So Summoners! Someone posted this on reddit. Discuss!

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

So: First result that I know you all will not like.
Ifrit-egi > Garuda-egi in 2.1. This is because, with 1 fewer dot, the extra potency of Ifrit and extra 1 ability every minute (20 auto attacks vs 19 auto attacks of garuda) add up to be more damage than the raging strikes + dots + contagion extra DPS. This means that Ifrit-egi is going to be almost always our go-to pet, except for situations where pet target changes frequently, or mobs are moving a lot and its hard to put the pet in a safe spot immediately.... (continues see post if you want to read more)

^^^^^
Discuss!
ADDENDUM:
WHAT HAVE I WROUGHT, but yeah since i can't post thanks Torin for the BiS <3

That guy has no idea what he's talking about.

i'd really like to use ifrit,i like it and has his strong points,but (assuming al paper maths are right and i dont think because ifrit has a bug in cooldown,real one is 2.5 instead of 2 written) ,ifrit is melee,so in t1 ads will take a lot of dmg (i still have to try how ifrit takes dmg by aoe because pets has strange resistance compared to char,some skill are more effective and other are so much less,anyway we cant loose time to heal pet with sustain and psych every time
i always thought that ifrit is better ,but for trash mob hunt,you can redirect him as u wish and by my test in short duration fights (30\40 seonds) do much more dmg than garuda,but without looking at contagion+rs bonus so garuda will be remain the most usefull one,
in T4 Ifrit is usefull if you have only 1 melee dps because he can hit knights (but you have less dps on soldiers)

He's listing a 427 accuracy set as BiS while also recommending to use ifrit which has an even higher accuracy requirement than garuda does.

Quote:

---

Originally Posted by xardus

in T4 Ifrit is usefull if you have only 1 melee dps because he can hit knights (but you have less dps on soldiers)

---

Garuda can hit knights without any issues.

Hmm, the additional Potency of Thunder with Contagion is only 175 compared to 550 from the other 3 DOTS. It's nice to have fights where you use Ifrit over Garuda and vice versa, and perhaps even fights where you should switch pets for optimal DMG. But at the point where Ifrit can out DPS Garuda with Raging Strikes + Contagion on 3 DOTs (4 if you include Miasma II which is better than casting Ruin with Contagion boosting bonus Potency to 600) makes it very unlikely to find a spot for Garuda other than DOTs > Contagion > Bane.

Quote:

---

Originally Posted by OmegaNovaios

Hmm, the additional Potency of Thunder with Contagion is only 175 compared to 550 from the other 3 DOTS. It's nice to have fights where you use Ifrit over Garuda and vice versa, and perhaps even fights where you should switch pets for optimal DMG. But at the point where Ifrit can out DPS Garuda with Raging Strikes + Contagion on 3 DOTs (4 if you include Miasma II which is better than casting Ruin with Contagion boosting bonus Potency to 600) makes it very unlikely to find a spot for Garuda other than DOTs > Contagion > Bane.

---

we can always start with garuda,apply all dots+rs ->contagion->swiftcast->ifrit



another thing we are all forgetting is...ifrit stuns...and interfere with other stun make theme resist sometime

Ifrit dies too easily for my tastes. I'm more than willing to sacrifice a possible slight dps advantage for a more reliably alive pet.

Quote:

---

Originally Posted by xardus

we can always start with garuda,apply all dots+rs ->contagion->swiftcast->ifrit

ifrit stuns...and interfere with other stun make theme resist sometime

---

Yeah, I guess you could open up with Garuda and then swiftcast Ifrit in but you will need to resummon Garuda again on longer fights. But it still better than casting Ruin x3 during your rotation. But as long as Ifrit on obey is stronger than Garuda on obey then it's fine because TBH Garuda on obey is stronger than Garuda on Sic because her actions reduce her DPS since they aren't optimal for single target.

Quote:

---

Originally Posted by Mhaeric

Ifrit dies too easily for my tastes.

