 |

|
 |

| |
Opinion: Rethinking Player Death
by Brandon Sheffield [PC, Console/PC, Exclusive]
|
|
| |
|
November 20, 2009
|
| |
[In an editorial originally published in Game Developer magazine's November 2009 issue, editor-in-chief Brandon Sheffield considers whether traditional player death conventions have worn out their welcome, asking "why is death even part of the equation?" for games if meaningful consequences aren't built in.]
Arcades have brought us a lot of significant advances over the years. From the industry's beginning through the mid-'90s, arcades were still where you’d find the best game graphics, and the best local multiplayer experiences.
A great many excellent design rules and guidelines were forged in these fires. Every so often though, I notice a trope carried over from the arcade days that just doesn’t seem to fit anymore. One of those is the concept of lives, continues, and player death.
The thought occurred to me while playing Teenage Mutant Ninja Turtles: Turtles in Time Re-Shelled for Xbox Live Arcade. I played with three friends, and aside from none of us being able to tell which character we were about 75 percent of the time, I noticed that the game defaulted to unlimited continues.
Playing essentially amounted to mashing the attack button and hoping for the best, with no real consequences to death, though the game did keep a tally of who among my friends had died the most times.
It really stuck with me -- in a scenario in which death essentially means nothing, why have death at all?
Granted, this was a port of an arcade game, but a number of kids’ games operate under a similar basic principle. Dying either places you right back where you were, or it does so until you run out of lives, and then you continue and start at the beginning of the level, lives fully restocked.
The game is basically testing your ability to complete the same actions again and again, rather than your skill. Except in outlying cases, it’s testing your willingness to persevere, and not to adapt.
Rise From Your Grave
This extends in a mild way to the checkpoint systems in modern games. Most triple-A games have rid themselves of the idea of continues, or even the concept of limited lives, but death is still not so much a punishment as it is a setback—you simply lose a few minutes’ playing time, and probably learn some strategies in the meantime.
So why represent this as “death,” rather than in some other way? It could well be because we’ve always done it that way, rather than for any reason anyone spent time thinking about.
Demon’s Souls is going to be a hot topic discussion among alternative journalists and academics for some time, perhaps rightly so. The way that game deals with death is well thought out, and actually has an in-world reason behind it. If you die, your (weaker) soul must go out in search of demon souls with which to reclaim your physical body.
In Assassin’s Creed, “death” is explained as a de-syncing of the player from his host body in history. In Prey, players must fight their way back from the valley of death, to reclaim their place among the living. Even Silicon Knights’ drawn-out resurrection sequences in Too Human are conceptually relevant.
So many games employ outlandish sci-fi or fantasy scenarios that it seems death could be explained away in simple terms -- or, even better, with some entertaining gameplay.
Continues and their progeny are usually not a nuisance. They just seem unnecessary, evidence of the early framework around which games have evolved. When a developer intends it, player deaths can be entertaining, they just need to be given weight.
If you don’t care if you die, the stakes of play seem rather minor. In Left 4 Dead, you feel like you’re letting your teammates down, so death has weight. But in Gears of War, which I otherwise like very much, death is just an annoying setback, and I have to watch cut scenes again and traipse back across the same shattered landscape just to get killed again by the same stupid Troika.
Death can be a good time to provide positive feedback for players, and add tutorial hints. Considering the fact that games are meant to be won, the player is essentially invincible in the grand scheme of things, requiring, again, nothing more than perseverance and the occasional GameFAQs hint. Why not try to create a game that contains similar challenges to standard games, but completely avoids death?
If the end result is the same, and meaningful (or humorous) consequences aren’t built in, why is death even part of the equation? It isn’t really death, after all, it’s really just the proverbial “flesh wound.”
These days we have a buffer against death in games anyway: Most first- and third-person shooters have regenerative health, whether it makes sense for the world or not. Once the player’s regenerative health is depleted, they crumple to the ground momentarily, only to have time magically rewound, with players finding themselves transported to a very familiar (usually identically instanced) checkpoint save state from their recent past.
The Death (Of My Interest)
Of course if you’re not careful, alternate methods of “death” can be even more annoying. Lengthy resurrection sequences (in fact, cut scenes of any kind) will take the player out of the action. Death- or loss-related minigames need to be fun and relevant. Almost any time not playing the game is time waiting to play the game—and that’s generally time wasted.
