Jump to content
  • Sign Up

And just like that. 2 great builds have vanished. A loss for Build Diversity.


Recommended Posts

@Obtena.7952 said:You can't argue your way around the fact that diversity is not defined with some additional quality characteristic. If the differences exist, it counts, no matter what your opinion is on how similar it is to another choice.

It’s because it lies on a spectrum, which in itself is multilayered because you have to factor in competition and complex systems analysis in order to even get an accurate description of gw2. It’s not black and white. and we aren’t making a toy model...we are trying to describe actual behavior seen in the game.

Just to take the colors example, you can have 10 shades of dark brown, and change one of them to red...but now what? You have to factor in the fact that agents now have to make a choice (introducing a competitive element) to even emulate behavior in the game.

If Red is the only color people pick because it’s vibrant or whatever, then brown can’t compete when it comes to people choosing. Likewise if brown is the only color people want to pick, then red can’t compete and falls out of favor. In both situations you are siphoning out choices because the diversity is low.

If the diversity is high, in which all colors are unique...red, blue yellow etc, then the choice that people make becomes qualitatively more rich...because the diversity is higher here.

In the grand scheme, if autonomous agents have the goal of creating paintings to sell to market (aka viable builds) then qualitatively, autonomous agents will choose the colors that sell more. If the market is nothing but brown paintings and red appears, red will sell more, and people will make more red paintings. Low diversity=bad balance.

If people have a unique variety of colors...blue, pink, green, orange... and the market is full of all kinds of colors, then no color will have any distinct advantage over any other color in the market. this is high diversity=good balance

If you change the colors, you will effect the balance...and because it’s a spectrum, as you collapse the choices closer to being homogenous (less diverse), you will create imbalance...like I mentioned earlier the two are different but linked together inextricably, and diversity is not as simple as just number of choices.

Edit: about usefulness. It’s not subjective. There is a definable metric that determines whether something is useful or not. This is what is called the autonomous goal. If the build you make can’t achieve some meaningful goal, then why would you run it? Meaningful goal is simple like “can you kill an enemy.” But it has to have some meaning in a competitive environment.

You can’t have an autonomous goal that is useless like “ stand around and look pretty.” Because making a build to fulfill this goal has no meaning in a competitive environment where others are competing with their own autonomous goals.

Link to comment
Share on other sites

@JusticeRetroHunter.7684 said:

@Obtena.7952 said:You can't argue your way around the fact that diversity is not defined with some additional quality characteristic. If the differences exist, it counts, no matter what your opinion is on how similar it is to another choice.

It’s because it lies on a spectrum,

What defines diversity is not a continuous parameter that's on a spectrum regardless of whether the system is complex or not. It's simply a count of distinguishable different things. This is NOT disputable. That is how diversity is defined. If you are going to redefine diversity with these additional complexities to suit your argument, that's just not having an honest discussion.

@JusticeRetroHunter.7684 said:

@Obtena.7952 said:Just to take the colors example, you can have 10 shades of dark brown, and change one of them to red...but now what? You have to factor in the fact that agents now have to make a choice (introducing a competitive element) to even emulate behavior in the game.No I don't. I simply count the number of different things I had before and after the change. If they are the same, there is NO impact on diversity. That's exactly the case here with the rune change.

If Red is the only color people pick because it’s vibrant or whatever, then brown can’t compete when it comes to people choosing. Likewise if brown is the only color people want to pick, then red can’t compete and falls out of favor. In both situations you are siphoning out choices because the diversity is low.

OK, but the level of diversity didn't change when change one of the colours, even if diversity is low.

If the diversity is high, in which all colors are unique...red, blue yellow etc, then the choice that people make becomes qualitatively more rich...because the diversity is higher here.

OK, but again, the diversity doesn't change if we change one of the colours, even if diversity is high.

Again we are NOT talking about how much diversity we have, we are talking about the impact on the diversity if you change the choices. Even by the examples you outline ... as long as the choice changes don't result in a change in the number of distinguishable different things, the diversity is not affected. This is what we have with the rune change.

Link to comment
Share on other sites

@"Obtena.7952" said:

Just gonna put an end to this back and forth already by providing you a scientific paper from IOP (The Institute of Physics)

https://iopscience.iop.org/article/10.1088/1478-3975/ab6754

XbW2qv7.jpg

As you can see everything they say here agrees with what I’ve said. Explicitly pointing out that it’s a relationship between richness (which in our case is the quality) and evenness (which is in our case balance.) they demonstrate that a system with less components can be more diverse than a system with more components because of these two parameters...and they go onto include there are more parameters and these are fleshed out in the equations below the abstract.

I’m not pulling this stuff out of thin air, it’s well researched in other sciences

Link to comment
Share on other sites

That's nice and all but we aren't in a research institute, we aren't physicists and we aren't trying turn a simple concept like diversity into something you need to study to have a discussion about it. In your world or whatever field that is, that 'expansion' of diversity definition makes sense ... this is an MMO ... with average people playing it, with average people making it, with average people discussing it ... they AREN'T taking a deep dive into a scientific journal to reset what the definition of diversity that people are aware of ... because they want to be understood when they have a discussion and not talk over people. You can put away your journal, here is what Webster says:

the condition of having or being composed of differing elements

Like I said ... you got a whole different definition going there ... not sure why you can't see it or acknowledge it.

