Basing gear choice on simulations.

Guides and discussions of all things specific to the Mana Adepts of Azeroth.
arby
Posts: 5
Joined: Wed Jul 19, 2017 6:52 am

Basing gear choice on simulations.

Unread postby arby Wed Aug 23, 2017 11:42 pm

Disclaimer: I don't use simulation, so some of what I grouse about may actually be in the simulation tool.

I maintain that dps for arcane mages can be most accurately calculated by the following:

DPS = ((DPS[ProcPhase]*ProcPhaseSeconds) + (DPS[BurnPhase] * BurnPhaseSeconds) + (DPS[ConsPhase] * ConsPhaseSeconds))/ Encounter Time

where
ProcPhase is the time spent dealing damage with the various buffs offered by talents, trinkets and tier gear.
BurnPhase is the time remaining in the burn phase after the procs expire until you need to evocate and begin your conserve phase, characterized by maintaining 4 arcane charges.
ConsPhase is the time you spend waiting until you can reenter either Burn Phase or Proc Phase.

Where this is all leading is the rating of the Gravity Spiral headpiece in the guide. Contrary to the guide, I feel that it is at least the 2nd strongest legendary available for arcane.

My assumption is that dps is highest in the proc phase and lowest in the conserve phase. I feel that dps is maximized when the conserve phase is minimized/eliminated, or in other words, maintaining a burn phase rate (4 arcane charges) until the proc phase is again ready. To be truthful, there are times during the burn phase you don't maintain a full 4 charges, e.g., arcane barrage against 4 or more targets with 4 AC, arcane barrage while moving, etc. but these are anecdotal in nature. It also turns out that in most encounters, it is relatively easy to overlap the evocate while you are forced to move anyway, so the evocate time is not entirely wasted.

Just testing on the targets seem to bear this out. Just firing AB at a target with AM as available, no procs or anything, a burn phase was extended approximately 40 seconds. Adding additional mana from Aluneth would extend this as would the Kilt.
Naustis
Posts: 69
Joined: Thu Apr 06, 2017 9:24 pm

Re: Basing gear choice on simulations.

Unread postby Naustis Thu Aug 24, 2017 1:17 am

If you are trying to fit all arcane dps in one formula, you are doing it wrong friend :p GS is bad simply because it give you onle one Evo throughout the fight. It is like 10 arcane barrages when using kilt, and kilt gives you the freedom with using your mana.
arby
Posts: 5
Joined: Wed Jul 19, 2017 6:52 am

Re: Basing gear choice on simulations.

Unread postby arby Thu Aug 24, 2017 6:23 am

I agree GS might not be the best for short encounters, as in M+, in raids it essentially gives you an evocate cooldown of 1 min. In an 8 min raid fight, that's 4 extra evocates. Even in mythics, using evocate immediately after a boss kill and moving at the same time helps improve the overall flow of the mythic dungeon.
nickseng
Posts: 272
Joined: Mon Jun 09, 2014 3:52 am

Re: Basing gear choice on simulations.

Unread postby nickseng Thu Aug 24, 2017 10:23 am

in raids it essentially gives you an evocate cooldown of 1 min. In an 8 min raid fight, that's 4 extra evocates.
No, it doesnt'. The charges don't go on cooldown independant of each other.
arby
Posts: 5
Joined: Wed Jul 19, 2017 6:52 am

Re: Basing gear choice on simulations.

Unread postby arby Thu Aug 24, 2017 8:11 pm

@Ni -- May I suggest you put on a GS, create TMW icons to separately track Evocation charges and Evocation cooldown. The 2 min cooldown starts immediately after the first evocate, and you can go full out until you have to evocate again on the 2nd charge. That 2nd evocate does NOT reset the cooldown. Towards the end of the cooldown, you might have to go conserve for a bit, but not for long. The cooldown pops, you now have 2 charges. Pop it. You will see the cooldown timer is 2 min, but you still have 1 charge remaining. Rinse and repeat.
Haruichi
Posts: 42
Joined: Thu Sep 22, 2016 10:36 am

Re: Basing gear choice on simulations.

Unread postby Haruichi Fri Aug 25, 2017 9:19 am

I have; quite a few times since I wear GS. The recharge of Evocation is sequential. And you don't really need WA to test it, although I do use WA.

