Guild Wars 2

What can ANet do to improve the process it follows when updating a class?

ANet has started dedicating resources (over multiple patches) to improving Guild Wars 2’s less successful classes – this is a great initiative, and bodes well for the future of the game.

Unfortunately, their first foray into this space didn’t go very well (I won’t provide details – last time I did that the thread was moved to that class’s forum, which missed the point that this is about ANet’s development process not the class it was applied to).

Rather than blame the class that was being updated, it is wiser to look at the process that was followed (to design and develop those updates) and see how it could be improved for the other classes to which it will be applied. Improving this process reduces the risk of repeating the same mistakes with other classes and increase the likelihood that work on those classes will be successful.

There seem to be three key deficiencies in the way ANet went about changing the first class:

• Abstract design philosophies were given precedence over how players actually used the class.
  Impact: Changes were delivered that alienated sizeable sections of the player base for that class.
• The class forum contained the detail information on what was required to make changes effective, but this information was not used.
  Impact: Time and effort was spent delivering changes that were of insufficient value to be used in game.
• Changes were not trialled with the community (either on paper or in test environments) before they were finalised and committed to the game.
  Impact: Changes were made that a) did not work in game, b) were less beneficial to the class than the aspect they replaced, or c) had unexpected impact in other areas of the class that outweighed the intended purpose of change.

Here’s some suggestions that may help the process of developing future changes to ‘unfinished’ classes:

1. The player community are your best guide for how to improve a class – listen to them they know what they’re doing. (This applies both to selecting which aspects most need attention, and to determining what has to be achieved to make each aspect work in play.)
2. If the player community says it’s important, its important – ignore this at your peril.
3. Adapt your design philosophy to work with how players use the class (not the other way round).
4. Test changes with the player community while you still have time to modify or abandon them. (If you can’t do that until live, then accept that you won’t get everything right and will have to reverse some ill-conceived changes after release.)
5. If you don’t have a public test environment, then allocate resources to rebalancing changes after release and build this into your development process.
6. Err towards over-delivering rather than under-delivering on new buffs, and have resources available to wind them back if you’ve overdone it. (There’s a natural tendency in this environment to be over cautious when delivering buffs. An overly generous buff can be quickly reversed or wound back if required, whereas an inadequate buff is a complete waste of effort.)

What observations do you have about the process Anet is following to update classes?
How could this process be improved?

My thinking is that ANet’s eyes are firmly on SPVP. And so the majority of profession changes they do are in a step by step attempt at making multiple profession builds SPVP viable, while also toning down the impact of “bunker” builds.

by virtue of SPVP and PVE (and by extension, WVW) sharing the same skills (there are some differences, but the majority are the same) any changes done with a eye towards SPVP will impact PVE in various ways.

Problem is that what works in SPVP may not work in PVE thanks to the difference in the opposition. The AI of your average AI is as dumb as a brick. As such they are given more health (and from champion on up, a special ability to help against “control” effects) than your average character.

End result is that effects that can make or break a fight in SPVP are bordering on worthless in PVE.

^ Which isn’t a problem of updating builds for Spvp and wvw.
They should improve their PvE encounters so that what some professions can bring is actually valued.

Make it possible and viable to be something other than dps? Stop nerfing every class except warrior with most patches? Got rid of the useless hunter pets.

Balancing spvp, pve and wvw the same is a major issue but not sure if any of the balancing devs play the game or this would be obvious.

The Big Mistake that Anet made was removing Energy/Mana for all classes at Launch.
There is a main reason for some classes to have Mana.
Energy Management.

Another thing is Stealth is way OP as it made you lost your target which is a Huge Advantage than any skills available fyi.
And yes i have a Pvp Mesmer & Pvp Thief. I only run these 2 Classes in the Mist.
Cause not only i know how to play them but i got Stealth as my Ace of Spades

Yes, because if a person stealth and you can no longer see them, you should still be able to target them. Makes sense

Zenguy.6421:

1. If the player community says it’s important, its important – ignore this at your peril.

I do not agree. I have the professional experience that if the customer says it’s important, it’s probably not important at all. The problem lies somewhere else entirely but the customer feels it through something he thinks is important. Fixing the root problem rather than the symptom is far more efficient.

Furthermore, patients, I mean, customers always lie to get what they want rather than what they need. Believing clients is not always the best thing to do. Yes, I do have a Dr House attitude towards helping my customers and that helps far more than believing in blind faith whatever they say.

Sleel.8365:

Yes, because if a person stealth and you can no longer see them, you should still be able to target them. Makes sense

Balance issue, not a ‘is this realistic’ one? After all people can’t sling fireballs in real life so obviously elementalists should be removed from the game?

If they feel thieves should be balanced by allowing someone to target stealth, then it will be this way. I personally don’t think such a nerf is needed, but your argument is rather silly.

PvE Class Balance sucks balls – and gets even worse every new patch.

My opinion (not fact):

1) First of all, condition builds should be radically changed for PvE. The mechanic is broken with the current cap and any deriving balance layered out on top of it will be fundamentally flawed.

2) A distinction should be made for:
2-i) in-class balance, where all weapons, utility skills for a certain profession have equal opportunity to shine. (focus on weapons here)

2-ii) between-class balance. Can all classes provide the three role archetypes (damage, defense, party support) with equal efficiency? (this goes for both instanced or open content).

3) All elite skills should be revamped. The elite should not be tangential to a build, but the core element. Here the skill-bar changing elites (or shape-shifting) are the biggest problem. Who wouldn’t like to take a 4th utility skill instead of an elite?

4) All balance should be done separately for PvP and PvE.

