Realistic Farming DLC

[David]

Yes, as well as soil acidity. I was thinking about it, I think ranges are a way to go, an ideal range, an tolerated range

In your view, how should all these be presented on a UI in a way that the user won't get overwhelmed?

[Brain]

I personally would display them in a new Section of the Infoscreen, Farming Guide. The player has three select boxes in the top, similar to the corporate details screen, on the left they pick a crop, in the middle they pick a city/region, on the right they pick a month, or "Growth Season" (which would recalculate median and variance over the whole growing seacon for the selected crop)
Below that will then appear a section with six graphs, 3x2 grid, calculated on game start or when a new region/city is opened. The graphs will look like this one:

One curve is the tolerance for a given property by the crop, the other one shows what the city/region has to offer in regards to this property.

• Rainfall
• Humidity
• Sunny Days/week
• Soil acidity (constant but the graph would be based on the ratio )
• Temperature

Below the charts there will be a section explaining the visual representation in a table, showing:

• The recommended growing season (when players set up a Crop Growing Unit they should pick not only the crop but also when to start the crop, defaulting to the ideal month in the city/region. This is necessary, to simulate growing crops in northern vs southern hemisphere, where seasons are 6 months shifted, as well as to allow multiple harvests in a city if the conditions would be right)
• the maximum yield to be expected in % of maximum
• the maximum quality to be expected in % of maximum
• fertilizer needs
• regional pests to be aware of

This should give the farmer a good oversight of the possibilities in the city/region and also explains in a semi-scientific way why a region would or would not match well for a crop.

When the Farmer(aka Player) has chosen a city/region for his farm, he would be able to use the micromap (I mean minimap) to switch to a farming mode where he picks a crop and the minimap then shows in colors where the city provides the best conditions for a farm. Acidity, should be displayed here, I do not think we will have to simulate local temperature hotspots or rain stronger on hill sides. Not before CapLab gets a better geology model

[smith121362]

This is something I have been wanting implemented since the beginning of this series. I live in Texas and we grow a lot of crops here. Some areas are better for cotton. Other areas are better for corn. And down close to the border with Mexico (The Valley) is best for fruits and vegetables. The availability of water (rain/irrigation) is a big factor as to what we can grow in different areas of our state. Weather also plays a big role in growing crops here. As an example: the farmers around Temple grow mostly corn. Last year all the farmers were able to plant their crops at the right time of planting season and the crops got the right amount of rain at the perfect times of their growing season. The farmers had one of the best crops they have had in 50 years. This year it started raining in the middle of planting season. Half of the fields were planted very late. Then there wasn't much rain during the growing season. It is almost time to start harvesting the corn but now it is raining. The fields that got planted on time will do good (if they can get harvested in time) But the ones that got planted late will barely pay for the cost of growing them. Our cotton crop is in danger also because of a lack of rain and our wheat crop was a total loss.

[David]

on the left they pick a crop, in the middle they pick a city/region, on the right they pick a month

If we develop it as a DLC, as opposite to a new game, the regions are probably going to look like the mock-up screenshots I posted earlier at
viewtopic.php?f=49&t=5161#p23864

So there will be two questions:
1) The regions do not have names, so the UI idea of in the middle they pick a city/region probably won't work here.

2) What algorithms should the game use for generating the following data on a random world map (as in the CES DLC)?
Rainfall
Humidity
Sunny Days/week
Soil acidity (constant but the graph would be based on the ratio )
Temperature

---------------

On the other hand, hypothetically if it were going to be developed as a new game, it might use a real world map.
If it were going to be developed using Unity, then there are asset library like the following that the game might use:
https://assetstore.unity.com/packages/t ... -kit-55121

But then, how exactly the player should be viewing asides from the world map? A zoom in map or a region map for each region? And how exactly should we break down the regions from the gaming perspective? If you have ideas, please provide a mock-up screen.

