Connecting the DoTs: A Summoner Guide

[Havenchild]

snip

Just to set the record on this, the logic to use Ifrit is true, however in 3.0, he can't (or barely ) competes against Garuda in the current instances as she functions too seamless with the 3.0 Smn toolkit.

Would appreciate any advice, thank you!

I keep Garuda on the add and bring down to about 20%. At that point, I'll refocus her and prepare to feed immediately. In the last phase, it is better to let a melee handle the adds with carnage zero because they can bring it down faster and feed safer in that phase, however, if you must grab one, then assuming no sacrifice you will likely need 3 AF stacks to burn down that add with a 2fester/1 Painflare combo to feed it on time.

[Nichrom]

Hey, do any of you have tips for doing Jagd Dolls on A4S? At what % do you redirect your pet back to the boss/leg/Straf (assuming you're obeying it on the Jagd Doll on tag)? What % do you stop hitting the Jagd Doll at, and does it change in last phase when you have to wait for Carnage Zero? I've watched a few SMN streams but it's not really obvious what the pet's attacking and when.

All the times I've cleared have been as melee 1 (leg/boss/Straf only) + no sac. My group aren't happy about it seeing as it's pretty much a hard carry.

Would appreciate any advice, thank you!

[ShinryuReishiki]

Unless I've messed something up with stacks, I just leave Garuda-egi on the boss (or Straf Doll, we only kill one). I usually toss an Aerial Slash on my Jagd, but that's usually it.

In leg 3/4 I just feed the Jagds as soon as I can; I generally get one last tick of Bio II + Miasma as I'm feeding, so I leave a little buffer HP. I'd rather risk the damage be a little higher than desired than risk having a dead doll.

In final phase, I generally end up switching back to the boss for a few GCDs in the middle. I wait until the Jagd does its 3rd stacking AOE before feeding, so I don't need to be on it the entire time.

[audi5000]

i spend td, 2 festers and put garuda on jagd dolls. at this point rouse/spur are also being used so that helps a lot. it's usually enough damage to feed on time. i ruin ii while i move in to feed at about 14 percent, it dies at single digit percents at that point.

[Sleigh]

I don't even really notice a difference in HW to use Ifrit or Garuda on a dummy, which is bad for Ifrit since Garuda does offer so many advantages. Ease of being able to Ruin spam, heavy MP gains, ranged (which is not only a DPS boost but a survival boost), mega-powerful Banes.

Only time I do Ifrit these days is trash pulls in dungeons, because Contagion's worthless in a good group. Also A2S possibly, in a good group ... still experimenting with that but the theory is solid.

[Sunako]

I think ifrit is doing weaker damage now because it increase more tank damage with radiant shield. It is pretty hard test how effective radiant shield is against dummy. Potentially tank get more dps and treat from it = less aggro combo = more dps combo = more damage. Only problem is current encounter right now that is mostly magical damage.

[UBERHAXED]

I think ifrit is doing weaker damage now because it increase more tank damage with radiant shield. It is pretty hard test how effective radiant shield is against dummy. Potentially tank get more dps and treat from it = less aggro combo = more dps combo = more damage. Only problem is current encounter right now that is mostly magical damage.

Radiant shield affects everyone in the party including other pets (lol selene and eos). I also believe that (based on logs) Ifrit does a counter attack whenever other targets get hit. It doesn't seem right that Ifrit is using radiant shield only when he is hit when in some fights he'll take literally no damage and still counter attack (take A3 for example).
http://www.fflogs.com/reports/7zZ4TR...-done&source=2

It goes without saying that no pet have an attack called "attack" (pet's don't auto attack, they use their base attack, in this case burning strike every 3 seconds). In the log I linked, Ifrit also used "attack" more times than burning strike, which leads to believe that he used attack for every physical hit received under the duration of radiant shield. Further digging through that log reveals that Ifrit took no damage during the encounter outside of splash and (aside from splash not being physical anyway) radiant shield was not active during the damage intake. Looking at the log though it seems that he will continue to use attack well past the duration of radiant shield.

In other words, any damage done to party members during increase the DPS of summoner, not just the target. That said, training dummies obvious cannot proc attack from radiant shield so if Garada + everything ~= Ifrit + everything - radiant shield, then there's an obvious reason to use Ifrit over garuda. In the linked log, including enkindle attack, "attack" contributed 28% of Ifrit's damage.