Try this: Put on GS, press Evocation twice in a row so both charges go on CD right away. You'll see that once the first charge comes off CD, only then does the second charge starts it's recharge/cooldown cycle.

Meaning: If you press Evo twice in a row (obviously you never would, but just to test it), you'll have 1 charge available after 1,5 minutes and 2 available after 3 minutes.

As such, GS effectively gives you 1 extra Evo per encounter.
nickseng
Posts: 272
Joined: Mon Jun 09, 2014 3:52 am

Re: Basing gear choice on simulations.

Unread postby nickseng Fri Aug 25, 2017 9:22 am

What Haruichi said.
I do have GS.... it was the first Arcane Legendary I got and after extensive playing with it, I was disappointed.

There is no way you get 4 extra evocations on a 8 minute fight.
arby
Posts: 5
Joined: Wed Jul 19, 2017 6:52 am

Re: Basing gear choice on simulations.

Unread postby arby Fri Aug 25, 2017 9:07 pm

Let me try again: When you are out of mana after the burn phase, you evocate. Without GS, you are 1.5 min away from your next evocate, and you can continue a burn phase for awhile, and then you have to run a lengthy conserve phase. With GS, you have an extra charge, so you can continue your burn phase until you are again OOM and then evocate again. The cooldown from the first evocate is not reset, so you are roughly 45 seconds away from when you can next evocate. That is the first extra evocate. When the CD expires from your first evocate, you can then do it again. But lo and behold, the CD to the next evocate is not the full 1.5 min, it is 1.5 min minus the time you spent burning the mana from your second evocate. So with careful usage, you are able to extend your burn phase to either the next proc phase, or until the evocate CD expires. If the latter, then evocating again does not reset the CD from your 2nd (extra) evocate, so in 45 seconds you will have another (extra) evocate. That's why having TMW or WA icons for the evocate charge and evocate CD are important.

Thus what Haruichi said is true, except for the last statement, because in actual use, you are overlapping the burn gained from the second of two evocates with the remaining cooldown from the first, and this cycle repeats. Note: if you look up Gravity Spiral on wowhead, and read the comments, you will see that they agree with what I am trying to say.
seoh
Posts: 64
Joined: Sat Oct 08, 2016 1:22 pm

Re: Basing gear choice on simulations.

Unread postby seoh Sat Aug 26, 2017 2:02 am

Regardless or not, arcane's burn dps outside of ap/rop is not as powerful as you think. The reason why Kilt/Soul and Soul/Shard are powerful is because they capitalize on maximizing damage in the ap/rop burn. Soul increases ab casts under cds, shard the same as well as amplifying TSM as it scales with haste and kilt just eases the rotation while providing the ability to burn with MoA + rop + pom in-between APs. I'm not heavy into math, but here's a heroic Kj log I had. I chose it because we don't lust on pull, so you can see some dps at 4 charges with a meaningful amount of time because otherwise I evo very quickly after second rop. My damage in the opener (ap, rop, tsm, pp) was 34mil over 14s. After ap was just pp, rop, tsm until evo (no arcane explosions until after evo) was 22.8mil over 16s. I'd argue this means arcane is front loaded, or cd/burst reliant, once AP ends, damage drop off is considerable. Logs for reference: https://www.warcraftlogs.com/reports/XW ... d=10106284 and https://www.warcraftlogs.com/reports/XW ... ffs=-12042.

Now you can imagine that AB spam without rop and tsm would be even less damage. So the question becomes, does that damage outweigh what kilt brings? Or, does arcane really rely on an long burn phase? This is where I trust the apl much more simply because they run millions of iterations and spend hundreds of hours, where I do not.
nickseng
Posts: 272
Joined: Mon Jun 09, 2014 3:52 am

Re: Basing gear choice on simulations.

Unread postby nickseng Mon Aug 28, 2017 9:27 am

Let me try again: When you are out of mana after the burn phase, you evocate. Without GS, you are 1.5 min away from your next evocate, and you can continue a burn phase for awhile, and then you have to run a lengthy conserve phase. With GS, you have an extra charge, so you can continue your burn phase until you are again OOM and then evocate again. The cooldown from the first evocate is not reset, so you are roughly 45 seconds away from when you can next evocate. That is the first extra evocate. When the CD expires from your first evocate, you can then do it again. But lo and behold, the CD to the next evocate is not the full 1.5 min, it is 1.5 min minus the time you spent burning the mana from your second evocate. So with careful usage, you are able to extend your burn phase to either the next proc phase, or until the evocate CD expires. If the latter, then evocating again does not reset the CD from your 2nd (extra) evocate, so in 45 seconds you will have another (extra) evocate. That's why having TMW or WA icons for the evocate charge and evocate CD are important.

