Swapping pets tanks your mana, ifrit seems better for anything except extremely spread out target swaps (due to travel times), and possibly AoE contagion bane.
Printable View
Swapping pets tanks your mana, ifrit seems better for anything except extremely spread out target swaps (due to travel times), and possibly AoE contagion bane.
Haha, I'm going to get to that part in the guide but whoever said phase by phase is in fact correct. There are special exceptions though.
However, you want to minimalize this to 2 swaps only in boss fight instances if you will do it..
So lets use T13 for example.
P1 and P2 you would use Ifrit.
P3 - You will use both. Garuda is the better for AoE here BUT there are limits of when to actually make the switch to Garuda. Lets say we have a few mobs next to each other. If you were to apply a set of DoTs onto one mob and bane them onto nearby mobs, If those mobs will survive for less then 18 seconds (a full duration of Bio, still use Ifrit. However if they will survive for longer then 18 seconds but closer to 33 seconds in total, use Garuda to maximize the effect on Contagion.
P4 - Ifrit again.
Using what I said for P3, the actual time to switch to Garuda would be slightly before when the Storm of Meracydia finally becomes targetable.
In terms of Dungeons, Garuda will be more useful for mob pulls on average (unless the 18 second rule applies), while Ifrit for bosses.
There's another case like T11, where the add phase requires a magic attack vs a melee on one node, so that's another way a switch will need to be made.
Another case is based off of my last example. In your party composition, if you have other magical DPS in your party who would DPS as well (SCH or BLM) and will beat the DPS of Garuda, still use Ifrit.
Again in T11 for example, If a SMN does not need help to kill the magic node in time because either a SCH or BLM is helping them or the melee is slow enough, Ifrit will still be better here.
*Super special case. T10. So while I said in my second case that the 18 second rule should determine the use of Garuda vs Ifrit, something like T10 actually disobeys this rule. In the case where there are SOLELY two mobs that a pet can hit while there are also multiple mobs more then two on the map, even should Contagion be a thing, Ifrit can STILL beat Garuda due to the 532MP to swap. The add phase but more specifically the given adds, do not last long enough as a whole to warrant the use of Garuda and 532MP, ontop of the fact that Gaurda cannot target more then 2 mobs at the same time. You can infact save the 532 MP to use for your own DPS later and win out here. The special exception to this case would be when Raging Strikes / INT potion DoTs can be baned onto the second target AND it won't take forever to die but at least long enough to take advantage of Contagion.
So basically those are the four exceptions roughly where even should Garuda technically be better in the enviorment, Ifrit will win out.
This reminds me that Pets need special attention as far as being able to help soak mechanics. (Repelling Shot, Fireball in T5)
Like your example of using Ifrit on T11. Even if SMN is the only mage for the Cube, using Ifrit for the melee damage on the Sphere doubles to help soak Repelling Cannons making the healing requirement trivial. Thus both healers could Clerics/Dot up the Cube for more potential damage on the Cube, where Ifrit has higher potential damage on the Sphere.
After considering the various cases, there is still the case when Ifrit can reliably die due to soaking, in which case you lose more DPS from having to resummon the Pet again. That would depend mainly on your healers AoE game.
I realize optimal time usage to using pets is a nasty grey area to describe.
Oh for sure there is that possibility. I just don't think a lot of SMN know they can use their pet as an extra body for things like that though. Most fights where it is an option, it seems like SE has balanced the fight around it being an option, so players should be looking be looking for those opportunities.
But yeah, in the case of T11 if Ifrit keeps dieing and your healers aren't taking the opportunity to DPS the Cube....then get new healers. ;p
OP are you sure about Shadow Flare being separate from the server? My current testing is showing it's like any other DoT and is completely tied to the universal 3s timer. There aren't double ticks available with good refreshing timing as you suggest, unless I'm missing something.
It is a separate DoT tick from B2/M/B but it's just delayed on the display and as far as I can tell it's still at the same rate in practice.
Edit: I'm not sure what's going on with it, been watching it the last few minutes.
