Yoshi P talks about Warriors- from Producer Letter Live @ TGS

[Kitru]

Because you decided that it is not possible. Because what you decide, is indeed a fact.

No, it's a fact of engineering and software engineering is no different.The basics of it, from an engineering perspective, is that you get two of the following: fast, cheap, and good. If you want it fast and good, it won't be cheap. If you want it fast and cheap, it won't be good. It's a bit simplified, since things exist on a sliding scale rather than explicit binary attributes, but that's how it works. The same principle applies to game design: you either get lots of options, high quality/balanced options, or profitable development. Since SE is a profit driven company, one of the two choices has been made already; all that remains is whether we get a lot of options or high quality options.

Also, it's not a fact just because I say it's a fact. It's a fact because I've learned it over years and so has everyone else who's actually dealt with this stuff. You may not like it because you live in some fantasy land where you can get everything you could ever possibly want for next to no effort, but I, and the development team at SE, live in the *real* world, where you have to play triage. You don't get to have everything so they played triage: they chose to provide high quality content rather than providing a whole slew of options.

Following your train of thought, WAR as choice for tank should not even exist, because it is clearly less optimal than PLD. Merely an illusion of choice.

You must not have read what I stated later on. If you had, you wouldn't have said this because I specifically addressed this.

The difference between a WAR and a PLD is not an illusion of choice. The illusion does not exist because they play *very* differently. The numbers on a piece of gear, on the other hand, are not. They do not appreciably impact the game beyond the mathematical. A WAR and a PLD will never play anywhere close to the same; gear with a lot of Determination and gear with a lot of Critical Hit Rate, on the other hand, will play effectively identical. There will be minute differences in stochastic and mean performance, but there isn't really a major difference.

Or I dunno, you could create bunch of well designed gear!

You *really* don't understand the whole "triage" thing, do you? Asking for a shitton of well designed gear means that you're taking resources away from other stuff that people want.

Numbers are everything. Literally.

Except that they're not. Numbers are the fundamental balance construct, but they do not control playstyle or theme or the massive number of other design choices that actually create differentiation. The difference between a piece of gear with 12 Determination on it and 18 Critical Hit Rate is numbers. The difference between a PLD and a WAR, not as to how well they tank, but the actual difference in how they play, perform, and everything *else* attached to them, is not. Changing numbers does not create appreciable differentiation.

Simple people don't care about choices.

And even simpler people act as if, when they make a choice, that it wasn't made for them beforehand. Choices are nice, but only when they are actual choices. Choosing between death and cake is not a choice. It may *look* like a choice, but it is not.

Choices *matter* but only when they actually *matter* and aren't effectively predetermined. If you were offered the chance to deal 5 damage every 2 seconds or 10 damage every 4 seconds, the choice doesn't really matter (which is what you're asking for). If you were offered the chance to deal 100 damage with an attack or 5 damage with an attack, the choice is effectively predetermined. Offering a plethora of differently itemized gear options doesn't create choices that actually matter. If the stats actually have an impact, the answer is predetermined. If the stats end up exactly the same, the choice doesn't matter.

Because of this, there's no real point in creating a massive slew of gearing options. The only reason you could ever get stuck on this is if you're too obsessed with the idea of making choices and deluding yourself into thinking that they actually *matter* that you don't give a shit about what is actually entailed in providing you those entirely redundant options that you're insisting be put in simply to assuage your ego so that you can feel like a fancy MMO player because you made a choice between two pieces of gear that are, for all intents and purposes, identical.

They cut that out because of the time constraints you mentioned earlier and also because it was simply less work for them. Lot of that useless crap, for example housing and vanity items, that will be steadily flowing back too, because that is what the large quiet masses want. More customization, more unique snowflake.

Except that there is a *massive* difference between player housing and vanity items. The *point* of those items is in the variety of pointless choices. You don't put plaid curtains in your player housing because they provide your house with +1 protection from sun. You put them there because they look nice. You don't choose to ride your Reaper rather than your Chocobo because the Reaper is faster. You choose it because you like riding magitek armor more than you like riding your armored battle chicken. Those choices do not impact the balance structure of the game.

Gear, on the other hand, is chosen explicitly *for* the mechanical advantages it provides. Offering identical options for gear means nothing because that choice isn't going to matter. The choice needs to have purpose. Huge amounts of gear options don't provide purposeful choices because the mechanical differences between them are either nonexistent or actually present, wherein the choice is predetermined.

If you seriously think that comparing gear options with vanity options makes *any* kind of sense, you've got no place being in this argument. It's completely nonsensible. Either you decided to make one of the most ludicrous strawman arguments possible; you're seeking refuge in audacity, hoping that going so far into left field that no one would want to follow you to continue the argument; or you seriously have no idea what you're talking about.

