Collective Economic Niggles

[Xit]

Not sure if this has been done before, or is being taken care of in the background already, in which case feel free to delete this thread.

Anyway, my idea is that whilst it isn't a priority right now, I think it would be good if as many people as possible could collect all the cargo/economical oddities they notice in the usual course of playing and testing, and collect them all in one place where solutions can be agreed upon. Then, by doing just a tiny bit at a time eventually all problem cargo types will be mapped out and relative values can be established or explained. I figure if we start now we could have a lot of ideas ready when there's enough time to implement them.

For example, from my last trade run:

Bulk seeds - should these really be cheaper at Atraxia than Atlantis?

Filtered water - should really be cheapest on oceanic planets and most valuable at small installations, yet most installations should be able to react their own from combination of hyrogen and oxygen? Not sure what the most efficent means of transporting or storing this is myself...

Seafood - should be cheapest on oceanic, not sure whether space installations or arid planets should find this most expensive.

Astral gases - ought to be cheaper in space than planets, surely.

Bonsai and livestock - should be at a premium in space due to the difficulty of keeping them, except at special installations.

EDIT : There already seems to be some control over the rarity of things like this, but I think there should also be a seperate buy/sell price, so to unload such things as fighter barracks you will get a very poor price, whilst buying them will be expensive. This is further complicated though by the premium of space on a station and the lifespan of what may be perishable goods. In the end selling price of odd goods at a location should reflect the convenience offered and the desire of trading posts to make a profit IMO
[/edit]

Also I believe prices should depend somewhat on proximity to sources and receivers of different goods, ie if in the same system as a source of x, then all facilities in the same system will have a cheaper value for this cargo, still keeping the normal trends in place. This could be extended between adjacent systems so that cargo would have a value proportional to the minimum distance it travels and the demand it experiences.

Edit 10/3/2008 - I was thinking that animal products shouldn't really be common, it would make more sense to go down the "vat grown meat" road, and leave animal products to be a delicacy only available from a few sources. This is based on the availability of modified bacteria, the huge amount of space, waste (and ethics) associated with animals, and the fact synthetic meat could probably grow anywhere with light and warmth, even on space bases.

[Coragem]

Thats a good post, I don't do a lot of trading myself but i really would like to know how the economics in VS work.

Also I believe prices should depend somewhat on proximity to sources and receivers of different goods, if in the same system as a source of x, then all facilities in the same system will have a cheaper value for this cargo, still keeping the normal trends in place. This could be extended between adjacent systems so that cargo would have a value proportional to the minimum distance it travels and the demand it experiences.

I agree. In system trading should exist, but be less profitable.

[Xit]

Another couple:

PAI Wetware - Should be least or at least rarely in demand at Fighter barracks, I think most planets could manufacture these but I don't know any back story on it...

Radioactive elements - Should only be found in real quantites on planets, and if anything should be more valuable to military installations for production of lithium 6 fuel?

Classical instruments - Really shouldn't be so valuable at fighter bases, but I can see that luxuries have special values in places like this.

I think we should have something like the bulletin board on Frontier Elite 2, where people can offer to pay extortionate prices for goods that simply aren't normally imported, or which have such high demand that there's never a surplus of them. This could be a random chance event or later on could also be caused by in system fluctations (say if the local trading post gets taken out the Aera) and by bases, industries and factions being able to consume so many units per time period. (Frontier was able to emulate this by having a normal baseline for all cargoes in a system, and current stocks would return to normal at a rate of so many units per day)

[whatch1]

Im fairly sure that the economy that has been inplemented so far runs on random numbers

To implement a universal wall street run by supply and demand would be difficult in the extreme 25 planet types 0ver a hundred cargo items, 40 some odd factions who supplies what. etc

A dynamic economy is an interesting possiblity currently getting mining equpment to serenity means going out of the system to get more that two of anything, mining bots, laser drills. even fitting out a new ship requires searching several systems for the right weapon sets.

This will go nowhere if the developers are not interested in doing this

mabee hacking up a good first start then presenting a kind of plan of how it should work would help

Bill
PS
Anyone remember Micro and Macro economics?

[ace123]

Yes, the economy is for the most part random numbers. There are certain random events that will cause surplus in certain systems and a shortage in others, but for the most part it is completely random.

This could be a really cool thing, and after a release, if we get some sort of multiplayer system running, it could be cool to add an economy into multiplayer.

Unfortunately, such an economy suffers from a scale problem: the main character is in no way powerful enough to affect the economy, so beyond certain trends, you can't really do so many interesting things with the economy.

[whatch1]

In a true economic system, thing are little more focused

Mines produce ore.
To produce Ore, mines need workers. workers consume food, drink, gasses, tools, and desire entertainment. All these have to come from somewhere else.