Seems to have its own duration independent of universal DoTs, this is true. It won't mold down or up to the latest DoT upon cast, as standard DoTs actively do. But, the DoT timer is not related to its duration (IE you can get a tick at any second of the duration unlike Bio which will always give a tick at 15s, 12s, 9s, etc of the timer remaining), and unlike B2/M/B it follows its own timer you can't tinker with by casting it or not, it's either related to the universal tick just delayed slightly or it's its own separate tick, still universal in a sense.
Regardless I've never gotten a double tick.
I will admit you can't reliably replicate it but like you said, it's separate entirely and procs randomly according to it's own timer. If I'm not mistaken, the maker of ACT stated it's NOT on a universal clock even on it's own timer but possibly adjusting per ShadowFlare cast.
So more like the last tick of the previous one procd, then after you recasted it, it procd immediately again within the same second.
Various reddit posts explored the fact that ShadowFlare in particular is quite a unique skill.
http://www.reddit.com/r/ffxiv/commen...parser_plugin/
http://www.reddit.com/r/ffxiv/commen...w_discoveries/
http://www.reddit.com/r/ffxiv/commen...ming_research/
If there is any known timing to replicate this, even at low reliability, please LMK. I've been trying both at the end of my Shadow Flare and just trying to reapply right before, during, and after the tick.
The common belief amongst those who repeatedly tested it was that the ability to replicate it fell with a personal timer from the user that was determined at the moment of loading into an instance.
Try reloading into your location to see if the timer adjusts correctly. Thats the only advice I could give here. I'll update the wording on that detail.
That is so dumb lol.
So its still a static 3 sec timer whose offset is predetermined, and not dependent on when you cast it?
All circle AoEs follow the rule that they go off a pulse of the caster. The 3 second timer follows after the *tick* of the users pulse. This is basically how sacred soil works, if you miss the actual pulse tick, you won't get the sacred soil buff until the next tick is supposed to occur post 3 seconds. The pulse in question is determined from when you load into an instance or area.
http://www.reddit.com/r/ffxiv/commen...ming_research/ -> Info here.
So ultimately it, if you were to reapply shadowflare on the tick exactly, the formal SF will tick and then the one that comes to replace it will tick as well in theory.
Sleigh was saying he tried that but it didn't actually do it, while I've seen other people claim that it's true and can be done.
EDIT: I'm starting to think this is under a perfect world scenario instead of reliably happening in game. Editted the OP regardless with a link and a disclaimer for that.
Gear specs:
Quote:
Time: 2:53
Int: 668
Det: 419
Crit: 495
SS: 371
----------
Myself: 385
Ifrit: 171
Quote:
Time: 2:55
Int: 670
Det: 394
Crit: 533
SS: 371
----------
Myself: 400
Ifrit: 177
Det vs Crit with relative Int build.Quote:
Time: 2:57
Int: 682
Det: 305
Crit: 602
SS: 422
----------
Myself: 387
Ifrit: 168
Parsed multiple times to get an average for each. I haven't tempered with the data.
Rotation was the exact same as well. I closed the parses couple of days ago and tried to retrieve but they seemed to be corrupted. They showed that ifrit was at 100 dps and I was 20 dps less than the original result at 3 mins which was odd (if you know a way to retrieve original parses for pets please advice)...
Also, please note that those results are -5/-10 from the original result due to me being a slow poke at ending the encounter.
The best thing to do, really, is gather A LOT of Data to determine DET vs INT on Pet Damage. Ifrit has an ability which is on a 6s C/D, 120 Potency or something right? If so, set up a controlled environment where your INT + WD is static and vary the value of Determination, then plot it on a graph, then do the same of Intelligence. At the same time, also manually cast Ruin and record the data.
Why?
You want to check whether or not Determination scales differently for Pets than it does to our actual player characters. You'd either want to multiply the recorded min/max values of your Ruin II's by 1.5, or divide Ifrits damage by 1.5 (so they're the same potency), before plotting on a Graph however. If the slope of both graphs are different, or if the Ifrit Graphs Slope is higher, then it means Pets would scale more with Determination.
