The Black Mage Rotations

[Synovius]

Why wouldn't you use Firestarter and Thundercloud prods as soon as they show up?

And why would you use Thunder II instead of III? When you open with B3 it's free...?

Firestarter doesn't proc until your Fire I has finished traveling to the target meaning that, to use Firestarter as soon as it's up, you'd have to wait after casting Fire I to see if you're going to get a proc and then use it. This results in a very clunky rotation and also a net loss overall in DPS.

As for Thundercloud procs, you are basically using them as soon as they come up. You're simply using them as soon as you are done with whatever cast you are in the middle of or immediately if you literally just finished a cast. The only caveat is that Firestarter procs take priority over Thundercloud procs if both happen to be up at the same time.

Using Thunder II over Thunder III comes down to timing. Thunder II has a shorter duration that fits nicely with one loop through the rotation such that when you cast Blizzard III to start off a new loop through the rotation Thunder II should have just or will soon fall off of your target (this is obviously dependent on RNG with Firestarter procs throughout the rotation). More importantly, Thunder II's 0.5s less cast time over Thunder III allows you to follow your Blizzard III with Thunder II and have enough time to cast the subsequent Fire III for free as you'll get a tick of mana exactly at that moment from Umbral Ice III. In fact, I'd call this the primary reason to use Thunder II over III. Do note, however, you should always use Thunder III with Thundercloud procs as it's instant and more damage.

[dragonflyseksparade]

The only caveat is that Firestarter procs take priority over Thundercloud procs if both happen to be up at the same time.

Why would you do this? If you use thundercloud first you could potentially get another thundercloud immediately. Whereas you will not get another Firestarter proc until you go back to Fire I spam.

[Synovius]

Why would you do this? If you use thundercloud first you could potentially get another thundercloud immediately. Whereas you will not get another Firestarter proc until you go back to Fire I spam.

This does make a lot of sense and, honestly, I hadn't thought about it this way so kudos on bringing up a great point! My concern would be that Firestarter does more damage and if Thunder isn't about to fall off your target, then that means you'll get one more tick of Thunder (and thus the chance for a Thundercloud proc) anyway some time between casting Firestarter and up until you're ready to cast Thundercloud. I'm not 100% certain but I think, mathematically, it may be a wash in that sense. If somebody wants to churn some numbers, I'd love to see where the math leads on this.

Thanks, Dragonfly!

[dragonflyseksparade]

My concern would be that Firestarter does more damage and if Thunder isn't about to fall off your target, then that means you'll get one more tick of Thunder (and thus the chance for a Thundercloud proc) anyway some time between casting Firestarter and up until you're ready to cast Thundercloud. I'm not 100% certain but I think, mathematically, it may be a wash in that sense.

It doesn't matter which proc does more damage. The two scenarios are as follows:

Thunder III -> Fire III
or
Fire III -> Thunder III

In both cases you've done the same amount of damage over the course of two GCDs. But the first of those two scenarios give you a chance at another Thundercloud proc immediately. The second of the two could result in a missed thundercloud if it were to proc again as you cast Fire III.

It doesn't really matter how much time is left on your current thunder because thunder will never fall off of a single target in a normal rotation unless you get some amazing luck and had Firestarter proc on every single Fire I during the rotation.

[NeoAmon]

It doesn't matter which proc does more damage. The two scenarios are as follows:

Thunder III -> Fire III
or
Fire III -> Thunder III

In both cases you've done the same amount of damage over the course of two GCDs. But the first of those two scenarios give you a chance at another Thundercloud proc immediately. The second of the two could result in a missed thundercloud if it were to proc again as you cast Fire III.

It doesn't really matter how much time is left on your current thunder because thunder will never fall off of a single target in a normal rotation unless you get some amazing luck and had Firestarter proc on every single Fire I during the rotation.

This is quite true, I have had 4 Thundercloud procs one after the other in T4. This is only possible if you ignore Fire III procs when you have Thundercloud up. While Fire III is definitely more dmg it won't proc until your Fire I lands on target while Thundercloud is just pure luck ^^ (and some cleaver tactics to get a better chance).

[YuriRamona]

Why would you do this? If you use thundercloud first you could potentially get another thundercloud immediately. Whereas you will not get another Firestarter proc until you go back to Fire I spam.

