(cross-posted from my blog)

I. PvE vs PvP

Ever since it’s advent in Doom, PvP (Player vs Player) has been an integral part of almost every major video game. This is annoying to PvE (Player vs Environment) fans like myself, especially when PvE mechanics are altered (read: simplified and degraded) for the purpose of accommodating the PvP game play. Even in games which are ostensibly about the story & world, rather than direct player-on-player competition.

The reason for this comes down to simple math. PvE content is expensive to make. An hour of game play can take many dozens, or nowadays even hundreds, of man-hours of labor to produce. And once you’ve completed a PvE game, you’re done with it. There’s nothing else, you’ve reached “The End”, congrats. You can replay it a few times if you really loved it, like re-reading a book, but the content is the same. MMORGs recycle content by forcing you to grind bosses many times before you can move on to the next one, but that’s as fun as the word “grind” makes it sound. At that point people are there more for the social aspect and the occasional high than the core gameplay itself.

PvP “content”, OTOH, generates itself. Other humans keep learning and getting better and improvising new tactics. Every encounter has the potential to be new and exciting, and they always come with the rush of triumphing over another person (or the crush of losing to the same).

But much more to the point – In PvE potentially everyone can make it into the halls of “Finished The Game;” and if everyone is special, no one is. PvP has a very small elite – there can only be one #1 player, and people are always scrabbling for that position, or defending it. PvP harnesses our status-seeking instinct to get us to provide challenges for each other rather than forcing the game developers to develop new challenges for us. It’s far more cost effective, and a single man-hour of labor can produce hundreds or thousands of hours of game play. StarCraft  continued to be played at a massive level for 12 years after its release, until it was replaced with StarCraft II.

So if you want to keep people occupied for a looooong time without running out of game-world, focus on PvP

II. Science as PvE

In the distant past (in internet time) I commented at LessWrong that discovering new aspects of reality was exciting and filled me with awe and wonder and the normal “Science is Awesome” applause lights (and yes, I still feel that way). And I sneered at the status-grubbing of politicians and administrators and basically everyone that we in nerd culture disliked in high school. How temporary and near-sighted! How zero-sum (and often negative-sum!), draining resources we could use for actual positive-sum efforts like exploration and research! A pox on their houses!

Someone replied, asking why anyone should care about the minutia of lifeless, non-agenty forces? How could anyone expend so much of their mental efforts on such trivia when there are these complex, elaborate status games one can play instead? Feints and countermoves and gambits and evasions, with hidden score-keeping and persistent reputation effects… and that’s just the first layer! The subtle ballet of interaction is difficult even to watch, and when you get billions of dancers interacting it can be the most exhilarating experience of all.

This was the first time I’d ever been confronted with status-behavior as anything other than wasteful. Of course I rejected it at first, because no one is allowed to win arguments in real time. But it stuck with me. I now see the game play, and it is intricate. It puts Playing At The Next Level in a whole new perspective. It is the constant refinement and challenge and lack of a final completion-condition that is the heart of PvP. Human status games are the PvP of real life.

Which, by extension of the metaphor, makes Scientific Progress the PvE of real life. Which makes sense. It is us versus the environment in the most literal sense. It is content that was provided to us, rather than what we make ourselves. And it is limited – in theory we could some day learn everything that there is to learn.

III. The Best of All Possible Worlds

I’ve mentioned a few times I have difficulty accepting reality as real. Say you were trying to keep a limitless number of humans happy and occupied for an unbounded amount of time. You provide them PvE content to get them started. But you don’t want the PvE content to be their primary focus, both because they’ll eventually run out of it, and also because once they’ve completely cracked it there’s a good chance they’ll realize they’re in a simulation. You know that PvP is a good substitute for PvE for most people, often a superior one, and that PvP can get recursively more complex and intricate without limit and keep the humans endlessly occupied and happy, as long as their neuro-architecture is right. It’d be really great if they happened to evolve in a way that made status-seeking extremely pleasurable for the majority of the species, even if that did mean that the ones losing badly were constantly miserable regardless of their objective well-being. This would mean far, far more lives could be lived and enjoyed without running out of content than would otherwise be possible.

