Advertisement

What makes a good 4x space game economy model?

Started by November 29, 2009 04:02 PM
26 comments, last by Platinum_Dragon 15 years, 1 month ago
All I'm saying is "At the very end, currency will evolve into such that there is a universal medium that all value will be based on." The currency is based on this medium, and increasing the money supply creates inflation while decreasing money supply causes deflation.

Edit:
On Earth, gold outnumber Silver by five times in mass, but gold values about fifty times as much per unit of mass. From there, we see that globally, the real world based on Gold not Silver. There is not enough silver to back up M0 money supply, but there is enough gold to back up M0 money supply and in certain perspective, there is enough gold to back up M1 money supply.

Edit:
-Currency should be based on a universal medium that is limited.
-Regardless of how many types of currency there are, the total value of all currency is equal to the total value of the universal medium.
-Inflation and Deflation affect currency value locally not universally.
-Specific inflation and specific deflation occurs to the value of individual goods, though they can be dependent on the inflation and deflation of transport/energy goods.
--As an example, (scifi example) a planet under siege will have prices raise very fast because goods are limited, and there will be no imports during this time period. An appreciation of value for goods make a local depreciation to the currency aka inflation without a change of money supply.

Edit:
Why interest rates cause conflict!
Here is the list of logical steps:
Loan 100 currency X
Money Supply increase by 100
larger money supply => inflation
10% interest flat rate
When loan is over: Money Supply decrease 110
Total Change: Money Supply decrease => deflation.
Interest Rates is the cause of deflation in the long run.
Deflation is usually associated with recession.
If deflation follows a score (20 year period) of low inflation, then it cause recession.
The only way to leave recession is by war.
The deeper the recession, the larger the war needs to be.
The World Wars (I and II) are the best example of this cycle.

Real world interest rates are compounded. Its overall effect cause a negative money supply. Debt has to always be there; therefore, loans must not be paid off.

What is money supply after all loans are paid off?
Money Supply M0 = 2*M0 - M3
The M0 on the left hand side is when all loans are paid off.
The M0 on the right hand side is before the loans are paid off. You must remember this is an important fact.
If the left hand M0 becomes negative, then all values come from debt not intrinsic values that we assume. The only way to get out of this is by a government printing money without any assets to back it up and use its power to force the citizens to value the printed currency equally without causing inflation.
Economy can only function with positive M0 conditions.

[Edited by - Platinum_Dragon on December 25, 2009 9:59:02 PM]
I use QueryPerformanceFrequency(), and the result averages to 8 nanoseconds or about 13 cpu cycles (1.66GHz CPU). Is that reasonable?
I though that the assembly equivalent to accessing unaligned data would be something similar to this order:

  • move
  • mask
  • shift
  • move
  • mask
  • shift
  • or

    So it seems reasonable to say that it takes 14 cycles for unaligned data since we'll have to do the series of instructions once to access and once to assign?
Edit: This wiki on Money Supply helped me understand the M0~M4 classifications of money. In part, it says
Quote:
Laura takes [her remaining nine $100] bills and deposits them in her checking account at her bank. The bank then calculates its reserve using the minimum reserve percentage given by the Fed and loans the extra money. If the minimum reserve is 10%, this means $90 will remain in the bank's reserve, and the remaining $810 can be used by the bank as lending money. The M1 money supply increased by $810 when the loan was made (assume no further loans); money has been created. (M0 = $900, M1 = $1710, M2 = $1710)
Lot of numbers there, but my take-away is that putting a dollar in the bank causes the bank to mint dollars of new loan money. Seems like an impossible-to-maintain situation, where the number of dollars skyrockets on a positive feedback loop.

Privatized Money Supply
Quote:
In 1991 legislation was quietly passed that eliminated required cash reserves by mid 1994. The result? Figures from the Bank of Canada Review show that by September 1998 the ratio of the banks' cash reserves ($3.893 billion) to their total assets ($1393 billion) had soared to 1:358, a ratio that was never more than 1:15 in the first half of this century. That means for every dollar of cash in their vaults or deposited with the central bank (i.e. the BoC) the banks have conjured up $357 from the void which they've invested or lent out with interest. Hence the record profits. Meanwhile not one person in a hundred grasps the fact that our government permits private banks to create about 95 percent of our money supply bringing huge profits to them and endless debt to us.

So making a game economy work is probably best not modelled after the corrupted usury of real life.

[Edited by - AngleWyrm on December 26, 2009 6:37:25 PM]




--"I'm not at home right now, but" = lights on, but no ones home
Advertisement
Quote:
Original post by sitwind
I think the economic system in a game should not be to complicated, not too much effort from the player to understand it. And players who are not interested in real world economy will certainly not have an interest spike just because it is in a game.

If a perfect system is the goal, the real world is hardly a shining example. If a realistic system is the goal, then possibilities to "cheat" the system should be available.

About the inflation thing, it is certainly interesting. But all the other resources are produced without any limit, e.g. coal , steel. If there is a big surplus, price should go down according to the market mechanism. So the problem is no only about the money inflation but everything.

