User:Juval/Test2

test

Comm-Link:Portfolio - Dumper's Depot:
 * Ezekiel Chikamoto, better known as Burner Zeke, born 2875 in Fujin City, Saisei, Centauri was the founder of Dumper's Depot.

Life
He was born to a middle-class family. Zeke became early interested in recycling, repair and rebuy of goods. Due to the high-price level on Saisei in the 29th century, his parents often refused to buy him and his brother new purchases.

After achieving Equivalency, he opened his own scrapyard in 2894. Originally, Zeke never planned to expanding. His first employee, Dayton Ferro needed to relocate to Yar in 2902. Zeke gave permission to Ferro to use his brand, if Ferro agreed to use Zeke's business practices. More franchises followed and Dumper's Depot became a big player in the recycling industry.

Timeline:


 * 2875 Ezekiel Chikamoto, better known as Burner Zeke is born in Fujin City.
 * 2894 Burner Zeke opens his first Dumper's Depot
 * 2902 Burner Zeke's first franchise partnership with his first employee laid the foundations of the rapid expansion of Dumper's Depot

Imp. Candidates

Emma Thorne: Comm-Link:Tracker - Bounty-less Antward Lillard: Comm-Link:Aremis Post - Fading Fortuna

SXY, Standard Xi'an Year

1 SXY = 1,28 SEY ↔ 1 SEY = 0,78125 SXY

The current SXY is III.1229 (?)

1377 SEY - 2950 SEY(current date) = |1573| SEY x 0,78125 SXY= 1228,906...SXY ‡ The elimination of House Xy.ō in II.1706 (1225) by the Kr'Thak lead to an interregnum, called The Dark.

Other 'milestones'

From CitizenCon 2019 (Lore Makers Guide 2949)

"ISSUES OF TODAY


 * Continued military action against the Vanduul are making people scared that there will be a retaliation
 * The lack of military throughout the Empire has strained law enforcement which has led to an increase of crime
 * Synthworld continues to drain economic coffers
 * Expanded trade oppurtunities with the Xi'an has led to a much loose border
 * The first Tevarin senator has inspired a movement among the Tevarin to become more active in government and culture
 * A new wave of Terran seperatists (sic) have started petitioning the government again to move the UEE capital"

 'Big Issues '


 * Vanduul
 * Crime
 * Synthworld
 * Tevarin
 * HuXa
 * Decentralization
 * Terra Capital

KAIZEN IAE ANTICIPATION
Done. Ref: Prasad Naturals lawsuit.

Open: Third quartal 2949: "Anvil Aerospace quarterly projections predict a downward turn after a a series of supply chain disruptions forced them to temporarily shutter several manufacturing facilities in Castra."

PORTFOLIO: YORM
Gotlieb Yorm (namesake), a famous racer and designer of 'competition-grade components' of unknown origin. He founded the Yorm corporation with the help of his fiancee Adel Fansekar in 2822.

Career
In 2796 he won a race in the underground racing circuit in Baker with an old Aurora. He was invited by Adel Fansekar which financed his racing team, Yorm first refused and later agreed later with the words: "I’ll always take your money, but never your advice. Yorm was known for his eccentric personality and min-max- approach, i.e. he begann racing without clothings in 2814 until a requirement by the Safety Commission stated that all racers must wear flight suits

He achieved the 2nd place in the 2816 Murray Cup Blitz race one second behind the winner. He ripped off the design of his shield generator to an minimum output and won the Blitz in 2817.

Due to Yorm's popularity, he decided to release a shield generator design which became commercially available in 2822. //

CEO Silvio Halbrook tried to expand the product line by hull plates - accompied by mismanagement, bad design descision and manufacturing issues - which ended with a retreat from this market.Bao Ingram returned to the production of competition-grade components.

=CitizenCon 2949 - Building a Dynamic Universe=

[Introduction]
"Thus far we've been focused entirely on a single system: Stanton (...) four planets, twelve moons, 225 economic nodes, 44 trading outposts, 13 Rest Stops and 50 asteroid fields."

- Tony Z TZ: (...) Hi, I'm Tony Zurovec, I'm the Director of Persistent Universe for Star Citizen. (...) and over here on the right on the keyboards is Jake Muehle, our economy designer from Austin. (...) We're going to show you some interesting things tonight and I want to make sure to call out our guitar and bass players back in Montreal, Martin and Rivera Lizey [?]. They spent a lot of time working with us over the last 4,5,6 months very intensely on what were gonna revealing tonight. (...) You need some background information to truly appreciate it (...) and you'll see why we consider it a quantum leap.

So as everybody here obviously knows star citizen is a very ambitious project, our goal is to deliver a variety of incredibly detailed solar systems that behave in logical fashion and that they can give you ample freedom to follow your interests whatever they may be. Thus far we've been focused entirely on a single system, Stanton, which includes four planets, twelve moons, 225 economic nodes, 44 trading outposts, 13 Rest Stops and 50 asteroid fields. (...) The point of this presentation then is to give you a better understanding of how, when the first one's taken so long we're aiming to kick things into overdrive so that we can deliver new systems much more quickly. This involves obviously a lot of different areas: procedural generation of terrain, modular art sets, algorithmic detailing of buildings and interiors and a lot of other stuff.

