[Game: Spelunky 2, Mossmouth, 2020]
Motivations to Spelunk:

On Spelunky 2, and Three Corruptible Virtues of Implementing Achievements for Games

 

Introduction:

This essay begins with a confession, one that feels on-par with admitting that one collects rocks or baseball cards or a similarly useless class of artifacts: I like achievements. In games that I am already enjoying, I actively make an effort to get achievements provided it does not impede that preexisting enjoyment. In fact, far from being impediments, I have often found that certain types of achievements lead to goals and playstyles that enhance the experience of a game. And if nothing else, purposefully attaining 100% achievement completion for a game can be a method of paying tribute to a game of exceptional quality—or of feeling that one has reached a satisfactory conclusion in otherwise endless affairs like roguelikes, arcade-style games, or even normal linear games that one can not seem to cease replaying.

In this article, I shall be covering the three clearest ways that achievements can be used for potential gains in terms of player experience and engagement. My main example in making this argument shall be Spelunky 2, as I believe that it and its predecessor represent nearly perfect implementations of achievements across all three categories to be covered.

The garden of achievements, however, is not filled exclusively with roses. There are a great many weeds and poisonous herbs to be found growing there. Achievements are often an afterthought, tacked onto a game by weary developers at the end of long projects, and—even when implemented with intention—may nevertheless include tedious, unappealing, or even exploitative goals. Thus, in each section of this article, after presenting the possible virtues of each prominent achievement type (with reference to Spelunky 2), I will also cover the vices and corruptions to which each type is vulnerable (with accompanying examples from other titles).

Spelunky 2 screenshot with Pilot approaching golden idol in Tide Pool - achievements

For consistency’s sake, and because it is the platform through which I personally access the majority of games, all of the examples in this essay will be restricted to achievements on Steam. This will have the additional virtue of preventing me from appealing to extrinsic virtues that achievement systems have on some platforms, where they can be used to unlock cosmetic customizations and the like. Today, we’re just focusing on the utility of achievements themselves.

Achievements for Progression

Of the three most plainly useful types of achievements that exist, progression achievements are the most straightforward and the most common. These are simply achievements that unlock as one plays through the normal campaign content of a game, marking events like completing levels and defeating bosses.

Spelunky 2, for instance, has achievements denoting the first time one reaches each new 4-level zone of the game, as well as achievements for beating each of the game’s two primary end bosses. What these achievements are, though, is far more intuitive than how they can possibly be virtuous or laudable inclusions in a game.

Principally, they are worthwhile as a method of tracking for both players and developers. Through them, one can monitor progress through the game by oneself, one’s friends, and the Steam community generally. In longer games or for players who play infrequently, recently unlocked progression achievements can even act as a subtle reminder of what one had been doing in their previous session.

In addition, in some particularly difficult titles, even awards for mere progression through a game could be a source of pride or glory, although this possible virtue will see its full potential realized in the next section.

Spelunky 2 screenshot with Valerie approaching Tun and Moon Challenge in Jungle - achievements

Now, it will no doubt be thought by some that Steam achievement unlocks are inaccurate sources of the kind of tracking data described just now. A few folks have objected on each occasion in the past that I’ve used the tracking info provided by Steam achievements to back up a point, like when discussing Luftrausers, Darkest Dungeon, and Crypt of the NecroDancer. And, in truth, to some degree, they’re right.

Steam achievements can only ever give a rough idea of aggregate player progression, as they will always be erroneously pulled in both directions to some degree: skewed too low because some players idle or run a game without ever really trying to play it, and skewed too high because some players use cheat utilities to unlock achievements they have not actually earned.

I’m fully aware that Steam achievement data is far from perfect when it comes to tracking community progress through specific games, but there are two factors that have kept me using it in the past, and which will keep me using it in the future. The first reason is, assuming folks interact with the majority of games in relatively consistent ways, such statistics will still be useful when comparing different games to each other.

