Sudden death in games?

I need a bit of advice here.
I’m redoing my game “D-Day” and would like to know what people think about “sudden” death in a game.
Since it’s a game about the invasion in Normandy and at Omaha Beach at that, it goes without saying that getting killed will play a major part in the game.

One of the golden rules of IF is that you don’t kill the player right away but allow him/her to be caught by the game. The problem in this case is that the game begins at the beach and as history has told us, it was a pretty dangerous place to say the least.

My plan is to set up the game so as long as the player moves on the beach s/he will have a certain percentage chance of getting killed. Say in the range of 25 to 50%. Would that be acceptable or do you think the player will drop the game?
I will hurry and say that there is of course a way to move safe from the beach, and there are clues pointing towards that.

I can imagine that if there’s a safe method, which is clued, and there are indications you’re doing it safely when you are - and conversely, if you’re not doing it safely, that there’s a 25-50% chance of death each turn, and signs that you’re not moving safely (apart from being shot?), I wouldn’t have any objection to being killed and undoing even in the first move. In short, so long as you as the designer work out a scheme for this sequence that will work for most people most of the time. If it appears to be completely random to a majority of players, that perception is as good as the game just being totally random, which is presumably what you want to avoid.

  • Wade

Depends on the clues, I’d say. If I start the game, try to get my bearings and get killed every other move, I’d drop that game like a piece of hot coal and never look back. There are other – and better – ways to build tension and excitement than being realistic (and how realistic is starting over and doing the same stuff again anyway?). Describe the smoke, the noise, the confusion, the fear, the adrenaline rush and print random messages about bullets whizzing past, that sort of thing. Letting the player enjoy the experience is always a better bet than giving him realistic odds.

If you consider moving from the beach a puzzle (and it seems to me that you do), maybe having the player character take cover and refuse to move instead of killing him outright could work (for the first time, at least). Make the clues more obvious. If he still doesn’t pay attention to them and continues to move, then by all means kill him.

Always remember the killing the player too often is a nuisance, it doesn’t reellay help immersion, it reduces your game to a meta-gaming experience (you’ll end up with players undoing or save-restoring their way from the beach).

I like what MattW said.

In such a game I don’t think a raw chance of dying every turn would be necessarily wrong, but the percentage would have to depend on how long the player will be there on the beach. For example, I might want players to stay on the beach 15 turns before progressing to the next area, and I’d want 95% of players to die a maximum of 3 times before they could progress (i.e, 5% of players would be unlucky and die 4 times or more.) That would make the each turn % be… well I don’t know. It’s too late for figuring out stats like that! But do you get the idea? Percentage changes like this are cumulative, so if you want them to be in the area for very long you’ll need to start with only 1%, or maybe even less.

It’s really hard to explain properly when you don’t want to give away too much of the game :blush:
But the game starts in one of the Higgins boats on the way to the beach. The player disembarks and heads towards the beach. So far he is relatively safe.
Once he enters the beach is when the trouble starts. The beach consists of 3 locations and north of those is a sea wall also consisting of 3 locations.
What I had in mind was setting a percentage if the player moves sideways along the beach, and another if s/he moves across the beach towards the sea wall.
It is in these circumstances that the player has the biggest chance of getting hit. As soon as s/he is in cover of the sea wall and moves further in land things would run relatively smooth.

I like what I said :slight_smile:

We can already see in this topic two different attitudes of how the idea would be received in theory.

I don’t mind undoing from being killed. If the stakes are high and I can maybe learn or optimise from it, it can feel right. It gives me the sense that I didn’t miraculously survive the incredibly dangerous situation, even though I do practically get to progress thanks to game magic.

If you can’t kill the player during the beach landing of Normandy, when can you? Some will say ‘never!’ But I say when appropriate and when you want to, so long as you’ve got your reasons and have actually designed the game so the purpose is not just frustration and there is a relevant effect from the killing.

  • Wade

(emphasis mine)

I don’t mind undoing from being killed – once. If being killed once doesn’t give me anything to reduce the chance of being killed again, that’s a problem. Sections where you have to UNDO over and over again are lethal to a game. (Do some testing. See how many times the average player gets killed; adjust your probabilities accordingly.)

Questions of design don’t exist in abstract isolation. Whether you could get away with this would rely a great deal, for instance, on how smooth the play experience is on the beach; if the sort of actions available aren’t clear to the player, if a reasonable synonym’s missing, if a puzzle relies on a single abstruse solution, then the random-death thing becomes a lot worse.

