Saturday, July 11, 2009

Ten Thousand Players

Every once in a while I like to dip my joystick into something massively multiplayer to see whether my distaste holds up. I decided to re-visit Conan and also play some Perfect World, which I'd never seen before.

I have the same problems I always have. I hate gated content: the idea that I can't wear a particular hat until I'm level 941,513,015,059,104,000 is kind of... painfully stupid. But it's easy to program.

There are many ways of gating content that aren't quite so egregious, but they don't fit easily into the rubber mold asian MMORPGs use, so they don't get implemented.

I also hate... well, actually, it would be easier to list the things I like. But, anyhow, you get the idea.

One of the things I always notice in these games is the utter stupid stupid stupid stupidity of the static game world. There's a conceit that monsters should spawn in every two minutes ten feet from the city wall... and the city folk don't care. There are at least a hundred other heroes within easy shouting distance of me, and they are all useless, pointless, might as well be filler NPCs despite the fact that they're all five times stronger than me. There's this deep resentment in me that all these powerful heroes never bothered to get together and just clean up the countryside.

It's not just that they can't due to respawns. Respawns aren't really in the world's fiction. Every time a hero kills of Ogorth the Demon Kitten, it's played up as if this is the first and last time he's been defeated.

The idea of integrating respawn into the world's fiction is appealing to me, because it could be extremely interesting. "Oh, yah, built my house on top of a spawn point. Five gold coins and you can go in and kill stuff. Ten gold coins and I'll let you use my fireball turret to do it."

But that's besides the point, really. The problem lies in the fact that there are so many other players. There's this peculiar idea that you can just do whatever it is you're doing while more or less ignoring the fact that there's someone five feet away doing something similar and you could, in fact, easily do it together if this were reality. It's like there's an invisible wall between you and them: you can see them, but you can't really interact with them.

I think it's a combination of factors, but many of those factors can now be resolved using technology, and I think it's time to just take a quick look at them again.

1) "We're all heroes here!" problem. The population density is too high in 99% of the MMORPGs on the market. Or, rather, there are too many heroes. The population density is just about right if you have, say, 5% heroes, 5% researchers, 5% chefs, 5% tailors, 5%... well, you get the idea. But when they're all the same, you can't let the world bend at all, because it'll change the world too dramatically for everyone else, who happens to have all the same concerns as you do.

There are a lot of ways to reduce the weight of this. One is to simply reduce the population density, another is to allow players to participate in radically different domains. Above I gave the example of different jobs. It's not boring to have those jobs: the minigames involved and the act of acquiring the things you need can be made just as interesting as the mindless, endless fucking grinding of the "heroes".

But there are lots of ways to do it. For example, what if there were, say, ten overlapping dimensions. Any given hero was only really active in one of them, but was somewhat visible in all of them. Visible enough to talk, trade, and so forth, but not visible enough to steal your kills or feel like a stranger.

Flat-out reducing the population density by a factor of 100 also goes well, because you can then begin allowing the players to build and develop the world through their actions, because only the edges of their world touch other player's worlds.

These are ideas that have been around a long time, but we have the technology, now, to implement them.

2) The "Robot Dancing" problem. Although the world is full of other people, you never really interact with them as people. Even when you're using emotes, they are a painfully clumsy second-hand method of communication.

There are two pieces to this problem. One is the lack of body language. This can be solved either by the avatars having moods and personalities, or by somehow reading the body language of the player. Either of these is possible these days, but we do have the problem that almost every game engine on the planet uses static animations rather than allowing for recombinant or on-the-fly animations. Until we can arbitrarily animate an avatar, this is impossible.

The other half of the problem is one of vision. Even if the avatars are beautifully animated, we can't freaking see them. In reality, we're constantly looking around us, and our eyes have a much higher resolution than the screen. To imitate this in a game, we'll need some kind of zooming, auto-looking-around thing. Perhaps avatars could be inflated in "bubbles" on the screen, zooming in on them so you can see their body language. There are other options.