Thundercloud's effective potency is 17 potency per tick. Every tick of Thunder (and you are not already under its effect), you are adding 17 potency to your rotation.*

Scenario 1: Thunder has three ticks remaining and Thundercloud is used, refreshing duration for 6 ticks.
Total Thunder duration: 6 ticks.
Total Thunder duration with Thundercloud proc down: 6 ticks.

Scenario 2: Thunder has three ticks remaining and Thundercloud is saved until Thunder expires.
Total Thunder duration: 9 ticks.
Total Thunder duration with Thundercloud proc down: 6 ticks.

Immediately casting Thundercloud decreases the effectiveness of the initial Thunder cast. The only reason I can see for immediately using up this proc is if your rotation would be interrupted negatively by saving it until Thunder expires.

*5% of 340 is 17. There's a 5% chance that every tick is a bonus 340 potency and full 18s duration Thunder.

[Allyrion]

Snip

I would agree with that if Thunder didn't always seem to proc one right after the other. I've had procs for another one while in the process instantly casting a Thunder III proc.
The odds of that should be astronomical, but it happens quite a bit.
Or I should say, I don't get thundercloud procs much but when I do it almost always chains.

The people earlier were saying they had the same experience, so I'm not going to trust the probability in this case.

[YuriRamona]

I would agree with that if Thunder didn't always seem to proc one right after the other. I've had procs for another one while in the process instantly casting a Thunder III proc.
The odds of that should be astronomical, but it happens quite a bit.
Or I should say, I don't get thundercloud procs much but when I do it almost always chains.

The people earlier were saying they had the same experience, so I'm not going to trust the probability in this case.

Any anecdotal evidence, however common, means nothing if probability provides a sufficient answer.

[Allyrion]

Any anecdotal evidence, however common, means nothing if probability provides a sufficient answer.

It means something when it's common for mechanics like this to be throttled by internal cooldowns.
Everytime I got a proc right after the other because I used it immediately, I made the right choice, right? I'm 360+ potency ahead of the mage who didn't.
Plus you're really not losing much damage from clipping compared to that loss.

If we were still blindly believing the tool tips then we wouldn't even be using Thunder II and using Thunder III only for procs.
It's important for players to try to understand the underlying mechanics when it's so ambiguous and tool tips can be either misleading or outright wrong (despite being the only official information we have on the abilities).

Doesn't matter. As I mentioned, Thunder is always on the target in a normal ST rotation. You're not adding any 'ticks' of the dot, you're using Thundercloud to get the proc hit. It has nothing to do with the Thunder dot and everything to do with using a proc that does more damage than your Fire I spam.

They meant that if you don't use it right away, you give the thunder dot that's already on the target more ticks. Which is more potency on top of the thundercloud proc you use at the end.

They would be right that it's more damage UNLESS you could have gotten a proc while waiting for the buff.
The assumption here is if you get a proc, you probably won't get another for a while. So it's a dps loss to clip the current dot until you have to (a second before the thundercloud buff expires or so).

I don't agree with the assumption but the argument is sound.

[YuriRamona]

​

snip

Are you implying that the tooltip is incorrect, and that there is some other factor besides the 5% chance? While this is a possibility, I'm strongly inclined to believe that the tooltip is correct, unless you have some hard data about this.
​

Doesn't matter. As I mentioned, Thunder is always on the target in a normal ST rotation. You're not adding any 'ticks' of the dot, you're using Thundercloud to get the proc hit. It has nothing to do with the Thunder dot and everything to do with using a proc that does more damage than your Fire I spam.

Well obviously the issue here is then, how easy is it to keep Thunder up? In a standard rotation, when everything is going normal, I would tend to agree. For messier fights where you want to throw out support abilities or re-position yourself, then this is a little more dubious. In situations where you can expect your normal-cast Thunders to expire for extended periods of time without recast, then it will be viable to maximize their duration before using up Thundercloud.

I'm not really sure where you're going with this. If you cast Thundercloud, for example, near the end of a Thunder duration, do you recast it during the time you normally would? If not, then your Thunder expires during your next Astral phase. If so, then you're cutting the effectiveness of the Thundercloud.

If, instead, you'd used a Thundercloud proc in place of the hard cast Thunder, then you'd be able to fit in another spell instead (for example, Blizzard III).

In addition, doesn't Thundercloud eat up a GCD? Casting it immediately would make even less sense if it does.