You'd also want to do the same check with Intelligence. So have a static Determination + WD, but varying INT, to check if Ifrit scales Intelligence differently.
Ultimately, you want to see if there's a difference between "INT and DET" and this will affect how you should build your gear sets, as well as the summoner weights.
If I'm wrong, someone correct me and I'll delete this post, lol. (Never touched smn's, just read)
One difficulty in comparing is that Ifrit seems to use the Physical Damage stat of the weapon rather than Magic Damage, as it does about 70% the damage a spell would at the same potency. Since I'm not really sure how that should be handled, I'm just looking at how the damage changed as a percentage. I just picked 4 sets, tested 10 minutes each. Not enough to show any detailed relationships, but at least a general picture.
Int seems to have the same impact across the board. The damage differences vary within a small range, but that could easily be due to how the game rounds values; would need a lot more data to really say anything there.
As for Det, it seems to impact Burning Strike to the same degree as spells, but Ifrit's Auto-Attack sees twice the impact; Ifrit's Auto-Attack is half the potency of its Burning Strike, but the resulting damage gained from Det is seemingly equal. I expected Bio to maybe show a larger change than Ruin as well, but the impact seems about equal there.
Set A (http://ffxiv.ariyala.com/QMXR)
82 MD 52 PD 658 Int 388 Det
Ruin (80): 305-338
Bio (40): 152-169
Burning Strike (120): 237-261
Ifrit Auto (60): 129-142
Set B (http://ffxiv.ariyala.com/QMXS)
82 MD 52 PD 658 Int 202 Det
Ruin (80): 277-306
Bio (40): 137-150
Burning Strike (120): 215-238
Ifrit Auto (60): 108-119
Set C (http://ffxiv.ariyala.com/QMXQ)
82 MD 52 PD 616 Int 388 Det
Ruin (80): 286-317
Bio (40): 142-157
Burning Strike (120): 221-244
Ifrit Auto (60): 120-133
Set D (http://ffxiv.ariyala.com/QMXO)
82 MD 52 PD 616 Int 202 Det
Ruin (80): 261-288
Bio (40): 131-143
Burning Strike (120): 201-222
Ifrit Auto (60): 100-111
Set A vs Set B (Int equal @ 658, Det -186)
Ruin: 321.5 - 291.5 = 30
Bio: 160.5 - 143.5 = 17
Burning Strike: 249 - 226.5 = 22.5
Ifrit Auto: 135.5 - 113.5 = 22
Ruin: 321.5 / 291.5 = 1.1029
Bio: 160.5 / 143.5 = 1.1184
Burning Strike: 249 / 226.5 = 1.0993
Ifrit Auto: 135.5 / 113.5 = 1.1938
Set A vs Set C (Det equal @ 388, Int -42)
Ruin: 321.5 - 301.5 = 20
Bio: 160.5 - 149.5 = 11
Burning Strike: 249 - 232.5 = 16.5
Ifrit Auto: 135.5 - 126.5 = 9
Ruin: 321.5 / 301.5 = 1.06633
Bio: 160.5 / 149.5 = 1.0736
Burning Strike: 249 / 232.5 = 1.0709
Ifrit Auto: 135.5 / 126.5 = 1.0711
Set B vs Set D (Det equal @ 202, Int -42)
Ruin: 291.5 - 274.5 = 17
Bio: 143.5 - 137 = 6.5
Burning Strike: 226.5 - 211.5 = 15
Ifrit Auto: 113.5 - 105.5 = 8
Ruin: 291.5 / 274.5 = 1.0619
Bio: 143.5 / 137 = 1.0474
Burning Strike: 226.5 / 211.5 = 1.0709
Ifrit Auto: 113.5 / 105.5 = 1.0758
Set C vs Set D (Int equal @ 616, Det -186)
Ruin: 301.5 - 274.5 = 27
Bio: 149.5 - 137 = 12.5
Burning Strike: 232.5 - 211.5 = 21
Ifrit Auto: 126.5 - 105.5 = 21
Ruin: 301.5 / 274.5 = 1.0984
Bio: 149.5 / 137 = 1.0912
Burning Strike: 232.5 / 211.5 = 1.0993
Ifrit Auto: 126.5 / 105.5 = 1.1991
Auto-Attacks value Determination twice as much as they do for standard Ability Skills, keep that in-mind, so it could also apply for Egis.