3) The "Greedy Greedy Gateway" problem. Most games really focus on the grind, and in order to keep that intact, they need to follow some very specific and horrifying practices, gating and segregating content and players. Most players are so used to it that they can't even see it any more, but it's there.

Static content itself is an example of this problem. With today's technologies, it's quite possible to give everyone in the game a unique weapon. Perhaps based on one of ten or twenty base frames, sculpted using various settings, add some add-ons to it, color it uniquely, add some particle effects... with statistics it's a bit harder to do, but not as much as you might think.

I also think that most next generation games should do as Conan does: the clothes are actually a separate model applied on top of the game model. This has a lot of disadvantages, sure, but it has a lot of advantages as well, none of which Conan uses. One is that it allows for more diverse, customizable clothes and armor. Another is that it allows for easy visual representation of wear, tear, and damage.

Bah, that's all just aesthetics, though. The core of my gripe is that content is rigidly gated, and players are herded through like cattle.

I am very bad at being cattle.

And it's just not needed any more. We CAN develop games with other techniques. We're just too damn stuck in this gated mold. We print these games one after another, and the only differences are aesthetic. That's not different games! That's expansion packs! Every MMORPG released these days is just an expansion pack that doesn't let you carry over your saves.


What's your opinion?


DmL said...

Agreed, spent almost 6 months playing the original everquest fairly solidly (15ish hours a week) and was totally frustrated by schedule problems, the gates on content. I could spend 4 hours and never gain a level, because by hour two I'd be so bored, I would attempt to fight higher-risk enemies for more xp reward, but since the game as basically a glorified game of black-jack, I would never move forward substantially. Course I just chalk this up to not matching my play style, since so many other people seem to enjoy playing MMOs so much. I started the free 7-day trial of WoW, and after 4 days of patching and trying to login, I spent about 3 hours in the game. Nothing was explained, there wasn't really any setup. After about two hours playing, I died, and it never explained what you're supposed to do then. I could have made some assumptions or checked the forum, but I didn't feel like being generous or reading a website. Just seems like most MMOs these days, to have any chance of success in a reasonable amount of time requires you to do massive amounts of extra-game reading and planning, which just totally kills it for me. The pefect MMO would reward me for standing around and helping/interacting people when I felt like being more passive, and letting or engaging me when I felt like exploring the world. As it is, I get punished for both. Terrible design!

Craig Perko said...

I agree, of course. Both Conan and Perfect World are horrible at explaining things. For example, I was a wizard in Perfect World, and I could see the skill tree... but I couldn't USE it. I never did figure out how to get the next level spell.

Conan is similar: you'll find "binding" weapons. "Are you sure you want to equip that? It'll bind to you permanently!"

What the hell does that mean? Is it cursed?


But really terrible tutorials is a relatively minor problem in comparison, for me.

Ellipsis said...

I'm all for cutting down on the massiveness of massively multiplayer. It's really not clear right now what I gain by being in a world with thousands of other characters.

I generally want to do two things multiplayer: play coop with my friends or play pvp. If I'm going to meet up with my friends to play coop, then the game will be much better if it's a story about me and my friends being a party of adventurers in a world that doesn't have many others in it. If I'm going to play pvp, then I can just take my character and pop them over into an arena or whatever, which is like a glorified online lobby for setting up pvp matches with whoever I want.

Anything more complex than this that you can potentially get from massively multiplayer interactions either don't happen, or happens really poorly, so I'd like to see someone get the multiplayer RPG right first.

Aaron said...

I have a problem with the many heros issue as well. The solution I would like to see, is to up the NPC count until the player/hero to citizen ratio is reasonable, and then make the NPCs scriptable by the players. I don't mean that NPCs should be blank slates which just do whatever players tell them to. The players scripting would be more like suggestions that the NPC may follow based on player stats, NPC goals, the environment, etc.

