Summoner / Scholar sort-of in-depth

[Shinryu]

T10 has a lot of adds, which means a lot of time spent on movement for Ifrit and overall less uptime for DK/SE/DE because of the shifting targets, the adds also means more damage potential from Contagion+Bane.

This week I decided to try Ifrit-egi on every turn instead of just T11 and T13. We don't have a MNK, so no benefit from Dragon Kick. Had about a 20 dps increase on T10 and T12 using same gear as the week prior. With a little planning, Ifrit-egi doesn't have to waste any movement time on the add phases.

[Niiro]

I find that very perplexing, I was bored and did a comparison of Ifrit and Garuda just wailing on a test dummy and over 5 mins Garuda came out to 89.2 DPS and Ifrit to 103.6.

I just don't see how Ifrit could be that much of a DPS increase compared to having Contagion, especially without DK.

Even more so if you get Requiem for any significant period of time.

[Taim Meich]

I can't see how Ifrit can be better than Garuda in T10. On one hand, there's the MP issue: Contagion helps to alleviate that. On the other hand, the add phases are a perfect opportunity to abuse Contagion + Bane, which is a huge DPS increase.

For T11, I see the merit of Ifrit, and now that we are going to tackle it (we just downed T10 this week), I might use it even without a MNK in our group. No opportunities to Bane plus very little movement away from the enemies (and the fact that Egg can only be hit upclose) look like a good thing for Ifrit. My only worry is MP management without Contagion.


P.D.: Back in the day I did try Garuda vs Ifrit, using all the pet buffs and attacks when appropriate, and it came down to Ifrit around 125 DPS and Garuda 100 DPS. That's the DPS I did with Ifrit in T6, so it translates well to a real scenario.

[Shinryu]

In my current gear, Ifrit is about 25 DPS higher than Garuda. Compared over 10 minutes, pet buffs on CD, 2 Enkindles, and no resistance debuffs. 149 vs 124. In T10, Ifrit did 147 the other night, Garuda did 120 last week.

The DPS gain from Contagion is very small; all it really let's you do is cast three more Ruins than you would have without. Its primary benefit is MP conservation, which is small enough that it's negated if coupled with Miasma II. Contagion + Buffs is a very respectable increase, but it only comes once every three minutes. In a 9 minute fight like T10, you get to do that three times, so that's an extra 45s of buffed DoTs over what you would have with Ifrit. On the other hand, Ifrit is providing a 20% increase over Garuda for the entire fight.

As for the adds, Contagion + Bane is nice, but the lion's share of the boost is from Bane. Aside from the last add to go down, Contagion is a bit wasted on the adds. Most of them don't live long enough to need DoTs refreshed. Even in T12, I would Raging + X-Pot + Contagion + Bane on the add phase because I'll actually get the full effect, but I still ended up higher overall with Ifrit because the majority of the encounter is single target.

For T11, which pet is better largely depends on number of casters in the group. I use Ifrit because we have 2, but if only 1 caster, I'd want Garuda helping on the cube. With 2 caster, Ifrit on the sphere and SMN on the egg. Egg would be a little annoying with Garuda, but you do have some wiggle room.

[The47thSen]

I'm about to hit 2000 tomes and I have no idea what to spend them on.
I'm thinking maybe pants and belt but then I lose all that delicious PIE.
Can't get jewellery because VIT.

Any ideas?

http://ffxiv.ariyala.com/OXJK

[Kaisha]

Poetic's feet seems to be the only real upgrade available (and are BiS when upgraded to i130) if you're not changing the melded stuff out. You get so much PIE from the 120-130 gear that it doesn't matter much.

You're also 100 acc under the cap for T13 with that linked set if you're working on that turn.

[The47thSen]

I got the pants and boots.
Thinking of rerolling my relic into all ACC. Like this maybe?

http://ffxiv.ariyala.com/OXO0

We're working on T13 but half the static won't be available for the new year holidays anyway.