[Dynamic Content]
"Dynamic Content responds to player and NPC actions in logical fashion, its systemic. This means that the universe is in constant motion which makes it a far more interesting place to explore."

- Tony Z For the purposes of this talk though I'm gonna focus on what I call the Dynamic Content. So what is Dynamic Content ? Dynamic Content responds to player and NPC actions in logical fashion, it's systemic. This means that the universe is in constant motion which makes it a far more interesting place to explore. Most missions should of course be dynamic in nature. They should only be offered when there is an actual need for something. This might take the form of a factory that needs some aluminum to create a retail product, a criminal that needs to be brought to justice or someone that keeping them from one location to another. At their core, missions or services provided by one party for the benefit of another and they are therefore one of the fundamental ways by which imbalances in the economy can be addressed. Likewise, economic nodes which include things like refineries, factories and retail shops need to be dynamic, since they must continually take into account the various market forces and adjust accordingly. When demand for something outstrips supply. Whether it's iron ore, bullets or ship repairs you expect prices to rise and conversely when there's an excess of something you expect that prices will fall. There's more to it of course. Concepts like marginal demand and price discovery require a fair bit of math, but in the end, it's all about the constant search for equilibrium as production and consumption levels vary.

I'll get around to explaining probability volumes in more detail in a bit, but suffice to say for now that they dictate what you encounter as you travel through a given area. As such, they should definitely be dynamic. So, i.e. if a new field of asteroids, rich and valuable commodities is discovered you'd expect to find a lot of NPC miners in the area, but as the resources are gradually depleted, their numbers should start to fall off. Pirates should gravitate toward areas with a lot of value to plunder and recoil as the amount of security ticks up. So let's look into each of these areas in a bit more detail and see why we've struggled to set up a single solar system and at the same time get a precise understanding of what needs to happen so that we can start moving a lot faster.

[Missions and Mission Design]
(...) A (...) mission is fundamentally just a container for code and data similar in concept to a class in any of a multitude of modern programming languages. Inputs can be passed into a mission from either the runtime environment or services to allow for customization. This allows designers to do things like say inject the type and quantity of cargo, whether the ship should be damaged or not where the ship should be heading into a generic transport scenario so that we can reuse the same framework for a lot of different situations. Missions can have content tags embedded within themselves to denote their composition or whether they contain, security, pirates, freighters, miners, asteroids or some kind of accommodation. Missions can also have somatic tags applied. These don't specify physical content but rather detail the headlines' story behind the scenario that a designer created. "A (...) mission is fundamentally just a container for code and data similar in concept (...)

Inputs can be passed into a mission (...) to do things like say inject the type and quantity of cargo whether the ship should be damaged or not where the ship should be heading into a generic transport scenario so that we can reuse the same framework for a lot of different situations. Missions can have content tags (...) to denote their composition. Missions can also have somatic tags applied. These don't specify physical content but rather detail the headlines story behind the scenario that a designer created."

- Tony Z So designers can differentiate between a straightforward pirate encounter versus one that includes an ambush. Once these descriptors have been assigned, the missions are injected into a database that other systems can access, so that they can be instantiated and customized depending upon the situation.

[Problems of Static Mission Design]
This however is where we start to run into some major problems. Designers have to explicitly link missions to various constructs within the game. An NPC mission giver, i.e. or a probability volume covering the area around Yela. This is time-consuming and greatly complicates maintenance if a content tag like 'Pirate' is split into subtags like 'pirate light' and 'pirate heavy' so that we can have more control over what we're aiming to instantiate - you have to go to every place you design the original mission category and individually determine which of the new ones should be enabled. Worse, this isn't dynamic at all. Mission content is fixed to do specific NPCs and locations so there's no evolution. The universe is completely static another major problem is that we don't have sufficient context to properly customize the missions at a macro level, the game doesn't really understand the concept of a route, much less what miners traversing that route should be carrying or what condition they should be in or which direction they should be if they're going (...) if they're empty heading towards a mine or full bringing the results of their efforts back to a refinery.

This lack of context means that we can't exploit mission's ability to have information injected into it for customization and that means that you see a lot less diversity despite requiring just as much work.

The last problem and it's a big one, is that there's no NPC regulation of the mission content. Missions are the basic units of work for the economy and if someone doesn't do what needs to be done then everything grinds to a halt. In a properly functioning system, NPCs need the ability to step up. If players aren't going to and the risk award justifies the effort and do the work themselves. If a commodity can be purchased at one location and sold at another for a dramatic markup and there isn't enough cost or risk involved in a transport to warrant the different, the differential NPCs should seize the opportunity and move the material themselves, until the price starts to make sense.

[Economic Nodes]
(...) An economic node is an entity within the game which might be represented via an actual storefront, a kiosk, a UI interface or an NPC that possesses an item manifest detailing the items that wants to buy or sell storage capacity and a fair bit of data related to determining prices. This includes refineries that process or factories that produce goods and retail stores, where you buy finished products. Factory inputs aren't explicitly denoted but are instead derived from the production formula of whatever they manufacture. Prices what economic nodes are willing to pay for their inputs and what they want in order to sell are determined algorithmically based upon the rate of change of their inventory versus the tangent of our pricing curve.