---

Yeah, Ifrit should be stronger than Garuda for the same reason Melee should be stronger than Ranged, in addition to having to move your own character moving the pet is a lot of work. So on fights where Ifrit WILL die then in most cases Garuda will be better.

Quote:

---

Originally Posted by OmegaNovaios

Yeah, I guess you could open up with Garuda and then swiftcast Ifrit in but you will need to resummon Garuda again on longer fights. But it still better than casting Ruin x3 during your rotation. But as long as Ifrit on obey is stronger than Garuda on obey then it's fine because TBH Garuda on obey is stronger than Garuda on Sic because her actions reduce her DPS since they aren't optimal for single target.


Yeah, Ifrit should be stronger than Garuda for the same reason Melee should be stronger than Ranged, in addition to having to move your own character moving the pet is a lot of work. So on fights where Ifrit WILL die then in most cases Garuda will be better.

---

No.

Stop talking about this as if the person who posted was right.

He/She isn't.

Don't spread bad information, we already went through this 2 months ago where people thought Ifrit was stronger.

He isn't.

Period.

Quote:

---

Originally Posted by Kevee

No.

Stop talking about this as if the person who posted was right.

He/She isn't.

Don't spread bad information, we already went through this 2 months ago where people thought Ifrit was stronger.

He isn't.

Period.

---

You didn't lose Thunder 2 months ago.

Only thing for certain here is you can't now until 2.1 hits and you go out and do some testing to confirm.

Theorycrafting based on tests performed in 2.0 suggest, however, that the loss of Thunder not only diminishes SMN DPS by small chunk, but also significantly diminishes the value and effect of Contagion (the skill that was the primary reason Garuda > Ifrit universally) and gives Ifrit the edge.

Quote:

---

Originally Posted by OmegaNovaios

Yeah, I guess you could open up with Garuda and then swiftcast Ifrit in but you will need to resummon Garuda again on longer fights. But it still better than casting Ruin x3 during your rotation. But as long as Ifrit on obey is stronger than Garuda on obey then it's fine because TBH Garuda on obey is stronger than Garuda on Sic because her actions reduce her DPS since they aren't optimal for single target.

---

on most boss fight you dont have to summon another garuda,and its a waste of mana do it 2 times,its just an opening chain with RS,after 3 minutes of RS cd bosses are near to death so is better focus instead contagion,you can set ifrit on full attack after you summon him and gain every bonus he can provide,but there is still the Stun problem to think about

Quote:

---

Originally Posted by NoctisUmbra

You didn't lose Thunder 2 months ago.

Only thing for certain here is you can't now until 2.1 hits and you go out and do some testing to confirm.

Theorycrafting based on tests performed in 2.0 suggest, however, that the loss of Thunder not only diminishes SMN DPS by small chunk, but also significantly diminishes the value and effect of Contagion (the skill that was the primary reason Garuda > Ifrit universally) and gives Ifrit the edge.

---

Ifrit straight out parses less than Garuda.

Like I said, we've already been over this 2 months ago.

Quote:

---

Originally Posted by OmegaNovaios

I'm not acting like that, I'm taking it into consideration. Simply put if you can use Contagion from Garuda every cooldown and use Ifrit for the rest then it SHOULD be stronger.

Also, if not switching pets mid-fight and also not fighting more than one mob (because Contagion > Bane is really strong) then Ifrit SHOULD (this means im speculating on how the SMN should work and not how it may or may not actually work) be stronger than Garuda for the same reason that Melee are getting buffed (IE Ifrit is going to die).

To say Ifrit SHOULD never be used is a waste of an action I'd rather push SMN and/or SE if not balanced, rather than have a job with 4 actions instead of 5. Or in this case more than 1 pet to use.

---

Can you even read?

This is one of the things SMNs were mad about.

All we got was stuff taken, and none of our issues were addressed.

Quote:

---

Originally Posted by Kevee

No.

Stop talking about this as if the person who posted was right.

He/She isn't.

Don't spread bad information, we already went through this 2 months ago where people thought Ifrit was stronger.