Link to comment
Share on other sites

@Obtena.7952 said:That's nice and all but we aren't in a research institute, we aren't physicists and we aren't trying turn a simple concept like diversity into something you need to study to have a discussion about it. In your world or whatever field that is, that 'expansion' of diversity definition makes sense ... this is an MMO ... with average people playing it, with average people making it, with average people discussing it ... they AREN'T taking a deep dive into a scientific journal to reset what the definition of diversity that people are aware of ... because they want to be understood when they have a discussion and not talk over people.

Like I said ... you got a whole different definition going there ... not sure why you can't see it or acknowledge it.

Because his definition is actually useful and meaningful for the discussion of balance.

What you are describing is variations, not diversity. Just because I CAN choose between some 1.2 billion available combinations before and after doesn't mean I will. Your choices are inherently influenced by your environment and thus you won't choose things that don't work. No one is running around with 14 different stat types in their gear.

You can't talk about balance without factoring in the environment because what works in raids doesn't work in fractals doesn't work in PvP. So despite there being a slew of different builds for each, the systems have inherently varying levels of diversity.

Anet adjusted the balance for an outlier by changing the system, and this shrunk the pool of viable choices. When they could have adjusted the outlier. It is a valid complaint, though I question if the rune change was as devastating as Justice claims.

Link to comment
Share on other sites

@God.2708 said:

@Obtena.7952 said:That's nice and all but we aren't in a research institute, we aren't physicists and we aren't trying turn a simple concept like diversity into something you need to study to have a discussion about it. In your world or whatever field that is, that 'expansion' of diversity definition makes sense ... this is an MMO ... with average people playing it, with average people making it, with average people discussing it ... they AREN'T taking a deep dive into a scientific journal to reset what the definition of diversity that people are aware of ... because they want to be understood when they have a discussion and not talk over people.

Like I said ... you got a whole different definition going there ... not sure why you can't see it or acknowledge it.

Because his definition is actually useful and meaningful for the discussion of balance.

Except in the history of the game, I've never seen Anet balance something because of diversity ... so the whole complaint is meaningless in the first place. Unless Anet starts reading IoP journals, I'm pretty sure that whatever 'useful' and 'meaningful' definition of diversity you want to argue is the relevant one for balance, it's not going to be the one that requires a deep dive and years of study ... it's going to be Websters.

Link to comment
Share on other sites

@Obtena.7952 said:

@God.2708 said:

@Obtena.7952 said:That's nice and all but we aren't in a research institute, we aren't physicists and we aren't trying turn a simple concept like diversity into something you need to study to have a discussion about it. In your world or whatever field that is, that 'expansion' of diversity definition makes sense ... this is an MMO ... with average people playing it, with average people making it, with average people discussing it ... they AREN'T taking a deep dive into a scientific journal to reset what the definition of diversity that people are aware of ... because they want to be understood when they have a discussion and not talk over people.

Like I said ... you got a whole different definition going there ... not sure why you can't see it or acknowledge it.

Because his definition is actually useful and meaningful for the discussion of balance.

Except in the history of the game, I've never seen Anet balance something because of diversity ... so the whole complaint is meaningless in the first place. Unless Anet starts reading IoP journals, I'm pretty sure that whatever 'useful' and 'meaningful' definition of diversity you want to create is irrelevant.

Well...

Here is where I go 'just because you don't see it doesn't mean it doesn't exist'

CMC at the very least is on record saying he'd like to see at least one build for every core/elite spec be viable. That's... By our definition at least, making it more diverse. The Feb balance patch was intended to lay the groundwork to assist in this endeavor and, here is where I disagree with Justice because, I think diversity has increased greatly since the patch. Though I'll grant that support hasnt changed much yet and it's mostly damage side diversity.

I don't think Anet is balancing for diversity with as much fervor as Justice would like to see. But they are balancing for diversity's sake at the moment.

Link to comment
Share on other sites

@God.2708 said:

@God.2708 said:

@Obtena.7952 said:That's nice and all but we aren't in a research institute, we aren't physicists and we aren't trying turn a simple concept like diversity into something you need to study to have a discussion about it. In your world or whatever field that is, that 'expansion' of diversity definition makes sense ... this is an MMO ... with average people playing it, with average people making it, with average people discussing it ... they AREN'T taking a deep dive into a scientific journal to reset what the definition of diversity that people are aware of ... because they want to be understood when they have a discussion and not talk over people.

Like I said ... you got a whole different definition going there ... not sure why you can't see it or acknowledge it.

Because his definition is actually useful and meaningful for the discussion of balance.