IV. Implications for CEV

It’s said that the Coherent Extrapolated Volition is “our wish if we knew more, thought faster, were more the people we wished to be, hard grown up farther together.” This implies a resolution to many conflicts. No more endless bickering about whether the Red Tribe is racist or the Blue Tribe is arrogant pricks. A more unified way of looking at the world that breaks down those conceptual conflicts. But if PvP play really is an integral part of the human experience, a true CEV would notice that, and would preserve these differences instead. To ensure that we always had rival factions sniping at each other over irreconcilable, fundamental disagreements in how reality should be approached and how problems should be solved. To forever keep partisan politics as part of the human condition, so we have this dance to enjoy. Stripping it out would be akin to removing humanity’s love of music, because dancing inefficiently consumes great amounts of energy just so we can end up where we started.

Carl von Clausewitz famously said “War is the continuation of politics by other means.”  The correlate of “Politics is the continuation of war by other means” has already been proposed. It is not unreasonable to speculate that in the grim darkness of the far future, there is only war continued by other means. Which, all things considered, is greatly preferable to actual war. As long as people like Scott are around to try to keep things somewhat civil and preventing an escalation into violence, this may not be terrible.


48 comments, sorted by Click to highlight new comments since: Today at 10:49 PM
New Comment

You don't run out of PvE content in games where players produce the content. The major contemporary example is Minecraft.

Another example relevant to this post is real life.

There are PvE elements early in some Minecraft game types, but once they're overcome, or if you pick a game type that disables them, the major challenge becomes building things that are impressive to you or to other players. If I had to classify that as anything in this typology it'd be PvP, but I actually think it's reflecting something orthogonal to it, more along the lines of the game vs. toy distinction. (Game: Doom. Toy: SimCity.)

One thing Minecraft does do to stretch its PvE content is procedural generation, elsewhere associated with the Roguelike genre and its relatives (Diablo, Torchlight, etc.)

Yes, the toy vs. game distinction is a highly useful one, though I read the OP as talking more about entertainment software rather than about games (not toys).

Procedural generation helps, but at the moment it's still limited. A different pattern of corridors with random mobs from a predefined set doesn't generate that much novelty. Diablo is widely acknowledged as a loot collection game, not a world exploration game. However there is a lot of room to grow so in the future I expect procedural generation to create much more interesting worlds.

See also Second Life.

When more computer resources are available, I expect to see games where part of the fun for players is remodeling the environment.

I don't want to accept that the best possible future must contain many unhappy people, because status contests have many losers and few winners. There has to be a better way, like procedurally generating PvE content. Of course if some people like PvP, let them play PvP.

The trick is to have lots and lots of status games in play at once.

Today almost everyone chooses to invest in PvP instead of PvE. Not just society but arguably the human brain is wired to engage in status games, often antagonistic and violent ones. Saying "let them play PvP" is basically saying "let everything stay the same".

But if PvP play really is an integral part of the human experience, a true CEV would notice that, and would preserve these differences instead. To ensure that we always had rival factions sniping at each other over irreconcilable, fundamental disagreements in how reality should be approached and how problems should be solved.

This strikes me as a non sequitur. People self-segregate into rival factions over stupid bullshit as well as irreconcilable, fundamental disagreements, so CEV could retain "PvP play" without having to preserve the deep disagreements.

Yesterday I was using YouTube as a jukebox to listen to Genesis while I was in the office (I'm cool) and in the videos' comments a few Gabriel-era Genesis fans seemed to be getting genuinely angry at the people who preferred Collins-era Genesis. That's the kind of triviality a CEV process could leave us to squabble over after it resolves all of our deep, fundamental conflicts, letting us fulfil our lusts for factionalism and status games. Should be fun!

Nice article. I was a bit unsure about your metaphor at first, but then the simulation part made it worth it.

