A Summoner's Gear Guide

Fair. Was just curious if there was something I was missing.

Anyone have a hybrid melee set for doing Blaack Halo + Flaming Crush spam?

With Shiva's crystal blessing, tp bonus earring, and /war tears 650 to bonus, which sounds like you can pump out some very respectable black halos and light SCs.

I feel like Black Halo > Flaming crush should be able to take down apex mobs? You could still swap staves for that too since your closing with the BP.

Thoughts?

Does anyone have an opinion on Draumstafir, the su5 dynamis weapon, specifically, Path C?

Route C
[1] Pet: Chance of double damage +50%
[2] Chance of doubling "Blood Pact"
status +25%
[3] Blood Pact Dmg.+15

What does #1,2,3 mean with respect to Volt Strike, Flaming Crush and also Ward BPs?

All I know is its weaker than Nirvana.

on the other guide frod did some testing with it and posted is fell somewhere in between jse and nirvana, but not 100% on which side of the spectrum.

Thank you!

useless for wards.

#1 is double damage on melee regular hits, unlike double attack, it doesn't work on bloodpacts.

#2 is double stat bonus to blood pact rages, we... don't really have much gear that gives base stats, so it doesn't seem to have much effect, maybe a damage difference of a couple thousand when procced.

#3 judging just this stat, it's behind even was. potentially with #2 this might be higher.

For my nuking set, I opted to go with pet MAB instead of BPD. I use glyphic horn and spats +3 and convoker body and feet +3 in addition to merlinic hands augmented with BPD MAB and INT. Is BPD a better option for nukes, or is MAB and MACC the way to go?

Avatars have naturally high MAcc, so you don't need a lot of it although some is certainly good.

As for BPD vs MAB, they're interchangeable, and the ratio depends on how much of each one you have. Generally, it's about 1 BPD = 2.5 MAB.

So Glyphic Horn +3 vs Apogee Crown +1 [A] is a wash. 25 more MAB and 10 less BPD. Should perform about the same, so you're fine there.

Glyphic Spats +3 are a huge downgrade from other options. Enticer's Pants are hard to compare because their main value is from the TP Bonus, but if we compare to Apogee Slacks +1 [A] then it's a big loss. You gain 16 MAB, and lose 17 BPD. It's also a loss against path D slacks.

Convoker Body +3 vs Apogee Dalm +1 [A] is also a loss, but not a total loss. You give up 35 MAB to gain 6 BPD, a downgrade to max damage potential, but you gain a decent chunk of MAcc in the process.

Convoker Feet +3 vs Apogee Pumps +1 [A] is a huge loss. You're giving up 2 BPD and 35 MAB to gain some MAcc.

If you're curious about more comparisons, I wrote a calculator in javascript that'll help you compare BPD against MAB: https://pastebin.com/eWurs5V1 (Save the file as ".html" and then open it.)

Thank you for the write up and the link. I guess I didn't realize that glyphic slacks were a downgrade from apogee +1, and I've really been avoiding apogee feet +1 because I worked pretty hard to get +3 everything. Guess I'll make those feet and fix my gear sets .

Thank you for this outline! I have not had very much success with the Gearswaps floating around, can anyone share theirs using these equipment sets?

Here's mine: https://pastebin.com/Fa5PtueC

Thank you so much, really appreciate it!

Where does Glyphic Doublet +3 fit in? I've been using it in my Physical BP set. Seems great, I'm doing fine damage wise according to SB. With that mixed in avatars have +47% DA and the 15% crit. I don't know how 20 DA, and 15 Crit stack against 16 BP on Convo +3.

I have +131% BP with the above for Physical.

15DA,20Crit

I think the formula aligns with the more BP you have the more DA will be worth and visa versa. With all things equal 15DA will do more than 16BP. However, AF has accuracy and from many many posts back I recall Crit does not do much for the BPs we use. So, I guess if you know you’ll be acc capped and the formula shows you DA is more beneficial for your equipment then Glyphic should win.

