Thread: Best in Slot for SMN after 2.1?

Originally Posted by xxalucard

Pretty sure it's not BiS.

How did you run tests for this? It's impossible for parsers to see crits on dots, which is the majority of our damage. Crit also holds a lot more weight for our pet than DET does, and the pet is a big chunk of our damage as well.

I sat there for hours, casting different potency spells (Ruin, Ruin II, Bio, Bio II, Miasma, Miasma II, Blizzard II) and wrote down each number that came up on the screen. I then took the average for each skill, and compared it against this formula: (WD*.2865894 + STR*.1001948 + (DTR-202)*.0220896 + WD*STR*.0036058 + WD*(DTR-202)*.0011015 -1) * (potency/100)

I verified that the averages for all skills matched the formula's calculated damage within an acceptable range of RNG variation. (since not every cast is the same damage)

One thing that I will admit isn't factored into this.. is that more crit on the SMN = more crit on the pet = more SS buffs from the pet = more pet and SMN damage. But, the effective DPS increase you get on yourself and the pet from these SS buffs is small enough that I don't _think_ this BiS set would be any different than what I calculated it to be. 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.

Originally Posted by xxalucard

I agree with 99% of the ideas you have, but gonna have to go ahead and say you're wrong on this one.

Crit is more important for SMN, even if it's just our class in particular. At almost any given moment we could simultaneously have 7+ chances of a crit proc; from DoT tics, pet attack, ruin, fester, our auto-attacks... it's a lot more important for us than what it would be on BLM.

The only unique aspect crit has to SMN, is that it procs SS buffs from the pet. For every other aspect of the class, CRT is simply another damage modifier, just like DET is, just like SS is, just like INT is, just like WD is. And from a strict stat weight perspective, it isn't even our best modifier. And really, it isn't even at it's best unless you have a moderate amount of DET and INT to go along with it. Simply put, outside of stuff that procs on crits (SS buff), CRT is no different from anything else that modifies damage.

Originally Posted by xxalucard

DET weight on auto attacks: 0
Crit weight on auto attacks: more than 0

Det affects auto-attacks, why would you think it doesn't? I mean, even something as simple as mousing over Determination on the Character screen:

"Affects the amount of damage dealt by both physical and magic attacks, as well as the amount of HP restored by healing spells. Auto-attack potency is also affected."

Originally Posted by xxalucard

You might as well tell bards not to add crit because "it's just a chance." Of course it's just a chance, but when things proc from it then that chance becomes real damage.

Mainly because their class is kind of build around crit procs, they get major damage spikes from it, whereas we have a 20% chance on a proc to get 80 SS on a crit. So 20% chance on a 20% chance (with crit oriented BiS) is a 4% chance every ~2.9x seconds to get an ~80 SS buff that drops Garuda's attack speed from ~2.92 to ~2.84 (for me) which amounts to a ~3% decrease in Garuda's recast time or a 3% increase in Garuda's DPS for 8 seconds. Garuda will get 20.54 chances to crit every minute, will proc 4% of the time, or 0.8 times a minute, and increase her DPS by 3% for 8 seconds. This amounts to a 0.4% increase to Garuda DPS every minute. Now, we're not even talking about getting rid of these procs all together, we're talking about dropping from proccing 0.8 times a minute to 0.6 times a minute, which amounts to a 0.1% loss to Garuda DPS. That is not even worth talking about, especially when you consider that the alternative is 11 int worth of damage.

Originally Posted by xxalucard

If we had no procs from crit at all, and DET affected our auto attacks, then I might be able to give it a chance. But imaginary weights that aren't class specific do not mean anything for summoner, or take into account the pet & burst situations, where crit festers / crit roused + spurred pet attacks deal huge chunks of damage.

That's the thing, we essentially have no procs, given the strength of the one proc we do have. 20% spell speed is a lot, but given that it happens so infrequently (less than 1 time a minute with full crit gear), it is practically non-existent. And DET does affect our auto-attacks.

I get it, big fester crits and big enkindle crits, etc. can seem like a lot of damage. But think about it like this... CRT is simply a damage modifier. Averaged out over time, you take CRT over DET. You deal more big hits, at the cost of consistently higher damage. Averaged out, CRT is just another damage modifier, just like DET. It just turns out that CRT is worth less per point than DET. CRT also is worth less per point the more you have. CRT is worth more per point with more INT and DET. So, having a mix is better, which is why the proposed set is better, because the weighted damage is simply higher.

I know, this is all kind of voodoo math at this point, but sometime in the next day or two I will show you some real world proof of why these formulas are real, and work, and most important can be PROVEN to be legit. This is not snake oil I'm trying to sell. I spent the afternoon mathing out all this crap because I had this nagging suspicion that I was missing something. And I was. Anyone can sit down and write down numbers popping up on the screen for DoTs and come to the same conclusions that I have. These damage formulas work for SMN just like BRD just like WAR just like DRG.

The only thing that varies from class to class, and gear set to gear set, are the weights, and the weights are derived from the stats you get from the sets. If you noticed in my 'New BiS sets' post a couple pages back, I included stat weights used for each set, because each set is different. I had to recalculate the weights for each set, and the total weighted int values for each set, to figure out which was better. Every set will have different weights, because the weight of CRT and WD are based on the total values of INT, DET and CRT.

Yeah… those have dropped for us three weeks in a row now… still no JF(@!U$V)(!*$!*!!! casting boots though. Game is just trolling me now.

On a more serious note, I’m curious to see the numbers and how you reached them once you finalize them as well. Good work though, I wouldn’t have the patience to sit there for hours transcribing numbers.

Originally Posted by Ragnirok

Not again. Crit has a diminishing return on gear and sure while its nice to have its still just a chance where as Det never loses its value and scales better. Now with all that being said the difference in dps is ~1 so if you prefer crit all the power to you but he has posted links to his discovery.

I don't know if you're understanding t0rin wrong, but crit doesn't have diminishing returns.

Every point of crit will give the same DPS. Crit damage is static, and the amount of crit you gain per point is static.

However, the relative value of crit goes down the more crit you have. Going from 5% crit to 10% crit is a bigger gain than going from 90% to 95%. You're still gaining the same amount of DPS, but the crit gained is HUGE--You're effectively DOUBLING your current crit rate, whereas going from 90% to 95% is a relatively small change at those levels.

Originally Posted by Ragnirok

Also where you been?!

I'm actually probably quitting.
My raid group fell apart, I don't want to continue paying 2 subs and 2 WTFast subs, and the times I need to raid are very specific.
Since the only drops I need are off of turn 4/5, there's not really room for me to pug it, either.