The music ramps up in tension as the fight continues, growing more and more stressful as you fight. A flurry of projectiles crosses the screen, even while the huge, looming figure makes themselves vulnerable for the first time in a while, bringing their leering, disfigured face forwards out of the shadows to watch you die. You leap between two obstacles, but it is the next obstacle that catches you off guard, annihilating you… and sending you back to just before the boss fight starts. The tension you felt just moments ago evaporates, and the player is left feeling not necessarily release, but a reprieve.

Death in video games serves a very different purpose across many different genres. For some genres and some games, death as a concept is an incredibly minor setback. Many modern platformers go for this approach. Even if the platformer has lives, the most you will be inconvenienced is having to start a level again. Death in those games is generally a minor punishment for a small failure. You are expected to quickly learn and adapt to your mistakes, and not to get stuck on one obstacle for too long – in a well designed platformer, that is.

Jak II, a game which remains highly ranked in my mind as a spectacular action-adventure-platformer game, has rather infamous balance. You have a circle of eight hitpoints, with most attacks taking off two hit points. This means getting hit four times is enough to kill Jak. Health packs, which are altogether too rare and can only be found by smashing particular crates (which often contain ammo instead), restore two hit points, canceling out one hit. This might not be such a problem if the checkpoint system was more effective. Most levels in Jak have one or two checkpoints – one at the very beginning of the level, and, if you are lucky, one right near the end. This obviously creates a problem – it is very easy to be killed, and it causes you to restart whole levels. Some levels are thus incredibly frustrating for first time players – especially certain escort quests. The final oversight in the Jak II death equation is that ammunition is lost even after death. Say you had twenty bullets in your red scatter gun, and used fifteen of them, then died. When you respawned at the start of the level, you end up with five bullets remaining. A patently unfair system, and a combination that makes death seem something much more serious than in other platformers. There’s something to lose when you die – a lot of progress.

This technically makes moment to moment gameplay more exciting. When there’s the thrill of something to lose, the game seems to want you to think that this makes the game more engaging. But, while this might mean the game can create some dizzying highs, it can also create some awful lows. Death in Jak II, unlike its predecessor or its sequel, slams on the brakes, causing you to have to replay an entire level over again, just to get to the bit that stumped you at the end. Levels are often wars of attrition, where you have to work hard not to get hit in the earlier sections just so you can afford to tank a few hits in the more ruthless end-of-level sections.

Genres like racing and fighting games don’t tend to have this issue either – at least, not out of the coin-operated realm. When you die in an arcade machine, you need to slot more coins in to keep playing. But in a modern game that you’ve already paid for in full, you can just hit the restart button. Sure, you might have to replay a race or a fight, but there’s little time investment lost. Death as a concept in video games has to be balanced with concerns and considerations of what you want the player to feel. Fighting, racing, or modern arcade-style games, or ones centered more primarily around an expressive or fun experience, tend to have death either be non-existent or as a lighter barrier to entry. Sure, you died, but you can always immediately jump in and have another go. Some fighting games, like Mortal Kombat X, even have a special currency you can use to skip fights in the story mode you are having trouble with.

In multiplayer, the punishment for death is often just a loss in rank, a way to try and fit you most accurately with your peers and skill level (at least, that’s the theory). This loss in rank is also a way to set you back, to make you lose a bit of time, but in a way that also encourages you to keep playing. You’ve lost some progress, but it is worth fighting on, trying to see if you can become even better than you were before.

The genre that probably has the hardest time with death is horror. Death is often the ultimate punishment in horror – if the enemy catches the main character, the threat of death is basically always the goal of the enemy. But in a video game, the horror basically ends when the player is killed. There is a definite kind of release of pressure, as well as a deflating of expectation. The player knows and understands, upon their first death, that what was just inflicted upon them is basically the worst thing that’s going to happen to them. That’s particularly why earlier levels in horror games have the easiest time frightening the player. There’s a lot of uncertainty around who or what the enemy actually is, and what they have the capability to do to the player.

There is also, in horror games, the question of time being lost, as I’ve already discussed. Games use that time being lost as a threat against the player, basically the largest actual penalty the game can dish out. Some games, mainly RPGs, might also ask for a gold or XP penalty on death, so as to encourage the player not to die. This always felt like a particularly nasty penalty to me – asking me to become explicitly weaker just so I can keep playing? A poor deal in my eyes, and one I’ve always avoided in games.