At least myself would not want to put so much effort into a "real" world economy model. Especially if a simplified model could be at least equally fun.





I agree, a game economy should be simple for the player to use. The main problem with real world economies that make them so complicated is all the laws and regulations surrounding them to stop fraud and to govern how companies (banks, etc) handle people's money.

In a game economy, the computer is all seeing and all powerful so no fraud is possible. This means that the system can be much simpler.

In games, you are not creating a bank or the legislation that governs its workings. What you are trying to do is to create a system of exchange.

Players can then create their banks and such if the want but that would then be part of player/player interactions, and not something you need to implement. It is because of a robust system of exchange that such emergent features (such as banking and stock markets) can exist.

In real life the robustness of money emerges from the fact that money started out as a physical object (or at least what you are exchanging is). Although new systems and non physical objects can now be exchanged, they were all built up from the fact that money was just a physical object.

In a computer, money has not been based on such a thing. Everything in a computer game is ephemeral and has not physical structure. You can create or destroy something in the computer game and the laws of conservation don't have to apply. If you want these laws of conservation then you have to specifically program them in. That is what my system does.

The premise that it is built up from is that nothing can be created or destroyed (but you can cut it into smaller pieces). I have taken this "object" to be the total value of the economy (what the metric is I have left up to the game designer as this can be many different things as the real world has shown - from pigs to gold to labour to shells to land - although for a game, land or labour are good metrics to use).

So when someone buys something the money is not destroyed, but instead it is moved to the seller's possession.

When a Authority (player or AI) creates currency, this does not create value, but divides the value of the economy between the units of the currency.

eg: if the total value of the economy was 10 units of value, and you had 10 units of currency, then each unit of currency would be worth 1 unit of value. But, if you then created 10 more units of currency (for a total of 20 units of currency), then each unit of currency would only be worth 0.5 units of value.

The value of the economy is maintained despite changes in the amount of currency, it just means that each unit of currency does not have as much value.

However, it is possible to generate more value in various ways (beneficial trade deals, increasing the supply of raw materials and labour, etc).

For any economic system to work in a game you must implement these two conditions. That is the law of conservation (of total value) and that of division (of the total value into currency).

If you look at my 6 rules you will see that these 2 principals are the guiding principals behind them (I included localisation of these effects as the middle two rules - which is needed if you are going to have multiple regions or currencies in your game).
If you look at my argument, then what I have done is criticize the real world economy. The only thing that I say is "There must be a universal medium for currency."

The most simple form of economy is a limited economy. Everything has a limit, and "physical" entities must follow a cycle of matter. This cycle needs to be design.
There is no such thing as growth, but the fact that more of the resources are in the hands of the player and other character instead of the environment. In the end, the resources must be limited.

Those who can make money control the inflation and prices. They will stay rich and the gap cannot close, so the only way to stop that is by not allowing anyone to create money.

We should all agree that "Law of Conservation" is the only way to have a fair and simple economy.

Edit:
After a few calculations, I end up with a conclusion.
If all loans are paid off, we will have about negative fifty trillion dollars (United States Dollars).
I use QueryPerformanceFrequency(), and the result averages to 8 nanoseconds or about 13 cpu cycles (1.66GHz CPU). Is that reasonable?
I though that the assembly equivalent to accessing unaligned data would be something similar to this order:

  • move
  • mask
  • shift
  • move
  • mask
  • shift
  • or

    So it seems reasonable to say that it takes 14 cycles for unaligned data since we'll have to do the series of instructions once to access and once to assign?
Conservation of Wealth and Conservation of Chaos
Mowing the lawn produces a pretty lawn and an unsightly bag of leaves. Breathing separates the air into used oxygen and unused gases. Even winning a fight seems to redistribute the contestants' health into stronger winner and weaker loser. Could this also be the case with wealth?

Perhaps there is such a thing as an organized and healthy richness vs a chaotic and diseased poorness, and those that successfully strive for wealth simply separate the rich from the poor.

That last paragraph just totally blew my morality sensor off the scale.




--"I'm not at home right now, but" = lights on, but no ones home
It is still undeniable that the only way to make money is through two ways: one, bankers create it from nothingness (aka giving out loans); and two, by getting it from someone else. The second state is what everyone believes in, and this condition does not violate the law of conservation. On the other hand, the first state does not violate the conservation by maintaining the same total value of the money. Therefore, the only way to make money is by getting it from someone's money either by getting the money itself physically or the value of the money through inflation.

Edit:

Off-topic:
I remember several years ago, I also believe in a "Universal Basic Income" that is enough to keep someone living, but does not have enough for them to enjoy. The problem with this is that many people will still choose to enjoy instead of living. An example of such case is in the States: our homeless still wears better clothes even if they don't have enough to buy food. That's why I become less decisive towards a "Universal Basic Income" because how can you tell if they are going to use it for living or for entertainment. If they starve themselves for entertainment then they cannot be saved. Those who learn to entertain themselves without anything of monetary value are the only ones that can be saved because these people are the only ones who will use the "Basic Income" to get their necessity.

