Why is Supernova the go to for single target?

[twistedmind]

Hey, I've been trying to come to some definitive mathematical basis for how to play the new arcane for legion.

The most popular topics that have be scratching my head are why nether tempest and supernova are the go to talents for single target fights. NT can be a topic for another day. But why is supernova the go to?

My understanding of it is that supernova is loved by the sims because it saves mana and procs arcane missiles at the standard proc rate. However, it's more dps to cast arcane blast after 2 charges or missiles after 1 charge.

The arguments I'm trying to figure out:

- Is it a net dps increase to build charges to get to 4 charges and then either entering a burn phase or barraging off the 4 charges? Or are you losing dps by delaying those cycles by 1 global to cast supernova? I've been watching this closely as my mastery % changes.

- Is it more dps overall from mana savings over the entire fight to cast supernova? And how does it change depending on fight length? The comparison could be made to charged up because taking supernova would mean a ramp up time you'd have to do on the pull.

- There is more "theoretical dps" to be yielded from casting Arcane Blast in the form of 1.) the increased proc chance rate for missiles 2.) and also the increase in dps from proccing touch of the magi. Quantifying this exact dps increase is extremely tricky because missiles also has total saved mana and because the cast time of arcane blast is always in flux from quickening, therefore making the different potential dps at every cast.

- Is it more dps to save supernova for those timing opportunities like the last cast during rune of power/touch of the magi/arcane power/int procs instead of smashing it on cooldown? You wouldn't be able to cast another arcane blast at the end duration of some of those windows of increased damage, so supernova is a good layering spell to add to the end of those windows of time where you're doing more damage.
But, you can also simply cast arcane barrage at these times and have close to the same behavior. But Arcane Barrage drops quickening. But maybe some more work needs to be done to compare using words of power and overpowered VS. quickening.

So you see my dilemmas. I can understand being in love with supernova when you have lower mastery. But frankly, the spell doesn't scale well. Outside of looking for cleave opportunities, I don't know why one would take supernova over charged up or resonance. And even then, resonance with the artifact talents will be better than supernova anyways, even during a conserve aoe rotation. Or at least that's the case with my current levels of mastery.

[x1xruex1x]

I'm going to leave a quick reply since I'm heading off for the night, but i'll give a more detailed response tomorrow.

Essentially, the way I see it, is Supernova does double damage versus the primary target. So whether its single target or multi target, there's always 1 target getting the double (unless you cast it on an ally or yourself accidentally).
Then that double damage can crit as well, so that's amplifying it even more so.

So that's the understanding I have on Supernova. The other two talents are Charged Up (just a quick way of getting AC) I think that it's good for pugs where you dont have a chance to get setup because people are racing through. Or if you're out in the world. Then you have Arcane Resonance which is nice for AoE because of the extra damage your Barrage does, but in a single target environment it's lack-luster honestly.

So yea, that's what I feel on it and get from the whole thing about Supernova. It's just a well rounded talent that doesnt have a single use, but multiple use.

As far as NT goes. I've been running tests in dungeons, switching between the three talents in that tier when I have a fight that i can sit still on and just focus dps into the target. I'm seeing 260-300k dps pulls with NT. while Unstable Magic had me about 220-240k dps ranging, and then Erosion just had me around 190-210k dps.
These are just roundings and giving a rough estimate. I've been screen capping the logs and data so I can compile something more detailed for the NT argument. But it does come with the task of maintaining another item on the fight, so if you're having issues with that, then I'd say Unstable Magic instead.

Hope this helps a bit.

[Suda]

I am curious, as I am seeing similar numbers and am constantly having same questions, are you guys using POM, Arcane Fam, or Words of Power for Tier 1. Are you doing different stuff for Mythic + and EN?

[twistedmind]

Let me clarify because there is a lot of unpacking that can be done here. I'm only concerned with single target right now. AOE throws lots of layers of complication that I can focus on separately.

