Economic Development Model - Opinions Please?

[Mark_Everson]

The latest-greatest version of the Econ model can be found here: http://clash.apolyton.net/models/Model-Economic.shtml


Please give it all the criticism it deserves

Don Weaver's (aka don Don) draft of an alternate / complementary model is here:
Proposed Clash Data Elements & Population Model
We'll have to pick between them quickly, or figure out some way to fuse them together to our advantage.

All the latest crop of models will be locked in soon (basically by 6/5 for sure), so speak up now if you see something you don't like. Remember to criticise them from the vantage of both playability and practicality to implement.

A special thanks to everyone who crashed to make the 5/30 deadline . I still owe a proposal on Diplomacy... Maybe tomorrow.

-Mark

[This message has been edited by Mark_Everson (edited May 25, 2000).]

[don Don]

(AKA Don Weaver)

Hmmm, yes; unfortunately I work and live mostly at night. As I look over stuff I haven't looked at for a couple of weeks (this project is moving faster than I thought) I see that my ideas are more of a population dynamics model, which naturally includes economic elements. I am at this very moment putting it into a form suitable for being picked over for any useful bits and pieces.

------------------
*a friendly note from your favorite heretic

[don Don]

A rough draft of my proposal is now posted on Mark's Clash site as Proposed Clash Data Elements & Population Model. It's kinda wordy, but I am trying to tie the whole population and economics model together.

Questions? Comments? Death-to-nitpickers flames?

[Mark_Everson]

don Don:

Your piece is chock-full of good stuff. I've been looking at your proposal especially with respect to the water requirements you set out. I'm trying to figure out the right scaling between the water available, and the area of the land you're talking about in the "Water Resources and Movement Costs by Landform w/Terrain Modifiers" table. Specifically say we had a city of 50k people. That requires 100 water in your model. In the table a river only provides 8 water. Is there some multiplication factor I'm missing to get from the 10-ish number for a river up to a 50k pop. city? And what is the terrain area for the terrain that you have in mind? Also, supposing I want to build an aqueduct to get to 100k population. What are the requirements (higher by x amount or whatever) of the water source for the aqueduct?

Thanks,

Mark

[Druid2]

donDon ..

I (think I) understand your ideas, and how they fit w/ Marks original proposal.

I forsee bitter land wars in early years over water rights... hmmmm.. that's about what happened if I remember...

For playability, I am concerned, however. This is *SO* much more complex a model than the civ-model... "good land, build farm, get food.. UGH!.. bad land, build mine, get minerals .. UGH!!"

Is there a way to simplify, even if we lose some realism in the process?

[Mark_Everson]

Druid2:

Thanks for the feedback . In terms of simplification, what are you looking for? Or what do you percieve is tough? The reason I ask is that this model can be even Simpler for the Player than in Civ2. If you only play with the tax rate, things will go along and you'll have... "big war, more taxes, crush enemy, Ugh", "war over, lower taxes, build economy, Ugh." Its only if you want to fine-tune the other stuff that it becomes progressively harder.

I agree completely that the model is 100x more complex than in Civ, but the complexity really only confronts the player when the player wants it to. This complexity may be an issue anyway...

For what its worth, in Clash v0.05, I have the old econ system, which like the one we're considering, made for possible hands-off operation,. When I'm feeling warlike in that
proto-game, I can go on a merry conquest phase and never even Touch the economy. How much more Un-complicated from the player standpoint can you get? Of course I'm losing some advantage with respect to the case where I'd be finetuning everything. But the 20% loss in effectiveness is made up for by being able to play several hundred turns worth in an hour or so.

-Mark

[Glak]

You do have to realize that some people will micro everything. My friends and I are some of those people. In the original civ I would check every city's production each turn. If I found a unit (I never made buildings, just settlers and chariots maybe with a few phalanx and catapults mixed in) one turn from completion I would check how many shields it needed and compare it to the city's production. So my city is making five and the unit takes only three? Switch from the forest to the water for this turn. Then the next turn I would switch it back. Yep I would have 20+ cities and I would do it just to get two more trade. I would also rename cities many times over the course of a game, just to make them fit my naming system.

So what I am saying is expect people to try to control every single variable. The game should be complex enough so that better players have an advantage over others but it should be simple enough so that multiplayer works (since multiplayer is all that matters)

I didn't read the economic models (connecting at 9600 bps is frustrating after being used to dl'ing 20 megs in three minutes) so I can't comment on them, however I just wanted to mention that grabbing new land should be more profitable than improving existing land, as it is in all good games.