Except in the history of the game, I've never seen Anet balance something because of diversity ... so the whole complaint is meaningless in the first place. Unless Anet starts reading IoP journals, I'm pretty sure that whatever 'useful' and 'meaningful' definition of diversity you want to create is irrelevant.

Well...

Here is where I go 'just because you don't see it doesn't mean it doesn't exist'

CMC at the very least is on record saying he'd like to see at least one build for every core/elite spec be viable. That's... By our definition at least, making it more diverse. The Feb balance patch was intended to lay the groundwork to assist in this endeavor and, here is where I disagree with Justice because, I think diversity has increased greatly since the patch. Though I'll grant that support hasnt changed much yet and it's mostly damage side diversity.

I don't think Anet is balancing for diversity with as much fervor as Justice would like to see. But they are balancing for diversity's sake at the moment.

I'm going to say we need a significant increase in the number of 'viable' builds to be convinced they are doing so for 'diversity'. I'm of the belief that they aren't going to achieve that with changes on the pool of choices we have, i.e., not through balancing otherwise we would have it. I think they are just going to continue to release new choices until eventually, some acceptable number of viable choices exist simply from the overwhelming number of options available to players.

To be frank, IF the question of diversity is so complex that we can't talk about it in simple terms, then it's absurd to complain that Anet shouldn't make changes that affect it negatively ... how can anyone possibly quantify diversity to claim a change is a negative impact on it? In otherwords, if measuring the impact on diversity is so complex that it renders it impractical to do so, then it's actually a terrible quality to assess the impact of balance efforts.

Link to comment
Share on other sites

Diversification is important but as God,2708 explains it's dependent on environment and how a particular build can fill a niche in that environment. What has unfortunately happened is only a few classes have become dominant because these classes have become to diverse filling to many niches . What Anet has to do is look at how and why these particular dominant classes are filling theses niches . Does the Scourge have to many effective AOE's in one single build ? Do the elementalist have far to much condition cleanse, Does scrapper have to much effective heal and utilities that effect to many people . Is Winds of Disenchantment far to powerful by disallowing the opponent from receiving boons . Is there far to much boon share. Are to few classes able deliver far to many effective boons ,cleanse , crowd control in a single build. Is Coalescence of Ruin still far to powerful effecting to many opponents when you consider the cast time and recharge being so low. Is barrier , resistance , retaliation over used and delivered by to few classes and needs to be cut back and spread out evenly with the other classes . Are similar skills and traits between classes actually balanced with the same ease of use and effect. Many are not .The core classes were fairly balanced prior to these releases with only a few exceptions that could have been easily fixed then but were not. The core classes are the foundation and ANET can't fix the problem without fixing the foundation or by weakening it further. The new releases should never have overwhelmed the core .If a problem exists in a single skill or trait then fix that skill or trait not the entire environment. You would not fill in a lake to eliminate one dominating invasive species of fish. Are some skills and traits over powered by doing to many things far to long or large that affect to many with to short of cast times and recharges ?ANET needs to take a really good look at what was introduced to the game when we consider that most of the problems started after Path of Fire.The power creep was a problem that was started with the release of Heart of Thorns and further compounded with Path of Fire and was addressed by lowering the bar for all classes in February. Now it's time to fix the issues that cause one class to dominate over other classes far to easily.

Link to comment
Share on other sites

@Obtena.7952 said:

@God.2708 said:Well...

Here is where I go 'just because you don't see it doesn't mean it doesn't exist'

CMC at the very least is on record saying he'd like to see at least one build for every core/elite spec be viable. That's... By our definition at least, making it more diverse. The Feb balance patch was intended to lay the groundwork to assist in this endeavor and, here is where I disagree with Justice because, I think diversity has increased greatly since the patch. Though I'll grant that support hasnt changed much yet and it's mostly damage side diversity.

I don't think Anet is balancing for diversity with as much fervor as Justice would like to see. But they are balancing for diversity's sake at the moment.

I'm going to say we need a significant increase in the number of 'viable' builds to be convinced they are doing so for 'diversity'. I'm of the belief that they aren't going to achieve that with changes on the pool of choices we have, i.e., not through balancing otherwise we would have it. I think they are just going to continue to release new choices until eventually, some acceptable number of viable choices exist simply from the overwhelming number of options available to players.

...In otherwords, if measuring the impact on diversity is so complex that it renders it impractical to do so, then it's actually a
terrible
quality to assess the impact of balance efforts.

My personal view, is that even if Anet has the intention of balancing for diversity, they have no idea how to do it successfully or have some very simplistic view of what it is, which is understandable. Any logical person, myself included would assume at first that increasing the number of choices or trying to equalize those choices via buffs and nerfs should increase diversity, but that is a fallacy.

The research I've been doing has been to try and understand why the changes that Anet has made, has made little to no impact on the diversity and by proxy balance of the game, and it is enough of a problem that it should warrant investigation. Given that there are probably 1.2 billion combinations for possible build combinations if not more in this game, seeing only 2 or 3 in the meta is a very large discrepancy, that shouldn't be ignored.