Innovative ways of dealing with player death aren’t just for the wacky indies, or the fringe titles, which is why I deliberately didn’t mention any. High-end games are starting to change the way they deal with what lies beyond, and it seems high time that more developers start to confront their own (in-game) mortality.
|
| |
|
|
Players are expecting a lot more out of a game and, because of the price tag, both the publishers and the end-users want to be able to see as much as they can.
The question that something like this brings up in my head is: With digital distribution methods in place, why aren't there more developers returning to the arcade/gambling model?
It is a good point though, I think death sticks around simply for those that refuse to accept games can be done without it. Bioshock's complaints spring to mind...
I think death is fine as long as there is an incentive not to die. Games like Demon's Souls and Dead Rising do it right. Games like Bioshock (which I loved as a whole) do it wrong, as there is really no penalty for death which is the point of the article. While many games do need to get away from the current death model, there will always be games where it will have it's place, most recently the new Super Mario - admittedly this is a "retro throwback" game, but it still feels refreshing to play something where a "free guy" actually means something.
I didn't play that game but I was thinking about it for a long time, such a game where you play many characters, also in both sides of a conflict in a net plot. And if you win or lose, or half-win, or escape, or can't save somebody, or can't steal or sabotage something, then the games and story goes on, with the consequences of the last facts.
Typically the objection is that one-time event driven programming would be prohibitive in difficulty, and scale, but I think thats not true, it is I believe possible to craft regenerative environments, npcs, etc that can add a sense of one-time rareness to a game, that makes it worthwhile as a story experience, and not merely a twitch-reflex experience.
Storytelling is about mortality, we remember, and tell stories because those things have passed on, and stories are about remembering and savoring the experience. Games that allow you to just reload and try again tend to leave you feeling that after all the effort to master the game, its all lost time, and has no meaning, since nobody cares about the story that ensued, everyone who 'beats the game', has the same experience more or less, and doesn't need to bother retelling their version.
I always wondered when a game would implement a "reincarnation" type of death feature? Changing player stats or even character types if a player dies a lot or dies trying a scenario much higher than his current stats should have allowed (ie: rewarding bravery).
Too many games are designed with death as a minor inconvenience with little if not any really consequence. Yet, with the casual market growing, people are complaining L4D2 is too hard now and death is too common on normal. We want to have fun and be challenged but don't want to be frustrated if death comes too often.
tom - baroque could have influenced demon's souls, but do keep in mind that baroque was by Sting, and demon's souls by From Software...they were both just published by Atlus in the U.S.
and to your earlier point, I agree that death with consequences is interesting and motivating - just hard to come across!
As a tweak of the "reincarnation" notion, what if, rather than being a punishment, dying imparted some kind of "wisdom from beyond" to the next iteration of the player character? This could become an interesting choice: do you reload and try again with no help? Or do you accept death in order to gain the minor assistance from your former incarnation? [Edit: This does sound a bit like Baroque, mentioned above.]
2. Related to the question of character death is "permadeath" -- the complete elimination of a character, with all his attributes and abilities. In RPGs, permadeath has become a "third rail" issue because no developer is willing to take away the progress of tens or hundreds of hours of play. But that's a side effect of thinking that RPGs must always include character advancement as a design element, and that's simply not true.
Traveller, for example, features a very detailed character building system, but virtually no improvement of attributes or skills during gameplay. Characters can possess resources such as knowledge or equipment, but those things can be "willed" to new characters if the original dies. So where are the CRPGs or MMORPGs that can offer permadeath because they don't mindlessly clone the convention of character advancement?
3. I can't resist this opportunity to promote a CRPG or MMORPG based on Paranoia. The notion of character death is already baked into this RPG: everyone has six clones; when you die, a new clone is (eventually) delivered to the group. This wouldn't be a general solution to player character death, but it sure would be interesting to see it tried in the specific case of a Paranoia MMORPG. I'd be very interested in seeing how gamers would react to that death mechanic.
I'm interested in permadeath as well, but it'd have to be very well measured. Valkyria Chronicles, and a number of other S/TRPGs have permanent death for subcharacters, with the understanding that you *can* save them, it just requires effort. Unfortunately that tends to just inspire people to save/reload when they lose somebody.
I guess in the end it comes down to finding an appropriate risk / reward scheme for playing good or bad. Does anybody have an overview of what has been done by genre in this respect ?