Also in case you were unaware, when radiant shield is used, it's effect appear to reflect the attack. So if it was proc'ed by "attack" then the target and Ifrit will counter with "attack". If it was proc'ed by "kamehameha" then the target will have appeared to use "kamehameha" in the log (and Ifrit will use "attack", as per radiant shield's description). The exact damage seems to be calculated strangely though. It appears to be a 50 potency auto attack for player, but in the case of SMN pets, the damage will be based on INT, rather than strength (SMN auto's are based on strength) and SCH pets will be based on what appears to be the SCH strength stat (although I suppose it could be INT seeing as INT for healers is low).

http://www.fflogs.com/reports/7zZ4TR...s=-2&source=24
Same report, selene uses fluid strike. Based on correlation, this appears to be the STR stat, but I digress.

http://www.fflogs.com/reports/7zZ4TR...done&source=37

Same report, but looking at the bard. Bard's don't use auto with WM so nothing to compare against, but this is clearly using the DEX stat for "fluid strike" and "shot". Interestingly, in the same encounter, Ifrit did not use "fluid strike" or "shot" (as pet's don't get hit by that attack) but still got the standard "attack" counter attack from it.

http://www.fflogs.com/reports/7zZ4TR...one&source=109

Looking at the main tank (the counter attacks are at the bottom since they weren't cast) we can pretty much confirm that it is about 50% of the regular auto-attack potency (the top attack). Going back to the report with Ifrit, attack will hit a bit less than half of Burning strike (burning strike happens to be more than 100 potency, so it also lines up with 50 potency).

You can fumble around the rest of the report, but the fact of the matter is that there is a misconception of the effect of radiant shield and a lot of summoners are missing out on large dps gains (and raid gains) by choosing to use garuda. By itself, the radiant shield should be able to cover the additional damage from contagion (especially, since you can synchronize with rouse 100% of the time and spur 50% of the time). It also increases raid dps by an amount proportional to the incoming physical damage (notably the MT is always getting hit by physical attack, autos or not, barring fights like A4). It also doesn't help Garuda's case that every single one of her attacks have less potency than Ifrit (making his DPS - Radiant shield > Garuda - Contagion).

Final determination is that if you have not trouble with keeping Ifrit with equal up time as Garuda you should be using Ifrit. As for the case of magical damage based fights, as a person pointed out above, Ifrit on a dummy (i.e. no radiant shield) is about the same as Garuda on a dummy (all skills including contagion) so it's not really a loss even then to just use him for the higher potency skills, given you have no problem with uptime.

Of course, Ifrit is melee, if you have him chasing things around, his efficacy decreases so there are obvious cases when the uptime factor counts. But in reality, enemy movement can be controlled a large percentage of the time.

Lots of controversial opinions here (particularly I'm challenging the opening post's opinion of Ifrit in post 58 content) so feel free to discuss any obvious oversights in the thought process, errors in the provided data, or contradictions in the logic.

[OneWingedSora]

It goes without saying that no pet have an attack called "attack" (pet's don't auto attack, they use their base attack, in this case burning strike every 3 seconds). In the log I linked, Ifrit also used "attack" more times than burning strike, which leads to believe that he used attack for every physical hit received under the duration of radiant shield. Further digging through that log reveals that Ifrit took no damage during the encounter outside of splash and (aside from splash not being physical anyway) radiant shield was not active during the damage intake. Looking at the log though it seems that he will continue to use attack well past the duration of radiant shield.

Ifrit auto attacks, and that is what it refers to in the log by "Attack".

[UBERHAXED]

Ifrit auto attacks, and that is what it refers to in the log by "Attack".

