Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

Kitru's New Class Idea

STAR WARS: The Old Republic > English > Classes
Kitru's New Class Idea

KeyboardNinja's Avatar


KeyboardNinja
03.26.2013 , 05:55 PM | #21
Oh, and I found the reference to Tertiary Stat: http://www.swtor.com/community/showthread.php?t=63362 It's part of the base stats calculation. Still looking for the reference indicating it affects the crit chance on Stockstrike.
Computer Programmer. Theory Crafter. Dilettante on The Ebon Hawk.
Tam (shadow tank) Tov-ren (commando healer) Aveo (retired sentinel) Nimri (ruffian scoundrel)
Averith (marksman sniper) Alish (lightning sorcerer) Aresham (vengeance jugg) Effek (pyro pt)

GeckoOBac's Avatar


GeckoOBac
03.26.2013 , 05:55 PM | #22
Quote: Originally Posted by KeyboardNinja View Post
KBN Wall of Numbers :P
Don't get me wrong, I love your work, but that's basically the reason why I dislike numbers so early in the design stage. They're INEVITABLY going to be wrong and they still don't represent the class. Imho the focus should be first on getting a nice design, with cool mechanics that make for fun and engaging combat. Once you get that down, you can start with numbers, otherwise any kind of overall design change is going to send you back to calculate everything again almost from scratch.
Light Knights: Gecko - Syed - Vor'sann - Joya
Nightmare's Legion: Anhess - Avilus - Wittard - Schroedinger

GeckoOBac's Avatar


GeckoOBac
03.26.2013 , 06:00 PM | #23
Quote: Originally Posted by KeyboardNinja View Post
Oh, and I found the reference to Tertiary Stat: http://www.swtor.com/community/showthread.php?t=63362 It's part of the base stats calculation. Still looking for the reference indicating it affects the crit chance on Stockstrike.
I may be wrong but I think it was just an early rumor due to Stockstrike being a melee skill in theory (as in, melee range). I remember reading *THAT* back then, but it was quickly debunked as being just false as stockstrike is a tech skill and not melee. I don't remember reading anything about confirmation of strength doing anything for troopers.

In particular, in the thread you cite, tertiary stats are just the "leftovers". The line
Quote:
Tertiary Stat = 10 + Level * 0.8 i.e. Strength for the Trooper
I'm fairly sure just means that all the non primary/secondary stats increase at the same rate, up to 50 at lvl 50, as can be seen in the table just above that statement. I don't think it implies anything more than that.
Light Knights: Gecko - Syed - Vor'sann - Joya
Nightmare's Legion: Anhess - Avilus - Wittard - Schroedinger

Kitru's Avatar


Kitru
03.26.2013 , 06:20 PM | #24
Quote: Originally Posted by GeckoOBac View Post
Personally while designing classes for games I'd rather go for the theme and the mechanics and perhaps some core, defining skills, than giving a complete run over. It's a better "sales pitch" too, as wall of texts are off putting (and, tbh, no matter your level of skills, it's basically impossible to balance stuff in a void like that so your numbers, which I'm certain you researched deeply, are still fairly meaningless)
Keep in mind that I don't think that the developers could just lift these numbers out of the post and put them in game to get something ready for release a week from now. As I see it, this is more "alpha" stage: it's not even ready for beta testing, imo. Hell, if you check out the talent trees that I put out, I don't think any of them are populated with the 36 talent points that are a functional *minimum* for a full talent tree and, iirc, fire is is only in the high-20s. Even then, for the listed numbers, the math I've done is *very* soft. I used rough approximations for most of it and used a much more intuitive approach to assigning damage numbers than I'm really comfortable with (the first round of my calcs was pretty much just assigning numbers that "seemed right" and then I determined what I expected the attack string to be to determine what the average DPS would be followed by revising those numbers to bring them in line with that I expect the proper DPS to be). It's not *supposed* to be a class that could be tacked on immediately. It's a starting point from which the class could be developed and further refined into a presentable form.

Essentially, to use a metaphor, the concept of an Echani class is a class "seed", this write-up is a class "sapling", and the full class is an outright class "tree".

Quote:
Btw I hope you take all these comments as I intend them: constructive criticism.
The primary reason why I posted this is to get the reaction from other players about something I've been working on since last November. As the people that I interact with will tell you, I operate *much* better when I'm not just operating in a vacuum. Since most of what I'm doing has ended up going over the heads of both my friends and people I know in game (*design* is a vastly difference skill set than play or analysis so most of them are simply dumbfounded), this is both a manner to get my ideas out there while also getting feedback on them.

