2018-12-01

Part 4: The importance of the core experience.
How critical are multiplayer, trading, and PvP to an ARPG, anyway?

Diablo III has RPG mechanics, but no RPG systems, and it added neither collision detection-based Action mechanics, nor a well-crafted, powerful story, to compensate for the lack. That rather begs the question: what kind of game is D3, exactly? What is D3's core gameplay experience?

I suppose that it would help to define exactly what we mean by "core gameplay." In Rules of Play: Game Design Fundamentals, Katie Salen Tekinbas and Eric Zimmerman describe the concept in this way [quotes by way of Karl Kapp]:
Every game has a core mechanic. A core mechanic is the essential play activity players perform again and again in a game. Sometimes, the core mechanic of a game is a single action. [...] However, in many games, the core mechanic is a compound activity composed of a suite of actions. In a first-person-shooter game such as Quake, the core mechanic is the set of interrelated actions of moving, aiming, firing, and managing resources such as health, ammo, and armor…
Sounds simple, doesn't it? We've already touched on this topic repeatedly: por ejemplo, when we examined the differences in mechanics between RPGs and Action games. We've only looked at this in general terms, though; Salen and Zimmerman are talking about something more specific.
A game’s core mechanic contains the experiential building blocks of player interactivity. It represents the essential moment-to-moment activity of players, something that is repeated over and over throughout a game. During a game, core mechanics create patterns of behavior, which manifest as experience for players. The core mechanic is the essential nugget of game activity, the mechanism through which players make meaningful choices and arrive at a meaningful play experience.
[...]
The notion of a core mechanic is a crucial game design concept, and one frequently taken for granted in the design process. [...] Game designers don’t just create content for players, they create activities for players, patterns of actions enacted by players in the course of game play.
Let's look at an example to help illustrate this: the Massively Multiplayer Online RPG. What do players actually do, moment to moment, in an MMO, which distinguishes MMOs from other game genres?

In this case, it helps that we're dealing with a product that really does contain what it says on the tin. MMOs are designed around player interaction; in most MMOs, you really can't get very far without joining a group, and most of your time will be spent interacting with the other members of your group. Yes, you'll kill monsters, and sometimes you'll "kill" other players' characters, too, but you'll do almost none of it alone. In an MMO, your core gameplay mechanic is other people. This has a few effects on the way the game works:
1. This isn't the same level of player interactivity that one normally sees in other games. Board games, for instance, are generally multiplayer affairs, but it's a rare board game that can be played by as many as six people at a time. MMOs, by comparison, routinely feature raids and dungeons that require dozens of players to complete successfully. That's the "massive" part in "massively multiplayer" -- and it show up both the strength and the weakness of games in the genre.
2. MMOs require very large active player bases in order to be playable.  They also require that the game's publisher keeps the game servers up and running, but that merely enforces the population requirement; a game with no players needs no servers, since it's not playable (or profitable) in any case.
3. Since all of an MMO's activities are built around larger groups, the pace of play is determined entirely by other players. Players who just want to explore the game world at their own pace, or who want to engage with the story and listen to all the voice-acted dialogue in the game... well, in an MMO, that's not happening unless the rest of your group also wants to do the same things. Quin69 actually commented on this while reviewing Lost Ark.
The effect on the core gameplay is profound. In some way, the idea is to take the RPG video game back to its roots a bit, with characters having pronounced roles, strengths, and weaknesses based on their class, and groups needing a certain number of player characters (PCs) in each of the various roles if the next n-player raid is going to be a success. Your group needs tanks, who can draw attention and absorb damage, shielding damage-dealers from harm; supports keep the group's members alive, and amplify their abilities, crucial since challenges are balanced to be extremely difficult for even a group of n players to complete without excellent coordination.

Compare and contrast to an ARPG, in which a single player normally expects to be able to complete the game solo, exploring the game's world and story at their own pace, engaging with as little, or as much, of the game's contact as they care to. In an MMO, the individual experience takes second place to the needs of the group; in an ARPG, the opposite is true. For ARPGs, multiplayer is not core, and this is true of all games in the ARPG genre, including the genre-defining classic which has served as the template for all of them. Sure, ARPGs often have multiplayer modes available for those that enjoy that sort of thing, but those are not the heart of the gameplay experience; they aren't what most players will spend their time in the game doing.


Diablo II was a very popular game, and it had a multiplayer mode that some players really enjoyed, but that's more a matter of correlation than causation.

That's going to be rather a controversial concept for a lot of people; after all, the common wisdom regarding D2 is that the game was popular because of its multiplayer mode. In fact, so common is that perception of D2 that the team at Grinding Gear Games, developers of Path of Exile, had basically planned their entire game around its intended multiplayer focus. PoE was going to include a whole laundry list of elements which the GGG team had loved in D2, including ninja looting, cut-throat leagues, PvP, and so on. And then they made the game, and invited people to play the game... and discovered that their potential player base didn't care about those things at all.