[Brain]

If we have regions that are unnamed, the Farming guide would instead of cities and regions show the world map, updated by selecting crops. We would calculate the bell curve of every world map tile based on its features.

I would like to learn about the current map generator capabilities. for example, it seems to be able to simulate plains, hills, mountains, lakes, oceans? can we adjust amounts of oceans, because at the moment it looks rather inverse to what it should be.

Accoding to the next image, it looks like soil acidity correlates with certain biomes, like dry hot/desert zones are more likely to be alcaline, forested regions and wetlants are more acidic.

Precipitation (rainfall) is stronger around the aequator and lower in the deserts biomes, higher near mountains when closer to oceans.

Humidity (RH, relative humidity) is mostly a function of rainfall and temperature, but also stronger near water bodies, and because of that fluctuates per year, see some nice represenentation here.
whereas temperature is mostly a function of distance from aequator, month of year and distance to oceans.

The chance for sunny days is negatively correlating with cloudiness, of course, which is kind of correlating with RH and height above ground, but also a lot with solar activity (sunspots)


If we use earth as real map, the data can be derived from the usual sources, MetOffice etc. Not sure what the Unity package provides, I am not experienced with Unity enough yet. But in case of random continental climate simulation. I haven't found a good climate simulator online, but this app might help: http://www.weathergraphics.com/climasim/. Maybe they have an CLI to feed in a continent model and they return you numbers for the clima per month or so.

[David]

If we have regions that are unnamed, the Farming guide would instead of cities and regions show the world map, updated by selecting crops. We would calculate the bell curve of every world map tile based on its features.

Do you think it would be a bit tedious requiring the user to click on each world map tile to find the ideal region? As you know, there are so many world map tiles.

I would like to learn about the current map generator capabilities. for example, it seems to be able to simulate plains, hills, mountains, lakes, oceans? can we adjust amounts of oceans, because at the moment it looks rather inverse to what it should be.

Yes, I believe that it is possible to adjust amounts of oceans, as the dev team told me that it is a fairly standard fractal map generator. But in the CES DLC, the player can only found a city on a world map tile that mostly comprises of plains. So when there is too much ocean, there will be fewer tiles to select from for founding new cities.

Accoding to the next image, it looks like soil acidity correlates with certain biomes, like dry hot/desert zones are more likely to be alcaline, forested regions and wetlants are more acidic.

Do the amounts of nitrogen (N), phosphorus (P), and potassium (K) in the soil needed to be taken into account in the simulation?

Precipitation (rainfall) is stronger around the aequator and lower in the deserts biomes, higher near mountains when closer to oceans.

How about rainfall distribution throughout the year? For instance, in a Mediterranean climate, the summer is dry. In a tropical climate, there is lots of rainfall in the summer. Does this have to be simulated in the game? If so, how should the data be presented to the player?

Humidity (RH, relative humidity) is mostly a function of rainfall and temperature, but also stronger near water bodies, and because of that fluctuates per year, see some nice represenentation here.
whereas temperature is mostly a function of distance from aequator, month of year and distance to oceans.

I clicked on the link you provided. It is a very nice web-based app.

The world map in CES which is derived from a fractal map, is nothing like a real world map. If we eventually decide to develop this, would you be able to provide more exact implementation details (something that a programmer can understand plainly) as to how a world map tile in the CES DLC can be initialized for all these different weather attributes?

If we use earth as real map, the data can be derived from the usual sources, MetOffice etc. Not sure what the Unity package provides, I am not experienced with Unity enough yet. But in case of random continental climate simulation. I haven't found a good climate simulator online, but this app might help: http://www.weathergraphics.com/climasim/. Maybe they have an CLI to feed in a continent model and they return you numbers for the clima per month or so.

My view is that a real map will have greater appeal to gamers, but using the current CES world map will be easier to implement.

