Tuesday, July 21, 2020

Always The Same

Several bloggers I follow have been weighing in with opinions and speculations based around information coming out of World of Warcraft's Shadowlands beta. Enthusiasm seems lukewarm at best.

The last expansion, Battle for Azeroth doesn't seem to be held in much affection. The only blogger I can think of who seems genuinely fond of it is Syp, who writes "I really, really enjoyed the questing and leveling through the Alliance zones". Other than that, most people seem to have had more than enough of BfA by now.

It's hardly surprising, given how long its been hanging around. As Wilhelm points out, with an official launch window of "Fall 2020", even should Shadowlands hit the front of its time-slot, this will be the longest ever gap between WoW expansions. That might make people really keen to buy a new box or it could just make them question whether the time has come at last to climb down off the merry-go-round for good.

One big caveat appears to be something known as "Covenants". Even after reading quite a bit about them I only have the vaguest idea what they are but I do know people don't seem very keen.

Flipping through the overview of the concept at WoWHead it reminds me of Orders in Guild Wars 2. That's a concept that once seemed crucial but now rarely merits a mention. If you followed your class story back at launch, as most people did, about half way through you had to make a choice about which of three "Orders" to join. It was painted as a big deal at the time and people did get a little angsty about it but like most things in GW2 it turned out to be mostly cosmetic.

Secret hideout of the Order of Whispers. Knock twice and ask for Charlie.
Shadowland's Covenants look to be a much more substantive gameplay decision. Marathal, in the post I linked earlier, suggests any imagined choice will burn away in the fires of pragmatism: "...if you are interested in playing any end game content, you will be forced to choose from one, maybe two options. The bonuses available will be so skewed that you will be forced to take the one that gives you the best bonuses so that you can clear content more easily". Kaylriene, discussing the issue at length, hopes that a meaningful beta process will allow Blizzard to tune the system until there's "...a real choice that allows you to pick the experience you want rather than picking the mathematically best spell and being stuck with all the dressing around it".

In a follow-up post he goes on to examine the wider issue of class balance in MMORPGs and the way players work to game any percieved deviations from the norm. Talking specifically about WoW, he says "World of Warcraft has, undoubtedly, nearly always had balancing problems with the viability of talents, abilities, and whole classes. One of the recurring themes of every expansion for over a decade now has been a discussion over which classes and specs are meta, which talent choices are absolutely necessary for top performance, and things like best in slot lists and group composition guides." You could almost certainly say the same for any remotely popular or successful MMORPG during the same period.

Comparing WoW with  Final Fantasy XIV, Kaylriene points up the very significant difference in design approach applied by the two development teams. As the parse charts show, FFXIV can make a convincing claim to have arrived closer to a point of equilibrium between classes than most games across the genre.

Keep ladling it on, guys.


And yet that difference still exists. They haven't eliminated it entirely, it appears. So, here's my question to anyone currently playing FFXIV: by reducing the parse gap to 10%, have the developers successfully avoided the issues of elitism that plague almost every other MMORPG? From comments I've seen in a number of blogs over the last few years, I suspect not.

My feeling is that in any MMOPRPG that becomes successful enough for there to be a surplus of players relative to the content, the perceived importance of any relative differential in efficiency will constantly adjust to match whatever difference can be measured. If there's only 10% between top and bottom then every percentage point will matter. If the developers could balance so brilliantly that there was only 1% difference, then every 0.1% would matter.

So long as people are able assign numbers they'll use them to make whatever points they want to make, be that about classes or the individuals playing them, regardless of the objective relevance of the numbers. In most cases, whether the content in question can be completed with different classes in various roles won't be at issue. Even if it demonstrably can, the question will be whether it should. There will always be a "right" way to do everything. For some, any other way is heresy. It must be shunned by all right-thinking players.

As Wilhelm rightly observed in a comment on Kaylirene's post, all of this is primarilay an issue of information flow: "With WoW Classic we have a solved problem issue, the way we have had on the EQ retro servers. People are able to take advantage of all of the meta gear and class discussions of the past that figured out the best path forward. There is no discussion as to what the path should be, it is just a matter of digging up what path was decided on back in 2006 or so and running with that".

Wisdom of crowds, is that it?
When games are new, no-one really knows anything. The goal is to complete content any which way you can. Whatever combination of classes, using whatever mish-mash of talents or builds, if it gets the job done, that's all that matters.

