There is something magical about collective suffering, which I think we have lost sight of in our quest to make everyone happy.EmoLine, it's a harsh, brutal simulation where great loss can be experienced.
And the mechanic of the game is world domination.It's about war.Obviously that is terrible, but there is something about that.Fighting for a limited set of resources against another tribe is pretty fundamental.
Welcome to Web3 with A16Z, a show about the next era of the internet.Our featured guest today is Hilmar Pedersen, the CEO of CCP Games, maker of EVE Online, a massive multiplayer online role-playing game.
In this episode, Pedersen shares his unique worldview and game-making philosophy, as well as a deep dive into the technology and economic design of his sci-fi simulation.
He also touches on how niche cults can break into mainstream culture, how slow databases can make for fun gameplay, and what to expect from EVE Frontier, a new blockchain-based overhaul of the space survival game that's now inviting people to apply as playtesters.
The other voice you'll hear on today's episode is that of Eddie Lazarin, A16zcrypto's Chief Technology Officer, and an avid gamer himself.Eddie speaks first, followed by Pedersen.
This conversation originally took place earlier this year at A16zcrypto's CSX Startup Accelerator program in London, videos of which are posted on the A16zcrypto YouTube channel.
Be sure to subscribe for more thought-provoking conversations and other insightful content. As a reminder, none of the following content should be taken as investment, business, legal, or tax advice.
Please see a16z.com slash disclosures for more important information, including a link to a list of our investments.
Maybe an initial question is, what does it take in your mind to create a virtual world, a virtual economy?
This is the core purpose of CTP, to make virtual worlds more meaningful than real life.And that obviously maybe sounds, well, it sounds crazy and it might also sound evil, but it has forced us a lot to think about what is really real life.
And to us that made the game, the game is like a toy almost.It's very hard to not see it like a toy.Every time I play EVE Online, it's like watching somebody do a card trick you taught them to do, but they're messing it up.
Always having this, no, no, no, not like that, not like that. It's torturous, but the players didn't feel like that.They felt it was real.And that sort of coincides with this crypto phenomenon, which is equally not real.
And that relates to our economy, which are equally not real. So we basically spent most of our waking hours in a game system that was designed by a Council of Economists under Eisenhower.
And their design premise was, how do we get us out of the post-World War II slump? And they came up with the idea of we should consume.This is documented.This is not a conspiracy theory.
This version of capitalism we live in was designed by people very specifically for the purposes of consumption.And it obviously succeeded. But it is a game system.If you trace it down, none of it is real.
Like the money isn't real, the jobs aren't real, the company aren't real.None of it is really real.And crypto proves that so well because out of like a new fresh foundations that nobody believed in, Companies were built, coins were built, meme.
But it obviously started on an agreed premise that it wasn't real.So sometimes we get fooled into this idea that our real life is real.
Our real life is basically buying a house that is too big, taking out the mortgage, so we always have to be at work, not at the house.We don't meet our kids because we're blah, blah, blah.
I think if I understand you properly, by not real, you don't mean has no effect or no consequence or something.What you mean is that it is synthetic, right?Can you unpack that a little bit?
Yeah, this obviously is very controversial, even inside CCP.And it's taken us a long time to figure out, OK, what is meaningful and what is real.This is very helpful.This is from Emmanuel Kant.
He came up with this sort of mid-1800s rules for happiness.Something to do, someone to love, and something to hope for.Something to do in the world today is basically jobs.Like, what is there something to do?It's jobs.
They're a fairly recent invention.Like, for most of our species' lifetime, we were not having jobs.But do we have them now? And they are lacking in many ways.
The stats are basically there are 8 billion people on the planet, 5 billion are working age, 3 billion would like to have a job, 1.3 billion have a job already, there's a problem, but only 200 million like what they're doing.So failed.
Real life has failed in giving us something meaningful to do.OK. Can we fix that?
Internet spaceships, for mostly nerds like myself, seem to generate happiness where people love their spaceships more than their cars, because their spaceships give them agency in navigating their social network.
They give them status that the people that they care about care about, whereas their car is just like a, I don't know, a Toyota that takes them to Walmart to buy more memory for their PC to play Marine Online.
And that might sound very dystopic, but what we have found is because of the designs of EVE Online, EVE Online is hideously complicated.It's a murder, death, scam simulator.I know it sounds a lot like crypto.