For a real map, the challenge is not only with obtain the data, but how exactly the UI and the different level of zoom levels should be available to the player. Honestly, I do not have any ideas on all these yet. I am just saying that in general a real world map will be more eye-catching and people can associate it more easily than a random generated map, when the gameplay centers around making the right decisions on planting crops around the world to maximize profits.

[Brain]

If we have regions that are unnamed, the Farming guide would instead of cities and regions show the world map, updated by selecting crops. We would calculate the bell curve of every world map tile based on its features.

Do you think it would be a bit tedious requiring the user to click on each world map tile to find the ideal region? As you know, there are so many world map tiles.

The world map would be colored according to the suitability of the selected crop. The user only would have to pick one of the more positively colored tiles.

I would like to learn about the current map generator capabilities. for example, it seems to be able to simulate plains, hills, mountains, lakes, oceans? can we adjust amounts of oceans, because at the moment it looks rather inverse to what it should be.

Yes, I believe that it is possible to adjust amounts of oceans, as the dev team told me that it is a fairly standard fractal map generator. But in the CES DLC, the player can only found a city on a world map tile that mostly comprises of plains. So when there is too much ocean, there will be fewer tiles to select from for founding new cities.

In a way the game would be a bit difficult if there are more then 20 cities on the map anyways...

Accoding to the next image, it looks like soil acidity correlates with certain biomes, like dry hot/desert zones are more likely to be alcaline, forested regions and wetlants are more acidic.

Do the amounts of nitrogen (N), phosphorus (P), and potassium (K) in the soil needed to be taken into account in the simulation?

Not on the map selection, because this is a fertilization issue that would be in the hand of the farmer. We might abstract it to a property "Soil Fertility" that the Farm would manage per crop unit by purchasing "Fertilizer". Producing fertilizers with different concentrations to mitigate different nutrient deficiencies and applying them manually might be a bit too micromanaging. I am imagining a 3x3 unit for the farm where you link simply a fertilizer supplier and link it with Crop Growing Units, which automatically takes as much fertilizer from it to reach its optimum soil fertility. Without such a unit linked or without fertilizer supplied the spoil fertility for that unit drops per harvest by some percent, impacting yield and quality. Setting the unit to idle will slowly recover it, or growing some cover crop. Some Farms might need additional watering, too, to mitigate lack of rain.Groundwater levels should be simulated per region, if someone builds water wells.

Precipitation (rainfall) is stronger around the aequator and lower in the deserts biomes, higher near mountains when closer to oceans.

How about rainfall distribution throughout the year? For instance, in a Mediterranean climate, the summer is dry. In a tropical climate, there is lots of rainfall in the summer. Does this have to be simulated in the game? If so, how should the data be presented to the player?

Indeed, Rainfall follows seasons, but there are averages for each month of the year with deviation known for each tile, so picking a random value from that gaussian random function would be good enough for simulation.

The world map in CES which is derived from a fractal map, is nothing like a real world map. If we eventually decide to develop this, would you be able to provide more exact implementation details (something that a programmer can understand plainly) as to how a world map tile in the CES DLC can be initialized for all these different weather attributes?

I might come up with a function for you, that would provide you with the necessary data, provided that you have object oriented calls, because that would be a pain to convert to and from byte arrays
The input data would be kind of this structure:

Code: Select all

function generateClimaModel(tile[][]) //latitude, longitude
tile:= {
	elevation:=measureType, // unit: meters. (please USA, go metric, pretty please)
	biome:=biomeEnum*, //output
	acidity:=double,	
	temperature:=measureType[]* //month array, output
	rainfall:=measureType[]* //month array, output
	humidity:=measureType[]* //month array, output
}
measureType:=
{
mean: double,
variance: double,
unit: String
}
biomeEnum:=[FRESHWATER,WETLANDS, MARINE, TUNDRA, RAINFOREST, SAVANNA,TAIGA, TEMPERATE_FOREST, TEMPERATE_GRASSLAND, ALPINE, CHAPARRAL, DESERT] // 

About Biomes

If we use earth as real map, the data can be derived from the usual sources, MetOffice etc. Not sure what the Unity package provides, I am not experienced with Unity enough yet. But in case of random continental climate simulation. I haven't found a good climate simulator online, but this app might help: http://www.weathergraphics.com/climasim/. Maybe they have an CLI to feed in a continent model and they return you numbers for the clima per month or so.

