Give me consequence or give me death!

I’m not speaking figuratively.

At the GameCamp unconference earlier this year, one of the talks given was on The Failure of the Failstate; a debate in which the speaker was arguing that gaming’s traditional punishment for failure – the Game Over screen – had been invalidated by save states and checkpoints. I didn’t completely agree – I felt that was far too much of a blanket statement – but it’s an issue worth addressing. This, several months later, is my contribution to the argument.

The standard failstate of a death followed by a restart works perfectly – for certain games. Challenging, twitch-based platformers like VVVVVV punish the player with a quick death and restart of each level, and allow them to gradually learn the techniques to continue on with each successive death. Similarly, endurance games like Canabalt send you right back to the beginning after each death, asking you to start again from scratch in your attempt to grab a high score. This is because these are games of skill, where the draw comes from proving your own abilities, and survival is the bar against which your skill is measured.

Where it falters is in games with a strong narrative flow, where threats are meant to be palpable. Much was made of Bioshock’s most dangerous (potential) enemies: the Big Daddies. These huge, shambling creatures were capable of withstanding and dealing out huge amounts of damage, and any action you had to undertake which might provoke them to combat was preceded by a great deal of preparation and tension on the part of the player… or at least, that’s what was supposed to happen. Unfortunately, the moment the player realised that each time they were killed they simply reappeared outside of a nearby Vita-Chamber, with their enemies still suffering from any damage dealt, any sense of threat was lost as all enemies, big and small, became nothing more than temporary nuisances.

Achievement Unlocked: no sense of danger in this game's combat.

Bioshock is an unusual extreme – not only is death not permanent, but you don’t lose any progress in dying; the setting doesn’t revert to even a moment before your death, it simply resurrects you. Most games at least have you go back a little in time following a death, but even here the cost is trivial. Battles in Call of Duty can be won out after a few retries, each retry letting you learn yet more of the enemies’ patterns. Gordon Freeman can face the combine as many times as he needs to vanquish them. And the triviality of reverting to an earlier save can do more than just neuter in-game tension; it can, for a certain kind of person (i.e. idiots like me) completely destroy any narrative impact. What’s that, an important NPC died/you trusted the wrong person/didn’t save the town? Well why not just go back in time and fix it like you normally do? Oh, I see, Mr Selfish only wants to turn back the clock to prevent his own death! Screw everybody else!

The latter may be an extreme view, but the dramatic discord between a character who can cheat death, and a story with enforced tragedy in it, can really undermine a dramatic narrative. The most obvious example of this is from Final Fantasy VII. Beloved by many, maligned by some, the most famous scene in this game requires from the player a huge leap of logic – the idea that your magical resurrection potions can work for any wound except for this specific one. Of course, for many it was a shocking and moving scene, but for many others the conflict between in-game logic and in-plot logic really niggled at them, undermining the whole scene. For a few, it destroyed their suspension of disbelief entirely (I was a niggler, rather than a shatterer. No, I got sick of the game purely because of its massive linearity and sodding random battles – but that’s a complaint that’s had its day on many forums across the land).

And to remind ourselves: even if dramatic discord isn’t an issue, there’s still the problem of games where quickly saving and reloading is so prevalent that any sense of danger is lost; this is a very real challenge that developers struggle to meet.

"What's that Aeris? I should give you a phoenix what?"

So how do we solve these problems? Easy! Well, actually, not easy. But if you’re a developer creating a game that you want to be driven by a strong, dramatic story, you really must live by the rule that gives this article its title: give me consequence, or give me death. Let’s break this down into those two options:


You have a great idea for a story! It’ll have twists and turns, and then there’s this tragic event that occurs that’ll really drive the player on! Gosh, this is going to be brilliant. Good for you, sport. But how are you going to fit this into the game? Are you going to shoehorn the player into this tragedy without any input from them? If you do, it had better either be so massive or inevitable that they have no conceivable way of stopping it. For example, let’s say you want to have some evil group kill close friends of the game character, giving the player motivation for fighting them. Don’t take this out of the player’s control, or worse, have it happen off screen. Get the player involved: one idea would be to have your town invaded by a huge force – don’t put the player in an unwinnable situation, or strip their control away. Allow them to defend an area successfully, some of the people they care about. Just not all of them – the player can’t physically defend every homestead. They can choose to focus their defence on a small area and save some people, but other characters will die. Or even let them try and defend everyone – and realise that because of their choice even more people have died. That’s a tragedy the player will engage with.