Paged through a lot of the summoner threads here, but never found anything definitive on Pet: STR/DEX/INT from augments -- like on Merlinic Dastanas for example.

Is it understood what the ratio of say Pet: STR or Pet: INT values would compare to with regards to "Blood Pact Damage" and/or Pet: Attack / Pet: Magic Attack Bonus?

Based on what I've read, I'm assuming this is a no. My understanding is that the specific details of how these are calculated, and how they compare to other stats is unknown, and what we currently know and have to work with are Pergatory's calculators that compare the BPD+DA, and BPD+MAB values. Great work on those by the way!

If my assumption here is correct, then would it be best to state that going for the Pet: Attack and Pet: MAB is likely to yield better results, in a general sense, than sticking with Pet: STR or Pet: INT on your Merlinic pieces? Any of those in conjunction with BPD and Pet: Attack / Pet: MAB are certainly welcome, but I'm more curious when you have to choose between say a low attack, high STR piece versus a high attack, low STR piece for Physical BP. Obviously, there are niche scenarios, but I'm going for a all-round general recommendation or guidance here. I'll gladly take any niche scenarios into account if you can provide or explain those as well, since I could always better myself as a SMN.

If it helps to clarify what I'm asking, then think of it in the sense of you have a BPD+10, Pet: STR+10 augment, but you get a roll that's BPD+9, Pet: Attack 28, Pet: STR+2. What would be the way of determining what would be better overall for Physical BP in those instances? Pet: STR+1 equals +2 attack or something like that, if this is even known?

I can also read any existing thread on this, if I've somehow overlooked where this was documented. Thanks!

No, the specific details of that aren't known. Not a lot is known about blood pact damage formulas, unfortunately.

MAB will be significantly better than INT for sure. INT is still nice, but MAB being a multiplier, it doesn't really compare.

STR is another story. I'd say it's likely better than attack. Probably not better than multipliers (BP Damage or Pet:DA) unless it's a large quantity. It's anyone's guess, though. You may have noticed in my Lua that my physical Merlinic aren't even 10 BP. I've gotten a few with 10 BP and some acc/atk, but my current ones have 9 BP, 10 STR, and 24 attack, and I've been unwilling to give up 10 STR for 1 BP Dmg. That may actually be foolish, it could be 1 BP is far better. No one knows, but I've always felt like Pet:STR is not to be overlooked ever since the atma days when it "felt" like Atma of the Voracious Violet and Stout Arm both gave a significant boost to BP damage.

The only thing we know on +stats is the chart SE gave awhile ago that essentially pointed out certain stats affected certain pacts like weaponskills. The issue with this is twofold, we have next to no gear that affects this, and the values are unknown and considered minor compared to straight bpd, da and mab, even draumstafir's aug didn't give all that much at first glance.

In order to test it we'd need someone insane enough to make full sets of merlinic with proper stat augments and thats just not gonna happen.

@Pergatory -- thanks for the insights! Very nice to know that the Pet: INT appears to be far inferior based on your knowledge and testing with the MAB multipliers! Keeping those thoughts on the Pet: STR will also be nice to have while I'm augmenting things.

@Frod -- thanks for the additional context on the stats! Happen to have a link to that stat chart that SE gave? I'd be curious to give it a read through myself to better understand the context and point you are making.

Since we have access to the new JSE neck it should be possible to test some, since you can increase the str/int augment bit by bit.
But im far to dumb when it comes to such things >.>;

Stickied in summoner section of OF

http://forum.square-enix.com/ffxi/threads/52561-Blood-Pact-Attribute-Data

the million dollar question though is why the *** is impact dex/chr.

@Pergatory, @Frod,

If I were crazy and tedious enough to do the testing in game, then what scenarios should I cover, and where should I cover them?