Quote:
Now, as for opinions... I didn't really read all the skills and talents descriptions, but the basic idea of two resources is nice. Personally I think the way to get out most of this kind of mechanic is making breath the normal resource as all classes get (ammo, force, energy), and keep intuition as a stacking self buff, kinda like centering and similar things BUT make it something you actually have to manage (and as this is going to be an additional class, we can make it more complex I think).
When I first saw Centering, I thought that it was actually going to be consumed as a secondary resource and was more than a little let down when I found out that there are only 3 Centering consuming abilities and they all consume all 30 stacks at once. Centering is more of a tacked on extra than a fundamental aspect of the class.

In designing the dual resource system, I recognized that it would be more complex than any of the existing resource models, which is bound to happen when you tie 2 mechanically unique resources together, but I didn't want to make it so complex as to be unplayable. The basic set up I went with is that one of your resource is your "standard" resource (Breath, since it's pretty much the simplest resource setup I could imagine while still being a resource mechanism) and the other is your "better" resource (insofar as it's used to fuel your "stronger" attacks). As I developed the specs, this was mussed up a bit as I decided to focus more on the interaction of the two resources rather than elevating Intuition naturally above the Breath, but I'm actually happier with it this way.

It's important to mention that you *do* have to maintain and manage both resources. If you run out of Breath, you better hope you have the Intuition to buy you some time for your Breath to recharge. If you're out of Intuition, you're going to need Breath to generate it. I didn't want to make maintaining *either* of them a major chore since it would just end up with a majority of players annoyed and not having fun (like most Guardians were at release since there weren't a lot of guide for how to play a Guardian and most people wanted to just spam the hell out of their attacks).

Quote:
By that I mean: you can decide whether you use it to "fuel" skills (IE: they use intuition along with breath) or you keep it at certain thresholds that give you specific bonuses and perhaps unlock different skills.
In this way you can/have to make trade offs: Will I keep my intuition to the max so I get the bonuses? Will I use it to active skills x and y? Will I discharge it for a powerful buff?

Gives a lot of flexibility in what you can do. You could make skills react somewhat differently depending on intuition levels, and so on.
While that does allow for some really interesting tactical and mechanical options, you can pretty much guarantee that it wouldn't really end up playing like that more than a month after release. Once theorycrafters have had their chance to math things out, you can pretty much guarantee that it would be proven that, for this spec, you want to max out and maintain your Intuition at all time, whereas, for this other one, it's not worth it to get over 6 Intuition, etc. At that point, it just becomes a question of creating arbitrary complexity purely to make it harder to theorycraft and, as a negative side effect, confuse the living bejeesus out of players that don't math and research for fun.

I do kind of like that idea however. I think I might coopt and tweak it for the other class I'm working on.

Quote:
Just suggestions... One thing I have to say though: we already have 5 melee classes. Tanks kinda require to be meleeish so I understand, but at least one of the advanced classes should be ranged imho. Not sure about the specific skills, but there are "traditional" martial weapon fighters in the SW universe, so giving one of the ACs access to something like the Nightsisters bows would keep the "martial fighters" feel and get you a ranged AC.
The problem with this is coming up with a reason for the Echani and Thyrsians to use said weapons. They're an explicitly melee focused culture so it doesn't really make much sense for a second AC to be ranged capable and coming up with some obscure bit of minutia as a vague thematic construct to force the theme to adhere to an arbitrary mechanical desire rubs me the wrong way. If it were *necessary* to create a ranged option rather than a melee option, I could see it, but I just don't see it as being necessary: as far as most of the game is concerned, DPS is DPS.
Walls of Text? I *love* Walls of Text!
My New Class Idea
Shadow Class Rep - Suggest/Review Questions Here
Quote: Originally Posted by Fende View Post
Listen to Kitru. Kitru knows all.

Kitru's Avatar


Kitru
03.26.2013 , 06:34 PM | #25
Quote: Originally Posted by KeyboardNinja View Post
The only major component that I *didn't* include here is the stacking "Ebb and Flow" proc, which is absolutely important but extremely difficult to model (still a work in progress).
This is probably the biggest weakness I can see since Ebb and Flow are part of the fundamental design of the spec: it provides a 75% increase to the damage of the first attack of a Breath or Intuition resource consumption cycle (which, based on my design, should be your biggest hitter; i.e. Vortex or Avalanche Strike). Ignoring those contributions are going to have a *major* impact on the actual total damage dealt since those attacks, augmented in such a way, form the backbone of the rotation.