One thing that occurs is that that while status games (PvP) are an efficient way to keep people occupied (low cost simulation), they're a pretty poor way to optimise hapiness. The majority of writing on hapiness I've seen (psychology/budhism/philosophy of hapiness etc) seems to indicate that one secret to hapiness is to stop playing PvP (and status games). Also, my anecdotal impression of political people, and for that matter heavily-status-focused generally is not one of hapiness. Perhaps that's because in most PvP, like you say, everybody is losing most of the time.

Maybe PvP is like taking drugs: it makes you unhappy most of the time, but it is very difficult to stop. It is something you want at the moment, but would want not to want at the moment of reflection.

PvP skills are strongly selected for by evolution. Being bad at actual fighting and/or status games generally means you are at high risk for losing in the natural selection game.

Well I'd agree that's a force at an individual level. But there's also mechanisms to consider like group-selection (selecting against groups that spend all their time "playing" eachother) and negative reciprocation (punishing people that carry on like arrogant *s :-) )

I like this comparison.

An observation: chess is a PvP game.

In video games, I prefer PvE to PvP and, in fact, largely stopped playing video games because I disliked the focus.

Yet I strongly prefer competitive over cooperative board games.

I haven't figured out why my brain draws this distinction yet.

On further reflection, I like to have control of the pace I play a game. Real-time strategy games against human opponents are stressful, but I don't mind PvP with asynchronous turns. (Even in chess with a clock, I'm not forced to move faster just because my opponent does.)

PvE can stay interesting if there's a hard mode, or challenges, or different approaches, or a strong random element. I have won Nethack several times, but never twice in a row (or even close). The later games were not much less interesting than the first. I have played Final Fantasy 1 through several times with seriously different parties. That worked quite differently each time. For on-going challenge, there are speed-runs in a gradually changing environment (Kingdom of Loathing springs to mind). You can also mod a game much more simply than creating it the first time, and that can change quite a lot.

Now... how does this translate to Science?

Social science would be on a moving target, and so could stay fairly fresh. You can try taking different axioms in math and seeing what you get, that would be sort of a mod and sort of a different approach. I'm not sure what speed runs translate into. Mods would be alternate physics.

So if you want to keep people occupied for a looooong time without running out of game-world, focus on PvP

Or invest in "procedural content generation", where the game world is constantly generated or regenerated. The "roguelike" genre has made games that have been played for decades (like Rogue, Nethack, ADOM) and continues to grow (Ultima Ratio Regum, Dwarf Fortress). It's hybridizing into other genres like action platformers (Rogue Legacy, Spelunky, Risk of Rain). Games are creating new genres by starting with PCG (FTL, Minecraft). Civilization and the Maxis Sim games are classics in large part because of content generation.

For another perspective, game designer Dan Cook has written several http://www.lostgarden.com/2010/12/steambirds-survival-goodbye-handcrafted.html blog posts on PCG leading to better-designed game systems than handcrafted content. Similarly, Jonathan Blow has argued extensively against games that extend their use of systems (eg. across all the levels of a Super Mario, Modern Warfare, or Call of Duty game, the player will see few or no changes in rules, just new sets) rather than exploring a system once thoroughly (Braid, The Witness, Portal, Polarity).

I'll leave the comparisons to "Scientific Progress [as] the PvE of real life" for the simulationists and solipsists. But I've always seen the human obsession with status and gossip as a bug rather than a feature and endeavored to advance more interesting things in the world.

Some assorted thoughts:

  • Virtual PVP games with permadeath (or even progress permaloss) are relatively rare.

  • There's currently no virtual PVP game that allows destruction of the game world, e.g. restarting and wiping a server triggered by an in-game event.

  • Some real-world PVP games (e.g. racing or MMA fighting) have their risks, but injury or death are relatively rare because these games are regulated. The percentage of the population willing to compete in such games is tiny. There must be unregulated PVP games with permadeath, but I'm struggling to imagine them taking place anywhere outside a Colombian prison - and I don't think the participation there is fully voluntary.