But it so happens when you live in a world like that, you need friends.And you really need them.And if they're there for you, the friendships become real.So when we survey EVE players and ask them, why are you playing this game for 20 years?
Some people have logged into EVE every single day for 20 years.They say that is where my meaningful social network is.Like, these are my peeps.They come to my aid. when I really need it.My real life friends, the friendship, I've never been tested.
I've never had to ask them to back me in a work campaign for over a year.So anyway, when we think about this, OK, if it gives you something to do.
more meaningful than the current crappy jobs the world is offering or not offering, gives you somebody to love, all the friendships you make through the game, where people attribute more value to those friendships than in the real world, because the real world doesn't really need you as much as online.
And then something to hope for, which are the updates we deliver each game.Often underwhelming, sometimes good.So, okay, if this is our core purpose as a company, And that kind of parallels into crypto in the case of like something to do.
Sure, you are mining asteroids in space and making spaceships and vying for world domination, but you still have to pay for food and housing and electricity.Is there a way to make that sort of less convoluted than it is right now?
Our end user license agreement states Nothing here belongs to you.CCP owns all of it.You get access if you're in a good mood.It's like what I basically have is your sort of coat of Hammurabi.Nothing really has changed.
There are 282 rules you must adhere to for me.
As a murapi, being okay with you accessing your spaceship.And then what are the consequences of that?Does that limit the maximum amount of value that people can project into this?
Like, how do you think about the consequences between a crypto system automatically enforcing certain things versus a fiat style?These are the rules.
So EVE operates under a social consensus due to its ontology of having existed for 20 years So because we have not confiscated assets for 20 years we pull off a little bit of sort of London real estate trick
Assets in London are largely not confiscated, but if you add her to the coat of army in London, then they have appreciated, I think, for a thousand years.
So there's norms that have accumulated expectation between the players and in this case.
Yeah.And I think that. is frankly more important rather than, okay, here's a mathematical proof that it works because people just largely don't really care much about math.I mean, they sometimes care about the people that care about math.
So there's like a Maven effect of like, okay, if Etty is cool with it, it's probably okay.But E-Online in many ways, like Bitcoin, derives its value from the social consensus that it's okay and it matters.
And it took Bitcoin, I don't know when to count, 10 years for it to matter in some way.I mean, it's probably still on a journey to actually matter, but that is more a social idea rather than a consensus idea.That's right.
Yeah, so there's this like Lindy, this idea that something becomes Lindy to take for granted its trust properties.
But then also maybe with the mathematically guaranteed versus social norms, you're also limiting maybe the tail risk in an extreme case where like, you know, 30, 40 years from now, CCP decides they want to try something new.
Absolutely.And sometimes we do try something new and people have big problems with it.It's not all been roses.We literally have had protests.Our economy policies are not always allotted.
We were just coming out of a period of scarcity where we forced famine basically in the game.And while famine can be tolerated or scarcity can be tolerated in real life because in some way it's a property of nature or God or whatever is your
idea of that.But when it's just a game designer with an Excel sheet, it's a little hard to swallow.It's just like some person in Iceland changed the value in an Excel sheet and now I'm experiencing famine in my virtual world.
So given this, I'm beginning to see a little bit the role that maybe crypto plays for the future of EVE.Do you want to say more about how to actually make a virtual economy, like the primitive components of a virtual economy, for it to work?
Yeah, so the economy of how materials compose into products, how products compose into larger products, how products compose into services.It's a very much kind of emergent phenomenon.
We did start with pretty solid fundamental principles of just how we would implement
asymmetry of resources on a certain topology, that is the map, how we would create strategic corridors and bridges, etc., inside the star map, so that we would induce competition, cooperation to bridge the asymmetries, etc.
And out of that, this emerged.So we have been looking for an opportunity.If we were to remake it with hindsight, what would be the impetus to do that? And also, if we were to do that, could we build it on a different premise?
So we've been thinking about that since 2015, where somebody explained to me how you could store spaceships on Bitcoin.
There obviously were Bitcoin games being made as early as, I think, 2013, where people were doing like NFT-like just on the protocol, which is actually quite flexible.I mean, Bitcoin is in many ways an engineering marvel. and continues to be so.