In return for having been killed, however, the player is supposed to receive real life experience. This is often played upon in rogue-like or Soulslike titles, wherein death is basically the whole point. It is the crux of the genre – you die to get better. Provided the game is also well designed, with a good gameplay loop (and is also fair with its difficulty), death should just be a challenge to the player, daring them to leap back into the fray and try their hand again. Returnal is particularly effective at this. It’s quite hard to put down, because with death comes further knowledge. Sekiro is practically an addictive drug when it comes to this too. Being killed by a new move from a boss in a new phase just encourages you to learn how to beat that, too. Death in these games is like climbing a mountain – you might be set back one step, but you move forward two next time. Slow and steady progress provides a rewarding and memorable backdrop for challenging games.

Horror, however, runs a risk when it asks a player to replay a level due to a difficult section. Horror is all about novelty. The longer a player lingers in a specific area, becoming more and more frustrated with maybe a one-hit kill enemy or an environment that is poorly lit and filled with many pits or traps or something similar, the less frightened the player is going to be. This means that yes, the player can become emboldened and fight back against the terrors – a good feeling in a game like Resident Evil, where the fighting is basically the whole point of the horrors – but it can also mean that the game stagnates, and becomes a shooting gallery, with no frightening enemies. Resident Evil 7‘s difficulty in particular does a marvelous job keeping you trapped between “legitimately frightening” and “able to be overcome”. Players are kept on their toes while still being able to proceed, with no one obstacle stopping them for too long. If everything was too easy, however, the player wouldn’t feel scared at all.

You can see the tightrope that horror games must proceed on. The threat of time loss as a death mechanic is prevalent in the very early survival horror titles, of course most famously the original Resident Evil, with a use of ink ribbons to save the game. I used to be a big fan of this, thinking that the higher the tension is, the better the game is. This is somewhat true. Just like Jak II, it allows for dizzying highs when you survive a long stretch of the game without wasting a save item. However, if a player has spent ink ribbons poorly, then they might end up feeling frustrated. It’s not a challenge that they chose to instill upon themselves, instead an unfair challenge that players are forced to experience, just because they saved one too many times earlier in the game. Some players are even forced to restart the whole game because of such decisions. This is certainly not frightening – it is demoralizing. Here, death is not so much a thing to be frightened of, but merely a punishment to cringe from. Something I realized is that the fear is not more prevalent towards the game – the fear is simply “what if I lose the progress I made across this entire gaming session”?

Again, rogue-likes play on this fear a bit too. But there, death is supposed to be a part of the experience. Cyclical attempts at randomly generated runs or enemy placement still rewards you, giving you money, weapons or XP to spend in between your runs. But in Resident Evil, there is nothing but a hollow feeling in your gut as you realize that very soon you will have to replay the last forty five minutes of progress simply because a new enemy type caught you off guard. Yes, you’ve learned, but forty five minutes is too great a punishment for what you’ve learned. Because you made it through the first half an hour fine, it was only the last quarter of an hour, say, where you were inconvenienced.

Death, then, is so hard to get right in video games. It should remain both a tantalizing threat and yet never so omnipresent that your game is hair-tearingly difficult, unless your game has many checkpoints or is specifically and meticulously designed around a cycle of failure and recovery. Horror most certainly has the hardest job. I remember being confronted by two machete wielding maniacs in Outlast and immediately being killed by them, with no chance to resist. One hit, and I was dead. There was no fear, no dread, left in the encounter. Every time I was caught, I got a machete in the stomach, and was killed. But because it was, after all, just a video game, death being so prevalent made me frustrated with the section, even if it was short. It should have been petrifying, being able to be caught by these maniacs, and allowed a short window to wiggle out, even if it was just once or twice per life I was afforded the opportunity. In horror, the player should brush with death, only sometimes experiencing it. Otherwise, the bubble has a habit of bursting spectacularly, leaving a player playing a game that suddenly has much less thrill, much less excitement.

Everyone remembers their first jumpscare in the first Five Nights at Freddy’s, at least if you’ve never seen or played the game before. That’s because the tension beforehand is crushing, and you’re not quite sure what those dastardly robots are going to do when they catch you. When you die that first time, you are most likely quite shocked, quite frightened – even if just for a single brilliant moment. Afterwards, however, there is generally a feeling of tightening one’s helmet around their heads and diving in for another go. Novelty is that beautiful, elusive thing in horror games, the thing that makes it so frightening. When a player begins to feel that novelty wearing thin, a designer is best served by giving them something new. This is why lengthy horror games are so hard to sustain. Shorter experiences like Concluse rule the indie scene for a reason – both because they are easier to create, and because fear is easier to sustain over shorter periods of time. Games like Resident Evil Village, with such a lengthy running time, often rely on other kinds of gameplay experiences or moods to once again later sustain the horror. Death erodes novelty, and novelty is key to horror. Sustaining a game after an initial jumpscare is harder than you might think.