He isn't.

Period.

---

I'm not acting like that, I'm taking it into consideration. Simply put if you can use Contagion from Garuda every cooldown and use Ifrit for the rest then it SHOULD be stronger.

Also, if not switching pets mid-fight and also not fighting more than one mob (because Contagion > Bane is really strong) then Ifrit SHOULD (this means I'm speculating on how the SMN should work and not how it may or may not actually work) be stronger than Garuda for the same reason that Melee are getting buffed (IE Ifrit is going to die).

To say Ifrit SHOULD never be used is a waste of an action I'd rather push SMN and/or SE if not balanced, rather than have a job with 4 actions instead of 5. Or in this case more than 1 pet to use.

Quote:

---

Originally Posted by Kevee

Ifrit straight out parses less than Garuda.

Like I said, we've already been over this 2 months ago.

---

This topic is about 2.1.

Until there are parses from 2.1, you're 2 month old parses don't apply. They are nice for theorycrafting, but are not evident of anything.

Quote:

---

Originally Posted by NoctisUmbra

This topic is about 2.1.

Until there are parses from 2.1, you're 2 month old parses don't apply. They are nice for theorycrafting, but are not evident of anything.

---

Yes, what is changing about Ifrit in 2.1? Or Garuda?

Oh, nothing?

Those parses were just straight Obey on the dummy.

Garuda did more.

Garuda has contagion.

Ifrit has nothing.

Quote:

---

Originally Posted by Kevee

Ifrit straight out parses less than Garuda.

Like I said, we've already been over this 2 months ago.

---

by my self test,using only pet in full attack mode ifrit is better in <40seconds fights
my test went in this way

mob lvl 38 ,cast garuda at her max range and pressed attack
Same mob, cast ifrit at same range of garuda,pressed attack

i did the same with a lvl 32 mob (for lower hp so shorter fight)

i dont remember right timing,i did it lots of time ago but can give you a qualitative near rappresentation
on lvl 38 mob fight lasts for 43 second for garuda,39 seconds for ifrit
on lvl 32 mob fight lasts for 19 second for garuda,14 seconds for ifrit

obviously as i said before i counted time ifrit need to reach mob at max range of garuda

a little strong point of ifrit in boss fights is for his dmg reflection skill (im in doubt if it is on titan and not on ifrit,lol correct me if im wrong) so MAYBE and i say MAYBE he can have a little more dmg when hitted by aoe

Quote:

---

Originally Posted by xardus

snip

---

I also tested this.

Ifrit does more DPS on Sic, rather than Obey, and it puts him even with Garuda.

That doesn't factor in that Ifrit is a MELEE pet, needs DRAGON KICK instead of Foe Requiem, and has nothing like CONTAGION.

Quote:

---

Originally Posted by Kevee

Yes, what is changing about Ifrit in 2.1? Or Garuda?

Oh, nothing?

Those parses were just straight Obey on the dummy.

Garuda did more.

Garuda has contagion.

Ifrit has nothing.

---

That sounds like a very accurate parse. Funny how parses suggesting the opposite exist. Oh, the dilemma. Perhaps your Ifrit has bad AI?

I disagree. They didn't change the potency or cast times or anything at all about Ifrit or Garuda. They changed the effective value of Contagion. The *actual* pet DPS between the two has not changed based on ANYTHING in the patch notes. So Ifrit wasn't being used because 1) no contagion, 2) less pet DPS and 3), less survivability in melee. 2.1 changes none of that besides the relative value of contagion, which is still far better in any situation where you can Bane.

Unless you have anything that refutes the above statement, then Kevee is correct.

Quote:

---

Originally Posted by NoctisUmbra

That sounds like a very accurate parse. Funny how parses suggesting the opposite exist. Oh, the dilemma. Perhaps your Ifrit has bad AI?

---

I'm pretty sure a 2 hour parse of pets(obey) on the dummy is more valid than whatever parse you claim to have.

Hey All!

