Det, SS, Crit

[thealphafirm]

That's complete bullshit.

Here comes Easymode with his douchey, elitist attitude. Yay!

[T0rin]

Would you like to provide any data or evidence demonstrating that DTR interacts with Potency differently than any other stat in the game?

I don't have any data to provide, I suppose all my 'learnings' have been anecdotal and are an accumulation of everything I've ever read on the subject.

As I understand it, DTR applies a % modifier to skill damage. So, if for example by gaining 10 DTR, you gain 1% damage to a spell, that 1% would apply in various ways. On a 200 potency spell, 1% damage would equate to 2 potency, increasing the spell from 200 to 202. However, on a low potency spell (like many SMN DoTs) like Bio, which has 40 potency, a 1% increase would be 40.40 potency, which would get truncated down to 40 damage. Not only is the potential damage gain from DTR lower on low potency spells, but the potential for the added DTR to simply have the damage add truncated is higher. On spells with > 100 potency, this is less of an issue.

Now, exactly how much of an increase is applied is more of a mix of the primary stat (INT, etc.), weapon damage and DTR, but at the core of the dynamic, that analysis is why I conclude that DTR is simply less effective as a stat as compared to crit, which will apply a fairly flat modifier to damage, since it is increasing the opportunity to deal 150% damage at a flate rate. (~13.4 crit for 1% increase in chance)

[Arkista]

It would Be nice if someone from the DEV Team would step in and let everyone know what the values of attributes are.

[EasymodeX]

It seems like your perspective boils down to truncation of small increments on DOT ticks.

Assumptions:

1. Damage is actually (not just visually) truncated for each ability separately.
2. You only invest in DTR in small increments.
3. Damage is truncated for individual ticks without crossover for "remainder" damage.
4. Damage without the delta DTR is an integer.

The first assumption is 50/50. By "separately", I mean with regards to all DOTs since they are stored as a single piece of data in memory. If Bio is getting +0.4 per tick, and Miasma is getting +0.3 or whatever, and Bio II gets +0.4, then the total would be +1.1. For your scenario of "DTR gets truncated out a lot", each separate spell tick would have to be separately truncated.

Also note that the "visually" part is important. There are already other data values in the game that have been demonstrated to be truncated for the UI but not in reality (specifically spell speed re: the GCD).

In addition, you are assuming your base damage per tick is an integer (re: #4 above). E.g. 40.0000 damage. This is highly unlikely since your WD is like 6x and Int is at >400. Whatever the actual damage formula is, calculating that down to "40" damage per tick is not going to result in a flat integer. It's just as likely that your 40.9 (truncated to 40) dot tick is boosted to a truncated 41 by 10 DTR (again assuming truncation is real and it's truncated per dot per tick).

The second assumption is a significant assumption, since no one makes gear strategy choices based on +/- 10 points, except for ACC since it has very specific thresholds. Generally if you're going to change your gear strategy based on stat valuation, they're going to swing in the 50+ range depending on gear availability.

The third assumption is minor. The fourth is pretty major and already discussed above.

[T0rin]

It mainly just suggests that optimizing for DTR is based around optimizing around these damage breakpoints, which nobody really knows where they come and how exactly they are implemented. But that the higher potency your abilities are, the easier it is to hit them, due to the nature of [potentially] truncated damage.

I'm mainly pulling on the work of people who are much better at this sort of thing than I am, and am assuming that the work of Eein (chocobro.com) and the like is fairly accurate. It all points to the weighting for DTR being majorly inferior for SMN vs. the work of someone like Valk, who draws a contrary conclusion, where the major distinction between the two classes is potency.

[Ragnirok]

Here comes Easymode with his douchey, elitist attitude. Yay!

Its not elitist if he's refuting a point that has absolutely no data. Making a claim which has gone against everything that has already been spoken about and proven and providing ZERO data to back it up is whats dumb.

[T0rin]

Its not elitist if he's refuting a point that has absolutely no data. Making a claim which has gone against everything that has already been spoken about and proven and providing ZERO data to back it up is whats dumb.

Everything in your small little world of BLMs. There has already been a lot of work done that comes to completely different conclusion about SMN damage and DTR affect low potency abilities. Just because Valk says that DTR works one specific way for BLM stuff, doesn't make it true, and doesn't mean the same applies to everything, even if it were.

The simple assumption being refuted here is that everything works for same for everyone, as it does for BLMs, which is something that was stated very early on in this thread.

And to say this goes against 'everything that has already been spoken about and proven' in reference to Valk's work is super laughable. But keep it up, your contributions are as always, amusing.

[EasymodeX]

Basically the DTR-DoT discussion on chocobro is inconclusive and he cites that more testing is needed. He stated that a SMN who did specific testing reported gaining +1 damage per tick for 30DTR.

Doesn't seem like much, but a sanity check needs to be done to ascertain what a player should expect from DTR. As follows:

1. For a moment, assume that SE is magic awesome and all stats are equally as strong as each other.
2. Assume 14 CRT for 1% CRT just because.
3. Ignore all baselines because this is super casual.
4. 42 CRT gets your 3% crit, which is an increase in damage of 1.5%.

How much DTR gets you 1.5% damage?

Assuming SE is magic awesome, DTR is itemized on items as roughly 67% of CRT. Therefore, 28 DTR should also result in 1.5% damage.

How much is 1.5% damage of a damage DoT? ...

Borrowing Puro's BLM numbers for a moment, 220p results in roughly 540 damage, so a 35p DoT tick should result in ~86 damage DoT ticks. 1.5% of that is 1.3 damage.


So for deltas of ~30 DTR, players should basically be expecting 1 damage on a tick. For clear results, you're going to have to test with +/- 80 DTR, which is a rather significant investment in resources.

The problem is that even then there's no conclusion on whether the damage is actually truncated or only visually so, and likelihood that the DTR could still push the tick damage over the next integer.

As such, it's still a lot of speculation and superstition at this point -- all centering on when the numbers are being truncated and for what purpose.

[T0rin]

I cannot honestly say if I am more or less confused about the role of DTR in terms of item prioritization. Still, I'd think it is a lot less likely that damage is 'rolling over' between DoT ticks, than the possibility that it is. Occam's Razor would suggest that they are simply truncating the decimal during the damage application process, and by that merit alone, DTR is worse on lower potency abilities.

I've also never seen any bulk of work on SMN damage formulas that would conclude that DTR is anything more than half as beneficial as crit is.

But yes, admittedly, it is all speculation and theorycraft at this point, I suppose I'm merely representing the majority of efforts made to try to attempt to explain stat weights as I've been exposed to.

[Peptaru]

One person is using numbers based on testing, the other says nu-uh and uses speculative numbers. Guess who I'll believe for now.