Ninja looting is still in the game, but only as an option, and it's turned off by default; virtually nobody ever turns it on. The last "cut-throat league" event that GGG ran was a half-baked "battle royale" mode which they launched as an April Fool's joke; the last one I could find before that was in 2016. Their big PvP expansion push, Forsaken Masters, featured Swiss matched tournaments and Alienware gaming laptop prizes, and yet ended up being their least popular expansion to date; they've never repeated it, and have no plans to.

GGG's Chris Wilson talked about this in interviews when they were rolling out the Fall of Oriath expansion, when he revealed that the overwhelming majority of the game's players play exclusively solo, and don't trade at all... which is why they're adamantly against adding a D3-style Auction House to the game. Re-balancing the game around friction-less trade would have excluded most of PoE's  player base.

Finding evidence of this in D2 is a bit tougher, since Blizzard don't talk about things like concurrent player counts at all, leaving us reliant on second-hand reports from people who were logged into the latest D2 ladder. According to those accounts, though, it's not unusual to still see 20K or so people logged into D2 at a time, with some 19K games running. Spread 20K people over 19K games, and 90.9% of those people must be playing solo... and since solo players can only play PvE in Diablo II, its core gameplay had to be solo PvE; it can't be anything else.

That's not a matter of opinion, either; it's just the the math. And if that sample is at all representative of the bigger picture, i.e. if anyway from 80% to 90% of people still playing D2 after eighteen years are playing solo PvE, then that had to be D2's core gameplay doesn't it? It's the thing that people are still doing; it's the thing that keeps people coming back. And this simple fact has other ramifications.


ARPG and MMO cannot be "hybridized" -- adding MMO elements to an ARPG turns it into an MMO, undermining the solo PvE focus that is the core of the ARPG experience.

ARPG-MMO hybrids have, naturally, been tried, but they normally don't work. Devilian, for example, attempted to combine the two core gameplay types into a single game, but it garnered lukewarm reviews from both critics and fans, and servers were shut down three years later. Marvel Heroes was designed by no less a figure than Blizzard North co-founder David Brevik, and boasted the Marvel Comics license to print money when it launched in 2015 to plenty of media attention, but it struggled to sustain a player base, and quickly cycled through several different expansions/versions before Disney/Marvel pulled the plus: servers were taken off-line in 2017.

The upcoming Lost Ark was touted as a "Diablo killer" when it's YouTube trailer launched a few years ago, but it's also more MMO than ARPG, and not likely something which will appeal to ARPG fans. Don't take my word for that, though; take Quin69's word instead, a noted Twitch streamer who's built a following streaming both Diablo III and World of WarCraft. This is someone who loves both ARPGs and MMOs, and who appreciates the difference between them, and who bluntly says during his review that Lost Ark will not compete with Path of Exile because they're basically different game genres (he really likes Lost Ark, by the way).

That's not to say that multiplayer game modes can't, or shouldn't, be added to ARPGs, of course. ARPG multiplayer fans tend to really like playing these games with other players, and also tend to be very vocal about it, which can lead to a lot of extremely valuable word of mouth for your game. This is why the common wisdom on Diablo II had over-estimated the importance of multiplayer modes and trading to ARPG format; D2 multiplayer fans had already been talking up multiplayer modes as D2's best feature for over a decade when Diablo III was starting development.

The same thing applies to trading. Few activities are as quintessentially human as trade; while interspecific cooperation is a thing in any number of other species, only our species actually trades items purely for their aesthetic or social value, and we've been doing it for hundreds of thousands of years. Put enough people and enough things in one place together, and you get a market, whether one is planned or not, so in a game with as many items as an ARPG, adding a multiplayer mode almost guarantees that some form of trading will emerge.

However, while trading in ARPGs is almost inevitable as long as they also have multiplayer modes, trading is not the core of ARPG gameplay. The core gameplay of an ARPG involves killing monsters, and being rewarded with loot; grinding for useless crap, which you then sell to obtain currency, which you then use to buy the loot you really need... well, obviously there are people who just love to buy and sell, and for whom that activity loop is actually more fun than the game itself, but most players would rather play the game. Even Blizzard themselves finally recognized this reality, and removed the Auction Houses from Diablo III, specifically because it undermined D3's core gameplay; Jay Wilson, who had overseen the creation of D3's Auction Houses, ultimately admitted that removing them was a good move.

Which brings us to Diablo III, which Blizzard claimed was "built from the ground up for online multiplayer." Remembering the concept of the core gameplay experience, and asking which elements of D3 are actually designed around interactions with other players, it becomes very difficult to see which elements of the game, if any, were actually designed with multiplayer in mind, at least for the game's PC version.