And that kind of unblocked this, OK, could we take it more fundamentally, where we could decouple the sort of Code of Hammurabi of like, there's no Hammurabi, there are no clit type rules, there are no police, there is no military.
Could we go all the way that down?So there was a collision of these two things of like, OK, currently we're building on SQL Server, hosted in the Docklands area somewhere. Could we extend that further?
And then could we take 20 years of hindsight and build from a different premise?And we went all the way here of just, okay, how can we take the organic output of EVE Online, backport it into a more sane structure?
Take this model, obviously our world is very emergent from particles
from quantum mechanics, you get atoms from atoms, you get molecules from molecules, you get compounds from compounds, you get like... But when you're doing biology, you don't need to learn quantum mechanics.
And there literally is a mathematical phenomenon called scale decoupling underneath all this.And we are now reorganizing our brain into a mechanic like this.
So that we also not only can build on better foundation, but we can have more emergent power from the emergent social economy.
So what's happening is that in the context of the game, there are different types of special systems, like there's the economic system, there's this
literal spatial system between solar systems and so on, and all these things, and each of them, even though they're all obviously part of the same literal object in the real world, there are little choke points between them where much simpler rules can be used by a higher layer to understand a lower layer or vice versa.
So can you give an example maybe of like... Like this is the tip of an iceberg to explain where the emergence become.So ultimately there's a bottleneck in what can stream into the database.
And there's a limit in the simulation frames of our physics engine.So what hits in the block of time increment in our physics engine matters a lot in whether a spaceship survives or doesn't survive.
And then as the game grew, people started to fight for this.And there is a mechanism built into our sort of supercomputing architecture we built such that there's a queue to get into the frames.
But the queue becomes very unpredictable of like, OK, when is my message going to go into the queue and meet into the frame to go into the simulation?
So while we could increase the block space by making the system more performant, it is never performant enough for the timeframe of the simulation step to onboard all the actions that people have put into the system.This is very similar to crypto.
So once we figured out this is a never-ending warfare, we implemented time dilation.So time just slows down because that also happens in our own life. If there's a lot going on, time dilates.
So our time perception is very much correlated with what's going on.
So by doing time dilation, we actually allowed people to strategize more because when you're interacting with 10,000 spaceships, like giving orders what to do at what time, if you had to do it in real time, there's only so much you can do.
So by stretching the increment, just making it go slower, we created like a meta game on top of the time dilation.So there is a fight for block space, like it ends up in a limit, which is kind of like block space.
It's very extreme in crypto because there needs to be consensus about the blocks and yada, yada, yada.But still, this is not a unique phenomenon of crypto.Any shared simulation involving shared state will have this.
Even our own universe works like this.There is a simulation step. I mean, the whole thing, are we living in a simulation?I mean, if you stretch the concept of a simulation enough, yes, we're living in a simulation.It's obviously ticking.
There's a time increment.So by doing time dilation, we actually both address the technical foundation and make the game in a way more fun to play.Fascinating.
Another element is we have the Guinness World Record of the most amount of people fighting online in one session, something about 8,000 people. The next game is less than a thousand, and most games are really sub-hundred.
So if you go to a typical Fortnite match, I mean, maybe 64, some FPSs have gone to like 200, 300, and Eve is at 8,000.
And we get a lot of calls from people which are building scalable computing for games where you could have like tens of thousands of people interacting on the same time using our technology because getting 10,000 people to do anything
is extremely hard.It's really just concerts, sports matches, and protests, where you have sort of 10,000 people agreeing to do something.Like, that is the level of emergence that we've seen.
Yeah, so the scale of multiplayer is itself an immersion phenomenon that requires many, many things to be layered on top for it to ever happen, regardless of strict technical limitations.
Yeah.So it's quite easy to reason through our world and the emergent properties of our world, how from nature we get culture, from culture, governance, infrastructure, commerce.
I think we have a fairly good understanding that governments can change, but it doesn't affect the core principles of nature.
Once you've established the premise that people feel the world is real, then people messing about certain aspects of the stack seems like updating nature.The real world.If somebody were to come, hey, mankind, welcome to quantum mechanics 2.0.
It's like the one you knew, except the Higgins boson is not here, etc. I mean, when Einstein came up with theory of relativity, there actually was like a protest letter about this being heresy and evil, and it should be banned.