[cute]

Does the MND reduce effect from Cleric's Stance affect my Faerie?

[Seravi Edalborez]

Does the MND reduce effect from Cleric's Stance affect my Faerie?

Nope. Fairy does not take into account any personal buffs from you. Just gear on summoning (and rechecks gear when out of range, Fey Covenant, or Fey Glow/Light) and where your 30 bonus is. And fairy's own buffs / Rouse of course.

[cute]

Can Titan-Egi parry and will Parry equip add to it?

[Grey Jorildyn]

What the fu...

[SamanosukeShiva]

yes it will, if you believe, you can do anything.

[Taruina]

Anything can parry if you try hard enough.

[cute]

I wanted to use Titan for a FATE to get a minion but it does a lot of damage so I wanted to use my tank rings for extra Sustain heals.

[Lokus]

So since the SMN poetic book is complete shit, I started my SMN relic. I plan on putting capping DET and then allocating the rest to PIE and CRIT. How much PIE is needed to get the extra stat when it becomes Nexus?

[SamanosukeShiva]

I wanted to use Titan for a FATE to get a minion but it does a lot of damage so I wanted to use my tank rings for extra Sustain heals.

you can pretty much solo every FATE with titan without any vit accs, maybe except for behe/odin. For FATE like Eyes Have It you will have to actually maneuver titan to avoid swings and stuff.

[Andradi]

you can pretty much solo every FATE with titan without any vit accs, maybe except for behe/odin. For FATE like Eyes Have It you will have to actually maneuver titan to avoid swings and stuff.

Not sure about summoner but I remember trying Lazy for You on WAR recently for a lark and got obliterated pretty quickly. Dunno if SMN could keep up with the heals and poison on something like that.

[Sute]

Not sure about summoner but I remember trying Lazy for You on WAR recently for a lark and got obliterated pretty quickly. Dunno if SMN could keep up with the heals and poison on something like that.

Just have titan as focus target and heal him with physic when sustain isn't enough.

[Taim Meich]

So since the SMN poetic book is complete shit, I started my SMN relic. I plan on putting capping DET and then allocating the rest to PIE and CRIT. How much PIE is needed to get the extra stat when it becomes Nexus?

I also thought on adding PIE to the book (I'm in the same situation as you, doing Relic since tomestones book is crap)... But then again, if they decide to actually address the SMN MP problem some day, I'll get stuck with a suboptimal stat, so I decided to go full Crit/Det.

Regarding your question, this should help:

http://www.reddit.com/r/ffxiv/commen...on_11_we_need/

[Public]

Perhaps this isn't the correct place to post my findings, but seeing as I'm a SCH and I've done all my testing as SCH I felt as though I should share what I've found. I cannot confirm whether or not these results extend to SMN, but testing from this baseline should be simple. Some of this will be pretty common knowledge, but bear with me. We're going to go on a journey through my thought process.

Starting out, Eos (or any pet) is obviously affected by the player's stats upon summoning, and is merely a snapshot that is refreshed every time SPUR is applied. What else do we know?

Spell speed, although possibly not intended, affects fairies. Notable examples are Whispering Dawn and Embrace.

Fairies do not benefit from the 30 bonus attributes that are obtained via leveling.

Fairies do benefit from the 3% party bonus applied to main stats (not including bonus stats).

Another interesting thing to note is that your fairy might or might not take into account your secondary stats from base, but that's inconsequential for the most part.

As we now know, spell speed and critical hit are both calculated beginning from your base stats, which is 341 for both. As PuroStrider found, each point in SSpeed is 0.000952s off the GCD, but this only applies for each stat beyond the base 341. So, simple enough, your cast time on a spell such as Physick will be (SpSpd - 341) * 0.000952. This also applies to your fairy, and seeing how your base spell speed in this matter does not affect anything, then your fairy will have equivalent gains in spell speed to you. However, Embrace and Whispering Dawn have greater cooldowns than Physick and thus proportionally each point in spell speed will yield more results than they do for the player. At 691 spell speed, Eos will have a Whispering Dawn recast of 52s and an Embrace recast of 2.6s.