5) Metrics should be used to balance skill/build efficiency, not just the subjective ‘feeling’. For example, for each class, measure how much damage you can dish to single/multiple targets per unit interval (30sec/ 60sec/ 90sec) when using the damage archetype (more than one build should be available for each class). Or measure how much damage a class can sustain before going down in the defensive archetype.

6) Between-class absolute balance assumes that they all perform the same. However, this might not be desired if the devs want to give each class a specific advantage (a class uniqueness). However, this advantage should be clearly evident and exploited.

PS: I don’t know why I wasted my time writing this. It’s because I like the game and wish to see it get better.

I think you pretty much nailed it with your three key points. It very much feels like ArenaNet are working on a separate game, to what players are playing. Quickness is a great example, as it is something players complained about since launch, but it has only just been changed now. At this rate, stealth should be fixed sometime in october..

As for solutions:

1. Have a little more faith in your community, and let us take part in solving problems with balancing. Tell us what you’ve tried, and whether or not it worked. Just because the community doesn’t work at ArenaNet, it doesn’t mean we aren’t capable of thinking up solutions. A hundred ideas are better then ten.
2. Make various pillars to balance after. Having each profession play by its own rules, makes it very hard to balance them side by side. You need a system that allows you to separately measure the damage, control, support, or survivability of each profession, according to each professions capability. So for instance, in the survivability pillar, you include everything from base armor and blocks, to boons and condition removal, that gives a profession more survivability. This would allow you to effectively measure which profession needs more or less damage, control, support, or survivability, based on how you want that profession to play.
3. Grow a little courage to take bigger steps in your development, instead of being so cautious about making a single mistake. Experiment more, anticipate less.

marnick.4305:

Zenguy.6421:

1. If the player community says it’s important, its important – ignore this at your peril.

I do not agree. I have the professional experience that if the customer says it’s important, it’s probably not important at all. The problem lies somewhere else entirely but the customer feels it through something he thinks is important.

That depends on whether you’re talking about the solution to be applied or the result that needs to be achieved.

When it comes to the result required, the customer is the best judge of that – if they say a particular result is important, then it really is likely to be important. (The exception is when developing a truly innovative product, in which case a visionary approach works best. GW2 is over 6 months old with a large and well established customer; in this domain that’s a long way past the visionary stage for its eight existing classes.)

marnick.4305:

Furthermore, patients, I mean, customers always lie to get what they want rather than what they need. Believing clients is not always the best thing to do. Yes, I do have a Dr House attitude towards helping my customers and that helps far more than believing in blind faith whatever they say.

Don’t throw the baby out with the bath water. The greedy ‘I want …" posts are pretty easy to spot and discount. It’s the more insightful posts and the collective wisdom of the player community that’s worth drawing on.

(FYI, “Doctor knows best” works well when dealing with known problems with well defined solutions, but struggles when faced with new types of problems. Just compare the medical profession’s success at controlling rates of infectious disease with its failure to control rates of ‘lifestyle’ diseases such as diabetes and heart disease.)

Jabronee.9465:

The Big Mistake that Anet made was removing Energy/Mana for all classes at Launch.
There is a main reason for some classes to have Mana.
Energy Management.

I think this is a big part of why some professions have ended up at the top of the PvE food chain, while others are boring but competitive at best, and a laughing stock at worst.

Attrition was always a big part of the wargame, and the original Gary Gygax vision of a dungeon crawl. And Guild Wars 1 knew this. Elementalists were your Vancian magicians, having a deep reserve of spells that eventually petered out. Necromancers fed off death, combatting attrition with dark sorcery. Melee fighters had adrenaline, keeping them going even when energy was expended. Assassins saved on energy by focusing their strength on vital blows.

There’s no such logistical strategy here. You have your autoattack damage, and a number of cooldowns, and generally the cooler ability, the longer the cooldown. This heavily favors classes who were given a notably high coolness:cooldown ratio on an ability, or a passive proc so good that it’s more important than what’s in your utility/elite slots.

My take is that Anet’s “babystep buffs, huge nerfs” policy with balancing is just ruining things without giving anything significant in return, leading to a meta dominated by a couple still-unfixed gimmicks and very few viable alternatives.

Also the focus on letting classes have one very strong build that vastly overshadows everything else ain’t very good.
It’s flawed when everyone has the same build in endgame sPvP.

PvE is irrelevant. Like other have said, it needs to be tuned so that something besides ‘zerkers builds and traits are useful for something, unless they want to change all the abilities between PvE and PvE/WvW so that you’re almost playing 2 different classes.

The biggest problem with the process is they care more for some classes than they do others. It’s obvious they only change engineer because they have to. None of the devs are playing them with regularity

If they were, they wouldn’t be slightly buffing turrets (most engineers hate) and nerfing everything else. They just wouldn’t.

How about this:

1. defiant needs to if not go, get drastically toned down. That way you don’t need to voice com time each and every application of CC to juggle a bigger mob.

2. make CC last longer on trash mobs so that it can actually be used to control a fight. What is the point of applying a 2 second CC on a 30+ second cooldown, when the thing can take a beating for a full minute and you can’t?

Those two would allow people to use something other than APM to determine the outcomes of PVE fights, without the need to turn every fight into a gimmick fight.

Wow! The title of this thread got stealth nerfed!
The new title is not only inaccurate, it misrepresents me and my original post.

The original title was “How to Update Classes?” (or something like that). Which is exactly what my original post was all about: discussing ways ANet could potentially improve the process they use when working on classes.

Now it’s “My Take on Balancing”, which is weird, because my original post does not touch on balancing between classes at all. In fact, I took extra care when composing the post to specifically avoid that topic.

What’s going on here?
Was there something about the original title that was unacceptable, inappropriate for this forum, or just inaccurate? If so, what?