Words: 2238 Approximate Reading Time: 15-20 minutes
When we fall in love with games, we often seek to extract as much joy as we can out of them. That can involve all sorts of things. Maybe we replay it over and over until we’ve done everything we can. Maybe we explore every story possibility available to us. Maybe we seek out every achievement or trophy offered by the developers. Maybe we revisit it every year or so to enjoy it again.
One thing that we might do is embark on some kind of challenge. Find a way to play the game differently. Perhaps you play through the Borderlands series using only guns from one manufacturer. Maybe you play through Dark Souls at Level 1 the entire time. Or you could try beating the game as quickly as possible. There are all sorts of possibilities, truly limited only by your imagination and your patience.
Similarly, another thing we might do is make use of mods. Aspiring developers and fans have created a wealth of content for existing games, from cosmetic changes to scripts that randomize where items and enemies are placed to even just overhauling the entire game to try and make a new experience. The possibilities are, again, pretty substantial.
At the meeting of these two concepts is the “Kaizo” challenge. That term technically has a specific meaning, being attached to the Mario series. But it has also become a catchword to describe a particular kind of mod to a game.
What kind? Well, it involves reconfiguring the game to be insanely difficult. Precise movement, traps, tricks, and so on. All the kind of stuff that most people look at and say “nah.” There is absolutely no doubt that they can be infuriating to play.
But when done well, they can also offer the kind of new experience that we crave. The one that pushes the limits of what we can do with the game and really tests our knowledge of its systems. Which is not to say that everyone should seek these kinds of mods out to prove themselves. These challenges should be seen as ways to have fun, not as a status symbol.
Now if all there was to do here was say “insane challenges like Kaizo levels can be fun,” there wouldn’t be much point to this essay. I could try to explain why, but that wouldn’t be that interesting.
Instead, I wanted to address a phenomenon I have occasionally seen within the community of those people who make these kinds of challenges. Specifically, the sort of “arms race” that can occur between creators and players that leads to challenges that…stop being fun and/or interesting. I want to explore what qualities make these kinds of challenges fun, and from there how creators might go wrong.
A Brief History of Kaizo
Difficult games are not new on their own. A lot of older video games were designed to be tough – sometimes frustratingly tough – for the purpose of extending playtime or extracting money from kids. But as video game consoles took hold in homes and we started to learn principles of what made games fun and not merely profitable, games got progressively most friendly. Hard content still existed, but players could more easily sidestep it. The hardest content in the game was usually at the very end – when players had already beaten every other challenge available – or was a secret reserved only for those dedicated enough to complete every last bit of the game.
The first iteration of Kaizo Mario World was released in 2007, constructed by someone for their friend to play. The game was built out of Super Mario World for the Super Nintendo, but with the levels created entirely from scratch. The point of the game was to be intensely difficult – you need to be precise with your movements and platforming, memorize movement patterns, react quickly, and die a lot. The levels featured plenty of pits to fall down with only small blocks to land on, plenty of enemies and traps that needed to be avoided (often to the point that jumping on an enemy would cause you to launch up into a set of spikes), and almost no power-ups to give yourself a bit of breathing room. That said, the game was beatable – in a pretty strict sense of the term.
As the mod itself took off, you had a variety of similar mods developed for other games. Many of these focused on the Mario series, with the same Kaizo Mario World developer releasing two sequels using Super Mario World as the base. Other developers made similar hacks using Super Mario Bros. 3 and Super Mario 64. And by no means was this limited to the Mario games. The Pokémon serieshas not only an “IronMon” randomizer challenge, but then a “Kaizo IronMon” to ramp up the difficulty. Super Metroid has two Kaizo mods. And Hollow Knight has its own special challenge mod called “Impossible Any%.”
It is not really an experience I would recommend. If you’re aware of these mods and have found the idea appealing, then you really don’t need my recommendation to try them out. And if you’re on the fence, then my recommendation would simply be subjecting you to torture. Which is not to say that they’re actively unfun, but that you have to go in with a particular mindset – and a willingness to admit that you might be unable to finish a level/game.
The Arms Race
So imagine that you’ve set out to design an extremely difficult level in a game, or an extremely difficult challenge.
The first question that of course needs to be asked is how you will make it difficult. What type of difficulty will you include?
Because you could, for instance, limit the player’s time: they not only need to pull off a complex sequence of inputs, but do it incredibly quickly, having only a moment’s leeway. Or you could test a player’s stamina – the longer you make the sequence, the more that needs to be remembered and pulled off, meaning the less likely a player is to succeed. Or you could just throw every possible hazard at a player and force them to react quickly.
Whatever the case may be, the core problem is going to be to make that challenge interesting. You can very easily make it deadly. But making it fun to play requires a careful balance of toughness and feasibility. It is that concept of “fairness” that I have talked about before. And while these kinds of Kaizo designs are inherently unfair, there is still a sort of fairness lurking within them.
