Vanguard - Brad posts about balancing the classes (HELLA LONG)

[Suineg]

I think he took the anti-SOE stuff to heart.

And just look at "How Crunk" he is so I doubt he edits his post.

Overall retarded move. Brad may not have gave us the exact design doc but do you see any other mmorpg designer giving any kind of information out like this ahead of time even if it is skimpy?

[Dis]

Originally Posted by Doc

That's exactly my feeling. This game is going to probably suck, he's getting his knob polished by some bitch pushing out his puppies and he's a happy man. He honestly doesn't give a fuck about the hardcores, I mean for anyone whose established in the industry why would they care about the minority? If V: SoH sucks donkey dicks and you attempt to shout him down at the next E3 you will be removed from the building and he will still receive great press cause he's trying to help keep the MMO industry afloat and they need everyone they can get. God the people on this board have I think unreachable expectations for his game. I think they will be disappointed no matter what comes out. Oh and any of you who test beta, it still won't release Live to your expectations and you whiney cunts will bitch. I'm betting this game will be marginally better than WoW, he will lie to the hards and play more to the casuals espcially if he's stupid enough to work with a big company near SOE"s size that only cares about the bottom line. This games gonna be ho hum.

Way to degenerate this thread to WhinePlay status....Kudos


Dis

[Ronne]

Originally Posted by Braen

In EQ2 (at least back when I was playing) you wanted a Guardian or Zerker over a monk or paladin because their abilities made them more viable. Also, a players perception also had a lot to do with it.

Play perception is an evil monster. Berserkers can't hold a candle to a good monk in the tanking department, but that doesn't matter. Berserkers wear plate, they must tank better! I think EQ2's tanking 'system' has worked out pretty decently, sure guardians tank best most of the time, but that's what they are there for, their dps is laughable. I especially like the compatability of the healing/tanking classes, a templar trying to heal a monk is really gonna have a hard time and be oom constantly, but a defiler/mystic can easily have mana to blow on nukes/debuffs with a monk tank because of the way wards and avoidance compliment each other.

[Zehn - Vhex]

Step 1) Eliminate the concept of DPS classes. It's poor design. You either end up with EQ style raids where you have 12 classes sharing 24 support slots and 48 slots left for DPS classes which, of course, you want the best of. Hence the 8 clerics, 39 rogues and 1 wizard comment.

Designers now are taking the easy way out. Instead of actually, gasp, making every class have relatively equal use in a raid, they just limit the number of people you can take. So now you have 8 classses sharing the 24 support slots and 2 classes taking up the remaining 14 dps slots. Not as bad, but a cheap copout.

Through whatever extensive testing you can manage, try to keep every class in the same ballpark for dps. Over a 30 second, 1 minute, 3 minute, 10 minute and 2 hour period, the DPS charts should all be roughly the same.

Step 2) Get rid of the main limiting factor, mana bars. IT's a fucking horrible concept. Get over it, it sucks. It worked for Final Fantasy, it should have began and ended there. Rage/energy bars are only the tip of the iceberg.

For a quick example, the "Arcane Flux" bar for mages idea. It's a scale from 0-100. As you cast spells, the bar increases, and as it incrases, you do more damage, a 0% bonus at 0 and double damage at 100%. However, past 50% you run an increasing risk of getting hit with feedback where in addition o doing your bonus damage, you also take any bonus over 50% in damage as well.

This opens up a whole new avenue of items. Rather then +mana on items, you can more exclusively customize them for classes. No chance of feedback until 60%, reduces chance of feedback, extends the bar to 120 instead of just 100 etc...whatever.

Step 3) Overlap responsibility. You -can- have classes that have almost identical abilities. One of the -biggest- problems that designers have is they try to make classes the 'best' at something. Here's a better idea. Take 12 classes, but give each class 2 fundamental roles, and most importantly, make sure they are equally effective at both roles.