And that in a way was an update to our reality.We didn't fully properly understand.And I mean, then with quantum mechanics, even Einstein had a problem.It's like the God could not design the world like this.
So we're kind of like, we understand through our own world that we are discovering more and more layers to it. So is there a way to replicate this into a game simulation such that it gives us more love and happiness than our current emergent reality?
Because the emergent reality is largely designed by people.It's not a creation of nature.And we see in the pinnacle of our creation, which are basically mega cities like London, People are generally more miserable.
And as defined by people in a city, usually have fewer friends than people in a rural town somewhere, especially America, somewhere in rural America.Friendliest and happiest people you ever meet.
That's true, actually.So what do you think the reasons for that are?And what are the types of things that you would recommend putting into a game or virtual economy to mitigate that?Like, what are the key things?
So EVE Online is It's a harsh, brutal simulation where nobody actually dies, but great loss can be experienced.And there is something magical about collective suffering, which I think we have lost sight of in our quest to make everyone happy.
Anyone that's been in crypto for a while, the crypto victors are the collective suffering. That's why there's such bonds between people that have survived many winters in crypto.
And this idea of collective suffering is an underestimated part of our biology.People seek this.They go on an adventure trip together.I mean, corporate offsides, they're not dangerous, but they're trying to be.
Like, we're going to do something like trust exercises and things.We're going to throw an axe.Yeah, throw an axe.
But Iwan Lain does a different way to simulate this, where I now have to trust a non-trivial amount of people to really say what they do and do what they say.And if they don't, it's going to be a serious problem.We prep for this for a year.
If it goes wrong, we will lose everything.And then we will suffer together.Something about this cycle strengthens and forges friendship.And this suffering has like a yin and yang to love being created.
It's almost like a Joseph Campbell-style hero's journey, but for communities of people.
That has been our conclusion, because our first discombobulation with the game was, why do these people love each other so much?And the game.And love and hate us, the makers.
And it was something, because of this collective suffering, that the game facilitates through its brutality, frankly, and often it is at the hands of another group, but you're still doing it more like it is a sport.
It's like Liverpool and Manchester United kind of rivalry.Sure, it can be quite extreme.There's still a layer of abstraction, plausible deniability almost.Yeah.
But there is something, I think, as we've made the world safer and safer and safer for good reasons, like that's a good plan, but the outlet still needs to be there.Where is our moment of collective suffering?
Yeah, just to put a finer point on it, does the suffering come from the fact that these are valuable things, the rules are always enforced, they're enforced beyond the control of people, they're known rules that can be exploited by others?
Like, is that where the suffering comes from?
In this case, the mechanic of the game is world domination.It's about war. To take the pinnacle of the extreme, people have cut power lines to their enemies' houses in the real world.
Imagine the madness that has to exist for somebody to find a group of people in another country, find the power supply to a house, time when it happens, and do it.Imagine the madness. And obviously that is terrible, but there is something about that.
Fighting for limited set of resources against another tribe is pretty fundamental.So the suffering comes from, here's a playground where the rules are clear. But our rules, which is our culture, is our own making.
There is no Tao that algorithmically enforces that I have your back if we're going to this fight.I'm going to have your back.There's no Tao to implement that.It's a trust amongst us.
And trusting other people and building that trust, and it does take time.There's actually a formula.It's proximity, duration, frequency, and intensity.CAA invented this formula.CAA gets most of its intel through friendship.
So they have mechanized it.And we have kind of retrofitted that on top of almost backtracing, interviewing EVE players and asking them, Why have you done this for 20 years?Okay, it's because you have friends.
Why do you attribute the friendship you have through him online more than the friendship you have in reality?It's been very fruitful to talk to veterans from militaries around the world.
And they usually equate it with this, the intensity of life and death, forces like relationship at a speed that nothing else does.And then they explain to us how our game does that for them, which is obviously very humbling.I come from Iceland.
We don't even have a military.We're extremely spoiled when it comes to that.
Can you say a little bit about specifically how you're thinking about crypto in the context of the next projects?
What I'm hearing when we announced we're going to make a crypto game, we announced that last year.We can make a crypto game.I promise it's going to be good.I promise. It was very similar when we said we were going to make a game on a database.