This means that they're smart enough to say raise the price of what they're selling even when their inventory is low if they detect that they're being resupplied at a sufficient rate to get where they want to be in a reasonable amount of time. So clearly there's some dynamism happening here, which is fantastic. Unfortunately, that's where the good news ends. As with missions, economic nodes knows don't have any real context in terms of supply and demand and there's no systemic flow of goods throughout the economy. Why does the refinery want ore? Why is an NPC willing to buy some drugs from you? Where does the shop get the goods that it's selling? Right now the answer to all these questions is the same. We fake it. Economic nodes conjure up their own supply and demand out of thin air. If a node is designated as wanting to purchase a particular item, i.e. designers dictate a formula that describes how that inventory will be gradually burned off so that they want more. These products you deliver to a node then, they don't actually go anywhere, there's no real demand. This is impossible to balance because, while production and consumption are fixed, the player count isn't. Sometimes a hundred players are interacting with an economic node and sometimes none... and I mention this because.. behind the scenes at the economic level all of the players despite being on different servers, really are connected to one single system. A real economy is a tangled web of dependencies and you can't expect logical results when its gears can completely seize up due to the action or inaction of players.

What we really need then is NPC regulation of the system. For NPCs to purchase items when they need them, thus keeping the demand side logical, while stepping in to help with the supply, when players are down and the risk reward warrants it. This has ripple effects in two other areas. If the demand for missiles spiked because there was a lot of combat. You'd expect that the commodities used to construct those missiles would start to rise in price and you'd also expect to see more miners and freighters working to alleviate that shortage. If the price got high enough that those transports would be carrying a lot of value (...) you'd expect piracy to increase at which point you should start to see a lot more security patrols and requests for combat escorts.

A simple change in demand should be able to alter not only what you see as you travel throughout the system but also the available missions. These kinds of knock-on effects in fact should happen regardless of whether any players in an area and that's a separate problem I'll talk about more when we get into probability volumes.

So let's take a quick look at what it means to set up an economic node. As I mentioned there are about 250 economic nodes within the Stanton system and many of them deal with a lot of different items. Each of these items can specify its own production or consumption formula along with a lot of other information: storage space, optimal inventory level, price offsets and much more. This information can be archetyped but of course, the world's a lot more interesting, if every economic node is a bit different, so there's still a lot of customization that winds up happening. Let's take a look at the economic nodes around a typical planetary system. And finally, let's take a look at the economic nodes around the entire Stanton solar system. That's a tremendous amount of data to set up and maintain as we're routinely adding new items and changing prices and the worst part is that it's all extremely rigid. There's no way to have an economic node increase how much of something it produces or consumes because those things are attached to formulates that have no understanding of external events.

So right now designers try their best to brute force this stuff to make it feel like it's a systemic world, even if it's not. If an area is configured to have a lot of pirates they increase the price offsets and demand for goods at nearby nodes. If the number of pirates is subsequently decreased or maybe more security is added to the area and the nodes are updated to reflect that fact. It's a ton of work and while this sort of hand tweaking does yield something that feels vaguely logical, if static, it's definitely not a real solution and it doesn't even begin to address the major problems.

[Probability and Probability Volume]
On (...) of the last of the major areas for which I wanted to illustrate the problems that we currently face (...). A probability volume is an area of space that contains information detailing what you should see if you pass through the area. It's an optimization of sorts in that it allows to achieve a desired effect, a certain number of pirates in an area or a particular likelihood of encountering a freighter filled with iron, without having to burn a lot of computational horsepower simulating things when there are no players around to see it. This is accomplished to be a linking mission content tags to probability curves that span all or part of the expanse of the volume. So for example you can denote that security ships are commonplace near Port Olisar are and gradually fall off over a couple hundred kilometres whereas it's very rare to see pirates who would prefer to avoid security close to Port Olisar but more likely to see them as you get farther away.

There are quite a few different types of curves denoting (...) because until a player actually accepts one of these missions as it heads to that location there (...) are a variety of probabilities... [shakes his head and stops] I think our notes got screwed up here.

Anyway, so one of the more interesting types of curves is the 'spoofing curve' which allows missions to skirt the normal rules requiring that they can only be created in the proximity of a player and this is done so that an NPC can request that you travel to them to provide something they desire in which case obviously we can't limit them to only existing, in your immediate and yours or in other players immediate proximity. This is how it is currently handled and we'll be adding some other things like NPC refuel and NPC passenger transport in the near future. The reason it's called spoofing is because we only keep a small stub active on the back end which is way more efficient.

We've got other types of curves there 'reinforcement curves' that allow designers to dictate how quickly backup tends to arrive when you've been identified as a felon and we also have the ability to modulate that time, according to how far away you are from the place you were last spotted and how long ago that was.

There are 'bounty hunter curves' which allow us to link to a probability (...) to your wanted level so that we can increase the encounter frequency when they'd actually be hunting you and otherwise keep them very rare. Let's continue on through and get a better understanding of how exactly probability volumes works. Here we have an area of space that's marked up as having a certain probability of pirates and freighters, security, asteroids and derelict ships. Now let's go ahead and activate all that content.

That's a lot of content to simulate when no one is around to see it. A lot of pathfinding NPCs, animations, vehicle physics, collision detection. Now let's go ahead and introduce a player and clear the board. Those pirates and freighters and everything else they still exists but only in a probabilistic state, as the player traverses through the volume we roll the dice and the various curves and force the Probabilities to make a choice.

