States, Firms, and Households cover
States, Firms, and Households screenshot
PC Steam
Genre: Simulator, Strategy, Indie

States, Firms, and Households

New Experiments
Check out these pages to learn more about the amazing world of economics that SFH helps to uncover:

https://www.kickstarter.com/projects/690843973/states-firms-and-households-the-search-for-equilib

https://www.kickstarter.com/projects/690843973/states-firms-and-households-the-search-for-equilib-0

Also check out the story of how the game was made:

http://www.fullindie.com/portfolio-items/denis-ivanov-states-firms-and-households-full-indie-may-2016/

With version 31, we will be releasing a cosmetic fix and soon the ability to edit simulation variables and to use your own building sprites and characteristics, if there is a continued commercial interest and success.

We need as many people to try out the experiments as possible to have a wide data pool. You can help by conducting experiments and discussing your results and plans in the discussion forum.

You can access the source code for the game here: http://github.com/dmdware/sfh

The principles of the game mechanics are very simple and are explained in the PDF manual available for download in the bottom right of the store page.

http://store.steampowered.com/manual/458820

You can do many experiments now (without editing sim variables yet), for example making supply-demand graphs with time-quantity-cost variables. You can do this for all eternity, filling in all the dots to make a complete graph and the more people doing this and sharing data the better. There is a lot to do. I already had a surprise looking at the charts using a mathematical model solution or half-simulating it, and there's probably more surprises in the multiverse brute force or full simulation. And you can do all sorts of experiments eg placing two supplier or consumer buildings at 90 degrees or on a line and various distances etc. So this is for anybody that wants to do experiments. SFH is a valuable tool for insights and ideas in economics.

Eg you can set up a map with ambient buildings which are also important and have to be controlled and have a warm up period to get the workers moving then measure times and variables and pause and write things down. And measure eg effects of variables on others, based on the input to output ratios of the buildings, relations to the bigger picture, etc.

I look forward to seeing all your interesting insights if you choose to do this, and to perhaps inspire you with more experiment ideas. We are continuing work on SFH 2 years later because we believe we may perhaps achieve a greater player base, in the tens of thousands or more (we only have about 2,000 copies sold so far).

Also here are some memories from way back when SFH was on greenlight:

https://steamcommunity.com/sharedfiles/filedetails/?id=362660305

Version 28
- bigger map size
- labourer animation
- ui improvement: have graphs and manager shown at the same time
- ui improvement: notifications now 12 lines instead of 4
- multiplayer matchmaker back up

Manual
A manual is now available here http://store.steampowered.com/manual/458820 and a tutorial will be available shortly.

Version 16
- disconnects from matchmaker on quit host
- building cycle minute
- new transport pricing strategy
- resource decay half-life
- fixed laggy buyout

Version 15
Quick update:
- pauses live join until map finishes downloading
- improved support for multiple devices behind router
- different stocking behaviour
- labour-hour to labour-minute

Findings
This model was originally developed to test out ideas and learn things. This was meant to see what would happen if agents made their decisions on profit maximization and price-distance tradeoffs. From this, it was found that a supply and demand curve could be found. It is probably a borrowed idea but few have probably tried to find out the resulting curves. The model was meant to reconstruct a view on economics from the bottom up. Elastic, inelastic, fixed, variable - these are all relativistic terms. Elastic graphs can look inelastic if the measuring unit of supply quantity is changed. And fixed costs are only fixed within in a certain range, as a single shipment may suffice for a given tonnage and considered fixed, but over the long-term may be variable. A supply curve may curve down, as fixed costs are outweighed by earnings or even if the output per input is greater than a unit. If 2 units are sold at $2 for 1 $1 work there's a $1 profit, but 4 units sold at $2 for 2 $1 work will be $6 profit - a decrease in marginal cost of 1/3. This was a problem a few days ago, but using a more reasonable production range and price-distance utility equivalance of $0.03-to-1 cm reasonable curves are obtained, as the marginal cost increased. Also, a firm that operates on profitability and marginal cost - the intersection of supply and demand - may gain greater profit by getting over a "cost hump" and the difference between supply as marginal cost and average cost is the difference between a profit-maximizer and a firm just covering expenses. A case where average is greater than marginal is a point after a decrease in average cost or slope, and a demand intersection with marginal cost in that case leads to losses. It is the important if hard to understand difference between marginal and average cost. Marginal cost is the differential of average cost. Marginal cost tells one how much one should sell at given what has already been produced or sold. The assumption is that everything leading up to the current quantity has been sold at the price it cost to produce. It is maximizing over short jumps. Production is continuous with time. At a single unit of production, the differential is equal to the average. The average and marginal are really the same - cost - except at different scales of quantity or time. It really depends on the point at which a price is set and sale made.