So the post on reddit was mine.

I did do my own parse of ifrit egi on obey vs garuda egi on obey, and it turned out in my ~5 minute test that ifrit egi won. That said, it was a quick personal verification, and since it matched my expectations, I moved on and didn't actually look at the crit %. It is very possible that the crits favored ifrit in such a small sample, and my test was wrong.

For that reason, as i stated on the reddit post, I will be testing this again as soon as I get back into the game on tuesday. Once I have a better estimate for pet damage, I'll get a better sense of the garuda vs ifrit argument.

That said, if ifrit egi fails to out dps garuda egi straight up, then yes, obviously garuda egi wins.

Quote:

---

Originally Posted by Hitokirinomad

I disagree. They didn't change the potency or cast times or anything at all about Ifrit or Garuda. They changed the effective value of Contagion. The *actual* pet DPS between the two has not changed based on ANYTHING in the patch notes. So Ifrit wasn't being used because 1) no contagion, 2) less pet DPS and 3), less survivability in melee. 2.1 changes none of that besides the relative value of contagion, which is still far better in any situation where you can Bane.

Unless you have anything that refutes the above statement, then Kevee is correct.

---

ifrit isnt so light armored as u think,i was using to take ifrit in most of 50dung with great results,castrum ak,wp...then popped out theory about garuda and contagion and yes its better combined with rs and dots,and helps you to take a breath for countdown,but ifrit is very very good believe me! of course you have to manage him better redirecting at right time and so on ,but does his work

Even if they were to remove contagion from garuda entirely, she'd still be better than ifrit.

I've also read accounts of ifrit having a higher accuracy requirement than garuda. I haven't done the testing myself because I never had any interest in running ifrit, but it you were planning on it it's something you should look into.

Quote:

---

Originally Posted by xardus

/snip

---

You realize that this is not an actual test of DPS right? And even if it was, no fight that is significant last <40 sec? I also don't think Ifrit is more lightly armored than Garuda, I think that being in melee puts him in more danger than Garuda is, and the extra micromanagement lowers your personal DPS. Also, fights where you pet gets targeted, having him right up near the tank is a no-no (WP, Chimera, Hydra immediately spring to mind, haven't taken Ifrit higher than that).

And Noctis, can you link to these parses? The reddit thread was a simulation and the OP in it has said he might have 'missed some things'. I haven't seen anyone else parse and post, and the 450k seems like enough time to be a good test.

Quote:

---

Originally Posted by Kevee

Yes, what is changing about Ifrit in 2.1? Or Garuda?

Oh, nothing?

Those parses were just straight Obey on the dummy.

Garuda did more.

Garuda has contagion.

Ifrit has nothing.

---

Considering that Ifrit attacks much more often than Garuda and pet Crits give the Caster a duration buff to Casting Speed (I think +30%) that allows for 1 additional Ruin (80 Potency) for that rotation. Which can't be parsed by looking at the pet individually. But from my testing Ifrit on obey V Garuda on obey the DMG is VERY close.

Also, I don't know which parser your using but from my past experiences parsers are horrible, particularly for SMN (they can't detect DOT crits, they can't detect AOE hotspots like Shadow Flare and Flaming Arrow, they seem to have issues not calculating non-critting DOTs on multiple mobs, and I'm sure there are other issues that I personally overlooked).

Because pets only use direct attacks, the parsers are very accurate for them.

The bonus from enhanced pet action is 20% increase to the spell speed stat and does not translate into a 20% cast speed increase, not even close.

Quote:

---

Originally Posted by OmegaNovaios

But from my testing Ifrit on obey V Garuda on obey the DMG is VERY close.

---

If garuda and ifrit egi were close in their own DPS, then garuda egi will win.

The threshold is at around 15% extra dps on ifrit egi. (eg. If ifrit egi did 15% more dps than garuda egi, then ifrit egi would globally be more useful than garuda egi). In my simulation, Ifrit-egi was 25% more damage than garuda egi, therefore giving me the result that Ifrit egi was better.