Below is a graph of Spelunky 2’s achievements (in brown), charted along the x-axis in order from least rare to most rare. The y-axis denotes the percentage of players that Steam reports as having earned each of them. Note the relatively smooth curve across the game’s unlocks, with only a small group of tough challenges close to the x-axis toward its end. Now take a look at the line for Crypt of the NecroDancer (in purple) which I argued dramatically overemphasizes absurd endgame challenges that are comparable in difficulty to I Wanna Be the Guy-style fangames and kaizo romhacks. And then there’s the line for Luftrausers (in teal), which I argued has a needlessly sharp falloff in player progress that is created by the sudden spike in both randomness and difficulty accompanying some of its final missions. Even if approximate, I believe that these kinds of comparative relationships have demonstrative value.

Spelunky 2 Achievement Graph

The second reason I’ll definitely be continuing to use achievement data in future projects is that, frankly, it’s the best we’ve got—until a better method of tracking how many players have or have not accomplished certain in-game goals comes along, it’s the closest one can possibly come to objective information on the relative difficulty, desirability, or accessibility of various in-game tasks. So, absent ‘perfect,’ I’ll gladly settle for ‘good.’

Now, if fuzzy accuracy doesn’t really bother me, then what about the ways in which this type of achievement can go awry? You were promised corruptions and vices, after all! After a lot of thought, I could really only come up with one that seemed important: that when progression achievements are overused, they almost immediately feel patronizing. Receiving a virtual pat on the back for every inch of progress, like the achievements unlocking at the end of each and every chapter of each and every episode of The Walking Dead games, just feels like amassing a shelf full of participation trophies. Basically, if a game only has progression achievements, or progression achievements are being given to most players more than once per hour, they are probably overrepresented and overabundant in a game’s achievement list.

Achievements for Challenges

This section is about achievements for optional challenges that go beyond what is expected of the player in the normal course of gameplay. When a person thinks of achievements, this is probably the first type that pops into their head; after all, an achievement for completing a challenge is an achievement where it was necessary to actually achieve something in order to earn it.

In Spelunky 2, for example, some of the rarest achievements are for beating the game under special circumstances: with over $1 million, with $0, and in less than 10 minutes. There are also achievements for making item deliveries that yield in-game shortcuts, and for completing each of three minigame-style challenges administered in-game.

This type of achievement has the most versatile list of virtues, but also the longest list of potential problems. On the virtue side, achievements for challenges can provide players with opportunities to demonstrate acquired skill; they can lend longevity to a game beyond simple completion of its main campaign; and, similar to the subsequent section, they can nudge players toward different playstyles they may otherwise fail to try and areas they may otherwise fail to reach. Along these lines, some of the most enjoyable and triumphant accomplishments in my entire gaming career have been spurred by challenge achievements, like those for beating the games Hollow Knight and Hyper Light Drifter without a single death.

Spelunky 2 screenshot with Guy in Vlad's Castle in Volcana - achievements

The possible vices of this type of achievement, however, are numerous:

They can be used to add artificial padding (like when Team Fortress 2 gives achievements for accumulating enormous total damage numbers with particular classes, and enormous total victory numbers on particular maps).

They can bloat a game with repetitive or frustrating max-difficulty and speedrunner-style content (like the over a dozen deathless achievements in Super Meat Boy).

They can ask a player to either repeat a lengthy title an unreasonable number of times or else pressure them into selecting a higher difficulty than they would ideally want to play (like when there are achievements tied to each of a number of different modes, narrative choices, or difficulty options, as in Spec Ops: The Line).

They can prompt monotonous grinding or farming well after a game has run out of other content (like the incredibly dumb achievement in Hyper Light Drifter for 1000 deaths, which literally required me to die repeatedly to a crush block for over two hours after I’d already obtained all other achievements in the game—don’t worry, I listened to videos in the background).

And they can ask players to do something unrelated to the main campaign or gameplay which no one would ever otherwise want to do (like the two achievements in the simulation and management game Stardew Valley that are tied to one of its difficult arcade minigames).

I end with that last vice because it is unfortunately where one can point to a singular small misstep in the achievements of my primary example game: Spelunky 2 has an achievement for completing its tutorial in less than 30 seconds. Even though it’s not very tough, it is—at best—redundant with the main speedrun achievement, and—at worst—just kind of asinine.