Another approach to death is the style used in games such as "Way of the Samurai" or "Dead Rising." The game itself takes place in a very limited time line, and it is understood that you simply cannot do everything in a single run. Death is death, and you do not come back from it. Instead, you restart the game and give it another go, but with increased stats based on what you have done in previous runs.
I'm surprised that no one mentioned "Prince of Persia." "Death" in the most recent installment of PoP wasn't death at all, but "death narrowly avoided, thanks to the princess." Whether that's a good approach or not is up to the end user.
On a recent kids MMO project I had designed a Death Mode, which you entered into ...by dying. While Dead, you existed as a Ghost until your Rezz timer wound down. As a Ghost, you could move freely in the game world without constraint of physics or collisions , giving you the ability to scout ahead or watch your friends. You could speak, but others would only hear a disembodied "wooo".
(A special talent for certain classes would allow them to shorten your Rezz timer, which might not be what you wanted.)
Your eventual Rezz would happen at the latest checkpoint your group had unlocked.
This allowed players to stay active, but non interactive with the game world. No doubt some players would be able to exploit the scouting ability of the Ghost effectively, making Death something they would seek out on occassion.
Even having said that though, there still needs to be elements designed within the game to keep you alive most of the time so that permadeath becomes a rare event. For example, if the game has PvP with nations at war then healing and raising the dead would be a critical support component of an army. BTW by designing this into the game you actually help to put barriers on territorial control which is a good thing. In effect, the more territory a nation tries to control, the less they are able to support and defend those areas, thus they become much easier target for attack. In effect, wars and death should be costly and have consequences to it which is why other means such as diplomacy and trading can be better alternatives.
All said and done, it's about creating interdependencies between the gameplay elements so that an initial perceived weakness can be design into the game so it becomes a strength. But ya, you obviously couldn't drop permadeath into a game like WoW because it's current design couldn't support it.
"If I knew when I was going to die what would I do?"
This is an interesting premise for a game; if death was inevitable then the challenge would be to enrich your life that precedes that final moment. The quality of your "living" game experience can change the value of your death experience.
Alternative structures not tied to character development can be compelling and it would be nice if more developers were willing to experiment.
If your goal is to live and you die, you're given a clear indication of your performance. As mentioned in the article, that is player feedback. And even if inconsequential, giving the player that method of performance evaluation is very important.
While those "Look at all dat juice" moments may be an annoyance or frustrating, they aren't by default a bad thing. They can be executed badly, but on the other hand, they can be executed well and even become charming. "I hate vans", anyone?
I am a proponent of death. Even if there is no real danger underlying, it is still akin to the old playground game "Don't step in the lava!" and will still continue to aspire some/most players to loosen up, try again and avoid the mistakes they just made. By learning what not to do, players will continue to learn what to do instead.
Also, regarding Valkyria Chronicles, I perceive the reset/reload as something different. I am a huge fan of Valkyria, and even though I was guilty of the reset/reload when a character died... it really changed the experience for me than, say, Shining Force where you revive the character after battle and continue on.
It helped me feel more like the prodigy son of the late, famed General. Pacifist turned brilliant commander Welkin Gunther, a young man who dons his country's uniform and follows in his late father's footsteps to lead his homeland to victory defending against an aggressive and oppressive invader. His service to the militia is solely focused on restoring peace to his home, to return things to the way they were. The desire for peace is the source of his strength.
It wasn't just the "Oh crap, my Scout just died and I can't save her. Guess I'll do it over again..." It became more than that. My actions were tightly governed by the understanding that my troops' lives were in my hands and that I could permanently kill them. So my strategies shifted to more align on the cautious side. The bottom line was no one would die, so advance troops forward in groups, do so intelligently and actually utilize cover while in control of the character. Leave no one behind. (Or in this case, too far forward!)
Valkyria Chronicles death changed how I played the game and did so for the better. I got the satisfaction of victory without the loss of my soldiers. My companions and fellow citizens were not unduly wasted as the army's General was prone to. I cared for them, and as a result, I found greater satisfaction through the new play style.
Especially my victory in Chapter 11... it could NOT have been any closer of a victory! It really came down to the riskiest play with the greatest payoff and the closest win in any game I think I've ever played. That was an amazing battle.
That's definitely possible, Luis. I sometimes wonder if we've reached a point in MMORPG design where too many players actually believe that character advancement is a defining feature of the genre, and it's just no longer commercially feasible to do anything else.