To be frank, IF the question of diversity is so complex that we can't talk about it in simple terms, then it's absurd to complain that Anet shouldn't make changes that affect it negatively ... how can anyone possibly quantify diversity to claim a change is a negative impact on it?

Diversity is complex...Very much so, and i believe it's one reason why MMO's have such a hard time balancing between classes. It's not so complex that we can't talk about it...we can quantify it and even come down with solutions on how to procure it successfully. I've done a lot of the legwork to get it to a place here, where we can talk about it in English....even if it can be difficult at times.

But even getting to that point requires understanding the problem and the whole problem to begin with, and i think most don't even know what the problem is. We have frequent post on the forums asking for nerfs to specific classes or builds a l l o f t h e t i m e to the point where it's irritating. But we have to go into the rabbit hole and start asking...why is the build overpowered? Is it because of a mechanic? or is it because there is no sufficient counter measure to it? If it's because of a lack of a countermeasure, does that mean we have to buff other builds? Does buffing and nerfing classes make more overpowered builds or less? Why do we have overpowered builds to begin with? These aren't even right questions to ask yet...because they all lead back to an essential element, that balance and diversity are separate things, that are inextricably linked together, with the other parameters we've discussed (HomovsHeterogenity, Autonomous Agents and Goals, Competition, Differing Balance Mechanisms...) , that in my opinion, seem to not be in the thought process when making balance decisions.

We could say that none of it matters anyway because Anet doesn't read forums and it's Anets game they can do what they want...but then again why say anything on the forums at all, if none of it really matters anyway? It's sort of a dead-end deal with this line of reasoning no matter how you look at it in this prespective, so why not?

Link to comment
Share on other sites

@JusticeRetroHunter.7684 said:

@"God.2708" said:Well...

Here is where I go 'just because you don't see it doesn't mean it doesn't exist'

CMC at the very least is on record saying he'd like to see at least one build for every core/elite spec be viable. That's... By our definition at least, making it more diverse. The Feb balance patch was intended to lay the groundwork to assist in this endeavor and, here is where I disagree with Justice because, I think diversity has increased greatly since the patch. Though I'll grant that support hasnt changed much yet and it's mostly damage side diversity.

I don't think Anet is balancing for diversity with as much fervor as Justice would like to see. But they are balancing for diversity's sake at the moment.

I'm going to say we need a significant increase in the number of 'viable' builds to be convinced they are doing so for 'diversity'. I'm of the belief that they aren't going to achieve that with changes on the pool of choices we have, i.e., not through balancing otherwise we would have it. I think they are just going to continue to release new choices until eventually, some acceptable number of viable choices exist simply from the overwhelming number of options available to players.

...In otherwords, if measuring the impact on diversity is so complex that it renders it impractical to do so, then it's actually a
terrible
quality to assess the impact of balance efforts.

My personal view, is that even if Anet has the intention of balancing for diversity, they have no idea how to do it successfully or have some very simplistic view of what it is, which is understandable. Any logical person, myself included would assume at first that increasing the number of choices or trying to equalize those choices via buffs and nerfs should increase diversity, but that is a fallacy.

The research I've been doing has been to try and understand why the changes that Anet has made, has made little to no impact on the diversity and by proxy balance of the game, and it is enough of a problem that it should warrant investigation. Given that there are probably 1.2 billion combinations for possible build combinations if not more in this game, seeing only 2 or 3 in the meta is a very large discrepancy, that shouldn't be ignored.

To be frank, IF the question of diversity is so complex that we can't talk about it in simple terms, then it's absurd to complain that Anet shouldn't make changes that affect it negatively ... how can anyone possibly quantify diversity to claim a change is a negative impact on it?

Diversity is complex...Very much so, and i believe it's one reason why MMO's have such a hard time balancing between classes. It's not so complex that we can't talk about it...we can quantify it and even come down with solutions on how to procure it successfully. I've done a lot of the legwork to get it to a place here, where we can talk about it in English....even if it can be difficult at times.

But even getting to that point requires understanding the problem and the whole problem to begin with, and i think most don't even know what the problem is. We have frequent post on the forums asking for nerfs to specific classes or builds
a l l o f t h e t i m e
to the point where it's irritating. But we have to go into the rabbit hole and start asking...why is the build overpowered? Is it because of a mechanic? or is it because there is no sufficient counter measure to it? If it's because of a lack of a countermeasure, does that mean we have to buff other builds? Does buffing and nerfing classes make more overpowered builds or less? Why do we have overpowered builds to begin with? These aren't even right questions to ask yet...because they all lead back to an essential element, that balance and diversity are separate things, that are inextricably linked together, with the other parameters we've discussed (HomovsHeterogenity, Autonomous Agents and Goals, Competition, Differing Balance Mechanisms...) , that in my opinion, seem to not be in the thought process when making balance decisions.

We could say that none of it matters anyway because Anet doesn't read forums and it's Anets game they can do what they want...but then again why say anything on the forums at all, if none of it really matters anyway? It's sort of a dead-end deal with this line of reasoning no matter how you look at it in this prespective, so why not?

