Reaping the Benefits of Procedural Content

When Chris Winters tells me I need to play something, I pay attention.  Now, you may not know Chris; he’s an excellent art director,but, he’s also an avid, passionate gamer, and when he really likes something, that’s worth looking into.  The topic of Chris’ latest recommendation was to go back into Diablo 3 after the 2.0 patch and expansion release, and it was definitely worth it.

I’m not nearly as much of a gaming fanboy as I used to be.  I don’t generally pay attention to release dates, for example, because my backlog is so large that getting a new game just puts it into a long queue behind other excellent games.  The Diablo franchise, though, has a special place in my heart, so my wife and I both got copies of D3 the week it was released, and I dove into it wholeheartedly.  Like most Blizzard products, this was a fairly safe bet, and we got our money’s worth, but the endgame just couldn’t hold me.  I think I was on the third difficulty setting, and the grind was both long and painful, and gradually, I just stopped firing up the game.

But, Chris is persuasive, so I picked up the expansion, installed it, and have now been playing it for about a month.  Like previous expansions, it added a new act and a new class; what really grabbed me, though, was the endgame overhaul.  Obviously, there were already a lot of great materials to work with – excellent art, a world rich with lore, diverse and complex character options – and the expansion pack team made the most of this through going back to one of the core gameplay philosophies of the franchise: procedural variation.

Content is expensive.  For designers and producers alike, getting the most out of a limited budget is a continual problem.  The Diablo franchise has always gotten around this by playing mix-and-match in various forms.  The equipment system, for example, generates random collections of attributes (and random stats for those attributes) by mixing packages tied to the affix system with additional modifiers.  So, you could get hundreds of versions of “Ascended Broadsword of Madness”, and they would all look the same on the character, but they could be of various levels with various stat packages and differing usefulness to different character builds.  Similarly, levels are built out of a palette of blocks and set-pieces which have a consistent visual style; after a while, you start to recognize the pieces, but the combination is never fully predictable.  This not only provides a lot of play time for the player, it also sustains that feeling of exploration and possibility that are so critical to RPG’s.

All of this was present in D3, but Reaper of Souls took this even farther.  In the Nephalem Rifts, the art sets are not consistent; you can segue from a hell-scape to a woodland by going through a portal; in fact, in some maps, they transition art sets while in one space (which would be crazy expensive to do with downloads but works because all the art assets are locally stashed on the PC).  Yes, you’re seeing the same pieces and fighting the same monsters, but the context is continually shifting, so it feels less repetitive.

The rifts are powerful in a second way, as well.  Because they have better drop rates for top-end equipment, it incentivizes players to earn the tokens necessary to go into the rifts.  Earning the tokens comes from going out into the gameworld and doing pieces of the existing campaign, and, of course, you’re earning XP, gold, and loot drops of equipment while doing that.  This indirection is a key tool for making a grind seem less grind-y.  Zynga used this to great effect in games like Farmville and Castleville; instead of telling the player to gather 256 mcguffins, they would give you a quest to combine 2 of X with 3 of Y, but of course each X requires 4 A and each Y requires 3 B and 2 C, and before you know it, you’re gathering 256 things, but it doesn’t feel that way; World of Warcraft uses similar structures, but largely by manipulating drop rates, and going in a different direction, given the crossover of WoW and D3 players, was probably a good thing.  By dividing the Adventure mode into relatively short “bounty” missions that lead up to a larger “rift” experience, the process of grinding the old content is no longer an end in itself.

Make no mistake: the endgame is a grind.  It always has been.  It is, after all, for that hardest of hardcore audiences that wants to continue to play the same game for almost endless hours.  You can run bounty missions all day to get rift tokens, run all of your rifts, and then reset the game and run the bounties all over again.  But, for players who are slightly less hardcore (like myself), it is also possible to run one act’s worth of bounties, get enough tokens to run a rift, and feel like you’ve made progress, even when you haven’t made any meaningful difference in your character’s in-game performance.  For me, at least, this is a key difference.  The content is largely the same, but the re-organization of it into smaller chunks, with the interlocking layers, gives the grind an illusion of structure, purpose, and direction.

In the meantime, if you want to just run through the campaign again and again, you can do that too.  While I was originally skeptical, the auto-leveling of enemies and the displacement of difficulty onto an explicit risk/reward slider actually gives players more valid choices for how to spend their time.  I’m still a fan of the progress/difficulty structure in Diablo and Diablo 2 that had its own self-balancing dynamic (the faster a player moved through spaces, the faster the game got harder, allowing your more skilled players to get to harder content faster, while slower players would end up resetting content between sessions, slowing down the difficulty progression), but the slider is great for advanced players.  You can pick your own comfort zone, and the carrot is still dangling out there to move up the tiers.

It’s a case of having your cake and getting to eat it, too.  The Reaper of Souls team expanded player options, while preserving the existing content and structures as a subset of the overall experience.  That’s quite a feat for an expansion pack of any game, much less one with such a rabid fanbase.  I’m not entirely surprised, of course.  One of the key players in the x-pack team was (and is) Josh Mosquiera (@joshmosq).  Again, you may not know Josh by name, but he is truly one of the best designers working today.  While I was at THQ, his team made a little game called Company of Heroes at Relic, and while Josh would be the last person to seek out the spotlight, you should pay attention.  When Chris Winters recommends a game, I take notice; when Josh Mosquiera makes a game, similarly, it’s worth looking into.

RPG Dynamics and the Freemium Revolution

Josh Bycer just said publicly on Gamasutra something that I’ve been thinking for a while: RPG dynamics are spreading to all kinds of other games.  While I disagree with a number of his arguments about the specifics of this trend, it’s clear that it is happening.  RPG elements and dynamics are making their way into everything these days, from shooters (randomized loot in Borderlands, character progression in Call of Duty), to puzzle games (quests are now ubiquitous in online casual games), to action games (skill trees as a core character development dynamic), and even builders (quests, randomized loot, and stat progression).  The question is, “Why?”

[Aside:  I noticed a similar trend recently with AAA action games now including stealth: Assassin’s Creed always had some stealth in its DNA, but The Last of Us, and even GTA V had significant stealth components.  Stealth is not easy; to do it right takes a lot of polish on the AI, character controls, animations, and UI.  The reason why developers/publishers are making this investment?  My guess is that needing larger sales numbers, they’re trying to broaden their appeal by giving players multiple play-styles through which to experience their content.]

Oddly enough, I would argue that the rise of always-online gameplay goes hand-in-hand with the proliferation of RPG elements into other genres.  Sure, there are outliers (Puzzle Quest had a minimal online component and was arguably one of the first to really hybridize RPG dynamics with radically different gameplay), but most of this has happened in the server-driven world of the last few years.  I don’t think this is a coincidence.  As games shifted from shelf-filler to online hobby, they needed to extend the experiences that were available.  PvP has always done this for online shooters and strategy games (other players are endless content), so why did they start dipping into the RPG well?

In part, they needed to incentivize players to create accounts and use accounts that were tied to the always-online world.  Not only could you not reliably make RPG dynamics part of a sometimes-online, sometimes-offline experience (infinitely hackable) and maintain the competitiveness of your online world, it turns out that giving players carrots is a great way to get them to do things.  Again, shooters already had some tools here, with leaderboards and clans, but as games became services, every available way to hook the player and to wrap them up for as long as possible became a key element of competition.  Progressions in general, skill trees more specifically, and especially experience grinds serve the purpose of keeping that carrot dangling in front of the player’s nose – easy to see, but hard to reach.

The advantage of online communities for players is that endless PvP content; the advantage for developers and publishers is that in a server-secured world, every player has to pay.  When we published Titan Quest, we could look at our online usage statistics and spot that out of every 100 players (online only, single-player sessions didn’t get captured), 95 of them were playing with a pirated copy.  When I wrote about that online, it created a minor stir (mostly because someone mis-identified me as the CEO of THQ), but I also got confirmation from a number of other developers and publishers that they were seeing about the same rates.  I don’t think it’s random that the percentage of paying players in the free-to-play, freemium model is quite similar, at around 3-6%.

In fact, there’s very little that’s getting published anymore that isn’t in that server-secured model, and of those models available, freemium is definitely accounting for the greatest mass.  So, when you’ve got players online and you want to keep them engaged for as long as possible, where do you turn?  Why, MMORPG’s have been doing this for years!  RPG systems can be cheap and effective ways to extend content: give players the same gear, but with better stats.  Stick a new color on it, or a different particle effect, and it’s a whole new asset!  Scale up an enemy, give them a new name, and now you have a new enemy!  Give players an endless grind that nets them a 2% competitive advantage, and they’ll do it.  Particularly with freemium, games-as-a-service, if you can maximize player lifetime while minimizing production costs, you’re more likely to break even.

It’s a little more complex than that, of course.  It’s not all about cynical manipulation to extract as much money as possible.  The reality is that all designers are beggars and thieves.  We take things that work from other games and re-deploy them, re-shape them, and hope that they work in our games.  Game developers have been ripping off RPG’s for as long as there have been RPG’s.  If you think System Shock isn’t a shooter/RPG hybrid, you’re fooling yourself.  Let’s not forget that Doom (the granddaddy of all shooters) came from an RPG campaign.

The ubiquity, though, has definitely increased.  Even as we get fewer and fewer actual RPG’s (or even Action RPG’s) – because those things are really hard and expensive to make – we’re getting more and more RPG dynamics in other genres.  Tencent made a version of NBA2K for China that wrapped all kinds of MMORPG dynamics around basketball, of all things.  In fact, the core components of RPG grinding (experience, quests, resources, slot-machine loot, repetition) have become staples of most freemium games out there.  At some level, this is because those dynamics encapsulate the things we want out of a game: we want a goal (quest); we want to be rewarded for accomplishing that goal (loot); we want to get better (stats); and then, we want to do it again so that we can see how much better we are.

What RPG’s do that other genres (action games, particularly, but also shooters and other genres) used to not do is to reward time investment without regard to skill.  Sure, certain levels of skill require time investment, and sure, RPG’s can require skill, but what RPG dynamics offer the freemium space is a way to map time into power.  It’s reliable, testable, and proven.  It can be done (relatively) simply with spreadsheets and graphs.  And, once you monetize time, you’ve got a business.  This is what is at the heart of this spread: RPG dynamics extend play time through simple, procedural progression.  Whether your’e competing in the freemium space or the AAA space, that’s a win, but in the freemium space, it’s essential.