If Ifrit always auto attacks, then there's something seriously wrong with using Garuda considering the already large difference of potency attacks between the two (120 potency burning strike per 3 seconds + the alleged "auto" 50 potency, which appears to be more frequent compared to 100 potency wind blade). If Ifrit auto attacks, then the attack will be close to burning strike in potency (80% to be exact), but the move is clearly less than 50% (which matches exactly with radiant shield). Also the alleged auto should then appear when radiant shield is not in effect. And furthermore should be included on the OP to not misinform users the damage of the pet (which may have been the purpose of such an omission). You pet will autonomously use their 3 second cooldown move, which serves as an "auto attack". I've not seen any further evidence that this is the not the case, also considering that if Ifrit does an auto attack at a fixed interval and burning strike at a fixed interval, than in every fight the ratio of "auto attack" and burning strike should be fix (or, giving you the benefit here, consistent). This is shown not be the case from the following logs:
(Using a person other than myself as a control variable, vs Faust so there should be no external movement, the rest of the cases will be controlled by User (myself))
Control Vs. Faust
http://www.fflogs.com/reports/vrGVfq...done&source=10
Ratio: 37 attack to 36 burning strike; 102.7%
Case 1 Vs. Thordan
http://www.fflogs.com/reports/L1NH6A...rce=1&fight=11
Ratio: 152 attack to 132 burning strike; 115.1%
Case 2 Vs. Thordan
http://www.fflogs.com/reports/L1NH6A...rce=1&fight=12
Ratio: 136 attack to 149 burning strike; 109.5%
Case 3 Vs. Faust
http://www.fflogs.com/reports/b3pxn4...e=231&fight=14
Ratio: 40 attack to 35 burning strike; 114.2%
Case 4 Vs. A2S
http://www.fflogs.com/reports/b3pxn4...e=231&fight=19
Ratio: 125 attack to 116 burning strike; 107.7%
Case 5 Vs. Oppressor
http://www.fflogs.com/reports/b3pxn4...one&source=231
Ratio: 113 attack to 102 burning strike; 107.8%
Case 6 Vs. A2S
http://www.fflogs.com/reports/HnLKC7...-done&source=1
Ratio: 157 attack to 136 burning strike; 115.4%
Case 7 Vs. Faust
http://www.fflogs.com/reports/fbynRF...-done&source=2
Ratio: 41 attack to 37 burning strike; 110.8%
Case 8 Vs. Oppressor
http://www.fflogs.com/reports/fbynRF...-done&source=2
Ratio: 104 attack to 92 burning strike; 113.0%

There's a lot of evidence to the contrary so you need at least some form of evidence barring word of mouth to support your claim.

[Havenchild]

snip

I'm not gonna try to be mean in any sense, however you are misinformed on some things and you are welcome to test anything claimed in this guide in a controlled environment in game.

Just to be sure you understand:

1) Ifrit DOES have an Auto Attack that is separate from Burning Strike (the skill it uses every 3 seconds; Default Attack Skill).


2) The only time Ifrit "skips" (delays) a default attack (Burning Strike) , is if several of his skills are used back to back/ come up at the same time (If on Sic).
So instead of something like

Burning Strike + Auto -> Crimson Cyclone -> Burning Strike+ Auto -> Radiant Shield -> Burning Strike + Auto -> Flaming Crush,

it will become:

Burning Strike + Auto -> Crimson Cyclone ->Auto -> Radiant Shield -> Auto> Flaming Crush -> Burning Strike + Auto


3) Garuda and Ifrit damage is calculated differently. Thus you cannot directly compare their potencies.

4) Radiant Shield only procs (and can proc continuously) when a user who has the buff takes Physical damage. The damage it deals is based of the physical damage a user can do. This is why casters (Both Healer and DPS) will do much less damage than a Tank or other DPS.

5) Is Radiant Shield better than Contagion in dungeons? If the mobs die very fast, then it is better. Otherwise if they survive longer than Radiant Shield (20s) than Contagion was very likely better in almost every scenario. In post 58, Dreadwyrm Trance is too much of a buff to SMN damage universally for Ifrit to keep up from then onwards. More to the point that at 60, no mobs die that fast usually which further makes Radiant Shield and Ifrit mundane. Think about Aetherflow usage as well.

With Ifrit, you will have to Dot ups -> Bane -> Painflare -> Dot -> re-Bane again likely before mobs die.

With Garuda, you Will DoT up a mob -> Contagion + Painflare -> Bane ->-----> Painflare

Contagion saves you both, GCDs going up towards redotting enemies that now can be used towards AoE Blizzard II use, and maximizing damage from Aetherflow usage.