Thus what Haruichi said is true, except for the last statement, because in actual use, you are overlapping the burn gained from the second of two evocates with the remaining cooldown from the first, and this cycle repeats. Note: if you look up Gravity Spiral on wowhead, and read the comments, you will see that they agree with what I am trying to say.
Right.
But I don't see how that gives you 4 extra evocate in an 8 minute fight.
Generally, you'd get ~5 evocates in an 8 minute fight. You're saying with GS you're getting 9.
Haruichi
Posts: 42
Joined: Thu Sep 22, 2016 10:36 am

Re: Basing gear choice on simulations.

Unread postby Haruichi Tue Aug 29, 2017 11:57 am

Below, I'll use the 0 secs-point as when you use Evocation for the first time; obviously, your first Evocation wouldn't be 0 seconds into the fight.

Without GS; 5 minute fight, evocation used on cooldown:

0 secs: Evocation
90 secs: Evocation
180 secs: Evocation
270 secs: Evocation.

4 Evocations.

With GS; 5 minute fight, Evocation used on cooldown.

0 secs: Evocation, Charge 1.
30 secs (ish) Depending on Mastery levels and so forth, meaning after your prolonged burn phase: Evocation, Charge 2.
90 secs: Evocation, Charge 1.
180 secs: Evocation, Charge 2.
250 secs: Evocation, Charge 1.

5 Evocations.

The recharge process is seqential: Once both Charge 1 and 2 are on cooldown (after the prolonged burnphase), Charge 2 will wait for Charge 1 to come off cooldown before it starts its own recharge. And Charge 1 will do the same afterwards.

However you twist and turn it, the following statement is true: Wearing GS grants you 1 more Evocation per encounter than you would have had if you did not wear GS.

What you're describing below comes across as poor Evocation management. I'll go over each of your statements to clearify:
When you are out of mana after the burn phase, you evocate. Without GS, you are 1.5 min away from your next evocate, and you can continue a burn phase for awhile, and then you have to run a lengthy conserve phase.

True.
With GS, you have an extra charge, so you can continue your burn phase until you are again OOM and then evocate again.

True.
The cooldown from the first evocate is not reset, so you are roughly 45 seconds away from when you can next evocate.

I'm not quite sure what you mean by the word "reset" in this context, but you're right that Charge 1 starts its recharge-cycle of 90 seconds when you use it. Thus, by end of the prolonged burnphase, you're probably around 45 seconds away from having Charge 1 ready.
That is the first extra evocate.

That statement doesn't make a lot of sense in this context. The first extra Evocate was Charge 2 that you used to have a prolonged burnphase.
When the CD expires from your first evocate, you can then do it again.

This is where it starts to get muddy. By the first Evocate, I assume you mean Charge 1 coming off CD. But you don't wait with your burnphase until it comes off cooldown. You start your second burnphase roughly 20 secs before Charge 1 comes off cooldown so that you can use Evocation on CD. This also where your Overlap-Theory starts to lose merit, because overlapping between burn and conserve in regards to Evocation is something you should be doing whether you have GS or not; i.e. starting your burnphase so that it *ends* when Evo is off CD.
But lo and behold, the CD to the next evocate is not the full 1.5 min, it is 1.5 min minus the time you spent burning the mana from your second evocate.

In light of the previous remark, this is simply untrue: If you time your burnphase so that it ends when Charge 1 comes off CD, Charge 2 will have 90 seconds minus the time it took you to Evocate using Charge 1 - not 90 secs minus the time from one burnphase, because that burnphase was carried out *before* Charge 1 came off CD. As would be the case without GS: Your Evocation CD would be 90 seconds minus the time it took you to Evocate.