D3's character class abilities don't interact or interlock in any meaningful way; the rare occasions when players found multiplayer synergies in the game mostly involved playing groups composed entire of one class, and were promptly nerfed by Blizzard. Tellingly, Wyatt Cheng, in the lead-up to multiplayer-focused patch 1.08, described the game's existing multiplayer modes as feeling mostly like a waste of time. Even post-1.08, standard features of multiplayer-focused games, like guilds, group management, and match-making, were entirely absent: clans and communities weren't added to the game until patch 2.0.1, two years after the game's launch, and match-making was only added to the game this month, with patch 2.6.4.

Again, I'm not saying that multiplayer, trading, and PvP modes aren't valuable, or that ARPGs can't or shouldn't include them. These features may only appeal to a vocal minority of the player base, but that isn't a problem, and the very vocal nature of multiplayer gaming communities has obvious word-of-mouth marketing value. Well-rounded games can include all sorts of mini-games and activities that aren't necessarily core, and adding those features can sometimes lead to some remarkable and even revolutionary developments that shake up the entire video game industry: just look at Fortnite, whose Battle Royale game mode was absolutely not the core of that game as designed, or Gwent, which started as a mini-game in The Witcher 3: Wild Hunt before being developed into a stand-alone collectable card game with a growing player base. Adding multiplayer modes, or trading, or PvP to your game can have significant benefits.

There's only one real caveat, here:


If you're adding multiplayer modes to your ARPG, then they absolutely must not interfere with the core gameplay.

The game can be architected around them; coding the game's mechanics to accommodate their existence can make it easier to add multiplayer functionality later, and I'd argue is probably good practice, just generally. But the gameplay cannot be designed around them; at least, not without undermining the ARPG gameplay experience that fans of ARPGs, specifically, are going to be looking for.

The Diablo III development team either didn't understand, or just didn't care about, the core gameplay experience of the Diablo series. They must have ignored the data from their own game servers, which would have clearly shown them that most Diablo II players were only playing solo PvE; and they must have ignored their own D2 sales data, which would have clearly shown just how many people bought the game, but never bothered to play it online at all. Instead, they turned to the bane of product development everywhere: focus groups.

Jim Sterling has a pair of videos on the subject of focus-group-driven game development, Perfect Pasta Sauce and Damn Fine Coffee, which go into the subject of why this is terrible idea in more detail. The basic thrust of both videos, though, is simple: while focus groups may be an effective marketing tool, helping companies gauge the effectiveness of various means of selling an already completed product, they're a terrible design tool, because a) consumers don't always know what they want until they see it, and b) won't always tell you what they really want in front of a group, if what they want isn't what the group seems to think is worthwhile.

D3 began development at the height of the MMO era, when online-only, massively multiplayer, and massively profitable games were in vogue, and the players who formed Blizzard's pre-D3 focus group all agreed that they'd very much like to be in vogue. Armed with this focus group result, Blizzard went on to build a game that sacrificed solo PvE gameplay and strong RPG systems for an online-only game that focused entirely on its potentially profitable trading marketplace, but which included none of the features that are bread and butter for MMOs, and which sacrificed most of the core gameplay that ARPG fans were looking for... but hadn't told Blizzard they were looking for. A certain amount of selection bias, which saw Blizzard pulling their focus group from their active online player community (who would naturally skew strongly in the online-only/multiplayer-only direction) only made things worse.

Now, most of what I've written so far has been exclusively about ARPGs on PC, and naturally aren't as neatly applicable to D3 on the console. While many of D3's mechanical and core gameplay inadequacies seem to be a result of changes made to accommodate the process of porting the game to consoles, there's no doubt that the PC and console versions of the game ended up being very different animals. The PC version was built around its Auction Houses and the online-only DRM that was needed to protect them, but the console versions weren't going to get the AHs, and so could be designed without the online-only/multiplayer-only focus in mind. The result is a game which made for a much better single-player experience right out of the box, and which included a couch co-op/local multiplayer mode which was a clear highlight. It even include some online-only multiplayer features, like Nemesis Kills, which the PC version of the game still has not received.

All well and good, you might say, and you'd be correct.... except that the core gameplay experience of the entire Diablo series to this point wasn't couch co-op on console, but rather deep RPG progression systems on PC, and the console versions of the game simply haven't garnered anything like the following of the PC games of the series. Data is, naturally, scarce, presumably since the numbers are bad and Blizzard doesn't want to talk about them, but all the available numbers released so far would allow for console sales of D3 to be less than a tenth of the PC versions'... in spite of the console versions being arguably the best available versions of the game.


How could Blizzard have done this better?

You'll notice that I'm not leading with the rhetorical question anymore; there's no question that this could have been done better. The only question is, what could or should they have done instead?