I've met plenty of people that think like you do. They think that they can magically have everything because they're completely oblivious. You're one of these people. You're oblivious, and, after having pointed everything out very clearly several times over only to have you to me that *I'm* a simpleton, the only possible reason that I could come up with for you to *continue* to be this oblivious is that you're willfully remaining oblivious so that you don't have to change your worldview to account for realities of real life.

[RapBreon]

And even simpler people act as if, when they make a choice, that it wasn't made for them beforehand.

No truer words were uttered on this day.

[konflikti]

No, it's a fact of engineering and software engineering is no different.The basics of it, from an engineering perspective, is that you get two of the following: fast, cheap, and good. If you want it fast and good, it won't be cheap. If you want it fast and cheap, it won't be good. It's a bit simplified, since things exist on a sliding scale rather than explicit binary attributes, but that's how it works. The same principle applies to game design: you either get lots of options, high quality/balanced options, or profitable development. Since SE is a profit driven company, one of the two choices has been made already; all that remains is whether we get a lot of options or high quality options.

Also, it's not a fact just because I say it's a fact. It's a fact because I've learned it over years and so has everyone else who's actually dealt with this stuff. You may not like it because you live in some fantasy land where you can get everything you could ever possibly want for next to no effort, but I, and the development team at SE, live in the *real* world, where you have to play triage. You don't get to have everything so they played triage: they chose to provide high quality content rather than providing a whole slew of options.

Game development at organizations as big as SE is not some zero-sum-game where if you develop dungeons you automatically lose resources from some other part of game development. Any legit game development firm will have different teams or different people focusing on different aspects of the game. Claiming that the reason for limited amount of gear choices is due to lack of resources is something you have come up by yourself and is in no way grounded on facts. It is far more likely that it is simply a design choice, instead of "reality of game development" as you state. I happen to work at one of these software developing companies and surprisingly, whenever something needs to be done with GUI of the software, QA team does not just drop their tools and walk into where the GUI team is and start fixing that.

The difference between a WAR and a PLD is not an illusion of choice. The illusion does not exist because they play *very* differently. The numbers on a piece of gear, on the other hand, are not. They do not appreciably impact the game beyond the mathematical. A WAR and a PLD will never play anywhere close to the same; gear with a lot of Determination and gear with a lot of Critical Hit Rate, on the other hand, will play effectively identical. There will be minute differences in stochastic and mean performance, but there isn't really a major difference.

But they do play effectively identically. They both hold enmity and try to stay alive. Only major difference is that WAR presses their buttons after damage, while PLD presses them before damage. There is difference in stats though, if you give them parry, both will survive for longer, if you give them crit, WAR will survive longer (well, longer than WAR would survive with no crit).

To me it is kinda confusing that you advocate difference between classes and jobs, but for some reason they need to have the same gear, despite the chance of making them more distinctive through gear too. This especially confusing because if you look at physical DDs, they have separate gear with no statistical variation, just different focuses.

Except that they're not. Numbers are the fundamental balance construct, but they do not control playstyle or theme or the massive number of other design choices that actually create differentiation. The difference between a piece of gear with 12 Determination on it and 18 Critical Hit Rate is numbers. The difference between a PLD and a WAR, not as to how well they tank, but the actual difference in how they play, perform, and everything *else* attached to them, is not. Changing numbers does not create appreciable differentiation.

And currently they require somewhat different numbers to tank as effectively as possible. PLD survival is not tied to the amount of damage they can put out, while WARs is. As long as this is the case, they also get different amounts of benefit from the same gear. You can address this either by having more options for gear or on job design level. Currently it does not happen on either.

Choices *matter* but only when they actually *matter* and aren't effectively predetermined. If you were offered the chance to deal 5 damage every 2 seconds or 10 damage every 4 seconds, the choice doesn't really matter (which is what you're asking for).

((except option one kills enemies with 5, 15, 25...hp 2 seconds faster than option two, so it is in practice a predetermined choice))

I am talking about the optimization of the class through gear, pretty much the same thing you are talking about, but for some reason it is not possible to for WAR to have this distinction in gear because of "glut" "triage" "predetermination". Choice between more passive mitigation and more damage output is not predetermined, it depends on the content of the game and should, in optimal environment, be switched back and forth for variety.

I don't understand where you are pulling this "huge slew of options" when I am talking about maybe two or three alternative items for a slot. For accessories, there is already six choices for one slot (although they are purposely made objectively bad for everyone except the right job).

