Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 ... 7 8 [9] 10

Author Topic: Modelling a Global Supply/Demand Economy (Please!)  (Read 14515 times)

Squirrelloid

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #120 on: March 11, 2009, 10:55:55 am »

Ok, as *not* a professional programmer, I can think of a fast implementation.  I can't imagine a S/D economy adding more than 1% additional time to world gen.  Heck, i just described a good finding algorithm for matrices in my last post. 

I am assuming that the local area is entirely accessible to a settlement, and that it uses/produces some predefined fraction each year.  A depth-related function for that fraction isn't unreasonable (although it would take more time).  That function could depend on race.

Also, do note I responded to relative prices in an edit to my last post.

Granite: he did specifically mentions the site finder, and did so because doing a settlement-specific audit of resources does require site finding.  That my proposal requires site-finding world gen just means that such time spent on site finding would contribute to world gen time. 
Logged

praguepride

  • Bay Watcher
  • DF is serious business!
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #121 on: March 11, 2009, 11:05:20 am »

As a professional programmer I can say: it depends.

DEPENDING on how the code is currently implemented...
DEPENDING on the strenghts/weaknesses to the language...
DEPENDING on the way the new code is implemented...
DEPENDING on about 1,000 other variables...

This could either be a natural extension and add maybe .000001 seconds to worldgen OR it could kill the game.

I haven't looked at the source code, I don't know what Toady knows.
Logged
Man, dwarves are such a**holes!

Even automatic genocide would be a better approach

Granite26

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #122 on: March 11, 2009, 11:15:24 am »

well, assuming resource counts at all sites, it should be pretty quick to calculate relative prices based on what's out there.  It's simulating the usage and decisions over time that eat up your cycles.

praguepride

  • Bay Watcher
  • DF is serious business!
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #123 on: March 11, 2009, 11:19:38 am »

well, assuming resource counts at all sites, it should be pretty quick to calculate relative prices based on what's out there.  It's simulating the usage and decisions over time that eat up your cycles.

I wouldn't assume too much. I was shocked when I learned how fluid dynamics worked. Its' neat and it works, but definitly not how I thought it did.
Logged
Man, dwarves are such a**holes!

Even automatic genocide would be a better approach

Granite26

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #124 on: March 11, 2009, 11:24:24 am »

agreed, but adding resource counts is literally a core item for development

praguepride

  • Bay Watcher
  • DF is serious business!
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #125 on: March 11, 2009, 12:02:10 pm »

True, but so is magic.
Logged
Man, dwarves are such a**holes!

Even automatic genocide would be a better approach

Mikademus

  • Bay Watcher
  • Pirate ninja dwarves for great justice
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #126 on: March 11, 2009, 12:15:58 pm »

About "management":
Spoiler (click to show/hide)

About "poisoning the well":
Spoiler (click to show/hide)

Ok, as *not* a professional programmer, I can think of a fast implementation.  I can't imagine a S/D economy adding more than 1% additional time to world gen. 

Never, ever make this claim.

About fun and economics: more is not always better. The key here is "realism". By being a meta-simulation of a hell of a lot of various things DF invites all kinds of appeals for "further" realism. Medically inclined individuals believe that proper simulation of all human physiological functions down to the mitochondrial level would make the game better. Medieval fight buffs believe that the game would be better if it implemented the collected corpus of all  German Fechtbücher (medieval military manuals). Some in the spoken magic thread would quite argue that a magic system where you manually have to repeat "owrds of power" or your wizard would forget them would be fun. Ecologists, geologists, zoologist etc wants DF to simulate convection, fluid dynamics, plate tectonics, evolution and natural selection etc etc as unconditional as possible. Would one of these make DF a better game? Possible. All of them? DF would be a friggin' nightmare.

More specifically about economics: anyone that remembers Master of Orion III? The two first games were marvellous. The expectations on the 3rd instalments were sky high. New developer, though... They decided to listen to the fans, and listen they did. One fan wanted more of that economy, another of that. All was provided. The game blowed. It was less of MOO and more of an excel spreadsheet. Nobody want to play a second job.

Any voluntary/recreational game is about fun. And quite unlike claimed in one of squirrel's posts above, I seriously doubt very many play DF for the management of it. The management aspects is just a tool to the rest of the game, the Rorschach inkblot test underneath.

