Jack Of All Trades: A Guide To Red Mage

Also if you're casting a spell not affected by potency with Stymie. 5/5 empy will have higher duration than mixing relic in due to set bonus when Stymie guarantees the spell will land.

Depending on the spell, you're not swapping out the relic feet

I'm curious to see any info/testing on the new Archduke's Sword. I'm sure it's still way behind most serious options, but the different wording ""En" Spell Damage" makes it sound like it's a different effect from the typical "Sword enhancement spell damage." My first instinct was that it was just worded that way to include Enlight, but Brilliance also uses the "Sword enhancement spell" wording for duration.

Can't get it until friday.

I don't think anyone has seen it drop yet

Also

The archduke event said in the event posting

"Please note that there will be no changes to the spoils you can earn from the battlefield on these three occasions."

But then again also weird because only data miners are gonna see that they added said sword

This could imply the actual battle field drops itself now that I think about it

Carry on ...

So, any insight on the Archduke's Sword yet? Specifically, wondering about potential as an /NIN or /DNC offhand to Crocea Mors C for a heavily enspell focused build.

Still significantly worse than other options for RDM.

If you were intent on keeping only swords equipped while nuking it might be useful for BLU? Really don't know why you'd do that though.

It does have good macc/MAB so it'd be easy to think it can pump up Enspell damage while not gimping, say, your nuking prowess, but the enspell damage only applies to hits from the sword itself, making the added DPS quite meh. At that point you're doing more overall damage by swinging a Demersal Degen +1 and switching weapons for nuking, even with the loss of TP.

The duration makes a max potency Enspell last 24min instead of 18, but... it's entirely marginal and at that duration nobody really cares.

It's not an RDM piece imo. Maybe if you don't have the other options, but none of them (outside Crocea) are hard to get anyway.

It does look very nice tho!

This.

For Crocea Builds offhanding Thibron (if macc allows) or Daybreak (if you need more macc and/or wanna spam Seraph Blade) are better options, imho.
Now if the enspell damage applied to both hands it would be a different story, but it's not and to my knowledge no single weapon boosting enspell damage applies to all hits.
In addition to Crocea and Archduke's sword there's also an Unity sword with augment Enspell damage, forgot the name.
But it's all the same story.

Imho it's a secondary option compared to the main two I mentioned.

Yes you covered the ws buildes with Crocea, these are the pure enspell builds:
Levante dagger with wind element enspell is the best offhand for enspells as long as wind is alright, the wind affinity applies to both hands.
Pukulatmuj +1 with full augment is the next best option for when wind isn't an option.

Archdukes is pretty far behind and its duration just moves you out of sync with temper II for RDM

Where I think Archdukes is interesting is Enlight/Enlight II on PLD where it is an option instead of Brilliance, +60 sec duration vs +15 sec duration and 15 skill. Enlight only has 3 min duration so the added duration is much appreciated but it does lose potency with each hit so the added time may not get you that far and you may prefer the skill for potentially higher potency.

For a build that is focusing entirely on enspell damage and nothing else, then you'd want demersal degen +1 or a Pukulatmuj +1 in the offhand anyway. The first is for the OAT and the second is basically a 1/2 potency crocea, which would do more than flat +30 to enspell.

Edit: Augments are +50% Enspell damage and +150% Enspell damage respectively.

It's a shame the offhands are so much worse

For those lucky ones with RDM+2 earring, which sets do you use it on?
I'm curious to see if I missed some potentially good use on some sets.

Also do you use it specifically on any WS? On TP sets I can see the potential, sounds worth it even if you lose the STP in the same slot, but I didn't math it out.

TP set, Savage Blade set, CDC multi-hit set. Its really nice if you are lucky enough to have rolled max stats on it too with +15str/dex and 8% double attack. The better the roll the more its plastered into its right ear spot lol.

My Savage blade set involves a flametongue along with the +2 earring and makes for some gnarly savage blades. Which also comes with some nice magic acc on it. (so you dont have to macro weapons in just for Magic acc and lose tp)

I don't use it in TP sets due to Temper II. I already use Sherida Earring, so I try not to add too much DA if at all possible, but it's a good earring for TP IMO. I'm usually /DNC on RDM (so my earring slots are taken up by Eabani+Suppanomimi; /NIN is a DW cape that frees up Ear for Sherida), but even single wielding I go with something that has STP (Telos, Dedition).