Except that there is a *massive* difference between player housing and vanity items.

That comment was mostly regards to your statement of them cutting out lot of the clutter and focusing on what is important. It also shows that the lack of resources you keep bringing up is far less severe than what you claim if they have time for things that have absolutely no impact on the core gameplay.

I've met plenty of people that think like you do. They think that they can magically have everything because they're completely oblivious. You're one of these people. You're oblivious, and, after having pointed everything out very clearly several times over only to have you to me that *I'm* a simpleton, the only possible reason that I could come up with for you to *continue* to be this oblivious is that you're willfully remaining oblivious so that you don't have to change your worldview to account for realities of real life.

Likewise, you are not the first person I have ran into who is so stuck with their own notions that they simply cannot realize that actual improvements could be made. Also I would not really pay much attention to what you could come up with, because you do not seem to be a very inventive person in the first place. The reason I am still in this conversation is because I am somewhat committed on it and it is not simply a game of being right or wrong.

[konflikti]

No truer words were uttered on this day.

If you want to go all philosophical on this, it is not like I had a choice but to take part in this conversation.

[RapBreon]

If you want to go all philosophical on this, it is not like I had a choice but to take part in this conversation.

Precisely my point :P In the same way I couldn't resist but making my smug, self-copulating comment :< (apparently I can't stop myself).

[Kitru]

Game development at organizations as big as SE is not some zero-sum-game where if you develop dungeons you automatically lose resources from some other part of game development.

Actually, it is. It's zero sum because the developers are not given an unlimited amount of money to buy absolutely everyone. There are different teams, but they have to focus on how much to provide to the different teams. It's also no as if those teams are explicitly separated from each other. In game development, there is a *lot* of person swapping and moving about. Just because It is most *definitely* zero sum.

I happen to work at one of these software developing companies and surprisingly, whenever something needs to be done with GUI of the software, QA team does not just drop their tools and walk into where the GUI team is and start fixing that.

Because there's a massive difference between QA and UI dev. Those aren't triage attributes; those are static attributes. On the other hand, *design* stuff, like developing new classes, new content, new mechanics, etc. *are* drawing from the same human resource pool. If the producers want new classes, they're going to pull people off of developing new content to do so. They're not going to take people off of bug fixing, back end tweaks, or anything else like that. Those are not the zero sum resources that they play around with. The *design* attributes are. Unless you actually know about game development, which is *not* like normal software development, your experiences are only tangentially appropriate.

Of course, you also forget that, even in the example given, if there are multiple problems or programs with the UI, the UI team has to play triage with those problems. They're no going to draw from QA, but it's *still* a zero sum game because they don't have infinite resources to devote infinite resources at every problem.

But they do play effectively identically. They both hold enmity and try to stay alive.

Way to take a hyper simplistic view of things just to support your own argument. The stat usefulness of a WAR and a PLD have more in common than their playstyle does. Saying that they hold enmity and stay alive is like saying that SCH and WHM are the same because they both heal or that MNK and BLM are the same because they do damage. WAR and PLD have *dramatically* different playstyles. I realize that you don't know this because you're only level 15 on your GLD so you don't have the experience to realize this, but if you honestly think that gear differentiates the class more than the actual way they play, you *really* are an idiot.

To me it is kinda confusing that you advocate difference between classes and jobs, but for some reason they need to have the same gear, despite the chance of making them more distinctive through gear too. This especially confusing because if you look at physical DDs, they have separate gear with no statistical variation, just different focuses.

Because distinctiveness through gear is nowhere near the same as distinctiveness through playstyle. The only reason that a DRG and a MNK have different gear is because it was decided that DRG should have heavier armor: scale instead of leather, effectively. That gear difference is, honestly, a minor consideration which you seem to be massively hung up on. Hell, MNK and DRG actually share all of their jewelry so it's pretty obvious that they aren't even as numerically different as you posit.

MNK and DRG are different because they *play* differently, not because they wear different armor. The differences in the armor mean next to nothing because they actually use the same stats in the same way. The only difference is aesthetic. WAR and PLD are different because they play differently.

And currently they require somewhat different numbers to tank as effectively as possible. PLD survival is not tied to the amount of damage they can put out, while WARs is. As long as this is the case, they also get different amounts of benefit from the same gear. You can address this either by having more options for gear or on job design level. Currently it does not happen on either.