  • A CEV implementer can set limits to human conflict. For example, status games, Red vs Blue, bickering and insults are OK, but hurting / killing each other or degrading / destroying the environment are not allowed or impossible. Or, players could simply set the limits of acceptable loss in real-world PVP - or even limit themselves to PVE-only. No doubt there would be 'hardcore PVP characters' of various extent, but I think they would be in a minority.

There must be unregulated PVP games with permadeath

They are usually called "war".

Otherwise I vaguely remember something about Russians selling tickets for a "cruise" during the peak of Somalian piracy. The cruise was on a ship full of small arms (up to bazookas and HMGs, I think) with some special-ops instructors and it cruised off the Horn of Africa with the hope of generating a run-in with the pirates.

They are usually called "war".

Doh. Yes. How could I miss that? War is team PVP with permadeath, but I think we can call it a 'game' only when participation is voluntary, where players join as mercenaries, professional soldiers or militia, not as unwilling conscripts.

Don't forget that "war" is wider than an armed conflict between governments. We can speak of a gang war, for example, which is also PvP with permadeath.

Yes. Another example that comes to mind is conflict between rival groups of hardcore football / soccer ultras.

This reminds me quite a bit of The Metamorphosis of Prime Intellect

I like PvP games because the experience tends to be more dynamic and varied. Even games with relatively solid AI often feel artificial.

This post makes an interesting comparison to Scott's latest.

Notably for purposes of your analogy, 'PvE' content in games in not truly 'environmental' at all, since these games are artifacts produced by humans. When you play PvE game, you are overcoming human-created challenges, and observing human creative output. The difference is that you are meant to 'win', that is, the challenges are designed to be overcome rather than to defeat you. Final Fantasy and Fallout are essentially collaborative experiences between you and the developers, a kind of friendly conversation mediated by technology and financial remuneration.

In other words, we can say that PvE game experiences are in the subset of human interactions in which mutual cooperation is advantageous for all parties. But they're still very firmly within the realm of interpersonal interactions, generally. As per Scott's post, we don't really expect these types of experiences to be finite, so that in itself is not an advantage held by PvP/antagonistic/political interactions.

From a CEV perspective, PvP and political goals are bounded in problematic ways. Two people might both want to play League of Legends, but they also both want to win, which is impossible. You can maximize goal-satisfaction by making a really satisfying victory, but your subjects will fundamentally get to experience it half the time, and their wishes will not always come true. It may be that in a far-future situation, this is a fundamental limitation in political relationships that much reduces their use in CEV value determination, in favor of human interactions that can be multiplied to mutual satisfaction without limit.

But they're still very firmly within the realm of interpersonal interactions, generally.

I disagree. Certainly, things like Fallout or Final Fantasies were made by developers with the goal of offering me a certain set of experiences, but that by itself doesn't make playing such a game an interpersonal interaction. Not more than drinking alone -- for the winemaker has crafted that bottle of wine to provide a set of experiences for me, too...

Someone replied, asking why anyone should care about the minutia of lifeless, non-agenty forces? How could anyone expend so much of their mental efforts on such trivia when there are these complex, elaborate status games one can play instead? Feints and countermoves and gambits and evasions, with hidden score-keeping and persistent reputation effects… and that’s just the first layer! The subtle ballet of interaction is difficult even to watch, and when you get billions of dancers interacting it can be the most exhilarating experience of all.

Can these people start wearing tags so I can stop interacting with them?

Parent comment is a move in a status game.

And so is public, self-aware meta-commentary.

"You are now playing my status game. If you decide not to play it, that's just you playing it badly." Yep today sounds like yesterday.

This comment is the exact reason I will weld the holodeck doors shut from the inside.

You can try to run,
but you can't hide
from what's inside
of you-ouu

Steely Dan


"Can we find a way to lower the status of those people?"

I think there are already plenty enough ways of distinguishing extraverts from introverts.

Is the insinuation introvert = doesn't play status games?

If so, I object on the basis of the internet chat forum community comment sections.