Obviously not all players will want to get into scripting. Some might dabble just enough to get sidekick, or have a business run while they are off adventuring. But those that really embrace it could become very powerful. Ruling over cities, writing laws, controling trade, and raising armies etc. could be possible. Add to the mix that all of a players scripting influence is lost when they die, and hopefully some meaningful politics would emerge and lead to real player generated content.

Of course there would be plenty of players who never script an NPC at all. And they will remain the simple heros going along for the ride. Hopefully it would be a more intresting ride then a serial list of authored fetch quests.

What do you think?

Craig Perko said...

Ellipsis: Yeah, I agree, but there are other factors that massively multiplayer provides. First, it creates a player economy. Second, it can, if done right, result in huge amounts of fairly good created content.

There are better ways to do that than just dumping everyone in the same place, though!

Aaron: I like it. Personally, I tend to think in terms of evolutionary algorithms rather than scripting, because anyone can "choose the best of these three", but only a few people can program better than a professional.

There's a lot to be said for player-generated content, but it has to be made useful and easy...

GolerGkA said...

Before jumping in to comment, I must strongly recommend you Eve online - the best alternative game design MMO. That's the game where the world, wars and all of the economics are created by players, people do business and run their own financial pyramids and stock markets, and there is NO experience-based grind. Anyone seriously interested in MMO game design should play Eve at least for a few months, I think.

So, "we're all heroes here" issue. It's just not so, in most cases. If you're talking 5 years old WoW server with registration already closed - then yes, most part of the characters at the square in Trade Discrict in Storwmwind would be 80 lvl epics. But in open, young server, and especially in F2P game like Perfect World (and if we're talking mass MMO market, we mean F2P) there is always a pyramid: a lot of newbies, most part of the population consists of mid-levels, and top-levels are quite hard to find. It is just typical for the game with average content quality and, let's say, 9 months of content to the "top-level" to have average lifespan of an active player (who have passed the "newbie" stage) of 4-6 months.
About "multi-dimension" idea - it looks very much like the "district" concept from Guild Wars (I have played the localized version, so I'm not sure how in particular is it called in english). In GW the "grind" zone with monsters and quests is an instance, created for the player or for his party, and public zones, where you trade and chat, are too created in several named instances, like "Europe 1", "Asia 3" and so on. There is always a switch on the screen, allowing you to easily go from one instance to another. And, of course, the auction house is shared between all this instances.
We're adopting this public zone instance concept for public "grind" zones in our game, so in a few months I will able to tell you much more about that concept :).

OK, let's go to the next one - "The "Robot Dancing" problem". It's true than in WoW and games with WoW-like game world and camera you're zoomed out the whole time (because it is convenient for the actual game process), and emotional animations and "social" characters costumes fail to deliver, and chat isn't the best way to express your emotions or roleplay, for example.
But I must state that big part of what MMO can deliver that other genres can't is actual gameplay interaction with other people, in many different ways. Buffing someone, saving some other person from the agro or killing him for no reason is much more powerful in expressing your emotions then voice or even video-chat could have ever been.
To be successful, MMO must allow and encourage the creation of all types of social relationships - being real friends, enemies, trade partners and lovers with other people is something that you can never do in single games. And believe me, beating the heroic instance with your guild mates created much stronger feelings than most advanced animation technology.

GolerGkA said...

And the last, the "Greedy Greedy Gateway" problem. The real problem lies not in a similar content, but in doing the same things over hours and hours of grinding. And that one is real and serious.
To get to that problem we must understand, why the game need hundreds and hundreds of your time. It is really simple to see when you remember the monetization models of MMO that created this stereotype. In 80s and 90s the first commercial MUDs were paid by per hour basis - for every hour of play you were charged, let's say, 5$ (it seems unbelievable today with monthly 15$ costs, but that is the real price people played to stay in the text-based world).
Of course, after you managed to immerse the player in the virtual world, you would like him to spend as much time in this world as possible - thus creating a goals in the game in order to obtain which he would spend hours. The same applies to EQ-like games - just instead of hours, you would be charged for months.
And the founding principle of F2P games is "time is money". You pay for not spending hours of grind to find some rare drop on the one hand, and on another you spend your money in PvP on consumables, so you can say that it is per hour pay again.
Even in advertisement-based F2P game you have to keep the user in your game to show him as much banners as possible.