Quote:
I further made a stochastic assumption w.r.t. True Strike (i.e. using it when statistically ideal from an average cost perspective, not on proc).
I actually planned for Water to take the 1 sec cast time reduction in the Spirit tree so that it's actually got a 2.0 sec cast time. The baseline stats on True Strike are pretty much rubbish; I think when I was doing the math for it, the damage was so bad that you would never want to touch it (akin to Shadow Strike and Project for Shadows).

Also, were you factoring in the differences in the relevant damage types? The stance proc is Internal damage, which is substantially more valuable than raw kinetic/energy damage. In addition, I'm not sure why you had Warcry in there. It actually does less damage for more Breath than Power Strike does without factoring in the talents that increase Power Strike crit damage and decrease its Breath cost. Warcry shouldn't even be used as part of the ST rotation for Water (it's a useful AoE attack though).

Quote:
Honestly, I think the problem here is that the basic attack is too powerful.
I don't really think that's the problem. Quick combination is, explicitly, a basic attack and follows the formula for *all* basic attacks: no cost, no CD, 1.0 coefficient, weapon damage. It's a pretty good sign that, if you think the problem is that *basic attack*, then your model itself is flawed (as I said, I'm pretty sure your flaw is in the fact that you forgot to account for Ebb and Flow, which are a *major* part of the spec).
Walls of Text? I *love* Walls of Text!
My New Class Idea
Shadow Class Rep - Suggest/Review Questions Here
Quote: Originally Posted by Fende View Post
Listen to Kitru. Kitru knows all.

GeckoOBac's Avatar


GeckoOBac
03.26.2013 , 07:51 PM | #26
[Edit: english is shaky in places, it's late :P]
Ok not going to make quote of quotes quoting quotes or I'm going mad, just some replies in random order :P

a) Yeah keeping things simple is ofc good and it's something you should strive for, but I think classes in swtor are somewhat *too* simple. A little bit of extra management is not bad as long as it doesn't make the class inefficient.

b) I absolutely agree about the theorycrafting issue, that's why imho the different uses of the resource should be something more out of the box than just, I don't know, damage buff, heal self, absorb, etc. So nobody could ever take the class and say: never use this skill, or always use that one.

Ofc the core "rotations" or priorities of the class should be well defined, but let's say something like:
If you get full intuition you get a breath regeneration bonus, so you would like to keep that up in burn phases, but if it's made in a way that it's not necessary during normal play you may want to use it otherwise.
Now to complete the scheme you have to add some skills which are not mutually exclusive, but have different, situational, side effects.
So perhaps you can decide to lose your regeneration bonus if you see that dps is not lacking but heals are, so you use an intuition skill that still does a good chunk of dmg, so not to nerf yourself too much, and works as a group offheal (kinda like zen but on demand and potentially spammable). So you trade off a chunk of dps for a chunk of offhealing (or anything really, the more creative you get the more interesting stuff gets and less subject to being shoehorned into a strict rotation/priority).

Unfortunately I realize that the class structure in swtor is quite rigid and adding something like this might simply be useless cause more heals aren't needed and you'd need to be full dps all the time. Still, being this mostly a mental exercise, I think it's worth thinking about it. And, tbh, even old classes could really benefit from some kind of paradigm shift.

c) Yeah I get what you mean about the sapling to tree path, it's just that I rather focus on the abstract concept and mechanic than get bogged in numbers. Comparing damage to base attack is already a good idea, personally I'd go for "major damage skill" "main damage skill" "filler", etc, representing more their roles than they're actual theoretical output. Just a different way of designing stuff I guess.

d) finally I understand that they hardly fit the Echani background, but I'm usually designing just the class mechanics, with perhaps an idea of what they are or roughly represent, trying to find holes that need to be filled or create new holes altogether, and then trying to get a concept that can wrap the class and work with that. As I'm mostly interested just in the gameplay/mechanical part of the design, I'd rather have THAT be the pillar upon which I build everything else, so as not to limit myself in what I can do. I mean, worst of all cases you can invent stuff from the ground up, as far as background goes (though ofc in this particular MMO it's best if you can find roots in the very large lore available)
Light Knights: Gecko - Syed - Vor'sann - Joya
Nightmare's Legion: Anhess - Avilus - Wittard - Schroedinger

