Black Mage Sanctuary - A Guide to DPS

[Maero]

Except back in 2.o determination was actually of higher value then crit for sometime
Testing also needs to be done if DH has any impact, i know Red Mage it plays more of a role

[soslinky]

If anything, most of the data I've read on the forums/reddit are of DH and how the numbers indicate it being the better stat. Until someone is able to crunch the numbers on Crit and Det then DH is the safest bet.

[Galvuu]

This is mathematical.
The scalings change with each expansion, so whatever was true for HW has no bearing on what's true for SB.
Leaving speed aside for a moment (hard to quantify, as usual), dervy and a few other folk has a reddit thread with preliminary numbers (quite a bit of data, but they wanted finer points for the interpolation).
These numbers showed that, for SB, just for a pure dps perspective, DH>>Det>CH.
This makes sense because crit scaling got cut down to a third (you need like 66 points of crit to grab the 1% crit chance).
We have nothing in our kits that further augments crits and nothing that benefits from it (we don't get a Firestarter for sure if Fire crits, for example), so all crit gives is raw dps.
Therefore, as far as we know at the moment for SB, Det might be the better choice, for our secondary stat range.
(Can't find the thread :/)

[HaroldSaxon]

This is mathematical.
The scalings change with each expansion, so whatever was true for HW has no bearing on what's true for SB.
Leaving speed aside for a moment (hard to quantify, as usual), dervy and a few other folk has a reddit thread with preliminary numbers (quite a bit of data, but they wanted finer points for the interpolation).
These numbers showed that, for SB, just for a pure dps perspective, DH>>Det>CH.
This makes sense because crit scaling got cut down to a third (you need like 66 points of crit to grab the 1% crit chance).
We have nothing in our kits that further augments crits and nothing that benefits from it (we don't get a Firestarter for sure if Fire crits, for example), so all crit gives is raw dps.
Therefore, as far as we know at the moment for SB, Det might be the better choice, for our secondary stat range.
(Can't find the thread :/)

The thread on Reddit? Yeah... go read the comments.

[SunnyHirose]

For BLM, there has been an almost negligible difference between weights based on a dummy rotation and weights based on a theoretical rotation of one GCD. Even though buffs mana tick stuff ought to be accounted for, these don't change the spell speed weight significantly. Nearly all BLM damage has to come from weaponskills, and therefore benefits from spell speed about equally.

Moreover, unless core BLM gameplay stops being mostly hardcasting, it is hard to dethrone spell speed as a practical personal preference regardless of other stats' effectiveness. If you factored in a reasonable estimate of how many casts get lost for lack of it, you might end up having to utility-weight it higher than INT if you're really bad at it (like me). Now for career BLMs, I feel it's more like by the time you hit the SS threshold where more won't make things a single bit easier, it has already hit the point where increasing returns on SS has it outweighing other secondaries.

[Galvuu]

The thread on Reddit? Yeah... go read the comments.

https://www.reddit.com/r/ffxiv/comme...o_stat_values/
This is the thread for reference.
I read the comments... what exactly should stand out here?
He has some ultra minor corrections and people asking about crit for BRD/MNK due to their job mechanics and some talk on tenacity VS CH.
You need to be a bit more specific...

(Note that dervy posted this. The guy that does the stat weights every two patches or so. I tend to be partial towards believing him, and having glanced over his numbers, it seems to look aight).

For BLM, there has been an almost negligible difference between weights based on a dummy rotation and weights based on a theoretical rotation of one GCD. Even though buffs mana tick stuff ought to be accounted for, these don't change the spell speed weight significantly. Nearly all BLM damage has to come from weaponskills, and therefore benefits from spell speed about equally.

This is true for any job. The reason we've always been able to get away with speed stacking is because of our infinite mana pool. That, and our dots frequently clip due to Thundercloud (and you can swap out T3 for something else (B4) in your UI cycle if the timer is still very long on the current dot).

[SunnyHirose]

Well, that is the dirty secret behind all weighting in FFXIV (so far). You can hold off on actual potency estimates for a lot of jobs (except bard) for months and still come up with decent gear recommendations so long as your estimates of overall crit rate and weaponskill time are good. Particularly because SE has predictably made main stat that much more powerful than secondaries. But abilities (being unaffected by SS) end up adjusting that a bit. Autos also used to dampen SS value... not so much today.

Come to think of it, tanks and healers also don't need a whole lot of damage estimation to get alright weights for damage purposes, either. Now, if only I hadn't spent all of HW meditating on the mysteries of SS rounding....

edit: Fermi estimation can get you really far sometimes

[paoweeotter]

I'm very late to the party and just read galvuus theory crafting on 3.0 vs 4.0 rotation and wow that is such an eye opener! I love the ease of using the 3.0 rotation and it is good to know that that is our highest DPS rotation.

I don't have any qualms with b4 not being part of single target at all in fact I would prefer it if it was only useful for aoe double flares. the 3.0 rotation is much easier and that helps with all the movement were gonna be doing in deltascape 3 and 4!

[Galvuu]

snip

Well, nothing stops us from gathering an arbitrarily large amount of data beyond time and/or patience.
We can do a fairly accurate fit with that, adding to the given certainties (potencies and other stuff on the tooltip).
When you have a million or two of points, variance simply statistically dissipates. And since it's all relational, we don't even really need the actual damage formula, either.

I'm not sure why we're discussing this, but I like it, even if it has almost no relevance to the topic.

I'm very late to the party and just read galvuus theory crafting on 3.0 vs 4.0 rotation and wow that is such an eye opener! I love the ease of using the 3.0 rotation and it is good to know that that is our highest DPS rotation.

I don't have any qualms with b4 not being part of single target at all in fact I would prefer it if it was only useful for aoe double flares. the 3.0 rotation is much easier and that helps with all the movement were gonna be doing in deltascape 3 and 4!

Interesting. The only two mechanics that annoy me in all of Deltascape are some V1.0 ball patters and V4.0 void Blizzard 3 (it's actual bollocks).
That said, I've been testing the 3.X rotation, and have >4.3k on all of them except 4.0 because of Decisive Battle. This is a slight bit over the 4.0 parses I was having (~4.2k), so there it is.
I'm glad I was of service, either way

[SunnyHirose]

I'm not sure why we're discussing this, but I like it, even if it has almost no relevance to the topic.

If someone is talking to you, you have something they want. I don't think it's cynical to say this... it's a law of negotiation! This was something going through my mind while questing SB, in fact; so much of it, especially the MSQ, is straight up gobbie busydeals.

Anyway, that being the case, at least what I think I want is for people not to be afraid of collecting and discussing the numbers, as if they must be treated with mystique and caution and wait for the experts to hand them down from an ivory tower. No way, guys. There are no experts; none but all of us. A lot of the open discussion on this (at least on the English-speaking side) kind of died down during Heavensward. What was left moved to Reddit mostly, where it gets talked in circles usually and must eventually fall off the front pages so there is not much followup in the end (that is the software's purpose, and I don't blame it or any people for that; I'd much rather not play any part in that game where people try to put down one forum or another). I miss that wrong-but-honest ARR attitude towards theory, with a lot of spreadsheets and point-by-point testing flying around, where you can track which insights lead to what conclusions. To me, that is a better way than what is going on now, even as the very last formulas left seem to be getting close to 100% accuracy.

Otherwise, as per paragraph #1, I'd avoid posting here (or anywhere) and attempt to get on with pwning raids instead of trying to share a single tip, opinion, or equation. Yet here I am at post #549, tryna do my best by as many folk as I can, contrary to my ornery nature.