I Hate Food

[Chenier]

Eh, one thing I used to look at once food was introduced to BT was gold efficiency.   A city that produces 1,000 gold with 30,000 people could be more favorable than a city of 36,000 (25% more food required) that produces 1,100(10% higher income).  I don't remember the average consumption per population anymore, but you still need to check gold to population ratios some times to see if cities are on par with each other.

And trolling is only an issue when you've just finished off your bottle of vodka, and can JUST type, but not well.

Sure, if two cities produce the same gold output, then one should favor the one with the least population... However, if the "gold efficiency" ratio was lower, but raw gold output higher nonetheless, it's usually still better to go for the "less efficient" one. On continents with ample food surpluses, that is.

[Psyche]

Rereading where this is going while sober I was quickly reminded of the Monty Python skit from Holy Grail about coconut/swallow migration, coming to all sorts of hypotheses.  I think this is where I bow out.

[Chenier]

Regarding seasons, Tom told us that on Dwilight, food production was 113% of what it consumed.

So I did some calculations... (Assuming no looting, no troops, perfect production, no zuma, no drought, no monsters, etc.)

With seasons, at the end of the year, you'd have 1.55 days' worth of surplus production over what you'd have had the year before, accounting for the rotting of 8% of your total production.

Without seasons, and thus a steady production, Dwilight would instead end up with 6,65 days' worth of excess production at the end of the year, thanks to only having 4% of your year's production rot.

Seasons do have an impact.

Especially when you count in weather. With seasons, half of your food is produced during Fall. If you have a drought during fall, you lose a quarter of your yearly production, whereas on continents without seasons, having a drought for a season only results in losing an eight of the yearly production.

[^ban^]

Food production is at 108% of what is consumed on Dwilight, across a full year.

Uhh...

[Eldargard]

For what it is worth, I also believe that some regions (cities for sure, possibly strongholds, mountains, badlands and some forests) should be high gold/low food while some regions (rurals for sure, possibly some forests) should be high food/low food. Some regions (Townslands for sure, possibly others) should be mid gold/mid food. Some can have a surplus of both or a lack of both, but that should be the exception to the rule in my opinion. I also think that this should all be based on average taxes run (actual averages, not expected averages). I also think that a region should not be considered making an excess unless they are making more that what it would cost to support a reasonable amount of buildings/militia.

I would expect something like:

City = High Excess in Gold, High Deficit in Food
Stronghold = Sufficient in Gold, Deficit in Food
Townsland = Excess in Gold, Excess in Food.
Badland = Sufficient in Gold, Deficit in Food, Minor Excess in Gold, High Deficit in Food
Woodland = Sufficient in Gold, Minor Excess in Food or Minor Excess in Gold, Sufficient in Food
Rural = Deficit in Gold, High Excess in Food or Sufficient in Gold, Excess in Food

High Excess = Makes a whole lot more than needed
Excess = Makes more than is needed
Minor Excess = Makes a tiny bit more than needed
Sufficient = Makes what is needed
Deficit = Makes less than is needed
High Deficit = Makes a lot less than what is needed.

[Solari]

Food production is at 108% of what is consumed on Dwilight, across a full year.

Yep. So the question to be answered is whether an 8% buffer is sufficient to account for weather, rot, other loss, and delays due to ambivalent rural lieges. Thoughts?

[vonGenf]

Yep. So the question to be answered is whether an 8% buffer is sufficient to account for weather, rot, other loss, and delays due to ambivalent rural lieges. Thoughts?

Is it really? Rather, I would ask whether it is sufficient for a lively market. The amount of food could very well be sufficient in a world of perfect distribution, but establishing a world of perfect distribution through in-game means does not sound extremely fun.

My take is that the mechanics of the markets should be changed, and not the amount of food. It is entirely possible that the markets would be more lively with more food given the current rules, but I think that is the wrong way to look at it. With easier distribution in those places that do not want to specifically withhold food, food tightening would not be a problem as it could create IC conflict. As long as the misdistribution stems from a broken system and not lack of food, it won't.

[Chenier]

Yep. So the question to be answered is whether an 8% buffer is sufficient to account for weather, rot, other loss, and delays due to ambivalent rural lieges. Thoughts?

Clearly not.

While a lot of the losses are hard to quantify and predict, rot is not, and merely factoring minimum rot in turns a 107,2% surplus into an actual yearly deficit with 0 starting bushels. At 108% production, you really wouldn't need much of a starting value to get a yearly deficit either. Depending on the various food production levels I've been given, yearly rot value has varied among 8%, 16%, and 20% of production. Ignoring rot altogether when decided what's an acceptable production value is simply a very big mistake. Some of the loss values are difficult to quantify, but not minimum rot.

Note: I've made a (rather) clean set of tables for the prediction of food values on a continent with seasons. How does one send attachments on these forums? Or else, anyone know where I could put a .xls for easy sharing with everyone? You could then all put in your own numbers or inspect my formulas to see if everything adds up.

Additional data: with 125% food supply in summer, it takes about 5 years (assuming starting warehouse value of 0) for rot to equal the food surplus. Otherwise, it just takes a warehouse starting value of 72,000 on the first day of summer.

[Gustav Kuriga]

Fun thing about equations, they assume perfect selling and buying of food to provide for all regions. Unfortunately we're human...

[Chenier]

Fun thing about equations, they assume perfect selling and buying of food to provide for all regions. Unfortunately we're human...

Indeed.

But wouldn't you agree there's a problem when, even under an impossible perfect scenario, there's an average of less than 250 bushels of food left in every region's warehouses at the end of spring?

And that's without counting the continental population that grows at a rate of about 3,370.52 peasants per day, which results in an increase of 24066 additional bushels consumed per year not considered by my previous numbers, which had also underestimated population by 100,000.

[Chenier]

I've just noticed that when trying to compensate for population growth, I inserted an error in my formulas that overestimated food production... some of my last batch of numbers may be overestimates.

Is there any historical data kept for the number of bushels kept in the granaries? I'd really like to see its evolution since the last manual changes to compare it to my model.

Also, I'm not sure where the numbers I was provided come from... the stats page suggest considerably lower food outputs than what you guys have been saying.

[Indirik]

As far as I know, the only historical data is what's recorded on the stats pages.

[Chenier]

As far as I know, the only historical data is what's recorded on the stats pages.

Any way to get that in csv or xls format?

Also, am I right to assume that the food production statistics do not consider seasons in those display values? Only weather? And that the values displayed are on a 7-day value, instead of on a daily value? I don't suppose all of the food production and consumption stats could be displayed on a daily basis, could they? It'd allow easier analysis...

[Indirik]

I think Tom is the only one that could answer those.

[Chenier]

Oh gosh...

I'd like to know if anyone knows a good way to share an .xls file.

After reviewing my stuff, I've found out that a in a couple of places, food production was being considerably overestimated, because I accidentally had it scale with population. Now sure, it'll increase some... but I'd be surprised if it grew at the same scale... Cities were the most affected by starvation, not the rurals, and as such they have the most to build up. And while, sure, Barca gained a few rurals no one had owned before, Falkirk's also losing a bunch of them, which ought to compensate.

And I'd really enjoy some confirmation on what's an average daily food production value for summer. Is 5500 about right?