Best in Slot for SMN after 2.1?

[T0rin]

Yea I just can't trust any of these "Simulations", these "theoretically calculated damage results" are just not backed by any precise data as the game itself doesn't give that much info out.
I'll keep going with what I feel is best for me.

So you're saying... you'd rather "trust your feelings", than provable math and simulated rotations that you, yourself, can verify look legit? Ok, that sounds like a great idea. The game gives out precise data, numbers on your screen. Take those numbers, compare them to the given formula. They match? Guess what, now you can prove what is actually better, not what just "feels best".

DTR BiS - 94904

CRT BiS - 94556

CRT BiS w/ Allagan head - 94670

how is it possible for the 17 determination from AF2 pants to outweigh the 34 crit from Allagan? Can someone explain how allagan is not BiS?

It isn't. But the reason myth pants are better in that set is because the set drops myth gloves for allagan gloves (which are a lot more damage) and uses myth pants as a means to make up for the lost accuracy. There is no such thing as BiS for an individual slot, because all that matters is the set as a whole. (because of accuracy)

It isn't, but a lot of people jump on bandwagons & theorycraft math without regard to rotation / class / purpose.

IN reality if you get even 1 more crit from the extra crit rate it would be more DPS than the 17 det. Det is not a guaranteed increase in damage, it's a slight raise to the lower/higher damage window ranges. The actual damage can fall into the same range as someone with less det (or even hit lower than someone with less det) many times throughout a fight.

Only the lowest possible damage hit from both windows would guarantee that more DET is a slight increase in damage, and the same is true for the highest possible hit (within a few damage).

Since both stats are essentially "chance" increases, I rather bet on crit than Det. Though some of the best crit pieces come with det anyway.

Why do you continue to simply outright ignore the math, the only one part of this that can be proven at all? Why do you continue to suggest RNG related reason for why one thing is better than another? You do realize, that over any substantial amount of time, all of this simply averages out, and any scenario you might imagine where RNG goes bad for one set and not for another, is simply NOT a good way to compare what is better than another. You're making very vague generalization about theoretical situations, that if you look at the underlying math (1 point of crit DOES actually average out to be a specific amount of added damage, just like DTR, just like SS, just like INT, just like WD), it can easily prove which stat is better. Take the formula, do the math. DTR is very easily a better stat for average damage added. RNG does not matter. Getting lucky on a crit and not on a couple damage rolls, does absolutely NOT make one stat better or worse than another.

The only bandwagoning going on here is bandwagoning on the side of provable math, which we've lacked for so long.

In reality, the 'old' BiS set (CRT heavy) is almost (within 0.5%) as good as the DTR heavy BiS. There is no reason to argue the merits of CRT in the current SMN itemization. A CRT heavy BiS is just as good as a DTR heavy BiS. But all this vague rationalization going on is doing absolutely nothing to contribute to the conversation, all it does is confuse the issue for people who haven't taken the time to read all the actual proof.

[Kimahri123]

DTR BiS - 94904

CRT BiS - 94556

CRT BiS w/ Allagan head - 94670

Sorry, for a noob like me who got lost in the 17 pages here, can you please help me figure out what sets these three are? Sorry! ><

[T0rin]

Sorry, for a noob like me who got lost in the 17 pages here, can you please help me figure out what sets these three are? Sorry! ><

Simulated single target rotations with different sets of gear, to try to show how spell speed affects damage output, and to put some 'reality' to all this theorycrafting.

[Kimahri123]

oh I meant like what pieces in each slots.

[T0rin]

oh I meant like what pieces in each slots.

DTR Heavy BiS
Allagan Grimoire of Casting
Allagan Circlet of Casting
Summoner's Doublet
Allagan Gloves of Casting
Allagan Rope Belt of Casting
Summoner's Trousers
Allagan Boots of Casting
Hero's Necklace of Casting
Tremor Earring of Casting
Hero's Bracelet of Casting
Allagan Ring of Casting
Vortex Ring of Casting

CRT Heavy BiS
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
Hero's Necklace of Casting
Tremor Earring of Casting
Hero's Bracelet of Casting
Hero's Ring of Casting
Vortex Ring of Casting

CRT Heavy BiS w/ Allagan Circlet
Allagan Grimoire of Casting
Allagan Circlet of Casting
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

[xxalucard]

Why do you continue to simply outright ignore the math, the only one part of this that can be proven at all?

In reality, the 'old' BiS set (CRT heavy) is almost (within 0.5%) as good as the DTR heavy BiS. There is no reason to argue the merits of CRT in the current SMN itemization. A CRT heavy BiS is just as good as a DTR heavy BiS. But all this vague rationalization going on is doing absolutely nothing to contribute to the conversation, all it does is confuse the issue for people who haven't taken the time to read all the actual proof.

You just said there's a 0.5% difference in damage between sets. Does that mean you're claiming the parsers you used and your own "tests" are so perfect that they are more than 99.5% accurate?


What "proof" is there? You have no proof. You've been wrong several times throughout this thread and admitted it. Let me highlight:

I don't know if they changed the pet affecting food issue, but I can personally attest to the pet accuracy cap being under 436.

WRONG

Here was you admitting that a parser led you to believe this incorrect information:

As far as I can tell, the pet cap (~450) is still the same. The only thing that had changed, was that FFXIV-APP started mis-reporting 100% pet accuracy, when it wasn't. This caused me and some others to believe that the pet cap had changed, when in reality it seems to have not.

Here's you explaining your BiS foundation of reasoning with "INT Stat Weights":

The cap between the current BiS set and this proposed one is so large (11 weighted int) that the minor DPS bump you see from the SS buff won't amount to 11 int worth.