Does that content exist at that location or doesn't it? Here the player gets a hit from the 'asteroid probability' (sic!), so the mission databases in query define missions with the associated content tag. Dynamic parameters injected and the mission is instantiated. And now a bit later there's freight when the player trips the 'freighter probability' and then eventually runs into a pirate. The key point here is that much of the world exists as a superposition of probabilities until a player gets sufficiently close that the wave functions collapse, at which point we instantiate them and start simulating the full-blown entities.

[Bending Probabilities]
One of the quirks with probability volumes is that they usually represent a variety of low-frequency events. Space is big after all and even in a pretty heavily populated area, you shouldn't be running into other ships all that often. The problem is that the likelihood (of) multiple events happening simultaneously is a product of the probabilities, meaning that it's very unlikely you'd ever see two or more at the same time. (...) [T]his means in practice: Here's a player traversing a probability volume...he makes it through a good part of it before the freighters probability curve determines that an encounter has occurred at which point the probability volume queries the mission database, with the freighter tag, and instantiates a compatible mission. The player continues on and shortly thereafter, the pirate's probability curve indicates a hit, which triggers another lookup into the mission database.

There's nothing fundamentally wrong with this. In fact, it's a pretty accurate representation of what you realistically expect to see. Two rare events happening on their own schedule: the systemic behaviour of the freighter and the pirate, even if spawned separately would allow them to logically react to one another, so if they were spawned in close enough proximity, the pirate might even attack the freighter.

The problem is that this approach, lets the algorithms dictate too much of the experience and if you're not careful that's gonna lead to the gameplay starting to feel very formulaic. What we really want is the ability for designers to craft more custom content and then have a mechanism by which we can trigger it, such that it all feels logical, if a bit lucky in the timing. What we do instead of basically allowing the algorithms to entirely control what you see after is... we defer the activation for a short period of time and bend the probabilities of other things so that we're more likely to get a combination that would otherwise be the case. (...)

Basically, what you've seen here is the player hit the freighters probability. Held on to it. We bent the probabilities to increase the likelihood of anything else being able to be attached. In this case that involved the freighter, they didn't involve the pirate rather. Why does this matter? Think about a pirate and a freighter being created separately. It's going to take a good while for the pirate to close the distance with the freighter, disable it, park next to it, and board it. This means that you'd never come upon a situation where the freighters already been disabled and the pirates are rampaging through the ship. You'd better hurry if you want to save the crew. That's actually a really interesting scenario though one that be a lot of fun to experience.

While we're fine with the systemic functionality, driving things by default. We want to be able to bend things towards more interesting situations without it ever feeling too random every once in a while.

[Problems with Probability Volumes]
The problem with probability volumes then all comes down to their construction. Each one and there are lots of them contains a litany of curves and associated data and is set up by hand, which is time-consuming and requires constant updating as other things change. One of the main reasons why they're so difficult to maintain is that many of the variables that a probability volume details should impact one another. More security in an area should of course impact the number of pirates, but because they're all handled independently, designers have to be very careful when modifying them.

Mistakes are inevitably made which results in a lot of sleuthing around to figure out what broker is now poorly balanced and how to resolve it. The worst thing though is that they're static. If an area has a lot of pirates, it will always have a lot of pirates. There's nothing you or the NPC population can do to change that. Rewards won't vary based upon the level of threat in an area, because the threat is always the same. A new commodity discovery won't cause miners to move in to exploit it and won't in turn attract pirates eager to prey on them.

Demand for security patrols and escorts for that area won't increase in response to the pirates and the pirate threat won't recede as a result. What we really need then is a way to dynamically derive the probability curves based upon what's happening or rather should be happening, when we manage to resolve all these problems I've been pointing out, across the solar system. I've called out a lot of very serious problems. Things that require enormous amounts of bandwidth to set up and maintain and it still failed to deliver the experience we want.

[Requirements For A Dynamic Universe]
What we actually want is a dynamic, logical universe. We want the demand for goods and services to be the result of a legitimate economic need and we must have sufficient runtime context so that we can customize these modular mission templates. NPC's must be able to contribute in helping to turn the gears of the economy from buying and selling goods to creating and accepting missions. The system needs to automatically balance itself so that we can focus on the high-level rules instead of the details.

All of this. too, has to be incredibly computationally efficient which means we need a more focused simulation engine one that's only concerned with what we need to achieve the desired effect. If we're ever going to reach a point where we can deliver new solar systems in a reasonable amount of time then, we're going to need to make a quantum leap.

[Introducing of CIG 'Quantum' Tool]
What's the solution? (...) This is Quantum and I know what you're thinking: I've seen this before if the visuals look you know particularly familiar. It's because we started with the solar system map that we released years ago. But this really is a totally different beast. The original Starmap reference static data. There was nothing dynamic about it, but Quantum is totally different. For starters it has access to all of the backend services and can pull whatever data it needs from them. Everything you see on the screen then that's real-time data pulled from the game. Let's go ahead and take a quick tour toward the system (...) and jump on over to everyone's favourite gas giant: Crusader.

And in the background there you can see Port Olisar (...) and see what the players are up to. Each of those red dots is a real player. Thousands of them across well over a hundred servers. To be honest: we recorded this earlier. Mainly because we didn't know how many people would be playing the game while we were doing this show. I figured it would be totally anticlimactic to say: here all the players and three dots showed up.