Spelunky 2 screenshot with Airyn in Temple near Anubis - achievements

Nevertheless, a strongly underrated aspect of the achievement lists in the Spelunky games is that they actually exercise considerable restraint when it comes to challenge achievements, especially when compared to other games that borrow mechanics from the roguelike genre. Each Spelunky title’s absolute hardest normal run type (a full Cosmic Ocean run in Spelunky 2 and a full eggplant run in Spelunky HD) has no achievement associated with its completion. And even the actual hardest echelon of achievements in Spelunky 2 is, as I said toward the start of this section, really just composed of beginner-friendly tests of different challenge run varieties.

The ‘No Gold’ run is not a Low% run; you can still pick up items. The ‘high-score’ run is not a true ‘high-score contender’ run; there are no achievements for getting highly difficult, but perfectly feasible scores like two million or even three million. And the speedrun achievement has a very lax cutoff at 10 minutes, when times below 5 minutes are still far from top-level play. Moreover, none of those achievements have harder variants that also require the player to reach the Sunken City or the Cosmic Ocean alongside reaching the stated goals.

These choices maintain a set of max-difficulty and/or secret challenges that can be pursued by the game’s most dedicated and tenacious players, without withholding an attainable and satisfying 100% completion marker from everyone else.

By contrast, this is yet another ‘best practices’ memo that Brace Yourself Games clearly never received when putting together the achievement list for Crypt of the NecroDancer, which has achievements tied to each of the types of runs that have still, years after release, been completed by fewer than 100 confirmed individuals out of the game’s hundreds of thousands of players. Now, to be fair, that kind of monumental failure of restraint in achievement design is relatively rare among prominent games. Even Dustforce developer Hitbox Team, whose game has just one achievement that is incredibly difficult to get, still managed to get this right by keeping their actual hardest challenge, Yotta Difficult, disconnected from that achievement upon adding it to the game.

Spelunky 2 screenshot with Margaret in Sunken City near Kali altar - achievements

After completing everything else that a game has to offer, some players will want to put in the dozens or sometimes hundreds of additional hours required to successfully clear these games’ final gauntlets. But realistically, developers should aim to never have a gap between even their toughest achievement unlocks that far surpasses 10 hours for capable completionists. Presumably, developers want players to walk away from their games feeling that they reached (or at least could reach) a satisfying conclusion. But rejecting the advice in this section means that a huge number of even a game’s most dedicated and interested players may walk away with the feeling that they just weren’t interested enough in the game to ever want to fully complete it.

Achievements for Hidden Mechanics

The last of the three most promising types of achievements that can be productively implemented is also the easiest type to get wrong: achievements for hidden elements, areas, mechanics, and collectibles.

Spelunky 2, for instance, ties badges to accessing certain concealed areas and items that are helpful for reaching the second and third endings of the game. It also has a couple of achievements tied to interacting with every entity and area in order to fill a journal, and for progressing through a couple other optional NPC and mechanical questlines.

I think the virtues for this section are fairly self-explanatory: achievements for uncovering hidden stuff can encourage players to search for hidden elements, can confirm that strange discoveries are intentional, and can even alert players to the existence of such elements in the first place (giving them something to search for).

To begin discussing this section’s possible issues, there are several that are shared between this type of achievement and the directly preceding type.

As with challenge achievements, it’s possible for achievements in the hidden mechanic mold to pad a game (like the achievement for finding collectible hats in Papo & Yo, which don’t spawn until New Game +).

It’s possible for them to request that a player make a potentially unwelcome number of back-to-back playthroughs (like the rare item collection achievement in Dark Souls, which requires the player to get part of the way through New Game + 2, which is to say a third playthrough, with a single character).

And it’s possible for them to ask a player to do something completely irrelevant to the core gameplay (like the many achievements tied to the rather odd trend of pure puzzle games putting arbitrary hidden collectibles in their levels; this is shockingly common and almost never a welcome addition; some examples include Death Squared, Lara Croft GO, and Thomas Was Alone).

Spelunky 2 screenshot with Manfred fighting Olmec - achievements

But this type of hidden mechanic achievement also has a couple possible corruptions that are not as common for the other types.