My view is that a real map will have greater appeal to gamers, but using the current CES world map will be easier to implement.

For a real map, the challenge is not only with obtain the data, but how exactly the UI and the different level of zoom levels should be available to the player. Honestly, I do not have any ideas on all these yet. I am just saying that in general a real world map will be more eye-catching and people can associate it more easily than a random generated map, when the gameplay centers around making the right decisions on planting crops around the world to maximize profits.

different zoom levels for different features. at the current state having one more zoom levels for farming needs would probably fine, as crop fields are so much larger in RW. But even that is more a nice to have.
Personally I would prefer RW vs generated, but mostly because the generated fractal map is so unrealistic. If we can come up with a proper model that simulates continents with natural features, that would be great.
I put up a poll on this.

[David]

Indeed, Rainfall follows seasons, but there are averages for each month of the year with deviation known for each tile, so picking a random value from that gaussian random function would be good enough for simulation.

So how should the sowing month of a crop be determined? Should it mainly depend on the temperature?

[Brain]

Yes, it should be based mainly on temperature. Plant seeds have a germination temperature, and when that is matched on the average month growing season starts.

By the way, while looking at the Food Mod efforts I noticed we have to be aware of different kind of Harvestables. Simple plants like Green Salad or Carrots can be planted at any time and harvested a few weeks/months later, as long as being sowed within temperature limits in their growing phase. Crop Growing Units should have a setting for when to start the first growing cycle, defaulting to the first month with expected acceptable conditions, and then repeat harvest and seeding for as often as making sense in the given climate - might be all around the year.
Some Harvestables require more than a year of to grow - for example apples come from trees that require a few years from seed to harvest. We can of course say that we plant almost grown Fruit trees, but that sounds a bit silly. Plantation fruits grow differently and we should maybe have them as separate Farm facilities. Plantation would maintain a number of trees or bushes and it might take a few years to harvest first time. Plantation does not maintain a sowing months, but will have a Harvesting season.

We also might add Greenhouses as another option, which would be independent of soil fertility temperature and rain, but more expensive to maintain.

[David]

By the way, while looking at the Food Mod efforts I noticed we have to be aware of different kind of Harvestables. Simple plants like Green Salad or Carrots can be planted at any time and harvested a few weeks/months later, as long as being sowed within temperature limits in their growing phase. Crop Growing Units should have a setting for when to start the first growing cycle, defaulting to the first month with expected acceptable conditions, and then repeat harvest and seeding for as often as making sense in the given climate - might be all around the year.
Some Harvestables require more than a year of to grow - for example apples come from trees that require a few years from seed to harvest. We can of course say that we plant almost grown Fruit trees, but that sounds a bit silly. Plantation fruits grow differently and we should maybe have them as separate Farm facilities. Plantation would maintain a number of trees or bushes and it might take a few years to harvest first time. Plantation does not maintain a sowing months, but will have a Harvesting season.

All you stated above makes a lot of sense. How would you suggest modifying the crop's attribute list to incorporate all these?

The next step would be for me to organize your suggestions into a document and propose it to the dev team. I would try to convince them that with the details you provided, it will be practical for them to implement an experimental version based on your details. (Doing so after the major features of Banking DLC is done) Subsequently, by making the experimental version available to you and other selected testers, we can hope to evaluate gameplay quality of these new features and whether it is solid enough to warrant a full DLC development, which will take a much longer time, due to the need to implement the AI, to polish all the new features and gameplay balancing and etc.