Let's go ahead and continue the tour of the solar system and jump on over to Daymar. And you can see a number of people affixed to the planet some of those obviously you're going to be at an outpost, some of them are inevitably mining. Over there on the right you can see the the current player list. Let's go ahead and see what a couple players up. Just go ahead and highlight them so we can see where they're at. Go ahead and jump to surcoworm (sic!) and see where's he at.

It looks like he's in Loreville. (...) You can see some deaths down on the planet there. Obviously there's been some conflict there (...) There's a contract to the right you have access to missions so you can see that Clipsos Norwang (sic!) is requesting personal transport to Loreville and he's willing to pay 4500 dollars. Jump on over to Arial and we can get a view of those probability volumes that I was talking about earlier.

What you're seeing here there as I mentioned there are a variety of different types and lots of different content and thematic tags for all these things so you can see everything from the default tags you can see the security reinforcement tags, the bounty hunter tags, the spoofing tags.

There are a number of these. We'll be adding more. In general what it's doing is, as I mentioned it's basically representing things in probabilistic state, because it's too expensive for us to fully simulate them (...) Jump over to a shop: let's check out the Hicks Research Station, that's on Cellin. (...) Open up that shop. (...)

There's outpost Karea. And you can see that there's a lot of a lot of death around there. On the right at the Hicks Research Station you can see all the things that it buys themselves and what those prices have been doing over time. Now I suspect I know what you're thinking at this point: this is all very cool but it doesn't really solve any of the fundamental problems that I mentioned earlier and that's totally true.

[Releasing the Quanta]
Clear the board and get to that. The primary reason for Quantum 's existence is to enable us to have one unified world where we can you know simulate millions of NPCs and feedback into that all the player actions into one unified whole. There's really no difference there are some optimized areas the probability volumes that exist in between these two you know these two realities, but really this is this is the full-blown simulation it takes into account all the players actions.

Start by bringing a thousand Quanta to life... Jake here is cranking the number of Quanta up to a thousand and... they're basically gonna come streaming in from another system... and into the universe. These are simulated entities. The real significance of this is when you're talking about... working out the vast multitude of details that we really need to allow this universe to evolve and feel dynamic.

We don't actually need all of the incredibly high fidelity super-computationally expensive stuff that we would get if we were actually simulating these NPCs. The traditional way on the server so that they look exactly like they would when you see them. These NPCs don't need to do animations they don't need to do physics they don't need to do collision detection. They don't need to do a whole slew of things that we couldn't afford to do would be incredibly expensive to try and in the end even if we did it it wouldn't make the end results any more accurate.

JK: [They] also have personality traits - that [?] has a little bit of something going on...

TZ: There are a number of different traits that were modeling. We're still working out the exact configuration. One of them is ambition and that specifies how far a Quantum can push themselves relative to others whether in crime or legitimate enterprise. Another one is intelligence and that dictates what sort of things they can pursue and how well they can do something. Happiness is a measure of whether a Quantum - the singular form of Quanta - is miserable or content and unhappy and want to change their situation... move to a new location, change occupations and that sort of thing.

Aggression controls the lens to which a quantum will go to achieve their objective. A business-oriented quantum with a lot of aggression might push their workers harder trading their happiness for more profit and as a result have to deal with more turnover. And lastly criminality that determines the amount of criminal behavior that a Quantum will consider. No criminality equates to a law-abiding Quantum. Having a little means that they might be willing to commit some minor crimes like smuggling or illegal mining. And a lot means that they're open to piracy and murder they can have possessions store up money own a ship or property have a home now all these Quanta have basically come to rest within the system and the reason for that is because there's nothing for them to do, there's no reason for them to get up up off of the couch.

[Quanta in Action]
(...) we're going to add a little bit of interest to the system by adding an aluminum mine to Delamar. We've now injected a mine on to Delamar and yet you see the quanta still refused to do anything and that's because there's no economic reason for them to do it. There's nothing to do with the ore that they could go and extract. Open up that card though that represents the mine on Delamar. You can see a couple of stats towards the top quantity and purity. What these are is basically telling you how much of that material is in that mine and purity is to some degree a measure of concentration. It effectively details how easy it is to extract it. There are a couple of parameters at the bottom they are temporary. Those will eventually be derived from those top two parameters, but for right now we can override it for some of the testing that we've been doing...and jump to the next. We're going to go ahead and add an aluminum refinery to Crusader and... immediately you will see the Quantas start springing to life.

What's happening is they've all figured out ... they've that refinery it wants to build up a stockpile of aluminum and in order to do that it needs some raw aluminum ore and thus it's willing to pay for it and so the Quanta or our seizing upon that economic opportunity heading out to Delamar doing the work, extracting that ore and taking it back to the refinery.

This will continue, until it develops you know until, until it's full so let's go ahead and open that card up again and we can watch the refined aluminum ore gradually build up. You can also see the number of workers a few of the Quanta so you see many of them actually acting as miners. There's also a number of them that are working in that refining factory Refineries can't process this material without labor and there's a formula that dictates how much labor is required in order to process this stuff and so they have to hire a sufficient number of workers and order to actually execute this operation. This would continue for quite a while. Speed up the simulation a bit (...) keep watching the aluminum inventory... that's finished, so they're now building up their raw ore to basically load up and after that the entire economy comes to a halt.