Like EVE Online is the first database game, believe that or not.And when I said we're going to use a database to make an MMO, people were like, weird.Some elves in Iceland are going to use a database to make an MMO.Silly people.
And I just intuitively, okay, we're going to store value.It's going to have to be transacted.We have to build it on good foundations.And whether it's slow or whatever, databases were quite slow 20 years ago.We'll just work around that.
So I'm hearing a lot of that played back to me now. It was like, okay, blockchains are very slow.Yeah.They're like databases 20 years ago.We, we have a trick for that.
And then, okay, economic value, all these, the game won't be fun, et cetera, et cetera.And I put them to player auctions in Hong Kong that sells you FaceApps in E-Online.Like there's no tomorrow.Like I was asking about non-transferability.
It's happening anyway.When I look at my own data, it's rampant.People are doing it.They claim they don't want it, but I have just facts to prove otherwise.There's about a million dollars per month of liquidation happening in EVE.
I've also calculated, if I look at EVE Online, the aggregated economy and at the assets on average that people have. And if I transpose that to the offshore market value in Hong Kong, then people's savings in E-Online are on average $1,300.
And I was thinking, OK, a lot of people play E-Online.They have more savings in E-Online than they have in their bank account.If something happens to them,
I mean, they should be able to pull on that if they need to pay a medical bill and they have 1300 in EVE.Who am I to prevent that really?
They've spent 20 years accumulating all this savings in EVE Online and maybe their children have a problem or they have a problem.Why would I like disallow that?What is the moral principle of like making that wrong?
And then of course you have to go through the whole stack of like, AML, anti-money laundering, KYC, la, la, la, la, la.Okay, we could do that.Second level's done that.It can't be done.MMO-like experiences have done this.
So a Swedish game called Protetentropia that's run on that premise for about 20 years.But there's an interesting liquidity problem in that concept is that there's no counter buyer other than people inside the world.
So this idea with like, okay, we need to solve for the liquidity. And okay, here are crypto nerds that have bottled up a trillion dollars of loose liquidity.
If somebody were to play an MMO for 20 years, accumulate thousands of dollars in saving, needs to liquidate quickly because they need to pay a medical bill.
Well, okay, if we merge these two things together, good things should be able to come from that. So, yeah, I started to have these thoughts after, like I said, OK, I am basically Hammurabi, clay tablets are a database.I'm not going to live forever.
I can't really promise that somebody is going to take care of it, claiming that I'm some saint.But still, it's hard.I've been in the meetings where we have to uphold the ontology of the world.
Like, it's very easy when you have, I mean, 60 million people that play DemonLion.There are extremely powerful psychological levers when you have a game.You can see sort of if you play a mobile free-to-play game,
These are powerful psychological drivers.You can basically print money and it just doesn't feel like, okay, if we are to make virtual worlds more meaningful than real life and we're just on the spigot of printing money for corporate profits,
I mean, which we also do, obviously, we are a company, but is that the best premise of building it?Is there another way to build the premise where it does truly belong to the people playing it?
And there is a collective aligned interest from everyone participating in the economy.Also us as a bit of an operator, but we have the same alignment with everyone.Our savings are also in the economic value of the game.
Yeah, that's a very powerful idea.Just out of curiosity, what were the original motivations for using a database to start with?
Nobody at CCP had ever made the game when we made Immolene.It's the first game made in Iceland, and we had never met a game developer.We had not even seen a game developer from afar.Like, we were just so not game developers.
But what we had done is that there was a company in Iceland founded in 92 to make the Metaverse. So we had made a VRML browser, virtual reality modeling language.It was the hot thing in the 90s.
It was supposed to be the 3D internet, very Neil Stephenson, Snow Crash inspired.We had dancing aliens.We were broadcasting Contrast, Internet World 98.We owned it.
We had like a motion captured singer in New York, singing live on the internet and full motion capture to thousands of people over like voiceover IP and God knows what. And nobody needed the metaverse.
But to make the metaverse work on 286s required some like serious understanding of network stacks and whatnot.So we more had enterprise software developers rather than game developers.
We were more coming from that background because we didn't have any game training.So it's just natural to us.I mean, we're basically doing a backing application with a fancy graphical UI. We just built it like a bank.