Similarly, critical hit works identical to what I've listed above. Completely naked, Eos has an identical critical hit rate to what the player has at a base 5.33% and from then on it seems to maintain equal to the player at higher levels. I'm not entirely certain about this aspect, but simple testing seems to prove this trend.

The real oddity is determination. Using Valk's calculator can easily show you the inaccuracies of your fairies heals. I've done most of my testing in the current gear of: 587 MND, 284 Det, 80 WD, and 617 crit (although crit doesn't matter). Embrace's potency is 300, leaving the predicted outcome at 1000.02 HP healed. My in game Embraces are capping out at 914 and bottoming out at 863, noncrit. Leaving my median Embrace (which I assume is better than average Embrace for theorycrafting purposes) at 888.5HP healed. Alright, so what's going wrong?

This question was brought up on reddit a while back, but was quickly dismissed with the assumption that Embrace does not benefit from the Magick & Mend II trait, because the trait is applied to the player, not the pet. Hwoever, this is incorrect. If it were true, the average Embrace in this gear would end up being ~770. So, what else could it be? By just messing around with the calculator, I ended up reaching a "close enough" number by subtracting 30 (bonus) MND and subtracting 202 (base) determination. This leaves Valk's calculator at 883.05 HP healed, which is pretty close. Next, time to confirm. By purchasing a stat reset, it was really easy to see that Eos did not care about the bonus 30 MND. And, to be honest, the determination was just a guess.

Alright, let's confirm this. Using a higher determination set, 80 WD, 572 MND, 414 determination. Based on just a quick check, my median Embrace is roughly 926 HP. What does Valk's say? 909.26... Not quite. With this gear set, my average Physick is about 1414, what does it say for that? 1359.66... Well, that's pretty far off as well. Well, when I go back to my 587 MND set, the estimation is not that inaccurate, it's less than 1% error, so there must be something missing with the determination calculation.

At the bottom of Valk's calculator, he states that he got some help from a user on reddit named Volpes. Volpes' formula states that the healing calculation is something in the vein of HP Healed = MagDmg(.0189*Mind + .00434*Det + .478) + .485*Mind + .0855*Det and states that he knows that the formula has errors, but it's a good solution for the time being. Well, he has since quit and this subject has kind of fallen on the backburner. I am by no means a great numbercruncher or theorycrafter (you're going to be pissed when I show you my proposed solution) but I am genuinely interested in this, so I decided to take a crack at it.

Knowing what I know about Eos, that was the first solution I proposed. Take determination and subtract the base 202 from the formula. This worked, but overall it made things worse. So, I only applied it to the latter half of the formula. Aka MagDmg(.189*Mind... ... + .0855*(Det-202)). This works a little better, but I can't say for certainty if this really even matters in the grander scheme. Eventually, after a lot of tinkering, I've changed Volpes' coefficients to make the formula MagDmg * (0.017925 * Mind + 0.007525 * Det + 0.375) + 0.52625 * Mind + 0.05625 * (Det - 202). This, of course, assumes a 400 potency spell, so any factoring of potency will be SpellPotency/400 times the formula.

Let's see what I've come up with. For the 414 determination, 572 MND build I now have an estimated heal of 1412.4, and an actual median heal of about 1414, give or take 1 HP. My Embrace at this calculation is estimated at 924, with my actual at ~926. At 587 MND, 284 det build I have an estimated 1337 for Physick, and which is almost exactly correct. And my Embrace? Estimated 889.5, actual being ~888.5. Doing some other testing across other builds and combinations this seems to be very close, with an error of usually far less than 1%, BUT I am looking to see if someone would be willing to either help me from here or work off what I've built off of to find a better solution. I am quite lost at this point and I don't know how to mathematically calculate the true weights of each stat, so I've relied on simply tinkering and testing, which has become insanely tedious.