I've met two monetization models that avoid creation of that problem. First, the "game box" model is applied to Guild wars - you buy a box and then you don't have to pay at all, until you've gone through all the content and you want the expansion pack. And the other is Eve online - in that game you don't get skill points from any in-game activity; whether you're online or offline, you level-up your skills on a constant basis - so, to achieve your long-term goals, you just have to pay your subscription and login once a few days to turn on leveling of another skill.

So, when game designer works on a game with some of the traditional monetization models, he has to create something that consumes player's time. Rob Pardo approached this problem with the team of AAA-quality content creators and hundreds million in budget - unfortunately, that's not a typical situation for a game developer, so you have to create so type of grind for the players.
Of course, there are several ways to prevent that grind to become the killer of your game, but I've never met the ultimate solution.

Craig Perko said...

That's a buttload of text... so this's a buttload of reply.

It appears you're working a game, although you don't mention which and your profile is empty. So you're probably coming at this conversation from the perspective of how it impacts your game in particular. Please remember that it certainly ISN'T commentary on your game, because I don't even know what your game is.

Regarding Eve Online... most of what you said is true only from a certain perspective. Although there is no experienced-based grinding, there is a lot of grinding for other reasons and there is an experienced-based long-ass-wait. Eve isn't The Solution, but it does provide design from another angle, so it's certainly worth studying.

Which I did, of course.

When it first came out.

I don't know what you mean when you comment on the "we're all heroes here" bit. The fact that everyone is a different level (and is therefore involved in slightly different gated content) has NOTHING to do with the situation. Sure, what you said is technically right, but you might as well say, "on the matter of nuclear power, it turns out that many people own pets!" So I'll pass on that.

Instanced zones are one solution to the crowding issue, but my instinct is that it's a temporary patch, and that a better (and not impossible) solution lies in a different technique.


I like the reminder that that much of the interaction in these games isn't social. It's easy to forget that gameplay should still be king.

But I don't like writing off the social aspect the way we do. There's a lot of meat there, and we're ignoring it because we're used to ignoring it. That's just bad policy.

And I STRONGLY disagree with your idea about social play. You seem to think that creating some kind of socializing system would be somehow separate from the "real" gameplay. Instead, think about integrating the two together. In fact, there's a lot of low-hanging fruit there that we don't bother to chase because we're too busy copying WoW.


Lastly, I HATE YOU.

Okay, I don't hate you, but you're treating me like a kid with your ultra-remedial lesson in "why games are long" and "how we make money".

I wrote that same thing five years ago. I played Eve when it CAME OUT and commented on how the skill-over-time system was a valid and often superior alternative to grinding.

The rest of the game is repetitive, high-barrier-to-entry, and fraught with irritating crap. But I won't go into that.

Anyway, yea, I know the challenges facing developers, and I know the inertia that drives them to clone the clone of the clone of the clone of that other clone that we cloned from a MUD which was based on that other MUD some kid built on a computer with less power than your cell phone.

But inertia is not proof that the way we're going is the only way to go.

Christopher Weeks said...

I think that GolerGkA thinks that a hero is someone at level-cap and you think a hero is someone doing heroic stuff -- filling the story archetype or whatever.

When you write "5% heroes, 5% researchers, 5% chefs, 5% tailors, 5%..." I'd be interested to hear how, as a dev, you'd incentivise that kind of play. What about the way SWG made it pretty easy to reach the pinnacle of any career track, but in order to master another you had to give up some of what you had? In Wurm Online, each character starts with a randomly selected skill or stat which grinds at three-times speed. I think that's great but doesn't go far enough.

And then speaking of Wurm, it doesn't suffer from some of the problems -- like gated content, that most MMOs do. But on the other hand, there is basically NO content. What do you suppose the role of a sandbox is? How little "content" would you have to add to such a game to make it a "real" game?

