Ramble III: A Perfectly Reasonable Somewhat Reasoned Screedy Diatribe
Preramble
Sometimes you want to indulge yourself and curl up with a pint of Häagen-Dazs White Chocolate Raspberry Truffle all by yourself. This little rant is one of those.
A Perfectly Reasonable Somewhat Reasoned Screedy Diatribe
I would like to put forward a candidate for the Worst Thing in Games award. Forget pay-to-win or buggy messes or QTEs or forced slow-walk sections. The category winner, by a country mile, is the invisible wall. Literally (in a metaphorical sense), the worst. A travesty. A sin against the gods. They create a visceral reaction in my guts. A fiery anger that can’t be sated. These mime-adjacent abominations are the most terrible thing that has happened in games in the history of ever. They’re pure evil.
And, unfortunately, they’re also fairly necessary.
Let’s start with the whole ‘necessity of evil’ aspect. [Philosophical tangent redacted for time and sanity. -Ed.] Aside from any metaphysical discussions, invisible walls let developers create boundaries, which often make games better.
Players gonna play, play, play, play, play (or so I hear), and they will definitely go off the path. They’ll do this for any number of reasons, from intentionally exploring with an enthusiastic curiosity to unintentionally wandering with an enthusiastic meander. Sometimes off the path is good and a core part of the game, but in other cases, off the path just leads to bad. Players get confused (and frustrated) or lost (and frustrated) or stuck (and frustrated). With the invisible wall, designers can help avoid these kinds of frustration and focus the player’s (and developer’s) time on the core experience.
So yes, grudgingly, invisible walls are an important dev tool, from a heavily linear COD level to the edges of the expansive province of Cyrodiil.
But while invisible walls are an important dev tool, a boundary that is just an invisible wall creates its own big ‘ol pile of frustration. They break the rules of the world, can’t be anticipated, and set up the worst kind of surprise – one with only downsides. Rather than playing the game, the player is left guessing at the intentions of the developer. A player may not even notice they’ve stopped moving forward, and when they do notice, they’ll often assume it’s their mistake for not understanding the world or a bug. They’ll spend unreasonable amounts of time trying to keep going, and then blame themselves when they finally realize it was just an invisible wall, as if that’s an obvious thing to expect in an otherwise visible world.
A simple solution is to include a world-appropriate physical boundary. Fences are great. Or buildings. Or a police checkpoint. Or a tangled alien growth. Or lava. Or a towering cliff. Or a tall hedgerow. Or a different fence. Or… [Yes, yes. Enough. -Ed.] With a visible boundary in place, the intended playable space is clear, and players can navigate the game with a bit less confusion. Once the physical boundary is there, having a tall invisible wall that accompanies it is reasonable and will help prevent players from getting Somewhere Bad™.
When building these boundaries, the simplest wall may be functional, but it won’t be particularly engaging or natural. A bit of chaotic variety goes a long way. A fence connected to a house with an alley between it and the next house that’s blocked by a pile of trash connected to… The irregularity in the boundary makes it more interesting and helps the world feel real while still bounding and guiding the player.
This variety can be pushed even further by widening the world and pushing the boundary back, just a bit. Consider a country road with classic country fences running along either side and classic open country fields on the other side of the fences. This will clearly guide the player down the road, but not much else. Instead, move one fence to the far side of the open field. The playspace boundary is still clear, but the open field on one side will break up the monotony and give players something to consider briefly. The world is suddenly bigger and more engaging without any more uncertainty on the intended path.
When creating these boundaries, a good minimum height should be about a bit taller than what players are expected to jump and climb over in normal gameplay (otherwise, it’s easy to confuse the boundary with the intended path). Simple low walls work well when the player can jump and climb like a moderately encumbered tortoise, but once they can spring or bounce, these start to look more like a fun playground. With higher mobility, taller works to a point, but can be hard to make natural. In these situations, boundary layers with increasing height can be a good option. A first boundary should be passable but not inviting (like a low rooftop or jumpable fence). This first layer still guides the player down the intended path, but it can also reward them for exploring (with a good flanking path or goodies to find, for example). The next layers can then be taller to create a clear gameplay boundary while feeling more natural. Bonus points, the layers create some great parallax!
Even with all of this, physical bounds aren’t always a reasonable solution. In a multiplayer map where you can fly up high into the air from a massive exploding robot (just a random example), invisible walls would greatly frustrate the joy of falling with style. Of course, landing outside the playable space would break the MP balance, which is also not good. In these cases, letting the player know they’re out of bounds and giving them a clear warning and chance to correct it is reasonable. Aside from flashing big red warnings and klaxons, here too, level design can help. Keep the edge of the playable space discreet and clear, and the out-of-bounds area simple and uninviting. A no-mans-land without cover or excessive detail won’t entice players and helps guide them back to the playable space.
Which, I suppose, all comes back to a plea – set clearly visible boundaries. [Good relationship advice, too. -Ed.] When built well, these visible boundaries invisibly guide the player, creating a natural border in the world that informs, focuses, and avoids my kids having to listen to yet another ‘dad’s in level design mode’ rant.
Postrambles
Art provided by the talented Cup O’ Soup.
Another boundary option is to embrace that it’s a game and make a clear (if somewhat immersion-breaking) artificial boundary. Grounded 2, in their pre-release, put up some very obvious ‘under construction’ caution tape boundaries paired with the invisible wall. While this may be less immersive, the intent is obvious and it’s easy to implement, leaving more dev time to work on the rest of the game.
Practical Advice: With higher mobility, a wall that is about 1.5x higher than the player can jump is about the height you need to stop players from considering it as a potential path. For distance, it’s more like 2x how far they can jump. At these distances, most players won’t consider that as part of the main path (though QA will absolutely find ways over or across…)
More Practical Advice: When creating invisible walls along the perimeter, make them much taller than would seem reasonable, and then a little taller still. Otherwise, you’ll just end up doing this later when QA inevitably finds a way past the shorter walls.
For built boundaries to work, they have to go hand-in-hand with the overall design of the level. The overall player objective has to be clear while also doing many (many many) little things to nudge and guide the player. Something I’d like to cover at some point, but this LD Workshop on guiding the player has a lot of good practical advice right now.