Then, as time passes, if the game is even moderately successful, two things happen. First, players learn what works and what doesn't through trial and error. Second, someone inevitably comes up with a means of measuring player efficiency. Numerically. The dreaded parse has arrived.

Once people know both what should work in principle and what is (or isn't) working in practice, they can start taking names and kicking ass. Kicking your ass out of their group for not measuring up, that is.

Anyone who plans on playing nicely with others is going to have to make some kind of peace with those who don't want to play nicely at all. Even those who plan on soloing, either from natural inclination or as a way of exerting control, may end up having to contend with DPS checks margined against an optimized meta they don't follow and never signed up for.

As has been discussed many times, balance is a never-ending, ever-failing sisyphean task. It's been with us for what seems like forever. I remember talking about the folly of "cutting the legs off the table" to try and achieve balance as far back as the original EverQuest forums in the early 2000s. Put simply, true balance is never going to happen.

Except, could it? Let's try a thought experiment. What if every class - or at least every archetype - Healer, Tank, DPS, for argument's sake -  was exactly the same? 

Role model.
What if there was no choice for players to make, none at all, other than which role they wanted to perform? Neither at character creation nor ever after? What if  all skills, attacks, talents, weapons and gear were held in common? What if every Warrior was always identical in all abilities and stats to every Paladin? Every Rogue to every Wizard?

Well, for a start, with no intrinsic difference between character classes, any difference in efficiency would unequivocally come down to differences between the players, not the characters. It might not entirely be explainable by personal skill and ability - things like hardware and latency would be factors (and, for sure, excuses) - but in most scenarios the expectation would be that if the player wasn't hitting their benchmarks the only explanation would be incompetence or inabilty.

It probably wouldn't be a panacea for elitism and social maladjustment, then. I suppose it might just push some of that to the margins. Bullies and jerks are always going to find a way to be bullies and jerks but I guess it could conceivably put paid to the arguably more difficult social issue of someone helpfully sending you tells explaining how to optimize your build while they somehow manage to imply criticism of your life choices.

But it would also be a bit boring, wouldn't it? Everyone being the same? Why even have classes if they all do exactly the same thing?

For flavor, that's what. As progression and rpg mechanics you could replace statistical advantage with cosmetic advantage. Plenty of games are half-way there already, GW2 more like two-thirds. It's not for nothing the game's nickname is Fashion Wars.

No-one tells me how to dress. Apparently.
The problem with that idea isn't that people don't find dressing up sufficiently motivating. The evidence is that they very scarily do. It's that placing high value on what your character looks like actually risks making that the meta. If it feels bad enough being told you don't know how to play, imagine being told you don't know how to dress.

Maybe none of this goes far enough. If we're looking for a truly radical solution, why not set a time-to-kill on every mob then adjust it proportionally to the number of people on its hate list so that TTK  never varies, regardless of who's there and what they're doing.

It's all notional, anyway, isn't it? In almost every encounter between players and mobs in almost every MMORPG the outcome is never in doubt, only how long it's going to take. It would certainly be easier to manage your game session if you knew, to the second, precisely how long every kill was going to take.

If all of this seems unlikely, even insane, it's no more ridiculous than believing we can have multiple  classes making genuinely meaningful choices that significantly impact gameplay and yet expect that there won't be winners and losers depending on what choices are made.

Balance is a lie. Unless Covenants turn out to be true magic, after all.

11 comments:

  1. Your post seems to hit this issue more squarely than others, in that the thing being balanced is more important than the act of balancing.

    Where Blizz and SE differ is in the counterweight. SE appears to balance the 90% mark of optimal play. Blizz instead balances, from the getgo, for 105% optimal play, then nerfs as time goes on to eventually fall into the 90% bucket.

    To use a literal balance as an example, over time the player side gets better numbers (and knowledge), thereby increasing it's weight. The dev side of the picture removes weight, typically with nerfs to the content. They sometimes add weight to the player side, with a zone-wide buff. This isn't new. Blizz balanced most of TBC raids under the concept that everyone had Bloodlust and drums, which was frankly ridiculous.

    The meta itself only exists for that time where the perception of achieving balance is difficult. As long as Blizz' design focus is on content for the most elite, and balanced for the most elite... this conversation will come back again and again.