So it was just absurd to do it in a different way.
Do you think if you were an entrepreneur, a young entrepreneur again, you know, what would you do differently given what you know now?And given what we have with crypto today, how would you think about it differently?
So it's very much like find 10,000 people that love what you're doing.Keep that in mind. I think with crypto, everything moves so quickly around.
So you have a lot of this sort of, there are a million people maybe liking what you're doing, but there are not 10,000 people that love it.I would find that path.E-Online is very much that.It's just, we're making a spaceship game.We like it.
And if you like it, come.We're taking that almost to the extreme. I mean, you're all invited to join.Just go to projectawakening.io.Then you can play what we're making.You can also sign up and whatnot.
That's very much made in the spirit that people are going to love it.They're going to hate it, but there are going to be no ambivalence.You're going to be in either camp.
And starting there and then diffusing into mainstream is an underestimated value chain.If you look at many things in the world, they started as niches. like Metallica was making, like they go from, I don't know, Masters of Puppet to what they become.
Like, even if you start small, it doesn't mean you stay small.Most things, I mean, everything about the Apple was a niche cult in the beginning.Still is a bit of cult, but it's not niche.
They figured out how to scale that.Yeah.Yeah.Are there other things that we didn't look at that are cool?I don't know.
We actually have a bit of an invention in what we're building is that we have this concept of energy, which is somewhat our way for addressing simple attacks in MMOs.So simple attacks in MMOs are basically botting.
People write extremely sophisticated software to have AI play EVE Online.It's very hard to fight because the cost of the botter inside the world economy is basically zero.
So that a focused human in the world has a very poor ability to compete with 10,000 bots.The game is not designed for that.So we introduced this idea of energy. which is every agency in the world costs energy.
It's a bit inspired by gas costs for opcodes, which we also have because we're also making our own blockchain.We're crazy people.But we also have this world currency, which is scale decoupled from the opcodes, et cetera, et cetera.
And that gives us a fundamental new tool to make the game fair for humans against AI.
Is the basic idea that there is a marginal cost imposed on specific actions in the game?
Yes, but think more about it. we're going to do like off-code accounting, but for agency.So every time you make an action, we have an estimation of like how much agency is this giving you?
If you take a starship, move it from one solar system to another solar system, it has huge strategic relevance. So it would have a higher cost.So it would have a higher cost, indeed.And then, OK, now we basically have energy.
And OK, what's a good proxy for energy?Obviously, oil is a good proxy.And then we started to talk to people in the oil industry and we learned everything about it.And oh, my God, that thing is well game designed.
So we have all transposed this into sci-fi.So you have to mine crude matter through rifts in the universe.It's all very fancy.Then you have to refine it into multiple sort of products.You make this hideous complexity, et cetera, et cetera.
And this complexity actually plays a purpose in building friendship and love and all that, because somebody has to understand it to make it easy for others.Totally.
No one person can handle all these pieces.So people have to specialize.And when they specialize around this need to economize, then together they can do something greater than individually.Yeah.Spot on. Yeah, that's great.Maybe one last one.
How do you think about hiring around new technologies where you're trying to do things that haven't quite been done before?Is there a way that you think about hiring technologists?
A lot of the technology and techniques we've already trained people how to operate on. And sure, the blockchain stuff is a little different.
I mean, it becomes maybe a lot different when you're into sort of serial knowledge proofs, which we are dabbling in quite a bit because you also have to be able to create information asymmetry.
Everything being transparent on a blockchain is not very good for a war simulator.Like you need to be able to have secrets.So when you go into that, then you're quite off script.
But what we have found is that as soon as we announced this and the work got out,
Just all the people that know how to do that came to us kind of on their own because all the CK people need almost like a simulated environment to test their ideas at some scale.And we both have scale and kind of an alibi for interaction.
Like most of the cryptography stuff involves having secrets.And usually secrets mean that people are trying to hide something from somebody.So that can often go very left turn to a dark place.
But in a game like EVE, it's all basic premise of the game. Now I need to maintain secrets from my enemy, and voila!It's now in the alibi of interaction of a game.
Now all the secret people are excited about, like, okay, can we try to make a cloaking field out of CK math?Yeah, let's do that.Amazing.
Amazing.All right, well, Hilmar, thank you again for joining us.