Also, the answer to any abstract “is it okay to do this?” IF design question: “how well can you write it?” Random death, however you spin it, is a reason for the player to leave. If you include it, you need to be damn sure that you’ve given them plenty of reasons to stick around – and since this is pretty early in the game, you don’t have much time to show them those reasons.

The other way that prose matters here is that the more immersive the experience, the more real the writing feels, the more likely players are to accept getting randomly shot. If the writing’s flat, I’m more inclined to treat the game as a game, and therefore regard random death as a mechanical annoyance and a bad design element; if the writing draws me into the world, I’m more likely to think of random death as something appropriate to the world I’m in.

I would suggest giving the player an increasing chance of death the longer they stay still. If they’re in a location one or two turns they’re good. Every turn after that they don’t move increase the chance of death by 25% and let the death message mention that they’re likely to stay alive if they keep moving.

I’m not sure but I’m guessing that in your game when your PC dies you restart. I just want to point out it doesn’t have to be that way.

You could do it like they do it in Mass Effect, where the player is taking damage until they take cover, then they recover. If they stay in the open, they die.

There is a certain balance between realism and playability. You may have to remove sudden death from your game to make it playable, or sacrifice replay value. Sudden death is going to make it very frustrating.

What if instead of death as most games do it, “respawn” the player to another location, as if they are suddenly a different soldier?

I think that doing this in a serious game about WW2 would be difficult to pull off. RPG and FPS games have an established (hence largely invisible) assumption that a single bullet probably won’t kill you, and that anything short of death doesn’t make you any less capable. In IF this is much less conventional, and thus more likely to look silly. I’d rather have the player magically protected by The Plot than give them hit points and the assurance of surviving the first bullet to the head.
And, not to defend a petty realism, but if you care enough about the feeling of realism to introduce the (problematic) feature of the player getting randomly shot, you probably don’t want to imperil it by having the player take several bullets and keep going.

Didn’t Duncan Bowsman do this in To End All Wars? He was talking about it at one point, although I don’t know if it made it into the released game.

Those are valid points. In Mass Effect, it’s largely based on “shields” that take damage for you, and armor technologies that can “heal” you. That’s not very realistic in a WWII-based game. Unless, of course, the player character wasn’t human or was some sort of time traveller.

Nope, he’s not… Just a plain GI heading towards the beaches. :wink:

You could set up game difficulty levels (much like I did in “Menagerie!”) where at the novice level, sudden death was completely disabled, at normal level there were “hints” to avoid death, and at hard level there were no hints and death could come at any moment.

All the death (that and a lot of other stuff) made Menagerie feel pretty silly to me. What’s the point of showing that the circus is mistreating animals? The place is going to get sued into oblivion for its blatant lack of any safety standards.

LOL. I never thought of it that way. I was trying to write a horror game.

As we’re talking about an introductory scene, I’m going to break a major ludological rule here and advise you put the opening scene on rails: make players’ choice appear valid but give (variations of) the same harrowing, barely-successful plot-unfolding setup regardless of what their input actually is. Y’know, over the next three moves the following three events happen, perhaps in a randomized order, making reference to the player’s choice without it actually having been relevant. CYOAs often do this.

I think the on-rails approach can work if accompanied by a rich and rewarding interactive framework- the player wont mind so much not being able to effect the outcome of events on the broadbrush-scale (reasonable in the context of the invasion) if they’re rewarded by lots of things to look at and good thematic responses to reasonable actions that they might try. I think if you’re going to sacrifice one kind of interactivity, you should play to the medium’s strengths and offer another kind.

I felt that it would be legit to have sudden death in this scenario.
I mean, the player is on the beach without much cover other than his dead comrades and the wrecked equipment.
Like I mentioned earlier, the beach is made up of 3 locations. As long as the player stays in any of these locations he is safe. He can do whatever he pleases here, examining things, getting objects etc. The danger comes when he moves from one location to the other. With the Germans throwing everything, including the kitchen sink, at the invaders I found it only natural that there should be a certain percentage that he would get killed.

Well, I can see the motivation: if there’s no consequences, the player could dawdle indefinitely on the beach and make a little holiday out of it. How about describing increasingly close calls when you move between locations? This would lend a sense of urgency to the matter without subjecting the player to endless instadeaths. You could still have them die eventually if they go back and forth too many times.