This is the simplest form. short one thing and get no output
Refining, Manufacturing, Distrobution and Commerce have similar requirements.

I have some idea i think could work insted of random numbers
some system could completly automated small ore haulers on a loop insystem as an example

Requests from others systems for tranport of need full or unique loads to generate missions

imagine going into a coal mine in west virginia and buying a star ship

i don't think so

Bill



[/quote]

[whatch1]

By trading on the businnes route i have completely depleted atlantis of coffee seafood and serenity of trekky gems and radioactive elements atlantis no longer stocks mining equipment the catagory is gone

These aren't replenshed untill you quit the gae and restart it

but on the mining, relay, fighter bases i can almost always find things such as personal tranport mining items and gold for god sakes

Radom numbers all for one and one for all?


Again i wouldn't think so

Bill

[javier]

IMO, the actual trading system is conceptually broken. It needs a complete rethinking. I remembered how Elite managed it and I still feel it was one of the best though systems I have seen. So, these are some ideas about how I think it should be:
- Import/export list could be created for each different planet or base from a template list, but introducing some randomization to make the list in same type bases similar but not a mirror of each other. A more realistic distribution of goods could be made, nailing some inconsistencies. The price of goods should be described as increase/decrease percents over the base price described in master_parts list.
- To calculate a base price there should be a system wide goods list. Each item of this list should receive a percentage discount over its base price. This percentage then could be added to the base type percentage to compute the actual price. This make in-system provided goods cheaper.
- There should be a production/consumption simulation ran at base level, at least in the actual system. At adjacent systems, the simulation could be done system level wise. There should be a maximum stock level for each base and good produced, and a threshold to reduce the price, say produced goods stocked after 80% of the maximum receive an additional discount of a 10%, needed goods with an actual stock of ten times the consumption rate receive a 20% price discount. This make inter system trade more valuable, and eliminated incentives to make the same run for an extended period of time.
- Randomly, there could be a news entry demanding some good at some base or planet, with a significant price increase and a time limit to deliver, a la Elite.

[Xit]

imagine going into a coal mine in west virginia and buying a star ship

i don't think so

There should be some random chance to encounter a ship for sale anywhere though, what if somebody comes to a mining base and then decides to settle there, they'll need to get rid of their ship (this might mean it'll come with a lower price tag). It's more like going to a coal mine and buying one of the worker's cars, it's unlikely on a day to day basis, but not impossible.

Also I think that running a full economy isn't really going to be 1000x more complicated than Wall street, as atm we don't have individual companies, and really there probably aren't more inhabited planets in the whole game than there are countries in the world today...

mabee hacking up a good first start then presenting a kind of plan of how it should work would help

The whole point of this thread is to get some rules together to decide how it should all work, which I think is the first step really, the current implementation is already a good first start for how it should work...

Unfortunately, such an economy suffers from a scale problem: the main character is in no way powerful enough to affect the economy, so beyond certain trends, you can't really do so many interesting things with the economy.

I agree, but if we were to ever go down the building bases and trading fleets, faction etc route that I've seen proposed for VS in 2011 then it would become a lot more relevant.

Also if a player could restrict trade down in a system, say by waxing lots of traders or bases, they might be able to secure a significant monopoly...

[javier]

Also if a player could restrict trade down in a system, say by waxing lots of traders or bases, they might be able to secure a significant monopoly...

Even that is a stretch of imagination. It's difficult to think one ship holding a lock in an entire system. It will be more plausible if you're hired to join a faction blockade fleet.

[Xit]

That is a possibility, obviously it wouldn't happen with one ship

EDIT: Actually, we would really need to see how many ships go through a system, and if the player can destroy maybe 5 of those due for the the day, how much of an impact does that make? We're inclined to assume there are thousands of ships but it should depend on the location of the system and the rarity of the goods being transported.

As whatch1 mentions, we really need goods to be updated with time for this to work. Maybe we actually do need to start at hacking something together and then tweak it to decide what's sensible, rather than planning it the other way round?

[javier]

From my point of view, the economic system first objective should not be to be fully realistic, but enhance the game discouraging routine.
One problem with any economy we agreed with is that it's only really important at the beginning of the game, when you're trying to build your ship or fleet. After that, some millions here or there doesn't make any difference and you're looking for some other objectives, and that's a real problem in VS due the actual lack of a back story or missions.
I'm thinking there should be some type of maintenance costs attached your fleet to somewhat prevent this effect.
In a wider sense, you could say the game will be fun as far as you feel you have something more to do, or have some objective to accomplish, and the economic system must help to present challenges to the player.