I would bet that most people who take more interest in "the subtle ballet of [personal] interaction" than in "the minutia[e] of lifeless, non-agenty forces" don't think of the latter mainly in terms of complex status games. Intricate status games are probably just the most fascinating aspect of personal interaction from the perspective of that minority of people who find personal interaction relatively uninteresting to begin with.

I now understand better what RomeoStevens meant, though. I agree that it would be wise to watch out for abstract thinkers whose primary fascination is status games or Machiavellianism.

I know plenty of extroverts that I would not describe as playing status games.

I wouldn't say PvP is an "integral part of almost every major video game", many major games, from casual games like Angry Birds/Candy Crush to games for "hard core gamers" like Fallout or Baldur's Gate, are single-player only (or mostly) and don't contain any PvP.

It applies to famous games, and also, perhaps more interestingly, to crowd funding, most of the biggest crowd funding projects are single-player or mostly single-player : Torment: Tides of Numenera, Project Eternity, Double Fine Adventure, Wasteland 2, or non-directly PvP multiplayer like Mighty No. 9. Only game in the top 5 crowd funded games (according to wikipedia) that contains PvP is Star Citizen - and even Star Citizen is mostly backed (AFAIK) by fans of the Wing Commander saga, which didn't contain any PvP, so I'm not sure PvP is primary reason to back it (I backed it for the single player campaign, not for the online mode).

There are definitely are many popular and commercially successful games containing PvP, like all FPS games or most MMORPG but I think the amount is largely overestimated by people focusing to a subset of games.

most of the biggest crowd funding projects are single-player or mostly single-player

That's probably because people fund the remakes/sequels of their nostalgia games and in that olden era most good games were single-player. And you don't need to crowdfund FPS PvP games -- there are tons of them available.

I enjoyed this post, but I almost didn't read it since the first part was about video games, which don't interest me.

I just learned that I should not pass judgement so quickly, but I'm telling you so you know that you almost alienated a reader (and maybe did alienate more, who aren't commenting here due to obvious selection effects).

Your metaphor of science as PvE is great, and I do find your narrative leading up to a necessity of PvP in a simulated reality (or any post-scarcity economy really) quite persuasive.

But PvP can take many forms: All sorts of competitive games could be in it, popularity contests (including all the arts), and even tried-and-true competition over mates. I do not see how any necessity for fundamental philosophical disagreements follows from them being part of the PvP set.

[-][anonymous]8y 3

Question: How would something like pencil and paper role playing games fit into this dynamic? On the one hand it's sort of like Lumifer's good point about "You don't run out of PvE content in games where players produce the content." Except in the case of games like Dungeons and Dragons, there's usually a person explicitly designated as a content producer and rule arbiter. (The Dungeon Master) that isn't really also a player in the conventional sense of the word.

I don't think PvE is necessarily scientific knowledge. It's more like experience (to expound on the analogy further). While we're currently in one environment -- Earth -- it might be possible for us to explore other environments in the future. But, like the analogy proposes, it would take an enormous amount of manhours/manpower to actually reach this new content.

Very interesting article. The level of PVP could be related to sports in terms that the top players (Olympics) are the best of the best but most of us are happy playing weekend sports or even watching it.

I don't see that PvE is somehow less war than PvP. Both are about fighting.

Fighting isn't the only kind of interaction between humans. When I'm dancing Salsa I'm not fighting with my dance partner or fighting the environment.

Puzzles, minigames, and survival-oriented content are also PvE in a general sense, but are not about fighting.

Similarly, I've encountered reputation and social conflict systems that are PvP in a general sense without an announcer bombastically proclaiming "HEADSHOT" at any point. It's a lot harder to get these figured out right, though, and most games don't bother.

without an announcer bombastically proclaiming "HEADSHOT"



PvP is fun even if you aren't good at it, otherwise it is literally just a status game. This is a lot of obfuscating dressing on the idea that human status games is where it's really at. Never mind how they're negative sum, promote perverse incentives, how people coordinate and warp perception to be unfair to people as much as the target's low status will allow, and otherwise 90% shitty in the particular, -pvp is fun, and not really like social status games, so apparently we all have to be bitches in the future.

New to LessWrong?