IE: THe 4 basic roles are Tank, Damge Control, Crowd Control and Debuffing. Your Paladin can tank and does damage control (aka healing), your Warrior does Tanking and crowd control, your monk does tanking and debuffing...your priest does healing and crowd control, shaman does healing and tanking, druid does healing and debuffing, or whatever.

What does this accomplish? You design dungeons around 4 person groups, with the need for at least one tank, one healer, one CCer and one debuffer. Now instead of warriors being the only tank, priests being the only healer, mages being the only CCer...you have 6 tank classes to choose from, 6 helaers, etc...etc...

You allow for multi-dimensional classes and life becomes so much better. Toss in a few 'unique' non-basic role abilities to spruce up each class and voila.

I know it's alot harder then it sounds, but fuck it, they get paid alot more then me so they can come up with the rest.

[Tuco]

Step 3) Overlap responsibility. You -can- have classes that have almost identical abilities. One of the -biggest- problems that designers have is they try to make classes the 'best' at something. Here's a better idea. Take 12 classes, but give each class 2 fundamental roles, and most importantly, make sure they are equally effective at both roles.

IE: THe 4 basic roles are Tank, Damge Control, Crowd Control and Debuffing. Your Paladin can tank and does damage control (aka healing), your Warrior does Tanking and crowd control, your monk does tanking and debuffing...your priest does healing and crowd control, shaman does healing and tanking, druid does healing and debuffing, or whatever.

What does this accomplish? You design dungeons around 4 person groups, with the need for at least one tank, one healer, one CCer and one debuffer. Now instead of warriors being the only tank, priests being the only healer, mages being the only CCer...you have 6 tank classes to choose from, 6 helaers, etc...etc...

You allow for multi-dimensional classes and life becomes so much better. Toss in a few 'unique' non-basic role abilities to spruce up each class and voila.

Quoted for truth.

[kohl]

Originally Posted by Zehn_Druidbane

Step 3) Overlap responsibility. You -can- have classes that have almost identical abilities. One of the -biggest- problems that designers have is they try to make classes the 'best' at something. Here's a better idea. Take 12 classes, but give each class 2 fundamental roles, and most importantly, make sure they are equally effective at both roles.

IE: THe 4 basic roles are Tank, Damge Control, Crowd Control and Debuffing. Your Paladin can tank and does damage control (aka healing), your Warrior does Tanking and crowd control, your monk does tanking and debuffing...your priest does healing and crowd control, shaman does healing and tanking, druid does healing and debuffing, or whatever.

What does this accomplish? You design dungeons around 4 person groups, with the need for at least one tank, one healer, one CCer and one debuffer. Now instead of warriors being the only tank, priests being the only healer, mages being the only CCer...you have 6 tank classes to choose from, 6 helaers, etc...etc...

You allow for multi-dimensional classes and life becomes so much better. Toss in a few 'unique' non-basic role abilities to spruce up each class and voila.

I know it's alot harder then it sounds, but fuck it, they get paid alot more then me so they can come up with the rest.

I think there is a lot of truth to this. Maybe part of the problem is that these concepts of warrior, cleric, paladin, etc are just plain old and need some rethinking or need to be tossed out the window entirely. I realize they come from old fantasy and D&D and whatnot, but it's been done in practically every fucking fantasy game and maybe its time for something new. Like you mentioned, design classes around roles and give them new names or assign existing names to those roles if it'll be more familiar to players.

Every single time you start talking about hybrids and pure classes, balance ends up being a disaster.

[Kether Am I]

Let me state here that I have never worked for a gaming company and perhaps I am wrong but from the many current issues with released MMOGs as well as posturing posts by the developers of upcoming games there seems to be a distinct lack of one characteristic in the industry.

Science.

The science of cold, hard, brutal mathematics and system modeling. Sure, it exists in the game engine, but I see very little evidence it is employed in any thorough fashion to game design and balance.