Edit:
off-topic?
I believe that Chesterton is correct.
"There is less difference than many suppose between the ideal Socialist system, in which the big businesses are run by the State, and the present Capitalist system, in which the State is run by the big businesses." -G. K. Chesterton

[Edited by - Platinum_Dragon on December 27, 2009 11:16:48 PM]




I use QueryPerformanceFrequency(), and the result averages to 8 nanoseconds or about 13 cpu cycles (1.66GHz CPU). Is that reasonable?
I though that the assembly equivalent to accessing unaligned data would be something similar to this order:

  • move
  • mask
  • shift
  • move
  • mask
  • shift
  • or

    So it seems reasonable to say that it takes 14 cycles for unaligned data since we'll have to do the series of instructions once to access and once to assign?
Advertisement
Off-topic: Maybe entertainment expenses are further up the needs chain than greedy influence would have us believe. Also, I'm pretty sure there is something wrong with the way budgeting is taught; it looks way too much like dieting. The healthiest thing I've found is to take the difference between what's supposed to still be in my pocket and what is actually in my pocket, and mark it down as spent on disarray. If I can keep that down around 10% then I'm satisfied. With more effort I can get it even tighter, but there comes a point at which the extra money saved isn't worth the time spent fussing over it.

On-Topic: There's something fun about deciding how Other People should spend their money, almost because it seems to go against the definition of 'their money.' As an example, not too long ago there was a thread on these forums about spending a character's level-up points, and how they shouldn't be allowed to just do as they please. Another example, in GalCiv2 I would sometimes give away various technologies to my enemies, so that they would bog down their production queues with petty upgrades.





--"I'm not at home right now, but" = lights on, but no ones home
Off topic:
Sorry greenpower, We just have a little debate over small trivial ideas so we can grow our mind about other people's perspective. Hitting ideas from a slight difference angles, we see that people can interpret things differently.

In strategy game, tactics and logistics are trivially the same abstract level of game play. They are the two most important aspect that makes up strategy. Being at the same level, if you simplify one, you should also simplify the other. Because you do not want the players to frustrate over micromanagement, you need an easy control.

How to simplify:
Use a hierarchy control structure with a limited level and limited action in each branch.
Beginner Mode needs 2 levels of 2 actions, 2^2 = 4 total actions.
Very Easy Mode needs 3 levels of 3 actions, 3^3 = 27 total actions.
Easy Mode needs 4 levels of 4 actions, 4^4 = 256 total actions.
Normal Mode needs 5 levels of 5 actions, 5^5 = 3125 total actions.
Hard Mode needs 6 levels of 6 actions, 6^6 = 46656 total actions.
Very Hard Mode needs 7 levels of 7 actions, 7^7 = 823543 total actions.
Extreme Hard Mode needs 8 levels of 8 actions, 8^8 = 16777216 total actions.
Hardest mode should be 9 levels of 9 actions, 9^9 = 387420489 total actions.

The best that an average player can handle is 7 levels of 7 actions, but we have to remember that some people cannot even remember 7 different things. The spread of 7 +- 2 in the past is not valid. The new spread is 5 +- 4 items that someone can remember. People do not train their memory as much as people of the past, so you have to give these players the time to redevelop their memory skills through different difficulty levels of control.

As seen, those example games you posted on your article are all about simplifying to within the threshold of memory. Actions must stay within the boundary that the player's mind can handle. A good game model, regardless of which kind, will take player's limitation into account, and build upon the limitation of the players as the central control concept.

Abstraction is a simplification of concrete. A game can have the greatest details and yet the most abstract controls.

The split of the game:
Logistics -> Production for more income or production for more units
Tactics -> fight to steal others' resources or to defend resources
There that's 2 level of 2 controls. Top level split into tactics and logistics, and the bottom level split into offense versus defense.
Then the game is about dynamically shift only 2 shifters during different time frames. And that is all about strategy game at the most abstract level (in my opinion). The dynamic change of these shifters at the right moment is the same as any other strategy game, the timing is what really matters, not the micromanagement skill level.
Oh yeah, I need Sun Tsu's quote: "Strategy without tactics is the slowest route to victory. Tactics without strategy is the noise before defeat." Timing is strategy while micromanagement is tactics/logistics. Remember this well and you will understand that micromanagement is not the way to win a game.

off topic:
Sorry for the long post, but I look at the big picture before the little picture and return to the big picture.





I use QueryPerformanceFrequency(), and the result averages to 8 nanoseconds or about 13 cpu cycles (1.66GHz CPU). Is that reasonable?
I though that the assembly equivalent to accessing unaligned data would be something similar to this order:

  • move
  • mask
  • shift
  • move
  • mask
  • shift
  • or

    So it seems reasonable to say that it takes 14 cycles for unaligned data since we'll have to do the series of instructions once to access and once to assign?

This topic is closed to new replies.

Advertisement