I'm also only focusing on the supernova talent row. The other talent rows should be examined (not in this thread). FWIW, optimal nether tempest usage and looking at only burn phase damage is a 1.2% increase to my single target dps. Erosion should be a flat 8% increase once ramped up. Unstable Magic is only a 7.5% boost to arcane blast's damage, but you're also casting arcane missiles so one can't simply say it's a 7.5% increase. If you average 1/3 of your casts as missiles, then Unstable Magic is a 5% increase in dps. If you play properly and pool missiles for burst phases, then unstable magic is even lower. Prior to earlier this week, I would have said that hands down that erosion is the default for single targets. But investigating the differences in how PPM trinkets interact with nether tempest have me taking another look at it.

===========

Back to the topic at hand:

Supernova on a single target is 380% spell damage (from your spellpower). Arcane Blast at 4 stacks is (60%*4) * 165%= 561% spell damage.
But, arcane blast has an artifact talent for another 6%, plus it procs touch of the magi and has double the chance to proc missiles. Also, the above doesn't show mastery. My toon is at about 40% mastery, so arcane blast is 693% spell damage.

380% / 1.5 sec cast time = 253% per second
561% / 2.25 sec cast time = 308% per second

So at that basic level, you obviously want to cast arcane blast or missiles during your burst phase. Casting supernova is a loss.

====

Conserve phase rotations for me are also showing it as a dps loss. So again, why is everybody taking it and claiming "it's the best talent for single target fights"? I feel like I'm seriously missing something.

[Adimera]

Conserve phase rotations for me are also showing it as a dps loss. So again, why is everybody taking it and claiming "it's the best talent for single target fights"? I feel like I'm seriously missing something.

It costs no Mana giving you some time to regenerate some,it's instant and it can procc Missiles. My guess is these 3 things are enough to take it over the other talents in this row.

Comparing Supernova now with the other talents,for Single Target it's best since Resonance is lackluster in full single target encounters and taking charged up doesn't really synergize with Quickening,or am I missing somethere with these two?

The biggest thing though is that with Quickening,you won't really clear the stacks that often and Supernova gives you some time to regenerate mana while staying at 4 Charges,unlike Barrage which would clear all the Quickening stacks.

[Unreal]

There is probably a remaining bias from wod, because super nova used to scale with your mastery and hit like a truck when used on crystal.

You mainly use super nova to get a last hit into the enemy while your rune is up. It's definetely only a minor dps boost, though.

[Joban]

I actually find the biggest benefit to Charged Up, besides having better openers, is that you can burn your mana on your initial burn faster. Some of these fights can take a while to burn that first bar of mana, if you get caught running around too much, and then your evo is staggered from AP and you're basically screwed for the rest of the fight.

Also, SN usually ends up doing like 2-3% of my damage anyways, I would not be surprised at all if its a dps loss to cast it, even though simcraft ranks it higher than AB in DPET.

[Blownt]

Greetings,

As previously states there are many good explanations here but lets begin to look at the nuances of the spec in its current form/mastery requirements.

As you play mana = damage, more importantly cooldowns = damage. So how can you spend you mana to get the most value within your cooldown windows to maximize your damage while minimizing your mana expenditure to get from evo to evo.

Ex. holding 2 charges of missiles before you put a rune down so you can maximize one of your highest DPET spells into that window.
Ex. Dumping all stacks of AM before AP so you can maximize you mana savings in that window.

The reason Charged up sucks for "RAIDING" you go from window to window to quickly right now. So should we ever get to the point where we can spam AB / missiles from evo to evo then I imagine supernova will leave the rotation as it is technically lower DPET. (also it procs AM which has its own inherent value due to its low mana cost)

NT however is one of our high DPET spell. So casting 1 NT (Which costs practically no mana) can be an efficient way to 1. generate AM 2. to maximize damage.

Caveat to this statement is if you get the Kilt legendary then you can use charged up to generate mana and minimize stacking time. So as of right now in the current state of mastery and mana return use NT and SN in raids to help get you from evo to evo.

[x1xruex1x]

Let me clarify because there is a lot of unpacking that can be done here. I'm only concerned with single target right now. AOE throws lots of layers of complication that I can focus on separately.

................

====

Conserve phase rotations for me are also showing it as a dps loss. So again, why is everybody taking it and claiming "it's the best talent for single target fights"? I feel like I'm seriously missing something.