There is certainly not only room for, but a true need for "Vision" in game design. But what I see is not Vision healthily supported by exacting internal analysis. What I see is vision supported by...more vision. Finally magically the 'vision' stops when it comes to actually coding and server management. Why? Because without some actual objective success with the code and servers the game doesn't function. You can 'vision' your way out of an infinitude of class and content design failures and inadequacies. Telling players that they should accept their client or server ineptitudes because the server 'shouldn't be judge against other servers and instead should be appreciated for the fun of its quirks' doesn't fly.

As an applied physicist I have dealt with complex systems from aerospace to mining and what absolutely appalls me is that some developers have to be drug kicking and screaming in order to recognize the most simple of mathematical realities of their own game. They are (to an almost unbelievable extent) repeatedly surprised by the results of the impact of their choices. Presumably this is because they have no one to cohesively manage the analysis of the impact of their design choices.

Just sticking to some EQ examples: elemental bows, wizard elemental pants, bard FM, the entirety of the GoD expansion, and DPS "evaluations" that apparentely are so complex that they will be complete sometime after the introduction of profitable nuclear fusion.

At the first EQ summit the designer of GoD stood up and said, "I fucked up." in some sort of display of contrition. No, he didn't "fuck up". What he SHOULD have said is, "I am a fucking idiot for not realizing the absolutely transparent ramifications that the PoS I shipped would have." There shouldn't have been promises to 'listen and do a better job'. There should have been promises to, "Hire people who know what they hell they are doing, and will tell us when our 'vision' fucks up the game because we are clearly too busy/unable to do so ourselves".

I know lots of people who like cars. They watch Nascar, they read the C&D et. all. They talk about cars. They drive them for commuting, vacationing, and just for the plain fun of doing so. If my car is having transmission problems I go to these people, right? No, of course not. I go to a certified mechanic who knows how cars work, why they don't, and how to fix them.

Game balancing is a complex affair, and requires complex analysis. No, it doesn't mean that X will be as good as Y in every situation, it means that X and Y modeled over the breadth and depth of the game give as close as possible equitable tangible and intangible rewards.

Brad's, 'we will err on the side of fun' is nice and in fact proper as far as the premise goes. It however unfortunately sounds like an admission that they lack the intent and/or ability to handle balance issues. This is doubly disturbing since his games direction appears to be one that maximizes the potential impact of any present imbalances. The truth is that I deny the premise that fun/balance are at all conflicting entities.





Kether
-Arete
-Eci(Tunare)

[Merklyn]

Originally Posted by Zehn_Druidbane

Step 3) Overlap responsibility. You -can- have classes that have almost identical abilities. One of the -biggest- problems that designers have is they try to make classes the 'best' at something. Here's a better idea. Take 12 classes, but give each class 2 fundamental roles, and most importantly, make sure they are equally effective at both roles.

More or less what Guild Wars accomplishes with the primary/secondary profession system. However, (good) healers are still in high demand/short supply.

I think every single class/role/profession/whatever should have some form of self-heal.

[Chaid]

Originally Posted by Kether Am I

Let me state here that I have never worked for a gaming company and perhaps I am wrong but from the many current issues with released MMOGs as well as posturing posts by the developers of upcoming games there seems to be a distinct lack of one characteristic in the industry.

Science.

The science of cold, hard, brutal mathematics and system modeling. Sure, it exists in the game engine, but I see very little evidence it is employed in any thorough fashion to game design and balance.

Take a look at Gemstone IV by Simutronics. You'll get your fill of mathematics in a MOG, to the Nth degree.

Look at "brutal mathematics" from a producer's standpoint. You want to avoid having a heavy math requirement for players to understand the fundamentals of the game. Too much math (like GemStone IV) is bad, as it limits your target audience. So designers need to express complex math equations into simplified mechanics like AC, to hit%, to crit%, etc. Math is definitely there, just not directly exposed to the players.

[Tuco]

Look at "brutal mathematics" from a producer's standpoint. You want to avoid having a heavy math requirement for players to understand the fundamentals of the game. Too much math (like GemStone IV) is bad, as it limits your target audience. So designers need to express complex math equations into simplified mechanics like AC, to hit%, to crit%, etc. Math is definitely there, just not directly exposed to the players.