They're the one refiner we've added has no more incentive to buy more. It's storage houses are already full and therefore it's not willing to buy. Since it's not willing to buy, the Quanta have nothing to do. What we're going to do now is at a factory, but before we do that [we] get head on over to the power plant formula ... and you can see here that building a power plant, just like refining ore, requires something. It requires ten workers working for a duration of sixty ticks and it also notably requires two units of refined aluminum ore two in order to produce one power plant.

JM: It's got to be more complicated than that I promise.

(...) go ahead and add the factory to Hurston. Once again you seem to see the economy spring to life and now you've actually got two different tiers going (...)

Open up the factory: the factory, just like the refinery, needs workers in order to build these power plants it also needs refined aluminum. You've got multiple things going on. The factory is willing to buy refined aluminum. Someone has to transport that aluminum from Crusader to Hurston. Some of the Quanta are basically working (...) as freighters to move that product from one location to another. Some of them were working at the factory on Hurston and some of them were working at the refinery on Crusader. Some of them are heading back to Delamar to actually do more mining.

You've got a entire little economic cycle going here. However, it's gonna wind up, coming to a dead end. It again just like what we saw before, because eventually the factory fills up all of its warehouses. It can't take anymore therefore (...) it has no more need for refined aluminum, that will eventually shut down the refinery, which will in turn eventually kill off the miners ... And you can see there that the inventory is still building on the aluminum, which is why there's still some activity. JW: But as soon, as it's done it starts flatlining so they see that idea inventories capped out.

TZ: And now they're almost done. The factory is basically (...) but it's it's got all it can handle. The refinery has all of itsto handle, therefore the entire loop you know draws too close again. Let's add one more node. Let's add a shop that sells power plants to our corp. Now the loop is a bit more complicated. The shop (...) it's a retail shop. It's looking to sell power plants in order to do that it needs to be able to buy them for one price, sell them for another, so it adds its mark upon and it just like the factory, that requires transport of the refined aluminum from Crusader to Hurston.

It needs the power plants transported from the factory to it and so, if you notice there if the individualquanta they'll actually have a little translucent circle around them if they're carrying goods and so you can easily differentiate whether or not the ones that are moving for example from Hurston back to the mine are empty, they're going to bloat up on ore and then when they're heading from or back to Delamar and then from Delamar back to Crusader. They've actually all got ore and so this loop will actually continue indefinitely and the reason is because we've cheated something for tonight which and it sounds very similar to what I was basically saying was a bad idea: which is the shop right now, is a simple consumer.

It's burn it's burning off that inventory. The difference here is that the solution to that is very simple int he context of Quantum and the reason is. These quanta the next step of what we'll be working on is, ... they'll actually require Ships. Ships require engines and so all of a sudden there's a real demand. For how many power plants do you need how many ships do you need it all follows those same illogical equations. Go ahead and add a bit more economic complexity.

[Experimenting with the Laranite Production Cycle]
That's far we've shown a very basic economic clue. One commodity, one refinery ,one factory, one retail shop. The real world of course is far more complicated. You'll have lots of different commodities, refining and production those retail shops and players in Quanta competing against one another for a limited supply of goods and services it's this competition. This economic natural selection that ensures that things remain in balance, that a logical equilibrium is reached.

Our current architecture basically has players connecting to a game server and the SIRT (?) and then we've got services behind the scenes that can feed information to these servers and receive information back in return. You can see here: we've got shop services and probability volume services and all this

sort of thing. The basic problem with this arrangement is that servers don't have any knowledge of what NPCs are doing or rather what they should be doing so that shop servers that controls the prices it's got algorithms, but it has no real understand, it doesn't understand that there are 10,000 NPCs that do or don't need this or what pirates are doing or any of that...

We don't simulate at that high level anywhere within the game right now. You have servers (?) that basically instantiate the stuff near you and you have the services that can run the you know these formula and that's basically it. Now they can implement these services. These algorithmic price calculations, but supply and demand only comes from players and hard-coded formulae and it's the same permissions in PV's. The data is completely static and can't evolve. The number of servers of course as we've seen here it varies depending upon how many players are in the game so let's go ahead add Quantum. Quantum changes this in [a] dramatic fashion.

It gives us a place where we can efficiently simulate NPC behavior and then feeds that behavior to the game server, so that players can experience it. Further it allows player actions to befed back into the simulation, so that there no meaningful difference between the real and simulated worlds. Quantum creates a complete loop. NPCs can be simulated in efficient fashion and actually meant to say that Quantum feeds the services which in turn feed the game servers.

Go ahead and get back to the simulation. And what we're going to do now is alter the powerplant formula. Let's add a bit of complexity by changing the production formula for power plants to require Laranite. We need to change now we need to change your production. (...)

We're gonna add a Laranite mine to Wala and then we want to go ahead and add a refinery for this Laranite.

We'll add that to the R and R HUR-L3 (...) and adjust that powerplant formula... and we're gonna change the powerplant formula. Right now it requires zero Laranite. We're gonna change that to two and what we're gonna very quickly see here is, that we're gonna start getting some Laranite flowing to the Hurston factory, so that the power plants can continue to be created.