That said, I may have been wrong / not had enough statistics when I did my own parsing checks. As i eddited to include on the reddit post, I will be re-checking these tomorrow when the servers are back up.

Quote:

---

Originally Posted by BlueCoconut

If garuda and ifrit egi were close in their own DPS, then garuda egi will win.

The threshold is at around 15% extra dps on ifrit egi. (eg. If ifrit egi did 15% more dps than garuda egi, then ifrit egi would globally be more useful than garuda egi). In my simulation, Ifrit-egi was 25% more damage than garuda egi, therefore giving me the result that Ifrit egi was better.

That said, I may have been wrong / not had enough statistics when I did my own parsing checks. As i eddited to include on the reddit post, I will be re-checking these tomorrow when the servers are back up.

---

Hmm, if that's true then Ifrit needs a potency boost. It's weird seeing Garuda do around 200 DMG with 100 Potency based attack and Ifrit doing 140 DMG with 120 Potency attack and a 60 DMG auto attack.

Garuda is better even if you don't include Contagion. This guy is trying to theorycraft from scratch, not knowing half of what he is talking about, and spewing incorrect results.

The new BiS set will be:

Summoner's Horn
Summoner's Doublet
Summoner's Ringbands
Allagan Rope Belt of Casting
Allagan Breeches of Casting
Allagan Boots of Casting
Hero's Necklace of Casting
Tremor Earring of Casting
Hero's Bracelet of Casting
Hero's Ring of Casting
Vortex Ring of Casting

448 accuracy. As a SMN, when you see a thread posted by someone saying 'this is the new best <whatever>' and his BiS set has 432 accuracy, consider that a huge red flag. Take everything else they say with a grain of salt.

Also, what do you guys think about this? http://na.finalfantasyxiv.com/lodest...m/aee840713e3/

Omnitome
69 magic damage
32 vit
31 int
37 acc
26 spell speed
+ friggin' epic skin

Too bad our upgraded Allagan tome blows it out of the water from a DPS perspective...

http://na.finalfantasyxiv.com/lodest...m/387cd387c23/

Allagan Grimoire of Casting
_71_ magic damage
35 vit
34 int
_39_ crit
19 det

But, I'll probably snag the new i90 tome to use until my Allagan pants drop, or my Allagan book drops, either one.

Quote:

---

Originally Posted by T0rin

Garuda is better even if you don't include Contagion. This guy is trying to theorycraft from scratch, not knowing half of what he is talking about, and spewing incorrect results.

The new BiS set will be:

Summoner's Horn
Summoner's Doublet
Summoner's Ringbands
Allagan Rope Belt of Casting
Allagan Breeches of Casting
Allagan Boots of Casting
Hero's Necklace of Casting
Tremor Earring of Casting
Hero's Bracelet of Casting
Hero's Ring of Casting
Vortex Ring of Casting

448 accuracy. As a SMN, when you see a thread posted by someone saying 'this is the new best <whatever>' and his BiS set has 432 accuracy, consider that a huge red flag. Take everything else they say with a grain of salt.

---

lol noob, smn needz no acc, l2p.

^ I swear I had this argument on my server with a T4 Coil SMN, worst he had support from other players.

Quote:

---

Originally Posted by T0rin

Garuda is better even if you don't include Contagion. This guy is trying to theorycraft from scratch, not knowing half of what he is talking about, and spewing incorrect results.

---

Did you read my post? That was the low accuracy version, I specifically stated that I made an assumption (since my parser came out 95% hit rate on garuda in coil with 432 accuracy). However, someone else provided theirs, and so I re-ran it with different accuracy numbers, and now give 3 different options depending on what you believe accuracy cap to be (shown below is the lowest one) and I agree with your 448 ACC. After that is the 459 ACC build.