There isn't a reluctance to talk about it, but to be honest, if you are the only one in the room who understands what you are talking about, no one is going to listen to you. I don't think it's reasonable to claim the changes have a negative impact on diversity, especially if you are going to talk over everyone with how it's quantified and I hasn't seemed relevant to the game anyways.

if Anet is making changes that negatively impact diversity, what makes us think it's that important to them? I think you are making this too complex to suit your argument and create a platform that only you can have a discussion in; by making diversity complicated, of course the only person in the room that can discuss it is the one that 'proved' it's complication. Honestly, I don't think Anet's criteria for balancing is anything more than Anet going "We don't like how this works ... it's changing". There is no diversity meter they check before making a change. You're just not speaking the language.

Anyways, GL with your math proofs and complex, abstract theories on diversity to compel Anet to not make changes you don't like.

Link to comment
Share on other sites

@Obtena.7952 said:

I think you are making this too complex to suit your argument and create a platform that only you can have a discussion in;

That's a pretty low blow...i mean now your just attacking my integrity.

Sure, you may not like me, but science is science. Sometimes we want the answers to be simple, but cant always have our cake and eat it. Do i wish that diversity was as simple as you make it out to be? Of course i would...but i would be purposefully remaining blissfully ignorant to how it actually works...We didn't come this far in human history by believing in ghosts powering our CPU's...we had to learn really hard and complicated science to even play this game at all.

So if you want to be blissfully in the dark then be my guest. You can do the research yourself, its not like i'm anyone doing anything special...i'm just some shmoe putting in my time to do legitimate research...sometimes that research requires you to learn how to do calculus so you can calculate that integral in those probability distribution functions. That's part of it...and if you aren't willing to go that far, that's on you...not me.

Link to comment
Share on other sites

@JusticeRetroHunter.7684 said:

I think you are making this too complex to suit your argument and create a platform that only you can have a discussion in;

That's a pretty low blow...i mean now your just attacking my integrity.

It's not a low blow ...we don't need some complex scientific theory to make these discussions because complex scientific theories aren't how these changes are made to begin with ... but you insist that's where we push the discussion. It's not an attack on your integrity ... but I am questioning the motive of claiming this complex version of diversity that you claim has been negatively effected is even true or relevant. If you are going to make a claim that Anet has made some error in game change because of some complicated theory on diversity that few people are going to understand, you might as well be talking to yourself. I think the truth is that it's not really reasonable to select one change, claim diversity took a hit and conclude they shouldn't have done it. If the only person that understands how that claim works is you ... it's pretty meaningless.

It's not about if I'm willing to apply myself to learn advance math or some ecological theories to enter the discussion. it's about whether or not it's necessary to do even GO to that level. I don't see the relevance of pulling out quotes from IoP and such ... the relevance of going to that level to make a case for Anet to stop making changes we don't like ... is very low. You are talking over LOTS of people here.

@Voltekka.2375 said:2 less builds to play equals less diversity. Can't get simpler than that.

Except these builds can still be played and those aren't the only changes that affect overall 'diversity' so ... /shrug. I mean, again ... if you can't quantify how ALL the changes in this patch affect overall diversity ... there is no claim that we just took some massive hit to diversity that's bad for the game. It's just sensationalism at this point. Maybe these two took a hit ... maybe some other change boost 5 others.

Link to comment
Share on other sites

Altruism in that state felt pretty degenerate to play, even with the risk of instant downing when 1 or 2 condi bombs happen. Though there was an elegance to playing it by mixing in the regular healing rotations that I found fun.

What Arcdps needs though is more advanced sorting options for interrupts and blocks (and other things found in the detail stats window). This is a lot of information so even a buff table style layout where we can list cleanse, strips, interrupts, and blocks in one window.

Anet also needs to send healing values to the server so Arcdps can pick it up.

Link to comment
Share on other sites

@Obtena.7952 said:It's not about if I'm willing to apply myself to learn advance math or some ecological theories to enter the discussion. it's about whether or not it's necessary to do even GO to that level.

So you are the authority on where that level is supposed to be? That sounds hypocritical.

I'm sorry but, if i walk into silicone valley, i better know how to at least understand computer programming at a basic level...it's just a courtesy...or should i stand there and berate them about how they don't know anything, and that i know everything there possibly is to know even though I've never done an ounce of research on coding in my life...that they should cater to MY needs rather than theirs?

Btw the math isn't THAT complicated. Its doable without anything more advanced than basic algebra. We can do it right here, right now if you like.

Link to comment
Share on other sites

@JusticeRetroHunter.7684 said:

@"Obtena.7952" said:It's not about if I'm willing to apply myself to learn advance math or some ecological theories to enter the discussion. it's about whether or not it's necessary to do even GO to that level.

So you are the authority on where that level is supposed to be? That sounds hypocritical.

If you think it's appropriate to bring math proofs and IoP journal paper quotes into a discussion about diversity in an MMO to make your "I don't like this change" complaint more valid ... then you believe whatever you want.