Craig Perko said...

I've never played Wurm, maybe I'll look into it.

When we're talking about incentivising gameplay, we're getting into a whole realm of stuff to talk about, which I may post about instead of try to cram it into this reply.

In terms of actual gameplay - the technical rewards of playing - that's not actually difficult. There's a lot of ways to make alternate jobs interesting in terms of gameplay, if your design starts from the standpoint of making those things interesting. But the narrative around that gameplay is a very important component, and it's usually left out entirely.

The problems with that kind of incentive are the same as they are with the main gameplay: it requires just as much scripting, just as much content, and so forth.

There are a lot of things we may have to do to short circuit that method of content creation, perhaps replacing it with some kind of player generated content. It's pretty clear that it's not tenable to script up ten times more content for ten more methods of play!

I could talk about content all day, probably all week, so, again, I'm going to not go into huge detail here. But there are methods of creating content that do not require excessive developer time and do not fall back onto the "empty" feeling you get from games like Secondlife.

Unfortunately, they're pretty much all theoretical...

Christopher Weeks said...

Oh, I'd also be interested hear your take on the disparity between players who live the MMO and those who play a few hours per week. Interestingly, that seems to matter less to WoW than it does to A Tale in the Desert. I think this is something the devs should be mindful about but it doesn't have to be a big problem. It's just kind of like the casual players are playing a different game than the hard-core. We just need to make sure both games are fun.

Also, I expect huge stride in procedurally generated content over the next five years. IIRC, you are unimpressed with Spelunky but I think it's basically the state of the art for PC.

Craig Perko said...

I'm not unimpressed with Spelunky, I just don't like it. It's not my style. Similarly, I'm not unimpressed with Eve Online, I just don't like it.

I think you're right on all counts, and I do love talking about the disparity between "casual" and "hardcore" players. I can write an entire essay on that topic...

In short, I think that disparity is a major reason we're having a hard time shifting over to a more player-generated world. You can't risk a casual player getting too lost or confused by a changing world.

Ellipsis said...

Actually, I tend to think it's the hardcore players that are getting in the way of UGC, not the casual ones...the more game mechanically important the UGC is, the harder it is to fit it in with players who are trying to game the system as much as possible.

Craig Perko said...

I just wrote a post on that which I'll post once I get home.

TickledBlue said...

Another commenter mentioned Wurm online, I'd second checking that out (as long as you don't have an ATI graphics cards which currently seems to have an unreasonable memory leak that allows about 5 - 10 minutes of gameplay before crashing out). Its got some nice ideas but is very poorly implemented. It reminds me of a multiplayer settlers variant where you're playing one of the little people cutting down trees, eating your lunch and building the world that everyone shares (kinda first person perspective sim city in a fantasy setting). Horrid graphics, criminal UI and it has managed to find new and frustrating ways of implementing grinding.

Man! I agree wholehearted with what you're saying and could write a post rivaling War & Peace both in size and ability to put you to sleep giving you my perspective. So I'll try and keep it brief.

1) We're all heroes here problem: I'd love more opportunity to pursue other jobs/roles than monster thwacker in a game, crafting in most MMO's shows how poorly this kind of thing would be implemented by our current crop of devs. I am, however interested to see what falls out of things like the upcoming APB idea of cops vs robbers where you'll get job offers based on what the 'other side' is doing offering a kind of semi-dynamic PvP. As you mentioned I think the technology and ideas are there but they have yet to be implemented well in any one MMO.

Not that I've played it but doesn't the Death Knight in the newish World of Warcraft expansion have a quest chain where the player quests within their own layer of reality, able to see others but not interact? I remember hearing a vague something or other about it in a podcast somewhere, but at the time wasn't interested so I don't remember the details.