The Witcher games use this technique particularly well – forcing the player to make choices throughout the game which will lead to non-ideal outcomes. Tragedies will occur, no matter what you do, but the manner of tragedy will be dictated by your actions. Allow the player to take responsibility for the events which occur in your story and they will be all the more involved.

The problem with this approach is the obvious, and depressing one: time and money. For any branch to the narrative, you’re increasing your costs for scriptwriting and voice-acting exponentially. Even if you only created a single branch, following binary plot arcs, you’ve still added cost and time to your development. But if you want to create a satisfying, plot-driven, dramatic game that takes advantage of its medium and doesn’t leave its players with a sense of disconnect, you cannot ignore the benefits. The trick is to convince your publisher that it’s worth the effort, and that, well, that’s a challenge I’m not sure how to help with, other than to point to the resounding success of games like The Witcher, and the Mass Effect series, and the loyalty of those series’ fans.

Geralt, probably about to make yet another decision he'll live to regret.

But what about the issue of danger? How do you make the player fear their own mistakes in game? You can leave them with consequences to failure beyond having to reload. The Fable games tried this, with your character accruing scars for every death, but of course they combined this with the ability to reload and annul those scars anyway, thus undermining the system somewhat. A game with consequences that are irreversible – either literally, or practically – however, can be a powerful thing. Again, the Witcher series does this, but only for the choices the player makes, rather than their ability in combat: and it makes annulling those consequences practically impossible by having the effects felt many hours of gameplay after the event. This technique can work well as a deterrent for failure – if you never know when your failures might have repercussions, every battle becomes more tense. This worked wonders for the (hugely) flawed gem, Alpha Protocol, where the way you engaged in your missions could have all sorts of unforeseen consequences: on people’s opinions of you, on global events, and on the ultimate conclusion of the game. Perhaps most relevantly (to those of you watching your bank balances, particularly), this is key to the success of grand strategy games like the Total War series. And to a little indie game you might have heard of: it’s called Minecraft.

Again, this will almost quite possibly involve extra writing and dialogue to implement (or at least make much of your dialogue be potentially missed out on by a lot of players if you’re killing NPCs off as a consequence), but it’s a fantastic way to add tension to your gameplay.

And should neither of those appeal to you, there’s a third, and very much final solution.


By far the technique with the most strings attached. This is a method of adding tension which will automatically limit your audience, and so is only useful for developers with a very specific idea of what they want to create, and who they’re creating it for. The technique is simple to understand: when the player is dead, their character is dead. Permanently. No reloading, the only option is to restart with a new character. This is the approach used by roguelikes, and remains the prevail of niche titles: games like the giant mech simulator Steel Battalion (best known for its gigantic controller of doom), and the Way of the Samurai series.

The aforementioned gigantic controller of doom. No, this has nothing to do with death or consequences. Except maybe the consequences of spending £100+ on a joystick, which are that you'll have an awesome joystick.

One thing is for certain: it works. There is little more terrifying to a gamer than the idea of permanent death for their character: it means that one mistake, one slight error, can cause the loss of hours, even days of gaming. As such, this is a system which works best on relatively short games, or games without an end – few gamers would be willing to play a 100 hour epic where the slightest error along the way would mean starting from the very beginning.

It also allow for interesting ways of dealing with the player’s death: the most obvious one is to simply require they start from scratch, as if they’d never played before. Roguelikes, of course, randomly generate their world each time, so each death means a completely new game. An alternative is to have a fixed world in which your new character is literally a new character, wandering in after the past character’s death: meaning that each character becomes a part of the history of the game world. More commonly, you’ll see a milder variant of this, where successive playthroughs unlock new opportunities, and/or make the game easier for later characters, encouraging multiple attempts.

This is both the easiest (in terms of development) and hardest (in terms of game balance) method of adding real tension to your game, and when done well it can have spectacular results. Just ask the many fans of Spelunky.

And that’s it, that’s my piece said. Just to reiterate, I don’t think these approaches are required for every game. But if your game lacks a little tension, or if you want a dramatic storyline that doesn’t go off like a damp squib – you might want to take these approaches into serious consideration, and help make your game even greater.


0 Responses to “Give me consequence or give me death!”

  1. Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


July 2011
« Jun   Aug »

%d bloggers like this: