A couple years back, I did a column about why procedural generation does not solve the content problems in MMOs. This remains a true statement. You cannot fix the fact that MMOs have a limited amount of content with the fact that procedural generation can produce a seemingly endless series of strictly unique bits of content, in no small part because intentionality in design is always going to make a bigger difference than random chance. I’m not going to restate the whole column here; if you haven’t read it, go do that first.

But I also do want to revisit the topic because I haven’t revisited since then and just a cursory glance at the stuff I’m talking about in that column will reveal that several of the games I cite are games I really like. Warframe and City of Heroes both run on procedural generation to an extent. Star Trek Online did at launch and I was a fan of it then. So it’s worth examining how procedural generation can still be a good thing for MMOs even if it can’t fix the fundamental content problem – which means starting by examining what the “content problem” is in the first place.

The vast majority of video games do not have a content problem. Single-player games have a certain amount of content, and when you are done with that content, you are done with the game. Yes, I’ve noted before that you can decide when you’re really finished with a game, but the point is that once a given company has your money, that studio does not care how much you play that game. Even multiplayer games like fighting games don’t have a content problem; at a certain point, there are no more new fighters, sure, but that doesn’t really matter.

MMOs have a content problem because the games are designed around the idea that you’re not done. You play this game forever, but you are never actually done with it. That runs into the basic problem that there is only so much content to do, and the fact is that it takes more time to make content than it takes to clear it. If it takes months to make a new dungeon in Final Fantasy XIV, that lines up with the development cycle, but I am going to clear it in about 20 minutes.

How do you generate content to match that pace?

Procedural generation is, at least theoretically, a fix for this exact problem. The development cycle can’t be shortened, but once it’s finished, you can just plug in a new set of variables and produce theoretically infinite-ish content. But as I mentioned in the previous column, it doesn’t really work. The content doesn’t hit the right level of bespoke and thus doesn’t fill the roles of an infinite content engine.

However, there is something that it does very well, and that’s producing an endless sea of mashed potatoes.

Let me be clear about something: I like mashed potatoes. I have always liked mashed potatoes. I will eat basically as many mashed potatoes are put in front of me. And as a dish, mashed potatoes can easily be scaled up by using an ever-bigger pot and an ever-larger number of potatoes and so forth.

Mashed potatoes are not the main course of any meal. But they’re tasty and filling and you can have a lot of them. If you’ve already had the main course, they can fill you up; if you haven’t, they can spur your appetite. I use this term because the obvious terms like “filler” don’t really get the positives of the scenario quite right.

Let’s use Warframe as an example. That game is made out of branching hallways provided by procedural generation stitching various map segments together. For most missions, no two runs are going to have exactly the same map. There are several missions (as well as quests) that have bespoke, specific designs you are going to be running through, but the vast majority are using procedural generation. So why does this matter?

Well, because for most of these maps the mission objective means you’re just racing through, sending insane amounts of ammunition downrange while spamming abilities. The reason that procedural generation is used here is to make sure that every given run of the same mission – some of which you will need to run dozens of times – has some spice and variety to it.

The idea isn’t that a slightly different map for this particular Exterminate mission is going to change your basic interaction with the material, but it’s going to mean that you cannot memorize a single best route through the mission and instead still take note of the scenery and the area. Sure, after a dozen runs you’ll recognize parts of it. But the main course here is the shooting and ability gameplay. The layouts are mashed potatoes. They’re support.

Are the CoH radio missions the best content the game has to offer? Absolutely not. But they do mean that if you want to just level instead of clearing specific story arcs, they’re there. The randomized tips mean that you can always do missions to alter (or affirm) your character’s alignment. Sure, it’s all assembled from pieces and randomized, but those pieces are variable enough that you can’t run out, and it’ll take a while before you see anything strictly repeat itself. It’s filler, but not in a bad way. There are actual deliberate and structured pieces of content for when you want that; this is for the times when you just want to get the group together and do something low-stress and fun.

Procedural generation cannot fully solve the content problem because meaning cannot be extracted from something that does not have meaning put into it. The parts of game design that make for thrilling moments of play require deliberate, sculpted effort. But there’s a lot of any game that is taken up not by thrilling moments but just by moments. In Persona 3, there are specific boss fights and story encounters, but the dungeon doesn’t need to have bespoke layouts since it’s there to just house monsters for grinding and a handful of treasure chests.

And that’s something procedural generation addresses very well. It provides the filler content without making you constantly aware it’s all filler, and when handled well it can avoid scenarios like winding up with all your objectives grouped in one space or whatever. It keeps you from becoming locked in to a very specific layout and style, adding just a little extra flavor and substance to a game so that you have… well, more mashed potatoes.

Mashed potatoes might not make or break the experience, but gosh, do you notice when you’ve got some good ones and when you have enough to keep you filled. And while it may never fix the content problem in and of itself because it literally cannot, it does still help ensure that you have more stuff to do on a regular basis.

Sometimes you know exactly what’s going on with the MMO genre, and sometimes all you have are Vague Patch Notes informing you that something, somewhere, has probably been changed. Senior Reporter Eliot Lefebvre enjoys analyzing these sorts of notes and also vague elements of the genre as a whole. The potency of this analysis may be adjusted under certain circumstances.