KeyboardNinja's Avatar


KeyboardNinja
03.26.2013 , 08:11 PM | #27
Quote: Originally Posted by Kitru View Post
This is probably the biggest weakness I can see since Ebb and Flow are part of the fundamental design of the spec: it provides a 75% increase to the damage of the first attack of a Breath or Intuition resource consumption cycle (which, based on my design, should be your biggest hitter; i.e. Vortex or Avalanche Strike). Ignoring those contributions are going to have a *major* impact on the actual total damage dealt since those attacks, augmented in such a way, form the backbone of the rotation.
I do think that's true, and I intend to factor it in. I'm not sure that's enough to overcome the issue that I saw with the listed coefficients though. Specifically: the basic attack does too much damage relative to the other attacks (or the other attacks don't do enough damage, whichever way you want to parse it), and so the ideal rotation is practically just spamming your basic attack.

Quote: Originally Posted by Kitru View Post
I actually planned for Water to take the 1 sec cast time reduction in the Spirit tree so that it's actually got a 2.0 sec cast time. The baseline stats on True Strike are pretty much rubbish; I think when I was doing the math for it, the damage was so bad that you would never want to touch it (akin to Shadow Strike and Project for Shadows).
That does help quite a bit, but you still never use the capstone ability and the basic attack is practically spammed (every third GCD):

Code:
{0.95556, {a -> 4.15295, b -> 6.74255, c -> 7810.24, d -> 11., 
  e -> 44871.1, f -> 6.98213, g -> 23.2282, h -> 4372.65}}
Quote: Originally Posted by Kitru View Post
Also, were you factoring in the differences in the relevant damage types? The stance proc is Internal damage, which is substantially more valuable than raw kinetic/energy damage. In addition, I'm not sure why you had Warcry in there. It actually does less damage for more Breath than Power Strike does without factoring in the talents that increase Power Strike crit damage and decrease its Breath cost. Warcry shouldn't even be used as part of the ST rotation for Water (it's a useful AoE attack though).
I didn't factor in the stance proc, since I was primarily interested in the rotation priorities. Factoring that in basically gives a flat damage increase, but nothing more. I did factor in the damage type difference from Warcry. I wasn't making any assumptions about which abilities would be useful from a damage perspective (aside from discounting the AoEs), so I thought it best to include it. As you would expect, it's basically never used.

Quote: Originally Posted by Kitru View Post
I don't really think that's the problem. Quick combination is, explicitly, a basic attack and follows the formula for *all* basic attacks: no cost, no CD, 1.0 coefficient, weapon damage. It's a pretty good sign that, if you think the problem is that *basic attack*, then your model itself is flawed (as I said, I'm pretty sure your flaw is in the fact that you forgot to account for Ebb and Flow, which are a *major* part of the spec).
Well, there are several ways you can resolve the issue. Reducing the damage of the basic attack (or more correctly, increasing the damage of every other attack) is one way. The problem that the model shows right now is that the ideal DPS rotation uses Quick Combination significantly more than any other attack (Power Strike is next, used once every 6.7 seconds, after that True Strike). Adding in Ebb and Flow will absolutely help here, but the ICD on the proc makes me suspect that QC is still going to be over-emphasized. Once I figure out a good way to model it though, I'll be able to say with more certainty.
Computer Programmer. Theory Crafter. Dilettante on The Ebon Hawk.
Tam (shadow tank) Tov-ren (commando healer) Aveo (retired sentinel) Nimri (ruffian scoundrel)
Averith (marksman sniper) Alish (lightning sorcerer) Aresham (vengeance jugg) Effek (pyro pt)

KeyboardNinja's Avatar


KeyboardNinja
03.26.2013 , 08:36 PM | #28
Oh, another problem is that Quick Combination appears to be the only multi-hit ability. As such, its value gets inflated quite a bit by the stance (much more than it first seemed). Just picking up the 9% boost talent throws the entire model into chaos, reducing the ideal rotation to QC spam, TS and AS on CD:

Code:
{0.995071, {a -> 1.84429, b -> 55403.2, c -> 101437., d -> 11., 
  e -> 82049.2, f -> 123568., g -> 29.9992, h -> 6165.12}}
Computer Programmer. Theory Crafter. Dilettante on The Ebon Hawk.
Tam (shadow tank) Tov-ren (commando healer) Aveo (retired sentinel) Nimri (ruffian scoundrel)
Averith (marksman sniper) Alish (lightning sorcerer) Aresham (vengeance jugg) Effek (pyro pt)