[Druid2]

True, Glak.. it is more profitable to grab new land than enhance existing land in the other game.

Not so in life, however, where you have to support the enormous cost of military operations and make your people unhappy caused you killed a bunch of their kids. AND you have to assimilate the new folks who wont be as useful as the prior citizens [Can anybody say "East Germany"].

We're going to have some of those features in Clash, so I think it's going to be *SIGNIFICANTLY* different than the "other games"...

------------
Mark,
I'm one of those who MicroManages as much as I can. But if it gets to the point that it's more fun to stay late at work instead of playing, 'cause the game is too complex, it'll just be shelf-ware....

We have to strike a balance, and as is usual for this design: if the AI is good, all will be well. R2D2 where are you when we need you!?

[This message has been edited by Druid2 (edited June 02, 1999).]

[Blade Runner]

I like the economic model. Quite complex and different than the original CIV style.
IMHO the water and agriculture stuff are realistic, but probably to difficult to implement and enjoy. My second problem with the water stuff is this provide to big adventage to a civ with good resources. Maybe we can stick to an easiest model even if that is not so realistic.

Blade Runner

[Kull]

Economic Model Comments:

First, my compliments to Mark and Don. Your models are detailed and very high quality. You have raised the bar, gentlemen! With that, let the nit-picking...uhhh...constructive comments begin:

MarkE - "How the game model for infrastructure works":

Infant Mortality: The commentary links low ancient world life expectancy (@40 yrs) to "education shrinkage". And on the face of it, this appears to make sense. But isn't the average largely a factor of HIGH INFANT MORTALITY? Past the age of seven, most city dwellers could expect a lifespan which would carry them into their sixties. As for the "Short, nasty, brutish lives" of the country folk.....well, they're not big education consumers anyway! It's also true your shrinkage factor is very low (3%), so my entire argument is pretty close to a moot point.

Population Growth and Healthcare: Later in the section you note that improved healthcare will increase life spans, thus reducing the education decay rate. A corollary would be to greatly increase the population growth rate. All those children are now growing up to have families of their own.

Slave Labor: At the end of this section, it's noted that within a province "the people" will do most of the road building, while outside it the construction will be handled by "engineers". What about slave labor? Slaves were utilized extensively throughout the world until relatively recent times, as both production and agricultural labor. No economic model can be complete without the inclusion of this critical component.

MarkE - "Miscellaneous - Fisherman"

Overfishishing: As we are learning to our sorrow, the technology of fish extraction CAN lead to serious overharvesting problems. Should this be added to the model?

Don - "Population Unit Types by Occupational Class"

Clerical Class: I didn't see a "Religious" grouping in the table. It's quite likely that this class existed as far back as cave dwelling times (shaman), so it would seem appropriate to model them. There are corollaries with the "Official" class, so the easiest thing would be to group them there. But there are important differences too.

Don - "Water"

Bolsons/Acquifers: Not actually a section, but I wasn't sure where this observation belonged. Most of your discussion on water refers to surface or "ground water" type sources. In recent times (possibly earlier?), man has learned how to tap extremely large and deep sources of water. I'm very cognizant of this because the city in which I live (El Paso, Texas) draws most of it's water from the Hueco Bolson. This deep water source provides enough water to support a city of 600K plus another 1.5M across the border in Juarez, Mexico.(And probably another million across the desert southwest.)

SeaWater Distillation: Many of the large Persian Gulf cities now gather much of their drinking water from this technology. As you may recall, during the Gulf War the Iraqis poured vast amounts of oil into the gulf in an attempt to destroy these filtration plants.

Bottom Line: Water truly is the fundamental arbiter of where people live, and how many can reside there. But technology DOES change the equation. Will we capture this fact in our model?

[don Don]

Mark: I had left the terrain references vague, but I made the direct connection between water loci and the "sites" in your econ model. I didn't want to make my model strictly dependent on the small square (or sub-square) ideas I had broached earlier. So in each map square there would be 16 terrain divisions, or a random number of generalized sites, with the town able to draw on water from any nearby. I do envision water control technology as increasing available water by factors of 2 (roughly) at each level.

The river/spring source is in addition to whatever the climate/terrain provides, and 8± is just a minimum to be worth including in the model. The table is a rule of thumb, and I assume that generating tropical and temperate rain forests or other unusual climates would be part of map generation. Any navigable river would be 2-4 times as big, and major rivers would be on the order of 100. I would imagine that continental watershed rivers (Mississippi, Amazon) could easily be many thousands.