On the other hand, maybe that means the time is right for a high-quality, content-rich MMORPG in which players don't have to grind for months to reach the "endgame." In a MMORPG without character advancement, it's *all* endgame. :)
But in a single-player CRPG, I'd love to see someone run with Carlo's idea. How would you play a game if you knew without question that at the end of all your struggling was the extinction of your character?
Would that existential crisis make you feel free to do anything, take anything, be as bad as you wanna be?
Or would you act through your character as though ethical behavior still had value?
Of course, when it comes to single player games, I'm reminded of Steel Battalion, where they erase your player data if you die (or don't eject in time).
Of course, the point of both of these examples are to point out that death does matter in these games, and make them a little more involving. However, one of things that has been lost in modern gaming, compared to the games of the 80's, is the idea of starting the whole game over if you do lose (yeah, they were more harsh back then).
I have thought about it before, the first thing that comes to mind is how the game is played. When things becore "real life", it starts to give items and weapons more value, actions, strategy, planning, coop... I opens place for hiring a soldier to guide to another city and protect you, also many evil plans and playing start to rise, lies and level-hidding, etc...
I think the best setting to make such a game would be something in the patterns of SinCity, or similar, with ground for social and political pvp, things we not see too often... Ethical and moral behaviour also take a high place.
Without the leveling up thing, money gets more value and becomes the reason why player go abroad face challenges. It starts to change somehow with the player becoming the prey, and the mobs become the hunters... stealth and lightweight comes on.
I remember playing a game where you advanced to the right, but when you died, your controlled your spirit which had to advance to the left a certain length to revive. In other words, you were forced to backtrack, beating more powerful enemies than the ones you would find while "living". I found the concept pretty interesting.
Well! Two games do that, one out now and one in a few months. Both by BioWare.
Dragon Age: Origins has that choice, trying not to spoil anything here but near the end you will be given an opportunity to avoid sacrificing anyone at all, and even later (if you did not take the opportunity) you get another, to choose someone to be sacrificed, and again later you can choose if it should be someone else or yourself.
This will affect the ending dramatically I assume.
The other game is Mass Effect 2 and the facinating thing here is that you can carry over your savegame from Mass Effect 1 with all your dialog choices, decisions and who was alive or not etc. And as I assume most knows, it is possible for the main character from Mass Effect to die in Mass Effect 2,
meaning all those choices and decisions and whom you romanced (if any) will be severed, but the Mass Effect 2 story will continue apparently. But you will still keep wondering if that romance would have gone further (kids/marriage?) if the sacrifices you made could have made a difference etc.
In both these games most veteran RPG players knows to branch their savegames at certain times so they can go and explore one path or another in the story. With Mass Effect trilogy due to the savegames being carried over (if you want) it's a mini logistic nightmare, and the "Origins" in Dragon Age likewise as who you are will actually impact the ending (by giving certain choices you can make or not make at th end)
I'm rather surprised that BioWare are behind both these, so I can only assume that BioWare are one of the pioneers in experimenting with things like this, I only hope others follow suit. Oh and BioWare has a knack for tying things pretty damn well into the story.
Though those are scripted deaths, both games still has health "death" which is a bit of a shame, being wounded and having to recover skills while and making sure to sleep/rest a lot and similar would have been interesting alternative.
Over the years I've cheated a lot by using developer commands like god mode or infinite health and infinite ammo, and if not available I'd hunt for a Trainer. (I'm a PC gamer).
Why? Because dying while fighting the same enemy for the n'th time stops being "cute" pretty damn fast, and the more frustrated you get the more shortcuts you try to make which ends up killing you faster instead.
Ditto with ammo, tired of ammo boxes and stuff laying all around. Would be nice if ammo was tied into the story, so if you are low on ammo or empty then the story changes slightly, likewise if you got a lot of ammo the story change slightly, and if fully empty you get a story "help" to get more.
This is more in line with movies or books where the "hero" always against all odds survive until the end (as per the story in the game obviously although BioWare is toying with that part these days :) and the hero always has enough ammo/weapons/tools to get the job done.
So why you have to try die and retry or go ammo box hunting in so many games is kinda silly.
Lets hope BioWares Dragon Age and Mass Effect series inspires other devs. (and looking at sales figures I think there is a lot of "inspiration" for other developers there :P
While I doubt it was ever thought of in these terms, I think death was used in early computer games as an easily understood risk/reward structure.