I use it for both attack capped and uncapped Savage Blade and Black Halo sets, as well as Requiescat. Precast all spells, midcast Impact and Absorb-TP.

I'd imagine after a certain point, ranked Alabaster Earring pretty much replaces JSE earring +2 in several sets unless you have a high roll (mine is only attribute+7), but there's no way I'm ranking that earring up at this time, so it fills a good number of sets currently for me.

I've been pondering on TP set for that same reason.
But being that in the Multiattack order DA comes after TA, wouldn't that mean that DA would happen only if TA doesn't?
This diminishes the "power" of DA, but it doesn't hurt TA because it gets calculated after, right?

Correct. RDM doesnt have so much TA that the DA becomes useless, but there's better raw TP earrings like sherida or the limbus ear if you can manage to max it out, or crep, dedition etc since STP wont get devalued as much imo

All things being equal, you can usually evaluate 1 DA to be roughly equal to 1 STP. I don't think this holds true for rdm though, because with Temper II you should be running around with 40+% TA between the buff and gear. That means STP is going to be greater than 1:1 vs DA.

DA is definitely not useless, but having an extra +5 STP on all those triple attacks is going to be pretty beneficial for a rdm. That being said, using +2 earring is still going to be pretty decent. I've got a mid range one with +8 DA, so it's fine. +5 STP from Telos/crep/alabaster (aug'd) will also be just fine. I don't know specifically where +5 STP falls in relation to 7/8/9% DA from earring. They're likely very very close though.

Its still generally true as both MA and STP have diminishing returns. Just like you have 30~40% TA, you also have 20~30+ Store TP, you aren't adding 5 STP to 0 you are adding 5 to ~30.

The real way is to figure out the percentile increase in average attack per round for the MA option, then figure out the TP per swing increase on the STP option, and go with which one is higher.

Handy spread sheet to figure out average attack per round (Option 1)

https://docs.google.com/spreadsheets/d/1Zv6C1D04LxwjEE_464C4_Hv9eYqtaezQKfCcfWafwZc/edit?gid=0#gid=0

From here
https://www.ffxiah.com/forum/topic/31534/math-oax-multi-attack-and-you/

For option 2, just take (Store TP +100)/100 of both sets to get the percentile difference.

It doesn't matter, there are no diminishing returns for STP.

Say gain 100 TP/hit.

At 0 STP, 1 STP = 1 TP/hit

At 30 STP, 1 STP = 1 TP/hit

At 500 STP, 1 STP = 1 TP/hit

You could argue that it's less relative to where you already are (like 5000 + 1 is less of a % increase overall) but that's just...how all stats work? Do people stop using attack because it's a lower % increase with every point of attack you gain on gear?

Attack doesn't have a breakpoint at "get 1000 and hit the button to do huge damage". You're not factually wrong but it seems like sophistry to insist on this.

STP is literally the exact opposite of diminishing returns. It is a 100% linear return.

The only way it's not linear is if you look at the numbers in a completely bizarre way, which makes no sense.

Going from 1500 attack to 2000 attack is a 33.33% increase.
Going from 7500 attack to 8000 attack is a 6.66% increase.

That doesn't mean there are diminishing returns to attack. That is, fundamentally, not how the concept of diminishing returns works. Otherwise all math would fall apart.

the XI community has always misused the concept of diminishing returns. It's generally understood within the community that when someone says diminishing returns in XI discussion what they're actually saying is diminished relative gain.

This comes up a lot in this game because they are a lot of linear or mostly linear stats that end up effectively muliplying each other and when that happens the more you have of one the more valuable the other becomes even if it is linear. So in this case the more multi attack you have the more valuable each point of store tp becomes, more hits = more tp gained from the store tp but the reverse is also true, the more store tp you have the more valuable each point of multi attack becomes.