I'm sorry but, if i walk into silicone valley, i better know how to at least understand computer programming at a basic level...it's just a courtesy...or should i stand there and berate them about how they don't know anything, and that i know everything there possibly is to know even though I've never done an ounce of research on coding in my life...that they should cater to MY needs rather than theirs?

So you think you're being COURTEOUS by presenting your math proofs and such? To who? Definitely not the average forum reader. Maybe you think Anet isn't listening to players complaints because we just don't talk at this academic theory level to them and they think it's rude? There is definitely a balance here, but instead of ranting which is one end of the spectrum, you've gone to the other end with this deep academic connection. I don't think Anet is making decisions based on rants ... but I don't think they are making it academic proofs either because neither of those is practical and sensible.

Btw the math isn't THAT complicated. Its doable without anything more advanced than basic algebra. We can do it right here, right now if you like.

Go ahead and do the math ... I don't see the relevance of mathematically proving whatever you want to show because a lack of mathematical proof never prevented Anet from making game changes they wanted in the first place. The game isn't an experiment in a university. The people that create and decide how to change the classes aren't reading IoP on their spare time. If they were, it would cost us ALOT more to play it.

Link to comment
Share on other sites

@"Obtena.7952" said:

So before going into the equation, we need to adjust the terminology a little bit, since this is borrowed from biodiversity. So their terminology is things like "species" and "ecosystems," and stuff like that but I'm going to replace those terms with the appropriate and relevant terms.

One of the more simple ways to quantify diversity is using the Simpson Diversity Index, which is the following equationD= 1-(∑ⁿn^2/N^2))

Where N is the number of Total number of Organisms of all Species (This is the total amount of players in the sample size)and n is the number of organisms of a specific species (This is going to be the number people that play a specific build, which for now, we will classify as 1 of a total of 126 possible build combinations.)

The Equation above, when we plug in numbers, will give us a value that ranges between 0 and 1. 1 Being Infinite Diversity (Complete Heterogeneity) and 0 Being no diversity at all (complete homogeneity).

I'm going to show you two examples, and both will kinda show you the range at which the values of the equation take on.

The First Sample -Let's say we go into WvW, and join whatever squads we run into to see the composition of their group. Let's say we also run into at least 1 of the total 126 possible build combinations at least once in these groups, but most of these groups are dominated by the 6 builds you can find on meta-battle. Let's now say that our sample size consists of at least 1000 players, where 121 of them are running those non-meta builds, and the other 879 were running some distribution of those meta builds. This is a close representation of what we see in WvW right now.

With all the details out of the way, we now want to find N^2. which is 1000^2 = 1,000,000.

Now we want to find n^2, which would look something like this -Build A= 186Build B= 299Build C=128Build D=100Build E= 96Build F= 70Build G1,G3....Gⁿ = ∑ⁿ1^2 (These are all the players we ran into that were running at least one of the other possible non meta builds. When you sum over 1^2 with respect to n, you get 121.

we continue to find little n by summing for Builds A-F.186^2 + 299^2 + 128^2 + 100^2 + 96^ + 70^ = 164,497

Now we just plug in:D= 1-(∑ⁿn^2/N^2))D= 1-(164,497/1,000,000))D= 1-(0.164497)D= 0.835503

D is the diversity index of this distribution. I'm now going to show you another example, that will give this number a sense of comparison, so that we can see the range of how diverse the system actually is.

The Second Sample -Again, we go into WvW and we join squads to see what they are running. This time, let's pretend that instead of 6 meta builds, there are at about 10 times that...60 meta builds that players choose to play. Again we run with the same sample size of 1000 players, and we see at least 1 of the total 126 possible build combinations where 66 of them are running those non-meta builds, and the other 934 running some distribution of the meta builds.

Again we follow the same procedure with some distribution, To save myself from writing so much im going to do a shortcutD= 1-(∑ⁿn^2/N^2))

Where N^2 = 1000

Build A= 15Build B= 13Build C=14Build D=17Build E= 13Build F1...Fⁿ= some number between 2 - 17 (I'm just going to repeat the same sequence of numbers from builds A through E since doing the calculation for 60 of them is a pain in the kitten)Build G1,G3....Gⁿ = ∑ⁿ1^2 (These are all the players we ran into that were running at least one of the other possible non meta builds. When you sum over 1^2 with respect to n, you get 66.

We now continue to find little n by summing for Builds A-Fⁿ.15^2 + 13^2 + 14^2 + 17^2 + 13^2 + Fⁿ^2 = Approx 11,496

Now we just plug in:D= 1-(∑ⁿn^2/N^2))D= 1-(11496/1,000,000))D= 1-(0.011496)D= 0.988504

The Diversity Index, when we increase the range of viable meta builds from 6 to 60, jumps by only 18%. So you can see that qualitatively, the difference between an index of 0.835503 and 0.988504 is drastic. The same behavior happens in the opposite direction as well, when you close in towards homogeneity, where collapsing the builds closer to 1 causes exponentially less and less diversity. The reason it behaves like this here is because the relation is exponential, where as you get closer to either end of this infinite spectrum between 1 and 0, you are gaining exponentially more increases or decreases in diversity. This is one reason, why changes that effect even just one little tiny build can have a HUGE effect on the diversity of the game.