And that's that loop of guys right there that's what they're taking... Now one of the important things here is that the...let's go ahead and look at the Hurston aluminum. ... (...) it was very low for a period of time and that's because they were actually burning off that aluminum inventory as fast as they could get it. And the reason why you see that spike there in the aluminum inventory is because as soon as we alter the formula, so that power plants now require Laranite, they were still receiving from all these freighter transports, they still had aluminum coming in, but they could no longer produce power plants.

It's stalled and all of a sudden, they started stockpiling... And then when you see that the aluminum started falling, that's because finally the they started receiving supplies of Laranite and they were able to resume the production process, (...) their aluminium started getting burned off a natural rate... We're gonna add a bit more layer out in the asteroid fields and Jake's gonna handle that be (with) a macro... And now let's go ahead and increase the supply of aluminum and see what happens. We're gonna add a few new mines, let's add an aluminum mine to Cellin first... let's add one to Aberdeen as well... and lastly we're gonna do Lyria, and if you remember, we already have an aluminum mine on Delamar.

They're telling us we need to hurry up. You guys want us to hurry up or keep going ? ... [Publikum: Keep going.] All right so it's like four hundred to one...

Let's add a couple more aluminum refineries as well to the Crusader L4 and a ArcCorp L1...

Go ahead we just added all of this new aluminum so let's head back to that Hurston factory and see what aluminum prices are doing there. What's that factory having to pay for aluminum with all this new supply? These new mines, these new refineries? ...And there you see that the price is starting to fall off. Speed it up, speed it up a little bit let's see what it does and what you see [is] it continues to plummet. Those prices are pretty much falling off a cliff now.

From all that extra supply that's been brought on without us having to go back and rebalance anything. Head on over to Cellin and you can see that Cellin is in very close proximity to Crusader, so you've got to refine aluminum refinery and aluminum mining, yet there's very little activity ...and the reason is because the aluminum at that mine is very difficult to extract. Go ahead and adjust the mining time on Cellin from 180 to 120 ... And with this again reflects is how difficult is to extract from that particular location

And so you see there now you see a few guys are actually willing to put in the amount of work necessary to extract it. Not all of them. It's still a lot of effort and this comes back to every one of these guys having their own individual set of traits. Some of them were willing to injure yeah more risk more repel, ...but you know more effort that sort of thing so.

Go ahead now and increase the Laranite consumption ... just go with normal pace. (...) Make the Laranite more difficult to work with and more rare and see what happens. There so we're gonna adjust the global Laranite refining time from 60 to 120.

JW: Adjusting the time on Wala...

We're gonna do both. The global refining time, we want to take it to 60 to 120 and Wala's extraction of time goes from 60 to 180. (...)

What we've just done is basically make it more time-consuming more difficult to refine it and we basically made it more difficult to pull it out of the ground. ... And apply a macro also (...) to reduce the default purity of Laranite from 50 to 16 percent, which means that refineries are going to need more ore to produce the same amount of refined product.

Take a look at the Laranite prices on the Hurston factory... we just did three different things to make it considerably more expensive for Laranite to be utilized and... speeded it up... and you can see that Laranite prices are going up. It looks like they're starting to skyrocket this could be a complete economic disaster. You notice they're just going up... What we're gonna do now [is] to bring that back in the control... Adjust the power plant formula to compensate a bit... adjust it from instead of two Laranite to point one.

We just reduced the amount of Laranite needed in the economy by 95 percent (sic!). Check out the prices again. What we should see here in a minute is Laranite should start to drop off pre-significantly... There you go ...

TW: ...crisis averted ...

TZ: And this will continue to drop if we watch it you always get it a little bit of turbulence because you've already got contracts you set up themselves.

Leave it in. I just wanna see it drop a little bit more (...) it's already down 40 to 50 percent and it'll keep, it'll keep falling.

Okay. Add some additional factories Delamar looks pretty dead, there isn't a lot of economic activity happening over there. Add a bit of life, by adding a new power plant factory... And yeah let's add a Hurston L4 factory for power plants as well to create a little bit of competition for aluminum. ... Head on over to Hurston Factory and let's check out their workers.

49:40.

You notice that their workers are falling off a cliff. what is happening? Well we just open up a new power plant factory and they're getting offered better wages for another place... And what are the wages on ... Hurston (...) that they were offering (...) Delmar offering like almost three times the weight is yep so there you see the differential which is that factory is more desperate for workers (....) basically the workers skyrocketed on Delamar and plummeted on Hurston and eventually, those wages, of course, stabilized ... they find a point of equilibrium (...) Let's go ahead now and make some major changes, we're gonna do this with a macro and you can see where we currently stand on slide ... JD: There's like a bunch of stuff that we were supposed to be showing here. Absorb it quickly everyone... TZ: There we go... We've added lots of aluminum, titanium and agricium and the entries that you see on the screen mark system ... all of the moons now have deposits... we've added a bit more layer night but it's still

pretty rare, we've also added a lot of new refineries. Qe've added cooler and quantum drive recipes to most factories and they require aluminum, laranite titanium and agricium. he demand for coolers in quantum drives has been introduced and we've also increased the demand for power plants and lastly we've increased (...) the number of Quanta from 1,000 to 2,000 so that we've got enough workers to keep the economy humming. Now previously we've seen pretty

obvious cause and effect. at this point though the economy is starting to get pretty complicated. We can look at a few graphs just see what's going on with some of the prices to see this. and this is one of the most interesting things about a really complicated economy, which is these changes in quantities and pricesm the purposeful movement of Quanta, this dynamism. these are all opportunities that you'll you're going to be able to exploit within the game and therein logical motion. ... JD: A shame if somebody came to steal from these people - huh. TZ: bring up slide (...) 52:49