First, most heinously, achievements can be tied to elements that are only hidden insofar as they are only accessible through paid DLC. Games that make this mistake are so incredibly numerous that it honestly doesn’t feel fair to single any out, and that list includes a great many titles I personally love—but it’s never going to feel anything other than exploitative for a game to require an additional purchase in order to achieve 100% completion. It’s a sleazy form of pressuring advertisement being delivered to a captive audience of people who have already bought a product (something akin to a ‘smart speaker’ that attempts to sell its user a subscription service any time they ask it to play music). The only time this practice can’t be categorized in this way is when the relevant expansion or update is distributed to all players for free. Don’t get me wrong: there’s nothing inherently wrong with charging money for additional work in the form of DLC; but unless it’s a standalone expansion or you have an extremely good reason, just keep it away from the achievements.

Second, and much more minorly, hidden mechanic achievements can spoil interesting discoveries for players through inelegantly phrased descriptions or even obvious art choices. Once again, I won’t be showing an example here, as in this case that would only spread the mistake of exposing a potentially unwelcome spoiler. Instead, I’ll take this opportunity to the end this section by pointing out a good example of this exact potential issue being avoided by Spelunky 2: the 100% journal achievement indirectly, and therefore in a potentially spoiler-free manner, requires the player to uncover and experience one of the game’s most wonderful and well-hidden secrets for themselves—a secret which has no dedicated achievement of its own.

Conclusion:

These are not the only ways for developers to use achievements, nor are they even the only good ways to use them. The Binding of Isaac and The Binding of Isaac: Rebirth, for instance, tie the majority of their achievements to in-game item and character unlocks, making every iota of progress feel like it is actively expanding those games. And at one point Portal 2 even manages to use an achievement to underscore a joke. But I do think the three methods detailed in this article are the most reliable methods of productively improving a game by including achievements, provided one remains wary of the potential corruptions to which each method is vulnerable.

One intending to provide a satisfying and utile set of achievements could do a lot worse than looking to Spelunky 2 and Spelunky HD for guidance. These games contain tight and balanced sets of achievements that overlap across the three categories I’ve covered in this article, with only a couple of (unimportant) odd achievements out.

Spelunky 2 screenshot with Ana near Ushabti room in Neo Bablyon - achievements

Folks seeking additional examples of games that get achievement design almost exactly right can also direct their attention to Slay the Spire, Celeste, and the four titles released thus far by the developer Supergiant Games, whose achievement implementations are nearly as strong as Mossmouth’s. In fact, on the off chance any game developers are reading this article for tips, we can now step back from these positive examples to extract just a few further guidelines to help steer achievement design:

First, avoid achievement bloat. The corruptions of all three achievement types listed in this article become more likely as the total number of achievements grows; I don’t think it’s any coincidence that the eight games I’ve now held up as paragons of excellence in achievement design have roughly between 20 and 50 achievements each. Maybe this is a good range to keep in mind.

Second, once you have some player data, aim for the kind of smooth completion curves represented by these positive examples. A smoothly sloping completion curve means that a game awards players in a fashion that is proportional to their time and effort. A jagged curve means there are sudden expansions in time and/or willingness between certain achievement unlocks, and flat lines hugging the top or bottom of the chart can indicate overemphasis on a game’s easiest or hardest aspects, respectively.

Third and most simply, include an achievement that clearly marks 100%. This can be either the game’s hardest achievement, or else an achievement that literally corresponds to unlocking all of the others. This makes it incredibly easy for completionists to display what they’ve accomplished, and thereby prevents muddying the sense of satisfaction and pride that accompanies full completion of a game. If no achievement stands out as both rarest and hardest simultaneously in a fashion that may be clear even from its description alone, then the latter path of simply having a dedicated 100% marker is the better choice.

In closing, achievements are very far from being the most important part of game design. Frankly, they’re one of the least important. But that doesn’t mean they can’t enhance the experience of a work the way that other types of finalization and polish can. They have the virtuous power to amplify the satisfaction and glory of progressing and improving in a game—while they also have the corrupt power to diminish that satisfaction and glory. They have the virtuous power to guide players toward distinct and interesting playstyles and mechanics—while they also have the corrupt power to enforce undesirable tedium or even act as manipulative advertisements. So, spare a thought for these oft-overlooked badges . . . because they affect player psychology, and because some of us care about them.

[Game: Spelunky 2, Mossmouth, 2020]
Motivations to Spelunk:

On Spelunky 2, and Three Corruptible Virtues of Implementing Achievements for Games

was last modified: February 10th, 2021 by Daniel Podgorski
Bookmark the permalink.

Comments are closed