Firstly, to explain the idea of the challenge being interesting, it’s useful to think back to your own triumphs. Whatever games you’ve played, you’ve probably been faced with challenges imposed either by the game itself, or by other players, or by yourself. Many, probably even most, of those challenges gave you a feeling of accomplishment once you succeeded. But you can perhaps recall some occasions where success came with a feeling of relief – you did not actually feel happy that you had won, but rather glad that it was over. An interesting challenge invokes the former feeling, and not the latter.
The sweet spot will not be the same for every player. Someone may enjoy the thrill of rising in the ranks of a competitive game, and others will feel like the process is akin to having your teeth pulled. Someone may chase the thrills of overcoming bosses in Dark Souls, and others will hate banging their head against those walls. Kaizo levels necessarily aim for a sweet spot that is going to be out of the range of most players. That, on its own, is still fine. These are designed for “hardcore” players, anyway.
Secondly, let’s cover fairness. Even though, as I said, these kinds of incredibly difficult challenges are inherently unfair in a bigger sense, they still need to have some kind of fairness. Usually the fairness come from the fact that they are genuinely beatable. A human being can perform the tasks needed to succeed. Although even that’s a pretty bad standard. It shouldn’t be the case that someone with infinite patience could eventually beat a challenge to make that challenge technically fair. We should be aiming for something a little more realistic.
But I state these two points to introduce the odd balancing act at play. Because on the one hand, you want to make something genuinely challenging. If too many people can complete it, then surely that means in a way that you’ve failed, because a challenge should theoretically separate out players into the “cans” and “cannots”…and an extreme challenge should have very few people who can complete it.
Which means, of course, that you will likely seek for more and more ways to add challenge. To test the absolute limits of players, figure out just how small you can get that success rate without making it 0.
And yet, your goal in designing something as part of a game should also be to make it fun. Even if that fun is not for everyone, it should be fun for the audience you hope to target.
Succeeding in making something so tough that almost no one can complete – or maybe even no one could complete it period – is ultimately a failure.
My interest in this idea was sparked by a recent release of a special mod in Hollow Knight. There are all sorts of challenges within the game itself, including the Pantheons, which are basically boss rushes against powered-up versions against the game’s various bosses. This is already very difficult content, and that’s just the game as provided by the developers.
But someone eventually decided to take the final boss of the game, Radiance, and make them even harder. This is already quite a tough fight for the average player.
But people beat that modded version. So then the designer decided to take it up a notch and make that fight even harder. And people beat that.
So then the designer made yet another version that was even harder. You can see a video of the fight (done with an invincibility mode on) here. It’s…a bit much.
And it was in watching that video that I could immediately tell that this fight would not be fun. In fact, it felt basically designed to be technically doable, but not realistically doable. A computer could beat it, but not a human. At least, not a human that wasn’t cheating.
This boss was actually not the first time I thought about this issue. Having played a fair bit of Mario Maker, I tended to gravitate to these Kaizo-style levels. And one thing I found was that some levels were designed to be tough, but also fun. And some were designed to just be tough, with fun not really being part of the equation.
And I think the cause of the latter challenges is this arms race. The idea that the goal of making a challenge is to beat the player, rather than for the player to beat it. To some extent I even worry that major developers like FromSoft have stumbled into this issue. You could even argue that the superboss trope from various RPG games steps into this category – extra tough bosses can easily wipe out your team even at the highest levels, usually requiring some kind of trick and/or a lot of luck.
But it should always be remembered that a game should exist to be fun. Even if not every game – or every mod – is aiming to be fun to everyone, there needs to be serious thought put toward the audience. Will people actually engage with this, and will they enjoy doing so? Do I intend for them to win, and to feel good about winning? If we cannot honestly answer those questions with a clear “yes,” then we need to step back and rethink what we’re doing.
Conclusion
I love a good challenge. I’ve done so many challenges, some of which I regret. I’ve played some of the original Kaizo Mario World. I’ve played a number of Kaizo levels in Mario Maker. I’ve done a Soul Level 1 challenge in each game in the Dark Souls series (I tried in Bloodborne, got about 10% of the way through, and then stopped because I was too frustrated). I have completed the Pantheons in Hollow Knight and am now working on the “bindings” – extra constraints on your characters as you go through the boss rushes.
And yet, by the same token I find my patience wearing thinner and thinner over the years. Not in the way that I avoid challenge, but rather I am less and less inclined to put up with bad challenges. With the kinds of designs that are simply meant to be thoughtlessly hard. Where you are simply meant to try a hundred times at something until you finally prevail.
Designing good and fun challenges is tough. Nobody should be knocked merely for failing at that. A good designer will be able to learn from their failure and make something better the next time around. What should be heavily criticized, though, are those who see this failure as the point. When a challenge is poorly designed and meant to be poorly designed, that is when it is not just a failure of the design, but of the designer.
Abstracting Games 