[Xit]

After that, some millions here or there doesn't make any difference and you're looking for some other objectives, and that's a real problem in VS due the actual lack of a back story or missions.
I'm thinking there should be some type of maintenance costs attached your fleet to somewhat prevent this effect.
In a wider sense, you could say the game will be fun as far as you feel you have something more to do, or have some objective to accomplish, and the economic system must help to present challenges to the player.

Agreed, we'd just end up with Elite where beyond getting the biggest ship there's nothing much to do - I think the economy should be directly linked to the success of individual systems, the factions in them, and the availability of services at bases. This could even extend to some product becoming unavailable because a system is too starved to be productive. Maybe some special high power laser or ship will only become available later in the game when you can ensure that system Xxxooogyyyoig is freed from Aera attacks on cargo.

Maybe this would help us to appreciate a little more that some areas can be war torn, and give players more of a vested interest in the progress of the war...

[javier]

That is a possibility, obviously it wouldn't happen with one ship

EDIT: Actually, we would really need to see how many ships go through a system, and if the player can destroy maybe 5 of those due for the the day, how much of an impact does that make? We're inclined to assume there are thousands of ships but it should depend on the location of the system and the rarity of the goods being transported.

If that is going to made a significant impact, you can bet on a weak economy, so there should be no big gains in doing it. Again, what makes more sense is you to be hired by some faction to disrupt another's business. And that's a mission.

This kind of actions, at an individual level, only make sense attached to the existence of rare/strategic or high valuable goods who only could be found in a few places across the galaxy.

[javier]

After that, some millions here or there doesn't make any difference and you're looking for some other objectives, and that's a real problem in VS due the actual lack of a back story or missions.
I'm thinking there should be some type of maintenance costs attached your fleet to somewhat prevent this effect.
In a wider sense, you could say the game will be fun as far as you feel you have something more to do, or have some objective to accomplish, and the economic system must help to present challenges to the player.

Agreed, we'd just end up with Elite where beyond getting the biggest ship there's nothing much to do - I think the economy should be directly linked to the success of individual systems, the factions in them, and the availability of services at bases. This could even extend to some product becoming unavailable because a system is too starved to be productive. Maybe some special high power laser or ship will only become available later in the game when you can ensure that system Xxxooogyyyoig is freed from Aera attacks on cargo.

Maybe this would help us to appreciate a little more that some areas can be war torn, and give players more of a vested interest in the progress of the war...

Wow. We're really crossing messages left and right
If you remember Elite, that wasn't the real problem, but the limited range of missions you are offered in your ascension within the ranks of Federation or Empire. Once you reached the bomb missions, there is all repetition until you got bored. But the economic system was very well thought to help the game objectives, simple as it was, and the building phase was entertaining.
In some ways, I think the ability to have a fleet of owned ships disrupts the game more than it helps. It prevents you to careful planning the best ship to accomplish your objectives, and then having to travel across the universe until your equipment is complete.

[shadow_slicer]

I've thought a bit on how to model the economy.

Currently vegastrike uses random numbers to generate its 'dynamic' economy. As long as bases are of the same type, they are all treated identically, regardless of faction ownership and distance to necessary resources. This is fundamentally wrong, and makes every star system seem generic. There is no motivation to break through a blockade or protect a valuable trade route.

There are several things that need to be done to remedy this. First as @javier said we need an import/export list, and model production and consumption. If we want to support the dynamic universe present in vegastrike, I don't think we can do this statically. If Atlantis was taken by the aera, it wouldn't make sense for them to still demand Trekky crystals. Instead the aera would slash prices on the current stock of useless items.

We need production/consumption information (and base price) information for each planet/faction combination. With this information we can heuristically determine the supply and demand of a good. Then all we need to do is model the interactions of all the planet's supply and demands.

To me this problem seems amenable to nodal analysis. We represent the universe as a resistor network with nodes being planets and jumppoints, and resistors being the space in between. The currents into and out of each node could be equal to the supply and demand. Inside a system, the resistance could be based on distance, faction approval of item (ie is item contraband), presence of asteroids or other hazards (pirates, battles, etc). Within each system all the nodes form a connected mesh. Between systems, the resistance could just be some constant to represent the cost of a jump. This would form a system of equations that could be solved through either Gaussian Elimination, Gauss-Siedel iterations, or some other method to find the node 'voltages'. These voltages could be scaled and become the price. Further the commodity flow between nodes should be equal to the voltage difference between nodes divided by the resistance. If the AI system has access to this information it could order freighters to follow the routes provided by this method.

[spiffariffic]

To me this problem seems amenable to nodal analysis. We represent the universe as a resistor network with nodes being planets and jumppoints, and resistors being the space in between.

Are you an EE? Your analogy works, but is... different.