I hadn't worked out the details, but I was originally thinking of a simple summation watershed runoff calculation in the initial map creation to get a rough size and distribution of rivers. Rivers are shown as "4W±" in Forest because forested areas have much lower runoff coefficient than unforested areas. Hills, plateaus, and mountains have higher runoff coefficients, so the available water is adjusted (but the runoff calc, if implemented, would use the raw figure). I didn't mention this because it may be too ambitious.

Kull: City population has to be pretty big before you'd have hundreds of religious officials/professionals that you might want to differentiate from the secular. The "Rivers/springs" adds to water available. Water control tech will enable tapping of deep aquifers (how to model that might be tricky).

[This message has been edited by don Don (edited June 04, 1999).]

[manurein]

Mark :

Here is what could be a simple model for demographic growth :

datas :

- G = generation length (or age of reproduction). Varies with culture, education.
- Nr = natality rate (number of children in a generation, between age of reproduction and 50 years). Depends on culture, education, contraception availability. Fe for modern epoch, occidental civs it would smth like 2 over 1 generation of say 25/30 years.
- Db = death at birth rate in %. Depends on general heath, and particularly on hygien. Depends on the availability of food.
- L = life length; depends on health
- Wp = Women population. If no difference is made between sexes, half the total population.

Bg = number of birth in a generation
Bg = (Wp/2)*Nr*(1-Dr)

Ny = number of births in a year
Ny = Bg/G

Dy = number of deaths in a year
Dy = 1/L

Ay = Added population in a year
Ay = Ny - Dy (may be negative)

Tl = turn length in years
A = Added population in a turn.
Then A = Tl*Ay.
You can round this to heads, and carry the remainder from one turn to the other.

The computings are at the province level.

Diseases are handled through generating a number of death which are substracted from the population, so are famines. If no famine, but food is not enough for the new population, Db is raised accordingly. I think we should allow for moderate pop excess to eventually have effects such as health decreasing because general nutrition, hygien (water)... is available. Will also allow for dispersion of people in new settlements and migrations for econ. reasons.

IMO mobilized population is substracted from province pop until it returns (or not...). Track is kept of this mobilized pop to allow for war unhappiness or eventually proudness, and other stuff like that. So in case of war death people simply dont return to the province. In case of civilian death (if the war wages in this province) this is handled like for disease or famine.

One point about the slaves : IMO they have had important economic roles in the history, for their labour, but also for the trade they generated.

See ya

[Mark_Everson]

manu:

Thanks for the details, this is about what I had in mind. On the slaves thing, I completely agree we should have the effect of slaves... But at least so far I think those effect could be pretty much contained in govt, technology and culture... Maybe I'm wrong, but that's my take on it for now. What is the connection between slaves and trade?

[manurein]

Trade OF slaves

[Glak]

Druid2:

By grab new land I didn't mean from other people, I just meant land lying around on the ground (I guess all land is just lying around on the ground because it is the ground but you know what I mean). For example if I wanted to double my farm production I go get some new land and make some more farms or I could force more production out of my current land. I am not referring to taking land that people are trying to keep from you. That should be handled by combat.

The reason for this is to force conflict. Conflict is what seperates games from entertainment (the sign of a decadent society). Which is more gamelike: two players struggling for availible land (not necessarily military conflict) or two players sitting at home digging irrigation ditches? CLearly the first is because it involves player vs player conflict (essential to the definition of a game). It is true that some people will run the program for entertainment purposes and that these people are over represented here however many others want to play a game as a game, a chance to make others lose.

Also remember that if realism and fun are found together it is pure coincidence. If realism was all that great I'd go outside and give it a try. Saying that Clash is different isn't a good reason. If Clash is different is is almost certainly either better or worse than other games. If Clash is better in this particular way then why haven't other game makers caught on? I am not saying that this or that particular idea is better or worse than the stardard but that different often (but not necessarily) means worse. In order to "prove" (the value of any particular game concept varies from game to game of course) that a concept is better you should be able to express why. Good game mechanics do one or more of the following: increase conflict, favor the better player but make the game as close as the skill levels are (example: a great Clash player should win most of the battles but the other guy should win some too), teach the player how to be better at the game, are simple but lead to complex interactions, etc..

Ok done