Remove one meta build and that index of drops 0.835503 to 0.6 something something...remove another and that 0.6 goes to 0.2 something something... Continue removing viable builds and that diversity collapses closer and closer to 0 until we reach a single meta build (complete homogeneity)

Link to comment
Share on other sites

So ... this is essentially a quantification of diversity with a weighted count. OK ... so you have a method ... where did you get your data to apply it to the current situation and what difference did you get in your index based on the rune change?

I mean ... sure, at the extreme ends of this index, there is wild behaviour because of the governing math ... but that doesn't say anything about the rune change and in addition, this index is measuring the diversity of a whole system ... but you're only focused on one change that you assume drives people to builds that already exist to reduce this diversity ... what is the basis for your assumption?

You know .. it IS possible that the rune change actually maintains diversity or even increases it (according to your definition) you know. Nothing in that method predetermines that when a build changes, the diversity decreases. If I got 10 people using the 'altruism rune' build and it gets nerfed ... nothing says each of those people don't pick up 10 builds no one uses. That would be a diversity increase according to this method.

Link to comment
Share on other sites

@Obtena.7952 said:So ... this is essentially a quantification of diversity with a weighted count. OK ... so you have a method ... where did you get your data to apply it to the current situation and what difference did you get in your index based on the rune change?

Well the above scenarios are entirely fictitious, but they are meant to represent an approximation of the current state of the game (Build A = Meta Firebrand, Build B = Meta Scourge, Build C= Meta Revenant and so on....) and can be done in practice if you want to go into squads and just see what builds people run, you can find the real index at this very moment right now in the game.

The equation is meant to illustrate the relationship that diversity has with respect to the other parameters. The most important thing to take away is the comparison between two systems, is logarithmic/exponential. It's like the magnitude of earthquakes and how they are classified on the Richter scale. A magnitude of 10 is 10 times more powerful than a magnitude of 9...and a magnitude of 2 is also 10 times more powerful than a magnitude of 1. But we can agree that both a magnitude of 1 and 2 is pretty negligible while the difference between 9 and 10 is catastrophic. The homogeneous end of this spectrum is like looking at the difference between 9 and 10 on that scale... and on the other end in a heterogeneous side the difference between a 1 and 2. where minut changes in a homogeneous pool are just catastrophic, and minut changes in a highly diverse system are near negligible because of the exponential nature of the equation...

And this makes sense when we talk about diversity. If we killed a species of bug today on the planet we don't really notice at all because the system is so diverse. but if we killed a species when there are only 2 species left on the planet, that's a big deal. Likewise if we added a species in a world with only 2 species...that's a big deal. So looking at guild wars 2, where the meta is only 5 builds, nuking one of them is a big deal.

For the rune, i could technically calculate that too. But i don't really have to because i know the result is just gonna be a decrease in the diversity index. Like stated above, If you have 6 builds to choose from because they are the only ones that achieve a goal, and one of them gets essentially removed and you don't replace that with anything, then people will only pick between the 5 builds, and we approach closer to the homogeneous side of the spectrum. The less builds there are that people are choosing especially because of balance changes, the more potent the drop in diversity becomes because of that exponential relationship mentioned above.

Link to comment
Share on other sites

@JusticeRetroHunter.7684 said:

For the rune, i could technically calculate that too. But i don't really have to because i know the result is just gonna be a decrease in the diversity index.

Actually you don't ... because you don't know if people moving from a nerfed build adopt established builds (lowering DI) or builds that no one else plays (increasing it). Again, the build isn't REMOVED ... it's simply changed and there are numerous builds out there no one plays; maybe a number of changes make some of those appealing to people. You could even get an INCREASE in the DIndex ... nothing in the math prevents the DI from going up if a build changes. No, I don't think you can conclude that if a build changes, the diversity goes down. That's not a given based on the math.

To be frank ... if the definition of diversity is based on how many people play whatever builds, how can you even tell if a change in DI is due to a specific effect? If you can't tell us how much ... why are we assuming it's a 'significant' amount or at all? DI can go down AND up. I mean, there are actually LOTS of reasons for diversity to change and a rune change is just one of them. The number of people play a game mode and choose a specific build for it are numerous and both of these things affect the DI ... how does Anet know if a game change they make is what triggers some change in diversity? DI can't tell you WHY it's changing and when there are so many factors that can change it, it's folly to think a specific DI change could be attributed to a game change.

In otherwords, it's really great we can measure diversity but it's pretty irrelevant to do so because game changes aren't made in isolation. Therefore, anyone using 'diversity changes' it to argue that a specific game change is bad is just making a contrived argument.