For the sake of simplicity, I would leave out interest and the time value of money (as time goes on the money becomes less valuable because of interest and inflation). Also trying to model a stock market would probably be outside the scope of a space-flight simulator.

This is not to say, modeling economics is too much, actually I think it is important, and the model above sounds like a simple enough way of doing it. (I'll have to review my econ, but I think it will work)

By the way, yes, I study business and finance.

[Xit]

@ javier - I ommitted the whole missions thing becuase I know VS doesn't have any ranking system right now. As for the fleets of ships, I think that really comes down more to the infancy of ship ownership in VS, all the weapons, the rules on SPEC, changing vessels across systems, and how useful this makes the different ships for a given purpose is all seemingly waiting for big overhauls. Maybe if we had to wait for the delivery of our ship that's completely-different-to-the-everyday-one but ideal for the mission we want to do today, we'd be more inclined to think about all purpose craft (I really hope this is what you intended in that post)

@ shadow_slicer - that sounds like quite a good way of putting it, how efficient are gaussian elimination, or gauss-siedel iterations, and would they scale up well enough?

[javier]

@shadow_slicer
Seems a good idea to me. The problem could be the design and extension of the database needed, as you need to relate goods with planets and factions. The relation between goods and planets/bases should map what goods could be produced there. The relation between goods and factions maps if one good is desired, not desired or deemed contraband by a faction.
There could be also a distinction between natural resources and manufactured goods. An acuatic planet owned by R'laan not necessarily produce entertainment, but for sure water, even if that faction doesn't like or need it.

[Breakable]

I would propose to do some actual work, and just describe how you imagine it in the wiki with some formulas and cargo grouping, because it is much easier for a developer to implement something what is clearly defined.
Even if the definition is 10% good it is better to implement it and then refine.


http://vegastrike.sourceforge.net/wiki/ ... The_Design

Regards

[javier]

@ javier - I ommitted the whole missions thing becuase I know VS doesn't have any ranking system right now. As for the fleets of ships, I think that really comes down more to the infancy of ship ownership in VS, all the weapons, the rules on SPEC, changing vessels across systems, and how useful this makes the different ships for a given purpose is all seemingly waiting for big overhauls. Maybe if we had to wait for the delivery of our ship that's completely-different-to-the-everyday-one but ideal for the mission we want to do today, we'd be more inclined to think about all purpose craft (I really hope this is what you intended in that post)

Couldn't agree more. I brought the issue only as an example, and because I feel Elite as a good example of how to have something not too convoluted but well adapted to its purpose. One thing we cannot overlook is how the concept fits with another parts of the game, so it's inevitable in this discussion some other problems will appear. Should be alright while we don't loose focus.

[javier]

I would propose to do some actual work, and just describe how you imagine it in the wiki with some formulas and cargo grouping, because it is much easier for a developer to implement something what is clearly defined.
Even if the definition is 10% good it is better to implement it and then refine.
Regards

I think it's a little premature. We are in brainstorming phase yet , and we are going to write about ideas that surely will be discarded later.

[Xit]

I started trying to draw out a plot associating the different cargo demands with different races, their uplifted servants and eventually also factions - but it's a long process and there's very little info in the Wiki for some of it. I could share what I've done but it's more likely comical than useful right now

[ace123]

It might be interesting if we were to have some areas of the game have vastly different resources than other areas of the galaxy.

The set of all systems is a pretty large graph, connected by jump points, where each node is a system.

While constantly going through each system at run time could work, it would be interesting if we built up a sort of two-dimensional list with an index of the number of nodes between.
This would also solve the mission jump routes problem -- it could find the distance between two systems, then follow the path that decreases the distance number.

Assuming we manage to store some ID in each system node, it will take up at most 65MB of memory with our current list of 5700 systems.

however perhaps it would be possible to define trade routes within each sector... Since Vega and Sol sectors each contain enormous numbers of systems, it would bring it down to about 10 MB...

Anyway, even though this might not work with one big 2D table, perhaps we could simulate a similar effect... basically populate each area of the milky_way with different sort of items.
For example, some things would be macroscopically equivalent and everywhere like food or natural resources, but other things like Entertainment or Tools or things that need to be produced, would only start out in various factory systems and propogate very slowly system by system.

Also, we could have events where experimental technologies are produced in a single system in human space and slowly spread.

Or, perhaps the Aera produce a cloaking device, and it slowly propogates closer to human territory until finally you are able to trade for it nearby.

[javier]

Another implementation approach could be through a light database engine like sqlite. There is no real need to load all systems data in memory if you could store the information in sql tables. It could help even in multiplayer as sql is usually portable enough so you can even support a more powerful database engine at the game server, if needed.
Hey! we even could name that VegaERP.