Perhaps what the OP wants is just some tweaks under the hood, but it appears he wants the game to revolve around economy. That smells MOO3. That smells dull.
« Last Edit: March 11, 2009, 12:19:14 pm by Mikademus »
Logged
You are a pirate!

Quote from: Silverionmox
Quote from: bjlong
If I wanted to recreate the world of one of my favorite stories, I should be able to specify that there is a civilization called Groan, ruled by Earls from a castle called Gormanghast.
You won't have trouble supplying the Countess with cats, or producing the annual idols to be offerred to the castle. Every fortress is a pale reflection of Ghormenghast..

Granite26

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #127 on: March 11, 2009, 12:39:06 pm »

Improving the level of simulation and the rationality of reactions does not mean making the game all about it.  I feel your pain about MoO3, and I think it's a valid concern.  I don't see a correlation between given the players freedom to play in a +-10% level if they want, and making the game revolve around the mechanic to the point of not being fun for anyone but experts.

Spoiler (click to show/hide)

Squirrelloid

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #128 on: March 11, 2009, 01:14:25 pm »

About "management":

(snipped)

Spoiler (click to show/hide)

About "poisoning the well":
Spoiler (click to show/hide)

Quote
Ok, as *not* a professional programmer, I can think of a fast implementation.  I can't imagine a S/D economy adding more than 1% additional time to world gen. 

Never, ever make this claim.

Um, huh?  I've written code far more processor intensive than that which took less time than I could measure.  I do matrix manipulation as part of data analysis.  I'm not a programmer, that doesn't mean I don't use some basic programming as part of my work. 

Could I prove it?  I'd need to have better world map output than is currently possible, i believe.  Which is why I phrased it as opinion rather than fact. 

Quote
About fun and economics: more is not always better. The key here is "realism". By being a meta-simulation of a hell of a lot of various things DF invites all kinds of appeals for "further" realism. Medically inclined individuals believe that proper simulation of all human physiological functions down to the mitochondrial level would make the game better. Medieval fight buffs believe that the game would be better if it implemented the collected corpus of all  German Fechtbücher (medieval military manuals). Some in the spoken magic thread would quite argue that a magic system where you manually have to repeat "owrds of power" or your wizard would forget them would be fun. Ecologists, geologists, zoologist etc wants DF to simulate convection, fluid dynamics, plate tectonics, evolution and natural selection etc etc as unconditional as possible. Would one of these make DF a better game? Possible. All of them? DF would be a friggin' nightmare.

I would say that there's an easy bright line test to determine what level of detail any given element would ideally be handled in.  I'd say that bright line is dictated by Scale.  The first important scale is unit of focus.  DF handles (or intends to handle) from the level of individual dwarves to fortress government to 'national' government, including interactions between such entities (trade, diplomacy, and war).  Those are the relevant scales.  Biology too far below that scale is irrelevant and can be ignored.  Markets are *part of* that scale and are thus important.  The scale at which the player participates is local government, which is why one step up and down in scale are relevant - those are the scales he interacts with.

The second scale is time.  Evolution doesn't happen in short enough timespans for most organisms for the player to notice, so its not especially relevant (although there already is natural selection amongst megabeasts and civilizations).  Cellular events happen too rapidly to model.  Market forces happen at the scale the game occurs in.

Basically, markets are scale appropriate for the game.

Then we have to talk about what they add.  I think they add to depth, immersion, versimillitude, and game logic.  I also don't think this is just a 'realism' claim.  Resources are limiting, people do want things they don't have and engage in trade.  S/D should follow naturally from that, no matter how unrealistic the game otherwise is.  It would be like 2+2 not being 4.

Quote
More specifically about economics: anyone that remembers Master of Orion III? The two first games were marvellous. The expectations on the 3rd instalments were sky high. New developer, though... They decided to listen to the fans, and listen they did. One fan wanted more of that economy, another of that. All was provided. The game blowed. It was less of MOO and more of an excel spreadsheet. Nobody want to play a second job.

Ok, I never played MoOIII, but i imagine part of the problem was a scale miss-match.  At DF we deal with the production and sale of individual goods.  That's so far below the scale of a MoO I or MoO II game that tracking good-specific supply and demand is ridiculous.  The classic MoO production model abjures anything about actual market forces and simplifies it for the higher scale the game occurs in.