Care to actually support this with anything other than conjecture? Show us how changing WAR gear so that it has less Parry and more Determination is going to magically make them better tanks than a PLD wearing the existing gear (here's a hint: it won't; Parry contributes more to a WAR than Determination does because self healing is that weak). Your supposition is founded upon your observation that WAR gets some mitigation benefits out of offensive stats whereas PLD does not. It does not actually take into account the minute differences in actual value of those stats. You're asking for different gear but, honestly, what's the point? The difference would only matter to you because you're actively *looking* for one because you want there to be a choice for yourself.

Seriously, I challenge you put out some realistic itemization constructs for gear that provide more end gains for WAR and they do for PLD. Do some of this work that you think the devs should be doing since, apparently, it's easy enough that they should have done it automatically. Prove me wrong that

((except option one kills enemies with 5, 15, 25...hp 2 seconds faster than option two, so it is in practice a predetermined choice))

You completely missed the point of that, didn't you?

You'll see a difference between using gear *absolutely* optimized for one class over gear *absolutely* optimized for another, but the difference is so minute that it's wasted effort to create said optimization constructs unless you're simply looking to create options for options' sake (which is what you want; you don't care about it being effective; you care about being able to make a choice). Killing something 2 seconds faster means nothing without a reference point. If something has 105 hp, the difference between their kill speeds would be a whopping 5% and that's talking about massive difference in rate of damage. A more realistic set up would be 5 damage every 2 seconds or 6 damage every 2.4. At that point, that same 105 hp target would have a difference of 2.8% (at just 100 hp, the difference is 2%). Even then, an attack speed disparity like that is flipping *huge* in terms of what the game provides.

I can give you another example of a pointless choice as well: a 10% chance to deal 100 damage or a 5% chance to deal 200 damage. They're both RNG, and they both add the same amount of damage over time. The choice is an arbitrary one that would exist purely to provide you with a choice. Providing one like that in a game is pointless because it simply doesn't matter to anyone that doesn't just want to be able to make choices because it makes them feel like they matter.

I am talking about the optimization of the class through gear, pretty much the same thing you are talking about, but for some reason it is not possible to for WAR to have this distinction in gear because of "glut" "triage" "predetermination". Choice between more passive mitigation and more damage output is not predetermined, it depends on the content of the game and should, in optimal environment, be switched back and forth for variety.

Except that increasing a WARs DPS at the cost of passive mitigation does not increase its actual survivability. You seem to have decided that it does so without actually recognizing that it doesn't. If it *did*, WARs would all be wearing DPS jewelry instead of tank jewelry. As it stands, tanks get the most out of the same gear that PLDs do. The only difference is that, while the tangential damage stats that PLDs apply purely to enmity/damage, WARs add those tangential stats to a little bit of their mitigation.

I don't understand where you are pulling this "huge slew of options" when I am talking about maybe two or three alternative items for a slot. For accessories, there is already six choices for one slot (although they are purposely made objectively bad for everyone except the right job).

What you seem to not recognize is that 2-3 additional alternatives for each slot means 2-3 times as many pieces of gear. You're asking for the people who design the gear to double or triple their workload. Unless you want the differences to exist purely in the arbitrary (this one has Determination instead of Crit once again getting into the realm of "why the fuck does this matter?"), you're asking for the team to rebuild the entire itemization construct just to assuage your desire for more choice. At which point, I can assure you, that there will *still* be a single on that is still the best because that's how gear differentiation works.

In WoW, they used to create 2-3 weapon options for each class in every end game dungeon. They stopped doing it when it got to the point where everyone figured out which one was actually *best* and started ignoring everything except for that one *best* option. Gear differentiation does not lead to more choices. It simply adds more work to the developers that just gets wasted because they ended up developing redundant items that no one actually wants.

That comment was mostly regards to your statement of them cutting out lot of the clutter and focusing on what is important. It also shows that the lack of resources you keep bringing up is far less severe than what you claim if they have time for things that have absolutely no impact on the core gameplay.

Not really. There's a difference between core gameplay and wonky little side projects. The pointless crap that SE cut out isn't the pointless stuff that exists because it's pointless. The pointless crap that SE cut out is the pointless stuff that existed just because that's how it had always been done and didn't add anything of quality to the game. There isn't a plethora of weapon or gear options because they recognize that it doesn't actually add anything to the game; people will use the "best" option and cast the other to the wayside, so what's the point of developing those options if no one is going to use them?

Likewise, you are not the first person I have ran into who is so stuck with their own notions that they simply cannot realize that actual improvements could be made. Also I would not really pay much attention to what you could come up with, because you do not seem to be a very inventive person in the first place. The reason I am still in this conversation is because I am somewhat committed on it and it is not simply a game of being right or wrong.