Well, for starters, they could have designed a game which stayed true to the core gameplay experience of the Diablo series: a solo PvE RPG experience, with plenty of depth and complexity, which then added multiplayer functionality on top like gravy. Since porting the game to console was apparently a big part of their plans for the franchise, a truly cross-platform design, which saw all versions of the game getting all the same features and functionality, and which saw all versions of the game being released "day & date" (i.e. at the same time) should also have been a priority, even if it meant pushing back the game's release date. This is Blizzard, after all, and while Diablo fans wouldn't have been happy about a delay, they would have accepted it, as long as a) the final product was worth the wait, and b) they had the option of buying the game on their platform of choice, right out of the gate.

Multiplayer functionality, like clans, communities, match-making, Nemesis Kills, and the promised team deathmatch arena PvP mode should all have been present in the game from launch, on all of the game's planned platforms. And again, the game should have been delayed until this was possible; the deathmatch arena PvP, in particular, should have been in the game from the outset, not because it was in any sense core to the gameplay (it isn't), but because it was central to the game's marketing.

And then there's the big Kahuna: the Auction House system, which I think actually could have worked, if it had been implemented correctly:
  1. It needed to be possible to find decent items in-game. The console versions launched with a smart-drop system which was later adapted into the 2.0 PC version of the game, and which increased the chances of useful items dropping. This is essential; having the AH as an option is fine, but it can't be the only way to find useful gear for your character.
  2. It needed to be harder to speculate on the AH. A simple way to accomplish this would be account-binding any purchased or crafted item, whether it had been bought from the in-game vendors, crafted by the in-game artisans, or purchased from the AH itself. This would prevent AH "flipping," i.e. playing the AH instead of playing the game, and return the focus to the game, by removing items from the AH item pool permanently.
The AH should not have been the focus of Blizzard's "long tail" monetization plans for the game. Building a robust game that kept people playing for years (or decades, in the case of Diablo II) can allow for long tail sales success in the form of a customer pool who will buy additional game content, but Blizzard got greedy with D3: they wanted all the benefits of "long tail" monetization without the need to create any more content for the game. That simply doesn't work; free-to-play games succeed by constantly adding content, not by allowing players to sell existing game content back-and-forth to each other, with the developer taking a cut of the proceeds. Path of Exile has released seven major expansions to date, with the eighth, Betrayal, coming in less than a week (and combining the gameplay from the last three challenge leagues: Bestiary, Incursion, and Delve, each of which included enough content to make a decent small game).

Unfortunately, the naked cash grab that was Diablo III's Auction House has pretty thoroughly poisoned the well for this sort of thing; Path of Exile's developers are resolutely refusing to even consider adding one, and Blizzard were so badly burned by the experience that they not only removed D3's AHs, they removed trading from the game entirely. And that's a real shame; not because trading it core to the Diablo experience (it isn't), but because there were players who really enjoy trading in their games, and who bought D3 specifically because it included a robust trading system.


And now, a quick word about ludonarrative dissonance.

We've talked a lot about the core gameplay experience, which may be slightly confusing to people who've heard various Blizzard developers talk about the game's "core fantasy." It's important to note that these terms, while ludonarratively related, are not synonymous.

I've mentioned the Diablo series' ludonarrative dissonance before. Long story short, the narrative of the first two games of the series, and of the Reaper of Souls expansion (i.e. Act V), has been one of horror-fantasy, in which evil forces are temporarily beaten back, but only at enormous cost, and only temporarily. Diablo [SPOLIERS] ends with the player beating Diablo but not destroying him, and ultimately sacrificing himself (or herself) to contain Diablo's evil. Diablo II ends with the evils once more defeated, but not destroyed, and only at the cost of the Worldstone's destruction. And so on.

But the gameplay of the Diablo series, which sees the players' characters grow into a one-person monster-slaying machines capable of killing the devil himself, is at odds with this narrative. Action mechanics, which rely on collision detection and on the player's own learned skills rather than on RNG-based RPG systems, might actually be better suited to the themes of the Diablo series, allowing the challenges faced by the player to be constantly balanced in a way that makes beating them extremely challenging: think Cuphead, or the SoulsBorne games.

Reaper of Souls actually suffers the least from ludonarrative dissonance, ending with an unknown (but presumably high) percentage of the world's humans dead by Malthael's hand, and the Nephalem (the player, and presumably others awakened by the apocalyptic events of the game) possibly having grown into a new, and even deadlier, threat. This is entirely consistent with the gameplay, which (naturally) sees the player's character grow into a one-person monster-slaying machine capable of killing Death himself.

So, what is the "core fantasy" of an ARPG? Well, my friends, that answer.... will be the subject of my next installment.

No comments:

Post a Comment