Quote
Any voluntary/recreational game is about fun. And quite unlike claimed in one of squirrel's posts above, I seriously doubt very many play DF for the management of it. The management aspects is just a tool to the rest of the game, the Rorschach inkblot test underneath.

Of course, I actually made an argument about why it was true.  You can't just dismiss an argument with a wave of your hand - deal with the argument or conceed. 

Quote
Perhaps what the OP wants is just some tweaks under the hood, but it appears he wants the game to revolve around economy. That smells MOO3. That smells dull.

I honestly think it would be more like some tweaks under the hood from the perspective of most users.  They'd interact with it three times per year, and some players might need to reign in particular egregious strategies of overproduction, but given the game specifically wants you to produce individual items of clothes for your dwarves so they can replace clothing as particular pieces wear out, this is actually less micromanagement than that (since you get specific information on what types of goods are currently 'short' in the world market from traders, not having to keep checking your stocks to see if you need more dresses or mittens).

Basically, its less of a hassle than some things the game already wants you to do, and more rewarding than those things as well.  (At least, I don't find having to specifically order socks, trousers, and hoods particularly rewarding).
« Last Edit: March 11, 2009, 01:18:56 pm by Squirrelloid »
Logged

praguepride

  • Bay Watcher
  • DF is serious business!
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #129 on: March 11, 2009, 01:58:05 pm »

I played MOO3 and I thought I enjoyed it. Then I played MOO2 and realized what was missing.

Do too much and it collapses under its own weight. That's why I like Toady's (apparant) design strategy of just faking the complicated stuff so that to the layman, it "looks" real, but it "really" isn't.

Poisoning the well? I think not!
Spoiler (click to show/hide)
« Last Edit: March 11, 2009, 01:59:49 pm by praguepride »
Logged
Man, dwarves are such a**holes!

Even automatic genocide would be a better approach

Granite26

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #130 on: March 11, 2009, 02:16:45 pm »

Spoiler (click to show/hide)

Nobody isn't saying that the correct level of detail/abstraction isn't important, there's just some debate over how much that is, as well as confusion on everyone's part about just how much additional management a more in depth economy simulation will take to be effective.

Taking the Micro/Macro example: 
Some people might find it fun to build their fort into an economic powerhouse, with their military might derived from the large number of steel clad warriors they can afford to field, with losses not meaning as much because dwarves still flock to your fabulously wealthy fortress. 
Others might wish to design their forts into objects d'art, complete with trap rooms and magma flows.
Finally, some players may wish to focus on the military simulation, ignoring the finer aspects of the finantial system and simply sending out more goblets in order to focus firmly on training their relatively fewer warriors into paragons of battle.

praguepride

  • Bay Watcher
  • DF is serious business!
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #131 on: March 11, 2009, 02:23:15 pm »

That's fine if a game can support it, but the game ALSO shouldn't cripple you because you don't have a Ph.D in economics/statistics/biology/chemistry/physics/etc. etc.

And making a game that can support the layman and the expert...good luck with that. Many have tried, many have failed.

Spoiler (click to show/hide)
Logged
Man, dwarves are such a**holes!

Even automatic genocide would be a better approach

Granite26

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #132 on: March 11, 2009, 02:35:17 pm »

DF already does that with Geology.  Minerals are only found in certain places and you need to know where to look.  However, most sites have enough stuff at a given embark that you actually have to look for places without enough minerals.

If you have to TRY to screw up your economy because the controls you have trend towards center stability, it should work.
Spoiler (click to show/hide)

praguepride

  • Bay Watcher
  • DF is serious business!
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #133 on: March 11, 2009, 02:37:15 pm »

I can agree with that. IF the game can stabalize itself on its own to certain degree, then it'd be fine for the rest of us. Just like how dwarves will (usually) flee from danger without being micro'd.

Spoiler (click to show/hide)
Logged
Man, dwarves are such a**holes!

Even automatic genocide would be a better approach

Granite26

  • Bay Watcher
    • View Profile
Re: Modelling a Global Supply/Demand Economy (Please!)
« Reply #134 on: March 11, 2009, 02:44:09 pm »

Spoiler (click to show/hide)
Pages: 1 ... 7 8 [9] 10