Then you diminish the value of INT stat weights (your entire basis for what's BiS):

You know, the more I think about it, the less value I think there is in using stat weights. Let's stop to think for a second... what does a stat weight really mean?

My entire argument throughout this thread has been that the stat weights you came up hold no actual value (weights which you yourself even question).

You've admitted to being completely mislead by a parser before, and then questioned your entire foundation for what's BiS (stat weights).

My preference for a Crit BiS is based on the synergy with the pet trait & potential to do the most damage *(and, outside of coil, you'd sub in the allagan gloves & allagan earrings to have even better synergy)*.

You don't work for SE, you don't know the actual damage formulas, and the authority and confidence you have with your theories (theories, not proof) amaze me when you've been wrong several times throughout this thread.


Subjective tests on a dummy based on parsing results that have led you to false information before are not rock-solid proof of anything.

[T0rin]

You just said there's a 0.5% difference in damage between sets. Does that mean you're claiming the parsers you used and your own "tests" are so perfect that they are more than 99.5% accurate?

I have never claimed parsers are accurate. I gave my own simulations in full as "proof" of the 0.5% difference. The one thing that is constant about my simulation is the rotation. It never changes. Every set goes through the exact same rotation, to ensure a fair comparison. I've said time and time again that there is only so much you can do to prove things out. I think we've gone about as far as math and simulations can take us, at some point, if you can't put your faith into that, there is nothing to say but "best of luck". An accurate simulation with full disclosure and no bias showing that DTR and CRT have a 0.5% DPS difference is at least enough for me to be confident in the results. There will always be skeptics, but there is only so much you can do.

What "proof" is there? You have no proof. You've been wrong several times throughout this thread and admitted it.

I make no claims of being infallible. In fact, I eagerly point out when I was wrong and go about what I can to correct those assumptions. The proof is in the aforementioned verification of the damage formula and the resulting simulations. If you can't accept that as proof, nothing short of perfect robotic parsing and SE provided in-game DPS metrics will make you happy. And guess what? We're never going to see that.

My entire argument throughout this thread has been that the stat weights you came up hold no actual value (weights which you yourself even question).

True, stat weighting as a means to determine BiS is a completed flawed method of thinking. Due to the relative nature of the weights, they mean nothing when applied in a fashion that would show one set to be better than another.

You've admitted to being completely mislead by a parser before, and then questioned your entire foundation for what's BiS (stat weights).

Yep, I have. But the being mislead by the parser was an issue of pet accuracy, not in any way related to BiS, except in that we may have had to rethink things if we didnt need the extra 18 accuracy to cap out the pet. Again, I admitted to it, and made corrective statements about it.

My preference for a Crit BiS is based on the synergy with the pet trait & potential to do the most damage *(and, outside of coil, you'd sub in the allagan gloves & allagan earrings to have even better synergy)*.

The pet trait is complete garbage. 20% spell speed for 8 seconds (2 GCDs) on 4% of every pet attack (2.9x seconds, assuming not moving, or contagioning) is a very slim chance of occurence (less than one per minute) for a very minor buff overall in terms of DPS. The amount you gain from that proc happening is absolutely not worth the CRT focus. (see the aforementioned simulations I linked to see how a more frequently occuring spell speed buff actually affects DPS, because I included it)

You don't work for SE, you don't know the actual damage formulas, and the authority and confidence you have with your theories (theories, not proof) amaze me when you've been wrong several times throughout this thread.

No, I don't. And if we leave it up to them, nobody will ever know. Thing is, the formula we do have is accurate enough to ALWAYS give reliable results that CAN be traced back to in-game data. You can sit there and spit out 500 casts of ANY spell in ANY gear set you want, and compare that to the damage formula we are using, and get the SAME results EVERY time. If THAT isn't accurate enough for you, nothing ever will be.

I'll always be confident in what I'm doing, because I'm going about it in the best known way possible. And when my results are disproven, either by myself or others, or even brought under questioning, I'd GLADLY look at the issue at hand. This is a community effort to try to figure out what is best based on the information we have and what we CAN correlate to in-game results. I have no reason currently to think that my simulation, the damage formula, or the results that came out of it, are in any way inaccurate. They give a fair shake to any given build, and use realistic rotations to simulate how things actually play out in game. They give the benefit of every stat where that stat is due. Nothing is neglected that would favor one over another.

If you can think of a better way to go about this that can actually be documented and proven out, feel free to bring it up. For now, all you do is say this is a bunch of bandwagoning and say CRT beats everything, with nothing but anecdotal or RNG based "proof". I've been completely open about my methods and math, how I came to my conclusions, and there has been a lot of GOOD conversation about it, deconstructing it, finding the flaws. That's how it SHOULD work. Simply calling it out for being rubbish and saying you know better and providing nothing to back up what you say, is a waste of everyone's time, your's included.

Subjective tests on a dummy based on parsing results that have led you to false information before are not rock-solid proof of anything.

Nothing was ever based on parsing results except pet accuracy, and that has absolutely nothing to do with the current discussion.

You know Alucard, people don't even walk to debate you on this anymore, that's how ridiculous your arguments are coming across. I'm glad you're very enthusiastic about the topic, but we need to step away from the idea of RNG having any effect on BiS, and talk about these things in their most elemental way possible. I'd like you to be a constructive part of the conversation, but as it stands, you're doing nothing to move the conversation forward. If anything, you're simply trying to take the 'science' out of proving anything, and debating what boils down to illogical arguments.