I'm not sure and I don't like to speak for people, but I think he was saying that devs don't use mathematical modelling to design their games' balancing/gameplay etc, rather than saying games should expose their playerbase to intense math.

A game developer's design process utilizing or not utilizing mathematical processes is probably dependent on each firm/studio.

[Zehn - Vhex]

Actually, I'd much rather they handle healing more like CoH, or fuck it, let's go back to DnD on this one. Clerics main role wasn't exactly to sit there and cast cure light wounds. FUck no. A cleric was a badass mothafuckah who tore shit up, and then after the fight rez'd people who had died, cast heal and moved on.

I'd rather see the bread and butter of the 'healer' classes be in -preventing- damage. I still don't see why I have to be a masochist everfytime I play a healer. It reminds me of the one VG cats comic where Link is getting humped to death and just before the sweet release of death, that fucking fairy raises him only to be humped to death again.

Why is there essentially no 'protection from evil' type spells? Let's take the AC buffs and make them more reactive rather then pro-active. Toss a fire ward on someone to absorb fire damage, prismatic shield to reflect the next magical attack, shield of shadows that increases dodge by 50% for a few seconds. PW:S shouldn't be the expection to the rule, it should -be- the motherfucking rule.

In my time playing a cleric and now priest, my favorite part of encounters has been the "FUCK IT, BURN, EVERYONE NUKE, CLERICS, NECRO'S, JUST FUCKING NUKE" phase when the mob is at 5% and all your MT's are dead.

But this could just be me. Maybe other priests enjoy hitting flash heal at 75% over and over and over again...

[Taelnayael]

Originally Posted by Zehn_Druidbane

Step 3) Overlap responsibility. You -can- have classes that have almost identical abilities. One of the -biggest- problems that designers have is they try to make classes the 'best' at something. Here's a better idea. Take 12 classes, but give each class 2 fundamental roles, and most importantly, make sure they are equally effective at both roles.

IE: THe 4 basic roles are Tank, Damge Control, Crowd Control and Debuffing. Your Paladin can tank and does damage control (aka healing), your Warrior does Tanking and crowd control, your monk does tanking and debuffing...your priest does healing and crowd control, shaman does healing and tanking, druid does healing and debuffing, or whatever.

The only real problem is that you then become more dependent upon everyone in your group to contribute to healing, debuffing, tanking, etc. I imagine if you look across the playerbase, there are plenty of folks who love playing a healer, but the thought of actually working in debuffs etc, seems rather distasteful to them. So then what you end up with is something like in FFXI where they had their dual job system, and where someone had the potential to help out in a much lesser degree through their second job, they didn't do it for no reason other than "I made an X class, I only leveled up the other to make my main job better"

I am kind of rambling, but I will try and wrap it up. If you divide up abilities and roles amongst everyone in the group you might end up alienating the playerbase, because instead of being a class they really want to play, you give them a class that is pretty close to what they want to play with totally unfun (to them) abillities thrown in. Not only do you make the real role they want to play diminished, but then for optimum quality in a group, you are requiring them to play in a fashion they may have never wanted in the first place. Coupled with the fact that some people really like the simplicity of remembering only one role in a group "Taunt X Mobs. Don't let the casters get hit." Or "Nuke, nuke, nuke, masturbate for 5 mins, nuke nuke nuke" whatever the role is some people just don't want to play more than one role in a group.

Personally I always liked the flexibility that a system like this would give. CoH did an ok job at it, I think FFXI's dual job system was probably the best version of it. I am just trying to think of the main detractors as to why something like this would be nearly impossible to get right for people who like to keep their games simple. Granted, VSoH isn't exactly being designed for the WoW crowd so I may have been completely off base.

[Merklyn]

Originally Posted by Chaid

Take a look at Gemstone IV by Simutronics. You'll get your fill of mathematics in a MOG, to the Nth degree.