Kitru's Avatar


Kitru
03.26.2013 , 09:42 PM | #29
Quote: Originally Posted by KeyboardNinja View Post
Oh, another problem is that Quick Combination appears to be the only multi-hit ability. As such, its value gets inflated quite a bit by the stance (much more than it first seemed). Just picking up the 9% boost talent throws the entire model into chaos, reducing the ideal rotation to QC spam, TS and AS on CD
Even when you ignore the stance for every other attack and include the stance *and* multihit functionality for Quick Combination (1.203 = (1 standard + (1 - (1 - 0.34) ^ 3) * (0.2 / 0.7)), QC only breaks even with default Power Strike (1.2 standards) and is still explicitly worse than default Rising Fist (1.35 standards). You're also ignoring the contributions of Rising Fist and Power Strike insofar as they reduce the CD of the 2 hardest hitting attacks you've got (Vortex and Avalanche), which are pretty significant when you consider that you actively need to get the CD on those 2 down in order to maximize their damage (they deal 2.21 standards of damage *before* factoring in the +.75 from Ebb/Flow).

Basically, you have to use 3-4 Breath consuming attacks to get to 5 stacks of Ebb (since you'll likely get 1-2 from stance procs) and 3-4 Intuition consuming attacks to get to 5 stacks of Flow (for the same reason). Even if QC was *equal* in damage (which it's not; it's lower as I've pointed out several times now), you'd still want to use Rising Fist and Power Strike if only to use up the otherwise wasted resources in order to get the tangential benefits.

For your perusal, with Crashing Wave factored in and, for simplicity's sake, Ebb and Flow (for Vortex and Avalanche), here are the damage values (not including the increased damage for PS and RF from the bonus surge, for simplicity's sake):

QC: 1.203 = (1 + (1 - (1-.34)^3) * (.2 / .7))
RF: 1.447 = (1.35 + .34 * (.2 / .7))
PS: 1.417 = (1.2 * 1.1 + .34 * (.2 / .7))
AS: 3.964 = (1.7 * 1.3 * 1.75 + .34 * (.2 / .7))
VS: 3.964 = (1.7 * 1.3 * 1.75 + .34 * (.2 / .7))
TS: 1.725 = 2.3 w/ 2 sec cast = (1.7 * 1.3 + .34 * (.2 / .7))

Even with the higher proc rate, QC is *still* the lowest damage attack by a *substantial* margin so I'm still gonna say you need to take a good hard look at your model since it makes absolutely *no* sense: if all you ever used was QC, VS, and AS, you would have a massive amount of wasted Breath (which would be best spent doing *more damage* for the *same amount of time* by using Power Strike) and wouldn't have the necessary Intuition (since QC doesn't generate Intuition and Avalanche only generates 1).
Walls of Text? I *love* Walls of Text!
My New Class Idea
Shadow Class Rep - Suggest/Review Questions Here
Quote: Originally Posted by Fende View Post
Listen to Kitru. Kitru knows all.

KeyboardNinja's Avatar


KeyboardNinja
03.26.2013 , 10:26 PM | #30
Quote: Originally Posted by Kitru View Post
Even with the higher proc rate, QC is *still* the lowest damage attack by a *substantial* margin so I'm still gonna say you need to take a good hard look at your model since it makes absolutely *no* sense: if all you ever used was QC, VS, and AS, you would have a massive amount of wasted Breath (which would be best spent doing *more damage* for the *same amount of time* by using Power Strike) and wouldn't have the necessary Intuition (since QC doesn't generate Intuition and Avalanche only generates 1).
Well, there's basically two possibilities: Mathematica is doing something really wonky with Maximize evaluation, or the constraints are just wrong. This same model does a good job of predicting rotations on live, so it would probably come down to typos and/or misunderstanding. As I said in my original post, please check my work. It would be nice to see something specifically wrong with the model, beyond the obviously strange implications. I'm hardly done working on it and refining it.
Computer Programmer. Theory Crafter. Dilettante on The Ebon Hawk.
Tam (shadow tank) Tov-ren (commando healer) Aveo (retired sentinel) Nimri (ruffian scoundrel)
Averith (marksman sniper) Alish (lightning sorcerer) Aresham (vengeance jugg) Effek (pyro pt)