So, if I elaborate a bit on the initial time table, but still rough time segments, it'd be:

Fight starts.
0-20 secs: Burn.
20-25 secs: Use Evo Charge 1.
25-45 secs: Burn.
45-50 secs: Use Evo Charge 2.
50-90 secs: Conserve.
90-110 secs: Burn
110 secs: Charge 1 comes off CD, Charge 2 starts its recharge.
110-115 secs: Use Evo Charge 1. (Cooldown on Charge 2 at this point = 90 secs minus 5 seconds it took to Evocate).
115-180 secs: Conserve.
180-200 secs: Burn.
200 secs: Charge 2 comes off CD, Charge 1 starts its recharge.
200-205 secs: Use Evo Charge 2 (Cooldown on Charge 1 at this point = 90 secs minus 5 seconds it took to Evocate).

And so on.

The time segments are just to illustrate a point. Depending on gear and so on, your burn and conserve phases could be longer or shorter, but regardless of GS, you'd time burnphases to start with 20-30 secs left on your Evocation cooldown.
So with careful usage, you are able to extend your burn phase to either the next proc phase, or until the evocate CD expires. If the latter, then evocating again does not reset the CD from your 2nd (extra) evocate, so in 45 seconds you will have another (extra) evocate. That's why having TMW or WA icons for the evocate charge and evocate CD are important.

This section is nonsensical.

To that end:
Thus what Haruichi said is true, except for the last statement, because in actual use, you are overlapping the burn gained from the second of two evocates with the remaining cooldown from the first

True; so far as the first two burn phases or first prolonged burn phase - whatever nomenclature you prefer.
and this cycle repeats.

False.
Note: if you look up Gravity Spiral on wowhead, and read the comments, you will see that they agree with what I am trying to say.

I'm not entirely sure what to do with a statement like that.
davesignal
Posts: 18
Joined: Tue May 23, 2017 8:47 pm

Re: Basing gear choice on simulations.

Unread postby davesignal Sat Sep 16, 2017 7:29 pm

I think the best Gravity Spiral was ever going to do was during Nighthold, with the T19 4P. I legitimately could not burn mana fast enough to keep up with all the CD reductions from 4P, and Evocation sat off CD for at least 15 seconds every time; Gravity Spiral gave you a buffer to make use of all that extra CD reduction, and leverage them into more than just one extra Evocation.

And even then.. it just wasn't very good.
Ilm
Posts: 9
Joined: Fri Oct 21, 2016 6:32 am

Re: Basing gear choice on simulations.

Unread postby Ilm Mon Nov 20, 2017 4:57 pm

I think the best Gravity Spiral was ever going to do was during Nighthold, with the T19 4P. I legitimately could not burn mana fast enough to keep up with all the CD reductions from 4P, and Evocation sat off CD for at least 15 seconds every time; Gravity Spiral gave you a buffer to make use of all that extra CD reduction, and leverage them into more than just one extra Evocation.

And even then.. it just wasn't very good.
Exactly. Even when using T19, with tier set bonuses that are pretty much the best fit we've had so far in terms of "synergizing" (I hate that word) with Gravity Spiral, it's still basically a trash legendary.

Part of the issue is just that we really don't want to be spending more time than we already do Evocating, because we're doing little to no damage during Evocation (generally about 4.5 to 5 seconds of cast time, depending on your haste). Putting in a legendary that encourages us to Evocate more often is just asking us to spend more time doing little to no damage, which works out just as you'd think in terms of DPS: it's not much of a gain itself, and compared to other legendary items available to us it's a big loss in DPS. Its power when taken on its own isn't very much to write home about, and relative to other legendaries it's extremely poor.

At this point in Legion, I assume no legendaries will get any further tuning or re-working. That's a pretty fair and solid assumption to make, I think. With that said, if Gravity Spiral had been designed differently from the start, they could've gone a million different routes with it and it would've turned out at least better than the current iteration.

Oh well. Legendaries will be gone in Battle for Azeroth, and for the rest of Legion we'll most likely be rotating between Prydaz (or maybe Belo'vir's? I use Prydaz...) for certain parts of high Mythic+ keys, Mantle/Kilt/Shard/Soul in some combination for all our DPS needs, and maybe occasionally Sephuz for some speed runs and niche fights or encounters here and there. The other Arcane legendaries -- the belt and the bracers and the helm -- will likely not be used much as we move forward in Legion (as they haven't been used much recently either).

The belt is sometimes a fun thing to toy around with, but even at 100 stacks it's only +100% damage, which unless you're gearing up for a massive pull of 15+ mobs or something, your Mark of Aluneth is just never going to do very impressive damage, even with the legendary belt at max stacks.

Return to “Arcane”

Who is online

Users browsing this forum: No registered users and 7 guests