@Twistedmind

So I'm reading what you have stated above, and it's making sense. I mean, I guess my curiosity lies in the proper usage of Charged Up. I'm not going to involve Resonance as it's very niche. So from what this thread is getting at, is that Supernova just have a very minimal increase to DPS when it does happen to have a positive impact. Whereas the majority of the time, it doesn't have much of an impact and instead turns more into a DPS loss.

So, if that's the case, then what's the ideal way of implementing Charged Up? I know I've personally used it during my initial ramp up at boss pull due to over zealous tanks, or as a means of keeping dps after a clearing of AC's during a conserve phase with about 50% mana.

I'm just trying to get the logistics to this and what the best overall implementation really is. It sounds like it just may not be used often, other than to offset the odd scenario really.


As for Nether Tempest, with regards to the other chatter about it. It's definitely a damage increase, providing you can remember to keep it up 100% of the time. I often times find myself missing out a few seconds in favor of casting AB or AM, especially during RoP/AP or Touch of the Magi procs.

I'm considering writing up a Weak Aura that'll do better at alerting me that NT is about to expire, or has expired. It's just another thing to micromanage though in my opinion. I really would prefer using Unstable Magic or Erosion, but I just find the damage comparison lackluster even with my missing proper NT maintenance.

Anyone got a suggestion on NT, or is it just "Get Good" and learn to suck it up and maintain it better (I'm feeling that's just the case honestly).

[a0kalittlema0n]

So I kinda skipped quite a few responses here because I have to point out the obvious here that I didn't see anywhere. The main reason SN is the go to talent is because the other two talents only benefit from clearing your quickening stacks which is something you rarely do/something you just don't WANT to do. Every 40 seconds barraging would be silly to be able to effectively use Charged Up, and Resonance is GREAT for 4+ targets and scales exponentially when there are more targets, but to have it be a solid DPS gain over AE spamming you'd need 5+ targets for a sustained amount of time to combat the increase of haste/casts from quickening. This is why I don't ABarrage on Spider boss, I just spam AE to keep my quickening stacks to continue burning on boss when the spiders die, otherwise I was at like 20+~ quickening stacks, and then I start single target back at like 3 when the spiders die.

I have logs you can look up if you want to (a bit skewed since I have Rhonin's, but still) I RARELY cast ABarrage. 6 Minute H Nythendra kill I used it like 3 or 4 times. I mean 360 seconds / 40 = 9 charged up uses. I wouldn't even drop 4 AC satacks enough to make that talent usable. As well, Resonance is useless since it doesn't benefit from extra targets.

https://www.warcraftlogs.com/rankings/c ... ic=bossdps

(I forgot to swap off of Resonance from dungeons 0_0 whoops lol

This is all apart from the idea that SN costs no mana, is usable on the move as a filler with NT for when we have to move w/o IF/Shimmer, but it ALSO procs AM and is off CD more than CU, actually does good damage, better to the main target, etc. It's just good for so many reasons and even better since the other talents are horrific (raiding-wise)


The best way to utilize CU as a talent would be to run it with the pants and even then we'd be hard pressed for an actual DPS increase (betting, not sure obviously). You could do something similiar to this

Opener:
Pre cast AB
CU
MoA
RoP
AP
AB spam + AM/w/e
go down to sub 10% mana
Evo with an AM charge
AM to continue Quickening
Burn to about 70% then
RoP down to like 40%~
Abarrage
CU
Spend AM proc or two
Barrage
build up to 4 ABarrage

This would be a building conserve phase, then you could do it over again. Even this though might be much worse than a slow burn since you would be at lower quickening stacks overall I think over the long run. This might be more beneficial in fights where AE is at use more (Dragons, Il'gynoth)

Personally from the Rhonin's Wrists I can barely burn my mana quick enough.