So now we've we've introduced some some pirates. let's go ahead and so you can see that there are four pirate crack and there's the nova rioters lowriders ninetails and dusters pirates have to return to one of their bases in order to refuel rearm let's go ahead if you notice they head to the areas of highest value if you let's go ahead and zoom in there and you can see that those are missions being created by the NPC is no different than players would in other words you see a lot of deaths on a route that's got a lot of value and go ahead and hover over some of those contracts

these are contracts that are being thrown out by the NPCs on that route what's happening is the NPCs are basically being you know some of the freighters are being picked on by the pirates the ship is being destroyed and now that NPC is has ejected and needs transport back to civilization and so they're requesting that and these are missions that would be fed back into the game that you'd see and you could wind

up accepting these let's go ahead and take a look at Hurston the cooler is upon us out this is where they're operating out of you --- what do you wanna hear to check out the Hurston let's look at what happened to the cooler and quantum drives yeah so since we've now got a lot of pirate activity on that route we want to see what's happening to the cost of some of those Goods for a cool half-a-million

UEC you can get yourself your own power plants it's actually normal piracy is really terrible for economics you guys so yeah so let's go ahead now and really see it over at the coolers or it started it like a really reasonable thing yeah that's that's peg it's just it's the scale so basically they were down around a thousand two thousand and now they just absolutely skyrocketed in price because so many of the freighters moving whether it's supplies are moving you

know the finished product from one location to another are now being intercepted and these cuantas I mentioned earlier it's like they have various traits and some of them are very risk-averse and they won't go down routes where you know there's a significant chance of death and so to counteract some of this danger let's go ahead and bring in someone to fight pirates so here comes some security and what you'll notice is that security is drawn to areas of conflict where there is a lot of deaths and such so they're

gonna wind up gravitating towards those areas where there's a lot of piracy and there you can see I'm going over to speed it up a little bit not 200 it's just 200 too fast difficult Oh what's going on and if you notice by the way the police will wind up patrol Ian's area so they're in constant motion whereas the pirates basically are enabling quantum interdiction fields that's why they're basically picking a route on you know they're picking a particular location on the route between you know to between two locations and they're basically you know lying in wait and the security which doesn't know exactly where they are is scanning the area looking for them and so what you notice here is that the pirate situations starting to look a lot better now that Security's shown up and you would see

---

this reflected over in the prices and so what you have here is a perpetual game of cat-and-mouse to where the pirates look for the areas where they can reap the largest rewards with the least amount of risk and security is then drawn to them and as soon as you get ?? enough pushback from the system in terms of security then the Pirates look fo you know more opportunistic you know pastures and again this is very much like what you would expect to see you know you know an illogical functioning system

So let's go ahead and turn on the grid and what you're looking at here is one of the big problems I referred to earlier is how we wind up getting samplings of this you know of how we generate probability volumes and so what this is showing you is we can look at one of these high conflict areas and it's basically tabulating exactly what's in that area so you can see that there are ten total quantum Freyr eleven pirates five security it also shows you the total value being transported you know exactly what's going down that out? let's take a look at a few other locations here you see these security and pirates are fairly even six pirates seven security you know a decent amount

of value six freighters so it's a there's actually an almost equivalent number of freighters pirates and security this is the information that quantum and it's a small piece of thewhole this is what it would wind upfeeding periodically to the probabilityvolume services so that designers nolonger need to go in and say oh wellwe're gonna hard code a certain amountof security a certain amount offreighters and it can never change weship it that's all that's ever going tobe in this case the simulation isrunning it's constantly ebbing andflowing and your actions over on thegame side get factored into this resultjust like any of the quanta in thissimulationso for years we've had just the backend services and the game servers and thefundamental problem was that we didn'thave a single area where the entire gamecould be simulated where NPCs couldrequest goods and services because theyhave a legitimate economic need and could buy with players to provide thosethings we didn't have a way for thisactivity to determine what you see asyou wander around the universe thuseliminating the need to spend enormous sums of effort configuring vastquantities of static data that couldnever deliver the dynamic experience wereally wanted quantum solves all these

issues it also provides us with a lot more context which means that concepts not previously understood like how much risk there is in taking an item from point A to point B can now be easily calculated information like this is vitally important to the price discovery machinery and as I spoke about earlier in a properly functioning economy such things can and will generate a lot of  ripples we also get the information  necessary to properly customize mission content meaning we can deliver a lot more environmental and mission diversity without having to do any more work I've covered a few of the large areas that  quantum will impact but the effects will be felt far wider how many NPCs you see at a landing zone for example and what they're doing can now change over time  manufacturing towns might boom and bust depending upon how much Commerce is passing through them a didn't give in a moment NPCs can grow and evolve separately from their interactions with players into powerful bosses potentially and you'll be able to witness their growth in lots of different ways criminal criminal NPCs with a price on their head can move around the universe just like a player making time to catch them that much more fun we've got a lot more work to do before a quantum is ready for integration with the rest of the game but it's one of the last big steps in setting us up for a completely dynamic and systemic universe.

thanks for listening