However, it's worth bearing in mind that there was no real penalty in the original "death" systems - games like Donkey Kong, Space Invaders and their ilk either dropped you straight back into the action (no direct penalty) or simply took you back to the start of the current level (small direct impact in terms of time). This began to change as games began to introduce the concept of checkpoints (i.e. levels) and powerups. For example, R-Type's "floating satellite" powerup is only available on levels 1 and 4: death on level 2 means that you lose access to this capability until reaching level 4 (and even then, you only get back your lost capability, rather than the dual-satellites!).
This trend has continued to evolve outside of the arcades and as a result, the size and complexity of the gameplay inbetween checkpoints has increased. A level in R-Type lasts around 5 minutes and is fully deterministic; in GTA3, a player may spend 30 minutes getting a car, acquiring weapons and completing a mission before returning to a checkpoint. As a result, death ceased to be an enjoyable element of the risk/reward structure and became an annoyance.
Fast-forward to today and (perhaps at least partly due to the expansion of the "casual" demographic) there's been a number of attempts to tip the balance back. Sands of Time implemented the ability to rewind time - a system taken to it's logical conclusion by Braid. The recent Prince of Persia reboot removed the concept of death altogether, as do Prey and Bioshock. Even on games where death is still present, the cost has been minimised, thanks to autosaves and the like.
However, this can have a negative effect on the gameplay: without the risk, where's the fun in the reward? It's notable that the recent PoP was heavily slated for it's perceived lack of challenge - as was Bioshock (albeit to a lesser degree).
Overall, I think death is a valid mechanism for providing a risk-reward structure, but care has to be taken to balance the pleasure of the challenge with the frustration of losing significant levels of investment...
The main alternative is some form of sports-style rating system (e.g. distance travelled, points scored, time taken, final position) - this doesn't map well to all gaming structures and generally imposes a delay between risk and reward which can be overly abstract and reduce motivation.
It wasn't until Legacy of Kain: Defiance, that Raziel was forced to take the bodies of the dead. It was a great mechanic and I love it
However, that defining (and at the time extremely well received feature) stopped appearing in the series or future spinoffs, and I believe it's because not dying was actually becoming too frustrating. When you can't kill the player, it was felt that you still need to challenge them, so the games that couldn't kill the player usually become more trap-laden and harder. Interestingly, games that "kill" you and force you to retry from a previous checkpoint/start of the level don't feel as punishing to most players. I think it has to do with the psychology/mindset of the player more than anything; a kill stops the action, gives the player a moment to collect themselves, then allows them to try again, while just failing and needing to retrace steps while still in the gameplay (for the nth repeated time, in some cases), just ends up feeling more aggravating.
The handing of lives in the New Super Mario Bros Wii seems to be handled quite well for multiplayer. When you die, float on screen in a bubble a few seconds later, and another player just has to touch you to come back. You still have lives, and when you run out, you don't respawn and have to wait until the end of the level to get another 5 back (using a "continue" that you have infinite supplies for, but it keeps track of how many you use). This, combined with the fact that you're still punished for having everyone dead at the same time (you'll get booted out of the level) means that getting 1 Ups are still meaningful, while minimizing player downtime from dying too often.
Still, it's good to see that there's many different approaches to either handling death, or avoiding it, as many games cited in this article/comments have brought to life. From the crushing pain of rogue-like permadeaths, to the press A to try again in platformers, to the "what's this death you speak of?" in games like Noby Noby Boy, it's promising to see that we'll always see a variety of approaches, and new ideas still come along every once in a while.
A game which I felt handled death extraordinarily well (while, incidentally, still conforming to the "limited lives" scheme familiar to all) was "Faery Tale Adventure." In that game, you played one of three brothers. You start as the eldest, strongest of the brothers. If you died, then you instead took on the role of the middle brother, the clever one. If you died again, then you instead played the role of the kind youngest brother.
Some people never played anyone but the eldest. In the case of my Dad, he died twice almost right off, then took the youngest one through to completion of the game. It was a neat mechanic. You had three lives with which to play the game, but each life would be a little different.
I think Nis Bojin's work is very relevant to this notion of concepts that "lock" in place within game design, even those detrimental to the experience.
http://www.eludamos.org/index.php/eludamos/article/view/21