For example, I was thinking that being naked with only a Sancus Satchet +1 could be a good baseline control for a 119 avatar. I could then populate values with Ramuh's Volt Strike, and Ifrit's Flaming Crush. I imagine others would be interested in the other Pet: Attribute besides just STR and INT for future reference. Ideally I would also do this with (1) Sancus Satchet +1 & Campestre's cape, (2) Sancus Satchet +1 & Nirvana, and (3) Sancus Satchet +1, Campestre's Cape, & Nirvana to cover all those possible Avatar + Levels. I don't have a Nirvana, however, so I could only cover half those scenarios that I'm thinking of.

How many samples would be meaningful to gather? 1000, 5000, 10000 samples?

Would there be a specific place to do the testing that would be appropriate? For example, would it be best to test in like Gustaberg on enemies that will die in 1 shot, or would it be better to test on an Apex enemy somewhere? Being naked for controls, I would assume the former.

From there I would probably just use the JSE neck, and slowly use heroism's to bump it up 1 Pet: Attribute point at a time -- thanks to Wotasu for that idea, and compare those samples with the same avatar's, abilities, and locations among one another to see if we can find anything significant. I could then also compare my testing with the neck to augmented pieces of have of similar Pet: Attribute stats to basically see if the damage of the sampling period is the same for a one-to-one trade via equipment slots, or if it has a variance.

I don't know if I'd actually go through with all of it, but the prospect of figuring out an unknown is enticing.

Honestly I think the Avatar:Lv+ stuff will just muddy the waters for you. Changing avatar level changes all sorts of other variables like acc/atk in addition to STR. So I would start by focusing on changes to damage from changing NOTHING ELSE besides Pet:STR. The collar testing sounds like a much better and more controlled way of doing that.

Possibly an even better way would be in Abyssea/Voidwatch, since Atma & Atmacite affect pets. Voracious Violet gives regain which will throw the numbers off, so probably don't use that. Stout Arm gives attack, so probably don't use that either. Something like Atma of Baleful Bones would be ideal, it gives STR+20 and nothing else besides accuracy. Also, make sure to account for the mods in the link Frod posted. For example if you're testing STR changes on Volt Strike, STR is a mod on that BP. So it will have a different effect on Volt Strike than it will on, say, Pred Claws.

1000 samples wouldn't be enough. With the size of increase expected from a 1 STR change (probably a small fraction of a percent), your margin for error also needs to be in the area of 0.1% which is insane. I'm no statistics expert, but I'm not sure even 10000 would be enough. Testing larger increments than 1 would make your life a lot easier but I think you'll still want a margin of error that's well under 1%.

'Tis a noble goal. Godspeed!

Hmmm from some initial glances into the formulas to calculate the margin of error, and some quick calculations, not even 10,000 samples would reach a margin of error less than 1%; 10,000 samples would only get me 1.395% margin of error when using a level of confidence value of 99.7%. Sadly, it doesn't really seem feasible for manual testing. If it was something we could sim, grab the output, and analyze, then perhaps we could figure something out.

Thanks for the inputs, and answer my questions!

I was curious about this myself. In terms of an physical accuracy set, could this weapon be more useful than nirvana? or does the LV+ outweigh the amount of acc it provides?

Draumstafir doesn't come close to nirvana.
it might skew a little differently with a finished +2 neck, but i don't expect draumstafir to even beat Was nq or +1.

The double damage effect is only on melee hits, so no DA effect on volt/pred. If this were straight DA, it would outpace all but AM3 nirvana.

The bpd is only +15, vs nirvana's 40 (and was's 20/25), again, falling vastly short of nirvana's stats.

the chance of doubling bp status is a 1 in 4 chance of doubling the stat modifiers on bp rages. We don't have that mathed out, and doesn't seem to be that big of an increase. It doesn't affect debuff potencies, it doesn't affect ward potencies, nor does it have any effect on duration.

To further clarify, if I remember correctly Nirvana gives some 72 accuracy from the Lv+2. If you have it at rank 15, that's another 30 for 102 accuracy total.

Gridarvor has 95 accuracy. Was+1 has 55 accuracy. Draumstafir has 50 accuracy.

So accuracy-wise, it's actually at the bottom of the heap.