Maybe Egis are quite literally like "external players" and are mechanically the same as we are? (In terms of stat-scaling at least)
Could you check Ironworks i130 chest vs Demon Robe i120 chest? I never did get the Demon Robe, mostly because I play BLM far far more than SMN, but I'm curious to see how much that DET is worth vs the 5 INT loss in terms of raw damage (ignoring the other benefits of the i130 chest, which are definitely there).
Det definitely scales with Int but Crit is kinda a chance stat other than being a damage amplifying stat. Need to do more tests on a Crit vs Det build with respect to Ifrit's damage, which is what my post was referring to.
Yeah, I'm not really sure how Magic VS Physical Damage works. I just thought that mobs had lower MDEF than DEF so magic hit harder. Then we get tossed something like Raiton which is magic and hits like a physical attack.... So maybe it is just based on MD vs PD and Maim and Mend. (Are pets effected by Maim + Mend?)
---------------------------------------------------------------------------------------------------------------
I think when I was doing my pet weights I did it wrong. I was comparing the stats to raise SMN DPS 1% and PET DPS 1% and not just the damage themselves. I don't know if that's a good way to do it, but I just figured since Ifrit was on a static 3s GCD VS SMN on a dynamic GCD it might be better.
Crit rate seems to translate directly, at least.
Hmm, that makes me curious about the impact on book smacks as well. That's the one scenario we have where dropping Int for Det is a straight gain rather than a balancing act, so maybe the impact is more significant than one would expect given the relatively low damage of book smacks.
Edit: posted here http://forum.square-enix.com/ffxiv/t...=1#post2956452
The problem is it would take a loooooot of testing. 3 minute parses are far too little data to come to any conclusion. I've had a set with 600 Crit produce lower crit rates in a T10 than a set with 520; RNG is going to be RNG, no matter how we stack the odds. We would need to test each set for hours, and in comparing full rotations, human error also becomes a significant factor.
A more reasonable option, I think, would be to find the raw damage ranges a set produces, then adjust the damage based on a perfect RNG crit rate. Especially in a case like Ifrit, since it attacks at a completely fixed rate (Unless we put on Skill Speed...).
For example, comparing Burning Strikes from my set A and B from above with the Crit rate in the sets.
Set A (477 Crit = 14.82%)
249 * 0.8518 + 249 * 1.5 * 0.1482 = 267.4509
Set B (602 Crit = 23.85%)
226.5 * 0.7615 + 226.5 * 1.5 * 0.2385 = 253.5101
DET probably wont be a big deal for Book Smacks.
With the Dreadwyrm Grimoir, 77 STR raised my Book smacks by a whopping 8 DPS
30 DET didn't seem to be enough to tip the scales towards a 1 DPS gain, although I think it was close.
That's with 100% uptime too, which isn't typical on fights.
STR is the book smacking stat ^^ we have higher str than BLMs and is why we auto attack and they dont bother.
Also, if I remember correctly, Casting spells disrupt Auto-Attacks, so they wouldn't get any AA's whilst fire spamming.
I theory it should, but I manage to see auto attack numbers going off inbetween when im casting spells I just have to kinda spam auto attack.