I'm not inventive? Have you actually looked at either of the classes in my sig? There's a difference between being "inventive" and being realistic about what actually matters. You seem to be struck by the notion that more options are always better and that the developers have all of the resources they could possibly need to make all of those options a reality. Those options that you're hung up on don't matter beyond your own little realm of thought because you value choice more than quality.

Secondly, you're so stuck in your own notion that choices matter that you honestly think that they would be an improvement. Your entire argument is that more choices are better because *choices*. You haven't actually pointed out how choices in gearing would actually create appreciable differences in performance from the current state. Prove that what you say is actually true rather than just blathering on about it without having any kind of support. If you want to prove me wrong that gear differentiation would actually make WARs perform better compared to PLDs than they do now because everything is built perfectly for PLDs, show us. If you *can't* prove that the options you so dearly believe would matter or even provide evidence beyond hypersimplistic logic statements that don't bear out with reality ("WARs get mitigation benefits from offensive stats while PLDs do not; ergo, WARs should have gear with more offensive stats on it so that they can perform better than PLDs"), then shut the hell up, get your head out of your ass, and recognize that creating options just so people can make choices is not a good idea.

What you might not realize is that I like choices. It's not a choice, however, when you're choosing between 2 options that don't mean differ appreciably or choosing between 2 options where one is the obvious better choice.

Choices need to matter. If you're choosing between whether your gun is ocean grey or military grey, it's doesn't. As soon as you recognize this relatively simple idea, you'll realize that creating numerous gear options isn't something that creates choices of substance; the only thing is serves to do is play to your ego about making choices because all choices matter.

[Phreak]

Check and Mate.

[CzeslawVarpulis]

This discussion has been very enlightening for me.

[konflikti]

I'm not inventive? Have you actually looked at either of the classes in my sig? There's a difference between being "inventive" and being realistic about what actually matters. You seem to be struck by the notion that more options are always better and that the developers have all of the resources they could possibly need to make all of those options a reality. Those options that you're hung up on don't matter beyond your own little realm of thought because you value choice more than quality.

Red Mage/Mystic Knight seems somewhat complex and interesting, but Dark Knight is basically just marauder/warrior with pieces of PLD thrown in, but better.

Secondly, you're so stuck in your own notion that choices matter that you honestly think that they would be an improvement. Your entire argument is that more choices are better because *choices*. You haven't actually pointed out how choices in gearing would actually create appreciable differences in performance from the current state. Prove that what you say is actually true rather than just blathering on about it without having any kind of support. If you want to prove me wrong that gear differentiation would actually make WARs perform better compared to PLDs than they do now because everything is built perfectly for PLDs, show us. If you *can't* prove that the options you so dearly believe would matter or even provide evidence beyond hypersimplistic logic statements that don't bear out with reality ("WARs get mitigation benefits from offensive stats while PLDs do not; ergo, WARs should have gear with more offensive stats on it so that they can perform better than PLDs"), then shut the hell up, get your head out of your ass, and recognize that creating options just so people can make choices is not a good idea.

I have never once claimed that the reason that PLD performs better than WAR is due to gear. On the contrary, I have said that the big divider between WAR and PLD is the discrepancy between Shield Oath and Defiance.

I ran some pretty quick maths on this (it is 3am, I am not in my best calculating condition) and swapping parry to crit on gear would lower the average mitigation of parry from 3.2% reduction to 0.7% reduction for physical damage. This parry calculation is kinda sketchy because I was not able to get any verified data, but I assumed the baseline system is similar to crit chance, since they both have the same default value and operate in percentages. Weaponskill damage gain from this swap was roughly 7.5%. With full crit gear, you get roughly 21% crit chance (29% with Internal Release), which would allow a big IB, far more often. These numbers are based on ilvl 90 gear and are mostly based on valk-dancing-mad. EDIT: the str parry mitigations listed on vdm seem outdated, considering that I hit 23% damage reduction with my measly gear. Proper maximum physical damage reduction average from parry is then likely to be around 6.25%.

I'll see if I can muster enough effort to answer the rest of your post "tomorrow", but I feel this whole conversation has bounced so far off what I was originally trying to say it is getting kinda laborious.

[Faction]

Being needlessly long winded and wordy doesn't change the fact that you're essentially a just a random person who posts game design fan fiction (I mean really, you're proud enough to put it in your signature?? Embarrassing) as well as apparently a self styled psychologist.

Well, except in the eyes of people that see a lot of words then say "This person MUST be VERY SMART AND ALSO CORRECT."

You can't win when you're arguing with someone who's reply to any contrary thought is 5 paragraphs that all amount to "PSUEDO INTELLECTUAL PSYCHOANALYSIS OF RANDOM FORUM PERSON "