2) The Robot Dancing Problem: I still think this is a long way off. Not from a technology perspective as we have plenty of examples of motion tracking (like TrackIR) that with little extension could provide body language feedback input, but who wants their real body language expressed in the game? On the animation front, I keep looking at hideous examples like Everquest 2's 'mood' system that allowed you to set a happy, angry, sad kinda mood that was really just some tweaks to the mesh of the characters face they all made me shudder with how vile they looked - either completely over the top or not at all representative of the mood to be communicated. Similarly with Saints Row 2 character creation, the facial expressions are just awful (I know they're meant to be funny but they're not even that).

3) Greedy Gateway problem: I see several reasons for this approach and none of them are good. I still argue with my friends that computer RPG's (although I shudder to use the term) are still all far to heavily based on D&D and its concept of the level/challenge ramp. It's also a sign of how hard it is for the designers to get out of the linear storytelling thought mode. Most MMO's have an almost "tourists bus ride" through the game feel about them. I also see it as a direct result of the D&D inspired, level based gameplay that most of these games seem unable or unwilling to let go of (I know that's an oversimplification and there's a lot more to it than that.. but I really am trying to keep this short... honest).

To be honest I don't expect too much in the way of change in this (I hesitate to call it a) genre in the mid to near future. What I'm looking at more and more is the cooperative based gameplay mechanics that are coming out. Left 4 Dead hints at how to approach dynamic in game content and encourage player interaction. I'm now interested to see what Borderlands will offer in terms of things like it dynamic firearm creation content.

Craig Perko said...

I agree with you on all fronts, including about the crippling memory leak in Wurm.

I disagree with your pessimism, though. It only takes a few years to change the industry. If a game comes out with these added features and they only moderately suck, it'll be only two years until a polished generation of games come out featuring the same thing. Two years after that, it's everywhere.

You can see this pretty much tracks with the release of these open world games, first person shooters, "ethical duality" games, and so forth.

It's possible that these advances are ten years out, but I think it's more like five or six.

I suppose we'll find out.

TickledBlue said...

I hope you're right. Any other game type and I'd be drinking your Kool aid but MMO's always seem so mired in the past. I guess we can bank on Blizzards success driving a lot of interest in the field for a few years to come.

Unlike my main game love - space sims: which I see as even more mired in the past and the sooner the current crop of devs move on or die and the new guys start going "Hey here's an unexploited genre" the better. There's still some fight left in her but in general unless its an elite clone or a poor mans X-Wing/Tie fighter then it doesn't get made (don't talk to me about Derek Smart)... OK OK deep breaths... I'll try be less pessimistic next time

Ellipsis said...

Of course a poor man's Tie Fighter is still better than a lot of other stuff that gets produced. :P

Craig Perko said...

Tickled: You may be right.

On an unrelated note, you may want to keep watching this blog, because I might be releasing a vaguely-space-sim game in a month or ten.

More of a 4X turned on its head, I guess.

TickledBlue said...

No argument from me Ellipsis. I'm still a dedicated enough fan of the genre to keep pumping my dollars into purchasing space sims - the least offensive at the moment is Evochron Legends, good indy effort with some nice leanings towards exploration but still missing something. The most offensive is Tomorrow War... avoid if you're sensitive to atrocious voice acting, incomprehensible plot and arcane controls. Although it does have some nice elements such as the space to surface transitions.

Craig: I'll keep my eye out :) I love 4x games almost as much and buy an unfeasibly large number of them with Sins of a Solar Empire, Space Empires V and Sword of the Stars being my most played.

"...turned on its head", hmmm now I'm intrigued. If you throw some other comments like 'borrows themes from Star Control 2' and 'puts the wonder back in sci-fi' I'll plonk down some cash now ;)

Although it's not a 4x or a sim I am really enjoying the 2 Space Rangers games, that are a wonderfully tongue in cheek abstract turn based 2D space shooter/trader/adventure/real time strategy. Its almost as if the developers decided to try shoehorn as many wildly different gameplay types into the one game. Strange thing is how well it actually works. If you've not checked it I recommend at least the 2nd one highly.

Craig Perko said...

Sure, I'll go try it out, thanks!