It is because we are comparing two different states to each other and choosing which one is "better". To figure that out we need to know how much of an increase option A is, then figure out how much of an increase option B is, then pick the higher of the two. Relative gain is what we are looking for, and most values in FFXI have diminishing returns such that the value of any one amount is going to largely depend on how much of that amount you already have.

I suck at math, but if I'm correct in math the definition of "diminishing return" is a bit different from the way we commonly use it here on FFXI.
Maletaru's definition of STP increase being pretty linear is not wrong at all.
Altough I guess this doesn't take into consideration the factor of breakpoints. When you approach those breakpoints (meaning one less melee hit to reach >1000TP) even a single +1STP can generate a massive increase in DPS. So that in itself is tricky to define compared to STP additions away from breakpoints, which are indeed pretty linear.

In the case of DA+8% well... what I was trying to say is that for a job that has zero multiattack, that 8% DA roll is gonna be calculated every time.
For a job who has 40% TA, that 8% DA roll is gonna be calculated 60% of the times instead of 100%, i.e. each time TA doesn't proc.
So from this point of view the influence of that 8% to your overall dps will be of course inferior to what would be for a job with no other multiattack.
It still matters though, and thanks to the fact that the game processes Multiattack in an order that is usually beneficial for us players (TA before DA) the presence of DA will not hurt your TA rate.
It would've been different if DA was calculated before TA, but thankfully that's not how it is!

The hits-to-1000 does not matter anymore with all current WS's having linear or near linear damage scaling.



This is completely irrelevant, order doesn't really matter.

The link I provided earlier is a online spreadsheet that does the Mult-Attack math to generate the two most important values, Average Hits Per Round (better stated as attacks per round) and Single Attack Rate. AAPR is number of swings you will get every attack round and is the result of all that Multi-Attack stacking. If you have an AAPR of 1.5 and it goes to 1.6, that is a 1.6/1.5 or 6.6% increase in TP gain. Store TP acts as a direct TP multiplier, going from 50 Store TP to 60 Store Tp would be going from 1.5 to 1.6 TP multiplier, or a (1.6/1.5) 6.6% increase in TP gain.

TP gain per round is (Base TP * Store TP * AAPR), because those two values are multipliers of each other we can directly compare them as a relative percentage increase.

A note on SAR, it represents the chance of having an attack round without any MA procs, which essentially stalls TP gain. This is why TA and QA focused builds can sometimes stall out with TP gain. Take something like 50% TA vs 100% DA, both have a high AAPR of 2.0 and will have identical TP gain over time. The 50% TA build would have a 50% SAR while the 100% would have 0% SAR, meaning the TA build will go two consecutive rounds without a MA proc 25% of the time. Both have the same mathematical gain rate but will play differently.

Yes, but also: diminished relative gain is a pointless term, and it's a waste of time to even think about it.

Every stat in the entire game has diminished relative gain.

MDT has a reduced value for every % you put in.
MAB has a reduced value for every MAB you get.
Attack has a reduced value for every attack you equip.
Block rate has a diminished relative gain for every % you put on.

I agree with the idea of X-hits...in theory...? Though very few people actually build for this, for reasons already mentioned above, plus never knowing what Samurai Roll you will have (or not have), plus the chance that you'll multi-attack straight past the X number of hits, and a dozen other reasons. Sometimes you're building to 1500 before you WS, or 2000, or you're saving up for AM3 and getting 3k...there goes your X-hit build.

All that said, there's certainly a discussion to be had about balancing STP and MA. Although once again...the "diminishing returns" of STP are not at all relevant to that discussion. Every STP gives the exact same amount of TP to your swings. Whether or not you want to target a certain breakpoint, balance a certain amount of STP and MA to get the "feel" or consistency of your build right, or whatever...it's no more a diminishing return than literally every number that goes up in the entirety of human civilization.

If your bank account gets $1 in it, you've gained $1. That dollar has the same value regardless if you have $1,000,000 in your account or $1. Saying it has a diminished relative value because of the status of your account is pointless. It's worth a dollar. Subsequent dollars aren't worth less because they were deposited into an account with a different balance.

While I generally agree with you about diminishing returns/relative diminishing gains, I gave up that fight 10 years ago. The community wants to use the term this way, and will not be deterred.