - 151203 DPS
Roaring Blaze/SC - 150582 DPS

Your latest analysis, together with the corrections by Dastey and Gaidax, suggest that CDF should be slightly better than SC if you choose Roaring Blaze. This is, again, correctly reproduced by AMR simulations (see above).

Which leads us back to Gaidax's original point. The last talent tier is supposed(?) to be cleave vs. ST vs. AoE/generalpurpose but currently CDF seems undertuned for ST. Even for pure single target its value is very very close to that of SC, slighly better or slightly worse depending on the other talent choices.

My original point wasn't damage-related at all (my AMR simulations were but a sidetrack since I wanted to shoot down the arguments that CDF was an obvious choice for single target :P ) Instead, my problem with CDF is that it is an awfully uninspired button-bloat nuke; it's just something that's been crammed into the talent tree without any thought of how to integrate it into the big picture or how to make it interact with the rest of the spec. It doesn't really bring anything new to the spec. I have long wished for it to extend, not refresh, Immolate with each bolt-hit (so that it wouldn't drop the Roaring Blaze buff). This would provide some synergy with these two supposedly single-target talents.

I wonder if Backdraft affecting CDF is intended? It's not mentioned in the tooltip. But if it's working as intended, it is just the kind of interaction that CDF needs, even though perhaps not enough. Another (very minor) thing favoring SC is that it can refund the GoServ and Doomguard shards as well. I wish they removed the shard cost of summons...

[Deleted]

- Snip -

Besides the few minor details that would bring SC pretty close to CDF, I think you are on the right track. Granted, I only did a little testing on dummy and the two were roughly equal, and that could easily be lack of skill on my part. My issue with CDF is that there are things that specifically will buff CB (Either throughput or availability), and nothing similar for CDF. So, a talent that only "works" for 100% single target DPS, and becomes outdated as soon as you get a specific legendary or 2-4 piece bonus (Or artifact traits + relics).

I'm just having a hard time seeing CDF keep up with SC for long, and that is even in such a specific scenario.

[striderZA]

Oh right, I can't believe I forgot how soul conduit works.

When you think about certain details for too long you start forgetting other details , or I'm going senile... both perfectly viable scenarios

It happens

I've had decent results in dungeons with
Shadowburn
Cataclysm
Fire and brimstone
Sacrifice
Channel Demonfire

That's a very AoE heavy build, don't you suffer too much on bosses/single target?

[Baconeggcheese]

Still, I didn't use AMR without criticism

The issue is it isn't optimized, so the difference between two talents that it shows isn't necessarily correct since the rotation its following isn't necessarily correct.

It does very rudimentary mistakes, if you look at the log it does things like pre-casting infernal, immediately dumping all 3 dimensional ripper charges, with CDF and roaring blaze in the one sequence it didn't cast a conflagrate until 7 seconds into the fight. As you continue down the log you continue to find these small optimization errors, which all add up and paint a picture that isn't accurate.

For instance here's the opener cast sequence with CDF and roaring blaze: http://i.imgur.com/BxaMwKT.jpg This btw is set up with the tier 2 / 4p and a 863 ilvl with around 19% haste. You'll notice it casts conflagrate twice in 19 seconds even though conflagrate likely has something like a 7ish second recharge time. Truth be told the next conflagrate after the 19 second mark doesn't happen until 31 seconds into the fight. Now obviously if you use this sim to compare the difference between roaring blaze and backdraft you're going to get wildly incorrect information because the sim is literally just letting conflagrate cap repeatedly and sitting on it.

These errors aren't the sims fault, this is suuuuper early in the sims life span. It's absolutely no different than early SimC sims, they simply aren't optimized at this point and will have errors in them and they need to be iterated on over and over especially as the tuning changes keep happening in order to actually hone in on something truly usable. This has been true for the entire lifespan of SimC, and its going to be no different with AMR. Unless you're actually digging through the cast sequence and fully understand what the sim is doing then you're just giving yourself incorrect information.

Your latest analysis, together with the corrections by Dastey and Gaidax, suggest that CDF should be slightly better than SC if you choose Roaring Blaze. This is, again, correctly reproduced by AMR simulations (see above).

CDF is affected by backdraft, so nothing should change between those talents as I already pointed out above. I'm not sure if the sim even accounts for that yet. That is again to ignore that the sim is using conflag incorrectly so you're looking at two talents that are based on specific conflag usage being used improperly and trying to glean any useful information from it.

Which leads us back to Gaidax's original point. The last talent tier is supposed(?) to be cleave vs. ST vs. AoE/generalpurpose but currently CDF seems undertuned for ST. Even for pure single target its value is very very close to that of SC, slighly better or slightly worse depending on the other talent choices.

You have to remember that my napkin math favored SC every step of the way, and that the gap would in fact be bigger than my napkin math. The point was basically that even unrealistically favoring SC there was still a decent dmg gap between the two talents that clearly showed CDF to be better.

Granted this is by default, I could see SC catching up with scaling later in the xpac or just outright being the choice with that legendary that gives shards on fire spell casts. Though in the event of CDF just getting outscaled I'd fully expect blizzard to either buff CDF or nerf SC to get them back to where they want them.

My original point wasn't damage-related at all

Yeah its not super exciting since its a meteor effect, but it is what it is.

I wonder if Backdraft affecting CDF is intended? It's not mentioned in the tooltip.

I gotta say I'm surprised that CDF benefits from backdraft as the talent clearly states Chaos Bolt and Incinerate.

It isn't at all uncommon for beta tooltips to be inaccurate. Impish incineration doesn't mention gosac or gosup anymore, it doesn't mean they don't intend it to buff them.

I'm just having a hard time seeing CDF keep up with SC for long, and that is even in such a specific scenario.

I think legendaries making certain talents mandatory is just something we'll live with, that's not terribly odd. As I just said to uzkin though in the event that SC starts outscaling CDF I fully expect blizzard to buff up CDF and or nerf SC. I'd imagine the latter more likely, but its not something I see as a huge problem.