443 ACC:

Allagan Grimoire of Casting
Summoner's Horn
Summoner's Doublet
Summoner's Ringbands
Allagan Rope Belt of Casting
Allagan Breeches of Casting
Allagan Boots of Casting (version 2.05)
Hero's Necklace of Casting
Tremor Earrings of Casting
Hero's Bracelet of Casting
Allagan Ring of Casting
Hero's Ring of Casting

Quote:

---

Originally Posted by BlueCoconut

Did you read my post? That was the low accuracy version, I specifically stated that I made an assumption (since my parser came out 95% hit rate on garuda in coil with 432 accuracy). However, someone else provided theirs, and so I re-ran it with different accuracy numbers, and now give 3 different options depending on what you believe accuracy cap to be (shown below is the lowest one) and I agree with your 448 ACC. After that is the 459 ACC build.

---

I read it, before the edits, before Kevee corrected the wrong assumptions. At first, it was the only version, that's when the red flag went up for me.

There is no 'believing' what the accuracy cap is.. for those of us doing Turn 5, we _know_ the accuracy cap is 449 or 450. We don't need a parser to tell us. I'm not trying to sound elitist or something, but you make it sound like this is something that is up for debate.

The DPS lost from using the 459 build is not worth the gains from Garuda (<0.5% accuracy) compared to the 448 build, and the DPS gained from the 443 build is less than what you would lose from Garuda. Simply put, the 448 build is BiS.

"For those of us doing Turn 5"

I consider myself among that group, as I have downed twintania 4 times as summoner, with various groups. And across all of my runs, my ACC has been 432 exactly, and garuda, on ACT parser was running at 95% hit rate when I bothered to actually look at the numbers.

I have been shown more data that shows otherwise, so that is why I am making an "exception" because enough people are arguing about different accuracy numbers. I will be honest, I didn't check it too frequently (probably about 3 times in the past 2 months of playing, and maybe my statistics for those runs that I looked at my garuda miss rate... was just lucky) So, because of statistics, I am giving the alternatives.

Quote:

---

Originally Posted by BlueCoconut

"For those of us doing Turn 5"

I consider myself among that group, as I have downed twintania 4 times as summoner, with various groups. And across all of my runs, my ACC has been 432 exactly, and garuda, on ACT parser was running at 95% hit rate when I bothered to actually look at the numbers.

I have been shown more data that shows otherwise, so that is why I am making an "exception" because enough people are arguing about different accuracy numbers. I will be honest, I didn't check it too frequently (probably about 3 times in the past 2 months of playing, and maybe my statistics for those runs that I looked at my garuda miss rate... was just lucky) So, because of statistics, I am giving the alternatives.

---

Sitting at 442 accuracy put me, on average, at 97.5% accuracy, but some attempts were as low as 81%.

Quote:

---

Originally Posted by BlueCoconut

"For those of us doing Turn 5"

I consider myself among that group, as I have downed twintania 4 times as summoner, with various groups. And across all of my runs, my ACC has been 432 exactly, and garuda, on ACT parser was running at 95% hit rate when I bothered to actually look at the numbers.

I have been shown more data that shows otherwise, so that is why I am making an "exception" because enough people are arguing about different accuracy numbers. I will be honest, I didn't check it too frequently (probably about 3 times in the past 2 months of playing, and maybe my statistics for those runs that I looked at my garuda miss rate... was just lucky) So, because of statistics, I am giving the alternatives.

---

Then I guess all I have to say is: Pay more attention to parser results.

Nobody is really arguing pet accuracy, because all of the data shows the cap is either 449 or 450. We know for sure that 450 is 100% accuracy. We know that 448 is _very close_ to 100% accuracy. It might cap at 449. It is definitely capped at 450. Saying there is an argument or suggesting any cap lower than that is simply bad information and confusing the issue for people that aren't in T5 and don't have access to parser data. There is no need for exceptions, they simply muddy the waters. If people want max DPS, they pick the 448 set. If they want to do sub-optimal DPS, they can use whatever hell else they want.

There is one BiS set, there is one accuracy cap. There is no arguing anything except stat weights, and even then, I don't think the BiS set would change if you value Det or Spell Speed higher than most people do.