Augmented (http://ffxiv.ariyala.com/QN6Y)
82 MD 52 PD 663 Int 343 Det
Auto: 79-88
Ruin (80): 301-332
Bio (40): 150-165
Burning Strike (120): 232-257
Ifrit Auto (60): 125-138
Demon (http://ffxiv.ariyala.com/QN6Z)
82 MD 52 PD 658 Int 374 Det
Auto: 81-90
Ruin (80): 303-334
Bio (40): 151-166
Burning Strike (120): 234-259
Ifrit Auto (60): 128-141
Auto: 85.5 - 83.5 = 2
Ruin: 318.5 - 316.5 = 2
Bio: 158.5 - 157.5 = 1
Burning Strike: 246.5 - 244.5 = 2
Ifrit Auto: 134.5 - 131.5 = 3
Adjusted for crit, the Demon robe still comes out slightly ahead between these two builds.
Augmented (17.42%)
Auto: 83.5 * 0.8258 + 83.5 * 1.5 * 0.1742 = 90.77285
Ruin: 316.5 * 0.8258 + 316.5 * 1.5 * 0.1742 = 344.06715
Bio: 157.5 * 0.8258 + 157.5 * 1.5 * 0.1742 = 171.21825
Burning Strike: 244.5 * 0.8258 + 244.5 * 1.5 * 0.1742 = 265.79595
Ifrit Auto: 131.5 * 0.8258 + 131.5 * 1.5 * 0.1742 = 142.95365
Demon (16.27%)
Auto: 85.5 * 0.8373 + 85.5 * 1.5 * 0.1627 = 92.455425
Ruin: 318.5 * 0.8373 + 318.5 * 1.5 * 0.1627 = 344.409975
Bio: 158.5 * 0.8373 + 158.5 * 1.5 * 0.1627 = 171.393975
Burning Strike: 246.5 * 0.8373 + 246.5 * 1.5 * 0.1627 = 266.552775
Ifrit Auto: 134.5 * 0.8373 + 134.5 * 1.5 * 0.1627 = 145.441575
Very very interesting discussion. However I think the Demon Robe is much better, not just slightly than Aug IW. I stacked Det as high as possible and see a huge increase in DPS. I think the reason is even though SS is a DPS gain even though it's marginal on the player, pets are not effected by it (til 3.0 maybe). So in theory the Aug IW is weighted higher but in reality, without the SS you will parse much higher. I currently have 371 SS and that seems perfect. Any higher would be a waste. So Avalen was right. :)
However, my Crit is much lower than what she recommends. I rather not swap the IW neck for Dread and have an over load of Accuracy since I can't afford pentameld crafting gear. When I tested, I did about 4 DPS higher. I think it was just Crit RNG. Oh and I recently got Apocalypse Zeta. My god it's delicious xD.
In terms of raw damage on each attack, it's only slightly better, but in terms of DPS, it's likely to be more significant. That little difference grows more prominent when you consider buffs, debuffs, the simultaneous nature of SMN's damage output, multi-DoTing, and so forth.
Tooltips don't update, but pets are affected by speed stats. (But still, not worth it)
http://forum.square-enix.com/ffxiv/t...l-affects-pets
I never understood why skillspeed affects Ifrit and Titan on a class that doesn't use that stat. They should code their skills separate from weaponskills.
Because ifrit and titan use skill not spells.
I would suggest use demon for now if you dont need he HP, but keep the AUG IW for the patch
Skimmed through the thread so sorry if this has been answered already.
Why is it a good idea to use flaming crush and crimson cyclone with Ifrit?
It's a DPS increase. I used to think since Burning Strike is 120 potency and on a 3 second cool down. BS + auto attack was all I needed. Flaming Crushing and Crimson Cyclone are 110 but those numbers add up in the end. Just think about it. 120 + AA, or 120 + 110 + 110 + AA.
It's good because Ifrit's attacks aren't on a shared GCD. Normally it's Burning Strike, 3s, Burning Strike, 3s, etc.; when you weave in another attack, it becomes Burning Strike, Crimson Cyclone, Burning Strike, Flaming Crush, Burning Strike, 3s, and so forth. The animations may delay Burning Strike by some fraction of a second, but it still ends up a significant gain to weave them in.