[Suda]

Little man. I use the exact same rotation with Charged Up. I am 869 Arcane with Kilt Legendary(I do a few more barrages in fight for mana. Which means I have more base haste but Charged Up lets me instantly get back to 4 stacks). We appear to be pulling similar dps. Good post

[twistedmind]

Hey, I basically found out from spreadsheeting that with my build (mastery>haste>everything) that Nether Tempest is a single target dps loss, even though Nether Tempest gains more from haste than our nukes. Basically, erosion will always perform higher in any single target scenario. However, if there is at least 2 targets, NT is best.

I also found out that supernova is also a loss in dps. I think at lower mastery, the total mana saved by using it was gaining more over time, but it's more damage to take charged up or resonance at my 40% mastery on single target. And that's also trying to avoid casting arcane barrage (which is optimal if you have quickening). In any cleave/aoe situation, resonance is best by a widish margin, even if you're burning through all your mana instead of trying to cast barrage.

====

I also found out that you will net more dps with quickening at my haste levels (20%) by holding your arcane power burst until after your first evocate. Basically my opener is Mark of Aluneth>Rune of Power>burn mana>evocation>refresh quickening>Rune of Power again>Arcane Power>burn until Arcane Power is done.

This is also more dps compared to using arcane power right away, even if you would get 2 burn cycles otherwise. The reason it's doing more overall is because you're fitting in more casts during the arcane power phase and you're evocating sooner, which means more total casts.

I mathed this at the 2 minute fight length and longer. Burning first, evocating, and burning with 1 arcane power was more total damage in 2 min than arcane burning right away, evocating, and arcane power burning again before 2 min was up.

Hopefully that helps.

[Komma]

Without looking at whatever spreadsheet you're working on, it's hard to comment on what leads to your conclusion of Nether Tempest being a single target DPS loss. That said, I'm pretty sure that isn't the case.

Going back to the original topic though, you're trying to compare Supernova against the DPET of an Arcane Blast casted at 4 Arcane Charges (or to use Kavan's notation, "AB4"). Obviously, AB4 will have a much higher DPET. The issue however, is that this isn't a proper comparison between the two because you're not taking into account the mana consumed. The casting time "replaced" by Supernova isn't going to be AB4, unless you are in a situation with infinite mana. Instead, you need to consider how the same amount of mana will end up spread over a number of Arcane Blasts casted at lower Arcane Charges, which have much lower DPET than AB4, and then compare Damage-Per-Mana of those two combinations. This is a nontrivial task with spreadsheets.

I'm not sure I follow you on how Resonance can be best in a situation where you don't plan to use Barrage.

[Joban]

I also found out that you will net more dps with quickening at my haste levels (20%) by holding your arcane power burst until after your first evocate. Basically my opener is Mark of Aluneth>Rune of Power>burn mana>evocation>refresh quickening>Rune of Power again>Arcane Power>burn until Arcane Power is done.

This is also more dps compared to using arcane power right away, even if you would get 2 burn cycles otherwise. The reason it's doing more overall is because you're fitting in more casts during the arcane power phase and you're evocating sooner, which means more total casts.

I mathed this at the 2 minute fight length and longer. Burning first, evocating, and burning with 1 arcane power was more total damage in 2 min than arcane burning right away, evocating, and arcane power burning again before 2 min was up.

This makes sense, assuming you aren't using lust/hero at the start of the fight. Would it change on a fight with lust at the start? By the time you use AP lust would be gone.

[Samath]

I also found out that you will net more dps with quickening at my haste levels (20%) by holding your arcane power burst until after your first evocate. Basically my opener is Mark of Aluneth>Rune of Power>burn mana>evocation>refresh quickening>Rune of Power again>Arcane Power>burn until Arcane Power is done.

This is also more dps compared to using arcane power right away, even if you would get 2 burn cycles otherwise. The reason it's doing more overall is because you're fitting in more casts during the arcane power phase and you're evocating sooner, which means more total casts.

I mathed this at the 2 minute fight length and longer. Burning first, evocating, and burning with 1 arcane power was more total damage in 2 min than arcane burning right away, evocating, and arcane power burning again before 2 min was up.

Would this be better than the standard rotation, if and only if the legendary kilt is equiped ?
MoA-rune-burn, evocation, AP-rune-burn, refill mana with kilt, repeat ?