Link to comment
Share on other sites

Well that’s where the other parameters come into play that this particular equation doesn’t describe explicitly, but implicitly. It’s described by the following:

The reason people choose builds is to fulfill some kind of meaningful goal. Most build combinations simply can’t compete and this is why nobody plays them. That’s part of the parameters of the equation, is the number of people choosing to play a build.

Again I can choose to play some kind of special snowflake build where I just look pretty during a fight...this option probably exists in gw2’s possibility space. This doesn’t mean that my build will be chosen by others, especially in a competitive environment where being able to kill someone or survive a fight is a goal that is meaningful in the game mode. This choosing is part of that equation, in which you might see one person in a squad running some kind of off meta build...

The reason people don’t play my special snowflake build can be for a multitude of reasons, but the reason is more than likely because it can’t compete with the other options, and consequently is why it’s not one of the meta builds.

This is where you have to go beyond the scope of just diversity, in that balance and performance plays a key role in deciding what builds can achieve goals and which can not, which by proxy details which builds get chosen by players and which do not. This is a parameter of usefulness, which again has an effect on the diversity Index (by effecting what people will choose to play, which is a parameter in the equation.)

Link to comment
Share on other sites

Well that’s where the other parameters come into play that this particular equation doesn’t describe explicitly, but implicitly. It’s described by the following:

The reason people choose builds is to fulfill some kind of meaningful goal. Most build combinations simply can’t compete and this is why nobody plays them. That’s part of the parameters of the equation, is the number of people choosing to play a build.

Again I can choose to play some kind of special snowflake build where I just look pretty during a fight...this option probably exists in gw2’s possibility space. This doesn’t mean that my build will be chosen by others, especially in a competitive environment where being able to kill someone or survive a fight is a goal that is meaningful in the game mode. This choosing is part of that equation, in which you might see one person in a squad running some kind of off meta build...

The reason people don’t play my special snowflake build can be for a multitude of reasons, but the reason is more than likely because it can’t compete with the other options, and consequently is why it’s not one of the meta builds.

This is where you have to go beyond the scope of just diversity, in that balance and performance plays a key role in deciding what builds can achieve goals and which can not, which by proxy details which builds get chosen by players and which do not. This is a parameter of usefulness, which again has an effect on the diversity Index (by effecting what people will choose to play, which is a parameter in the equation.)

OK but this doesn't address my point ... you say that diversity goes down IN THIS CASE because of the rune change ... but you don't know that. In actual fact, we can't attribute a decrease in diversity to ANY particular change because diversity is affected by MANY things at once.

Link to comment
Share on other sites

@Obtena.7952 said:OK but this doesn't address my point ... you say that diversity goes down IN THIS CASE because of the rune change ... but you don't know that. In actual fact, we can't attribute a decrease in diversity to ANY particular change because diversity is affected by MANY things at once.

It's not that i'm not addressing your points...what i'm saying is me addressing the point. Perhaps i'm not explaining it well enough.

Determining whether a change will increase or decrease the diversity index, is based on what people are choosing to play. So if 70 out of those 1000 people in the sample size were playing Altruism Engi, those 70 people will now play something else when we resample. That something else they decide to play can be anything, but is determined by what is most likely going to be the most likely scenario... being that they will choose something that can achieve an autonomous goal ie: one of those already popular meta builds. The only way is to really find out is to just do a new calculation on the index.

Now you are right that there is no explicit thing to say that people will choose the meta...that's why this function requires a sample size for data, you have to take another sample at a later time (before and after an event) to determine the change in the system. So the explanation here is an implicit one.

Just to show you another example of what could happen here...let's pretend all the parameters from the earlier samples apply here, but this time im just using way less variables, where there are only 3 meta builds (One of which is Altruism Engi) -Build A= 80Build B= 60Build C=56Build D=3Build E=1

Let's say build C is Altruism Engi, and Build D is my 0 energy Rev, and Build E is some other Non-meta build The DI Index equation...D= 1-(∑ⁿn^2/N^2))

Now N = 200, so N^2 = 40,000∑ⁿn^2 = 13,146

D= 1-(13,146/40,000))D= 1-0.32865D= 0.67135

Now Let's say Build C and D get's nerfed with a change to whatever...rune, skill...doesn't matter so long as what gets changed has an actual effect on the choosing of that build. IMO Altruism change is enough for people to not play this build anymore (they effectively killed it). So now, we have a deficit of 56 players in our sample... where are they gonna go? Are they going to go to some non meta Build E? It's not likely.

The most likely scenario is that players will choose what works, which is Build A and B. Even in the scenario where they all decide to all go play some new Build E, The Diversity will at best remain the same...so you can have a distribution that either looks something like this:

The most likely scenario (players flocking to the meta builds)-Build A= 112Build B= 86Build C= 1Build D=1Build E=3

Where D= 1-(∑ⁿn^2/N^2)) = 0.50125

Or the least likely scenario (Players flocking to some other non meta builds) -

Build A= 80Build B= 60Build C=0Build D=0Build E=60

D= 1-(∑ⁿn^2/N^2)) = 0.66

As you can see, the number never really gets above the first index of .67.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...