    ReplyDelete
    Replies
    1. Soe/Daybreak have always used the Blizzard version of tuning for difficulty - probably before Blizz did, I guess. In recent years in EQ2II it's become quite controversial, with solo content in expansions being pretty tough at launch and group content completely inaccessible without following a very defined progression path.

      It's rough only in the short term, though. It works pretty well across the lifetime of the expansion, which in EQII's case is usually exactly one year. At the mid-point, where we are now, as a fairly diligent soloist, I'm overpowered in all solo content on my main character and fit to start group content if I wanted. I personally thing that a progression arc that takes the full lifetime of the expansion into account is good design, but it often feels like a minority opinion. A lot of people seem to get quite miffed if they can't do everything they want to in the first week or two. Of course, they'd be the same people complaining bitterly that they were bored if they could...

      Delete
  2. Quite a dress, Bhaggy.

    As for this:

    What if there was no choice for players to make, none at all, other than which role they wanted to perform? Neither at character creation nor ever after? What if all skills, attacks, talents, weapons and gear were held in common? What if every Warrior was always identical in all abilities and stats to every Paladin? Every Rogue to every Wizard?

    That would singlehandedly blow up the entire concept of raiding. After all, boss fights highlight different characteristics within a diverse group --such as Boss X dropping a debuff that can only be removed by classes Y and Z-- and such differences make the concept of different player abilities important. If every class had the same toolset to handle all of the issues surrounding a boss fight, then why have the different boss mechanics at all? We then differentiate based on whether a player can hit a button fast enough, which is still a gatekeeping style, but instead of it keeping some players out of joining a few elite guilds the differentiation is then more widely disseminated among an MMO. I fear that would become a gatekeeping hurdle that would lock out older players from joining some guilds, as the older you get, the longer your reaction time is.

    ReplyDelete
    Replies
    1. Yeah, I started out making that suggestion semi-seriously but even as I was writing I could see how it would probably create at least as many problems as it solved, which is why I ended up giving the post a slight satirical feel.

      I would, in theory, quite like to see it done. I had a couple of paragraphs in there that I took out about how GW2 did begin with something not that disimilar, with the removal of the Trinity, having all classes able to heal, rez, tank and dps, supposedly roughly with equal facility, and with aggro mechanics almost completely impossible for players to understand. The problem was, a lot of players really hated it. Not all - some loved it - but it was clearly hampering the commercial aspect of the game and over the years we've moved to a soft trinity and more defined class roles. I kind of wish they'd gone the other way, just to see what would have happened.

      Delete
  3. Functionally retail WoW and many modern MMOs, feel a lot like the hypothetical game you describe to me. For example, assuming that you are in a DPS spec, the damage that different classes can put out is very similar. When you pick a class you are deciding whether you apply most of that damage within melee range or further away from your target, and how difficult the rotations are to get the hang of. You are also correct on how player perceived this state of affairs. They look at 5% differences in the DPS outputs of different specs in parses and use that to argue that certain classes and specs are completely OP or utterly gimped on forums.

    When you look at modern games through the lens even something like launch era WoW (much less EQ or DAoC), the relative parity of classes seems odd. However it is a natural side effect of the pursuit of balance. To be fair there are good things about it. No-one likes falling into a new player trap by picking the wrong class on the character creation screen. It really sucks to put 100 hours into a character and then find out you will never be particularly good at the class role you have chosen.

    However, it doesn't seem like we have ever had much of a conversation about whether this level of balance is a good thing or a bad thing. We just sort of ended up here as developers went out of their way to eliminate new player traps from the character creation screen.

    ReplyDelete
    Replies
    1. Those are some very good points. So much in MMO design seems to go back to developers trying to reduce calls on customer service and/or increase retention metrics. I get the feeling sometimes its as if they're trying to put out a forest fire by dealing with one burning tree at a time.

      Although I'd be curious to see a genuinely flat design in action, with a lot of classes that look and feel different to play but have exactly the same damage output or healing power, what I really like are wildly differing classes, as we had in EQ/DAOC and the games of that era. I made plenty of characters I never played very far up the levels because they turned out to be different than I imagined or because I couldn't get the hang of how to play them. That was all part of the attraction, though.

      Curiously, GW2 is still very much like that today. People are continually asking for advice on which class to play in general chat and the answers they get are wildly contradictory. I'm not convinced the classes are that different in practice but peoples' perception is that they vary hugely. I guess that's all that matters when it copmes to having some variety.