I've been going with the WD theory just because the difference between PD to MD appears to be proportional to the difference between damage between per potency between you/Garuda and Ifrit. I feel like Maim and Mend does affect them, as Wind Blade (100 pot) does appear to do about 25% more damage than Ruin (80 potency). The Japanese description for Maim and Mend makes sure to specify that Auto Attacks do not receive the 1.3x boost, but doesn't mention any other exclusions.
Oh hey I am featured :) #SMN4LYFE
i prefer to use bio first going bio - bio 2 - miasma.
this way i can cast bio twice within the time limit of rs, making opening looking something like this
RS - BIO - BIO 2 - MIASMA - RUIN 2 - FESTER - SWIFTCAST - SHADOWFLARE - BIO - CONTAGON
this way fester can be casted straight after miasma or bane in case of aoe and have bio put on for extra 18 seconds
Please remember that "snapshotting dots" is not in and of itself anything 'special'. There is nothing that "snapshotting" dots does that is different from any other attack during RS and other CDs.
With that said, simply treat dots as an upfront nuke when determining the value of cooldowns.
So now bio is simply a "nuke" with 240 potency. With RS, it becomes 298 potency. Simple as that. You'll note it gains 48 potency - this is slightly more than 1 tick's worth, so we'll take that into consideration in just a second.
The big deal with Bio first is that you delay your first fester, meaning its harder to fit fester into the rotation. So if we consider this, with 0 spell speed, it will take you
[Bio II] > [Miasma] > [Bio|Swiftcast|Fester] is the first rotation, so it takes you roughly 6.5 seconds to get a fester out. This gives you 16.5 seconds for the second fester, or if you think about it, 3.5 seconds of leeway.
If you bio first, it turns into
[Bio] > [Bio II] > [Miasma] > [Ruin II|Swiftcast|Fester] so it will take you ~9 seconds for first fester.
This gives you only 1 second of leeway. Now, in an ideal scenario, this will work, but...you're playing with fire here. This is my primary motivator in not Bio first. Now, if you remember our potency discussion, you gain 48 potency for Bio during RS. Since fester is 300 potency, you gain 60 potency for doing it during RS. Needless to say, if you miss out on the second fester, you will lose more dps, and Bio is not worth it.
We also consider whether or not it is worthwhile to clip Bio prior to end of RS, for the Bio II > Miasma > Bio opener. In this opener, Bio comes out roughly around 5 seconds, with lets say .5 second delay from RS animation and Bio delay. So with 5.5 seconds, we have ~14.5 seconds before end of RS, meaning we can fit at most 5 more GCDs (6 total including first Bio GCD). This is almost exactly 3 seconds left on the tick, which means we lose on average 1 tick. Since 1 lost tick is 40 pot and RS is 48 potency, this is tentatively good. However, this comes at the cost of losing a RS II instead, which gains 16 potency, so is a loss overall. Thus, do not preemptively refresh your Bio if you do the "standard" opener.
Back to the opener you proposed, again the sticking point is whether or not you can put in that second fester. Another note to point out is, if you did not precast shadowflare, this delays your sflare for another GCD. So, again, not optimal. Loss of an sflare tick + RuinII bonus will cost you any benefit.
Great points Pandabearcat.
The big thing that sold me on going back to the Classic Opener was the Pre-Cast of Bio 2. If you practice timing it so your cast goes off just as the boss is pulled you are essentially 1 cast ahead of having opened with Bio. That 1 cast will translate to an extra Ruin @ 80 Potency. So right off the bat you are further ahead. (before having to worry about clipping, fitting in that 2nd Fester before RS falls off etc)
The reason this never occurred to me while comparing the two initially on a training dummy was due to me opening with Shadow Flare and then going into RS Bio etc, and after so much practice I could edge slightly ahead of the Classic Opener. But on a practical fight with SF placed and pre-cast Bio2, the Classic is just too good.
I suppose it really comes down to the type of encounter and how comfortable you are. I do recall my opener didn't allow second fester to land due to boss becoming invuernerable or simply untargettable and if I can save that 2~3 seconds on first fester it will help me land on it.
Just a food for thought though :3 I always find alternative opening on different opening handy.