Look at "brutal mathematics" from a producer's standpoint. You want to avoid having a heavy math requirement for players to understand the fundamentals of the game. Too much math (like GemStone IV) is bad, as it limits your target audience. So designers need to express complex math equations into simplified mechanics like AC, to hit%, to crit%, etc. Math is definitely there, just not directly exposed to the players.

He wasn't saying you had to be a math major to play the game. He was saying that developers rarely truly understand the impact of their designs because all they are doing is projecting their thoughts and ideas. There is no concrete basis for those assumptions to be born out in their game. A gigantic sympton of this sort of design are frequent nerfs and situationalization (new word) of abilities.

eg. Blizzard clearly didn't understand their warrior talent tree and abilities. Frequent warrior posts were routinely ignored so Indalamar showed them. They should have been able to figure out when implementing their class that a certain min/max situation results in a huge imbalance. Instead they choose the approach to let players do the real testing and analysis.

Will such a well-modelled system be perfect? No, but it should cut down on issues. The only real arguments you have against doing this sort of thing is a) too complex and b) not enough time/resources.

[Dis]

Originally Posted by Chaid

Take a look at Gemstone IV by Simutronics. You'll get your fill of mathematics in a MOG, to the Nth degree.

Look at "brutal mathematics" from a producer's standpoint. You want to avoid having a heavy math requirement for players to understand the fundamentals of the game. Too much math (like GemStone IV) is bad, as it limits your target audience. So designers need to express complex math equations into simplified mechanics like AC, to hit%, to crit%, etc. Math is definitely there, just not directly exposed to the players.

But yet at the same time he makes valid points. Ultimately all systems regardless of subject matter fall down to numbers. In the end all the PC and NPC are mathematical models who exchange numbers between them in the form of stamina, mana, and hp whether they are giving and/or taking. This falls to items that drop off of mobs by ratio or chance of %, the DPS generated by spells/weapons/abilities, or the damage mitigation of spells/armor/abilities. The REAL problem, as I stated in a previous post, is you end up making all classes one and the same. In other words a cleric is a druid is a shaman, rogue is a ranger is a brawler, and a warrior is a paladin is a monk. If you balance to such an extreme based on pure mathematical formulas and use nothing that gives a class it's "uniqueness" from the others because the devs are too worried it will upset the balance of the game, then that would be an issue. I think this is where the devs have to "wing" it, because ultimately you could hire a billion mathematicians to help program a balance class system and players will find something that is overpowered by either exploit or some unforeseen event (not to mention just plain old perception).

My take on Brad's post is he realizes this (i think and hope) but he will be more worried about making each class distinct rather than making no one class stand out from the other all in the name of balance. In other words, he is more concerned with the fun factor of playing different classes rather than making a game that has one class for all people.

Then again he could make me stare at a book until level 35 or whatever it used to be, lets hope it is the former.


Dis


Disclaimer: I am no mathematician by any stretch of the imagination

[Taggle]

Originally Posted by Zehn_Druidbane

Step 3) Overlap responsibility. You -can- have classes that have almost identical abilities. One of the -biggest- problems that designers have is they try to make classes the 'best' at something. Here's a better idea. Take 12 classes, but give each class 2 fundamental roles, and most importantly, make sure they are equally effective at both roles.

IE: THe 4 basic roles are Tank, Damge Control, Crowd Control and Debuffing. Your Paladin can tank and does damage control (aka healing), your Warrior does Tanking and crowd control, your monk does tanking and debuffing...your priest does healing and crowd control, shaman does healing and tanking, druid does healing and debuffing, or whatever.

What does this accomplish? You design dungeons around 4 person groups, with the need for at least one tank, one healer, one CCer and one debuffer. Now instead of warriors being the only tank, priests being the only healer, mages being the only CCer...you have 6 tank classes to choose from, 6 helaers, etc...etc...