      Delete
  4. Yeah, that dress is… a thing.

    The only reasonable way to get balance in a complex interacting system like an MMO is by dynamic feedback. You need to build in mechanisms that *automatically* nerf the meta and buff out-of-meta options. That's hard to do in a "natural" way, so game designers lean on providing manual balancing, but it's pretty clear that in the long haul that leads to serious shifting instability.

    I've never played EQ or EQ2, but from yours and @Wilhelm's descriptions it sounds like they have come closer to dynamic balance than most. One natural mechanism for achieving dynamic balance is markets, interpreted broadly: not just "the best items become more expensive = worse", but "the best events become queued up for hours = worse". Instancing may be a mistake: making sure there's enough of some desirable resource for everybody creates a positive feedback loop that ends only when everybody is entirely glutted with said desirable resource.

    As long as we're imagining MMO utopiae, let's imagine one where the stats of an item, class, etc have a component that is inversely proportional to the number of people playing it. It's just literally built into everything: if you pick the meta sword, that sword gets that much weaker for everyone. Now there's a bunch of interesting details here that really matter. Does "number of people playing" mean over the entire userbase or active userbase or people playing right now? How strong should this balance effect be? Does it apply to things like levels — that is, should level bonuses be inversely proportional to the number of chars at that level? Etc, etc. The biggest problem, I think, is that any air of "realism" just goes out the window: this isn't how the real world mostly works at the mechanical level. Is there a way to capture such effects with real-world metaphors? Fashion is one great example: if everybody is wearing the fashionable style, nobody is. Commercial marketplaces, with their supply/demand curves, are another. But when it comes to gear are there better tradeoffs than price/quality? What kind of tradeoffs make sense for magic? For skillpoints?

    Anyway, as somebody who has built some homeostasis into small games this way, I'm a big fan of the approach. Let the damn players balance the game themselves: they're the ones who are always whining about it. Utopia.

    ReplyDelete
    Replies
    1. I love the fashion analogy. If you removed the functionality and made everything about looks then yes, following the "meta" would be the very definition of being uncool. On the other hand, that is how fashion trends work in the real world: trend-setters move ahead of the pack but, again by definition, trend-followers stay on meta. If they don't there isn't a trend.

      What might work would be for gear and abilities to have seasons. They're rare and difficult to acquire when they're introduced, they increase in power as more people acquire them, but they also become easier to acquire. At a designated point, when a certain number have been acquired, that power begins to decay until it falls to an eventual end-point, by which time a new season has just begun.

      That kind of happens already with expansion cycles and updates and balance changes between expansions but it could be baked into the mechanics at a more fundemental level. Of course, players would just figure out the cycles and the values and game it anyway and we'd be back where we started.

      I suspect the only way to break the meta completely would involve a lot of randomization, which is something players seem to hate a lot, so that's not going to be very commercial. Personally, I can't get enough of it but they aren't making these games to suit my tastes any more...

      Delete
  5. I wonder if something even more radical is required. I only just had this idea so it might fall apart with more thinking, but what if character abilities all the same potential maximum damage, but achieving that damage required skill in a kind of minigame? Like maybe you have to draw a rune with your mouse really quick, and the more accurately you do it, the closer to the spell’s maximum damage you do. Or perhaps solving maths puzzles, or anagrams, or matching light patterns, all sorts of possibilities for testing different types of skills, even a ‘playing the piano’ type game that would basically be the current MMO combat style. You would pick a class based not just on thematic preferences, but on what kinds of things you’re good at

    ReplyDelete
    Replies
    1. I vaguely remember a game where you had to draw runes with your mouse. It might even have been an MMO. And Puzzle Pirates, which is definitely an MMO, does indeed have puzzles as part of the gameplay.

      Mini-games tend to be quite divisive so that could be a negative but there's certainly a movement building for some kind of move away from violence-based gameplay, so maybe this would be the moment. Although if the end result is still the classic mass-murder rampage that's been the go-to for almost every MMO ever, then I guess changing the mechanics of how the slaughter happens isn't really the paradigm change the culture's supposedly yearning for.

      Delete
    2. I wasn’t aware there was any kind of movement away from ‘solve problems with murder’, but if that’s the case I'm certainly all for it – I kinda get why it’s been the default for so long, but it’s about time more diverse options were explored.

      Delete

Wider Two Column Modification courtesy of The Blogger Guide