That is a pretty cool concept...basically making DPS a nonfactor by making all classes the same, and having other roles. You don't need to have 2 roles for each class to accomplish what you are (I think) trying to accomplsh. You could also make DPS all the same and make each class still have one primary role.

I do disagree with your different roles as Healing, Crowd Control, Tanking and Debuffing are same role (mitigation). Remember, your debuffer cant add DPS using your example. While it's not a bad idea, I dont think it adds much to the "fun factor" of a game. Sure, it could be balanced, but so can having DPS classes. Instead of the EQ two prong analysis (unlimited mitigation + enough DPS to cover regen), you would have simply all mitigation to balance encoutners with. Easy? Probably, but sounds pretty boring.

There are more subtle ways to make classes balanced by the numbers, but play differently.

In fact, lets use what Sigil has already stated the four roles would be (Ill use tanks as an example):

Warrior
Paladin
Dread Lord
Inquisitor

They all have the role of tanking. Make them each tank the same from a numbers standpoint, but in a completely different way.

Lets make warriors a rage based WoW default warrior. As they get hit, they build rage, which opens up skills that allows them to tank. So as the fight starts, they get beat on more, but gradually, they become a stronger tank as rage builds up (I know thats not how WoW really works, but I am just using an example). Warriors would be sweet against multiple targets because rage is building up each hit. Warriors can get aggro by a standard taunt skill that uses a little rage each time and builds up.

Paladins are the mana based tank, lets call if "favor". They use favor to stun, instant self heals, maybe to hold person an add. A paladin will be able to tank well, but will tucker out as the fight goes on, as his favor depletes. They can have AoE stuns to handle multi encounters, but it will deplete their reserves faster. Paladins can get aggro by a single spell that has a casting time, but guarantees aggro.

Over the course of what the developers determien is a normal length encounter, a Paladin and Warrior should tank equally. Over long fights, the warrior will tank slightly better, while over a short fight the paladin will tank slightly better. Mix and match long and short fights in single dugeons and you have two classes that play differently but does the same role.

Think about playing a healer for those tanks...the warrior would likely need patch heals at the beggining of the encounter, and then can settle into longer heals at the end. The Paladin may need hardly any healing for the first half of an ecounter, but then spam heals at the end.

Lets look at the other classes. The Dreadlord can tank by being "offensive" not in a DPS sort of way , but in a tanking way. Every time the Dreadlord hits a citter, he steals some of its essense. He can then use the essence to do things that helps him tank...lifetaps, debuffs, call a spirit that abosorbs so many hps (rune type self heal). SKs can get aggro by a self buff that procs taunt on each hit.

The Inquisitor could be a reactive monk style tank. Its could have great perception skills that indicate which attack a critter is preparing for, which opens up counters that mitigate damage. Players that love to mash buttons and play whack-a-mole will be all over it. Players that dont like it, can still be tanks, jsut a different style. Maybe toss in some mind control to slow down the enemy as well.

Each of the classes above would fit into the tank role, but you would have 4 of them (or 8 or 10 or whatever you want). They all have roughly the same DPS, and tanking ability, but play very different. More importantly, the other players would have to know to adjust thier style to fit the tank as well.

So a rogue would know that a warrior gains aggro slowly over time, so a slow steady approach to DPS is key. When a Paladin is a tank, the rogue will know to start off with lower DPS, until the paladin casts his aggro spell, then go balls out for 30 seconds, b/c nothing he can do in that window will draw aggro.

Anyways, just some thoughts, there are plenty of ways to "balance" the game and still make it fun. Your idea isnt bad, but I dont see what it really adds, aside from taking one of the more difficult things to balance, dps, out of the equasion and reducing it, very blatently, single balance factor...mitigation.



Obviously, folks can shoot holes in the above examples. They were with about 2 minutes of thought, but the idea is that its ok to make classes have the same role, just make them play differently. Taken to extreme, make one tank gear based meat shield and another tank class a tekken style twitch fest (whose "moves" come from bosses, raids, etc, to allow advancing).