Assessing the value of workers time

[couerdelion]

I should add here that the food/gold conversion does indeed require several elements in place including

1) The ability to appoint specialists (including engineers)
2) Slavery
3) A granary.

But that aside, I also looked at what a +1 food difference made to a city that grew continually and that each new population was assigned to a cottage. Naturally, the extra food city grew more quickly and started working the cottages sooner, so it acquired more commerce than the city that grew more slowly. The slower city was assumed to have +2 food surplus and all growth was assumed to start from size 2. The projections were assumed to run for 200 turns.

After comparing the additional commerce (which I converted to gold at 1:1 rate), I discounted these all at a rate of 3% per turn and compared the result to the discounted value of 1gpt = 33.33g

There were four different results depending on two factors: were the cottages on a river and whether or not you have the financial trait. The results were as follows

A: Grassland – equivalent to +2.2gpt
B: Grassland river – equivalent to +3.4gpt
C: Grassland with fin trait – equivalent to +3.0gpt
D: Grassland river with fin trait – equivalent to +4.5gpt

The first thing to note is this sets a minimum value for the estra unit of food for a civilisation and there may be other ways of getting more gold per unit of food eg whipping a library. It is also very dependent on the circumstances; a large city will see slightly less benefit from the extra food since growth takes longer. Also the higher the expected return that we want to achieve, the lower will be the perceived benefit of gains made far into the future.

But I have to admit to being somewhat surprised at the size of the numbers involved. As long as there is no immediate need for commerce – ie to speed research on a particular tech, the long term gains from just one extra food are very significant indeed.

[fed1943]

I knew the difference was very weighty, but frankly never arrived so

big numbers.

Does the simmulation assumes the cottages are already there and

each pop go there imediately without other concerns?

Best regards,

[couerdelion]

Originally posted by fed1943
I knew the difference was very weighty, but frankly never arrived so

big numbers.

Does the simmulation assumes the cottages are already there and

each pop go there imediately without other concerns?

Best regards,

Yes, I did assume that cottages were ready as they were needed so it would imply that for faster growth the cottages would be needed that much sooner.

It also assume that there is almost unlimited space for cottages and that, in the case of river-side cottages, these spaces are widely available.

Fortunately, the discounting effect means that the 10th extra cottage tile has a relatively small effect on the overall result which is achieve not just be the 2-3 extra population but also by the fact that the cottages are worked that much sooner and therefore grow sooner.

I did achieve this through simulation rather than from a case study on the game itself. The timescale involved is just too long for this to be managed easily.

[uberfish]

The thing is the value of +1 food diminishes when you already have a good food surplus in a city (as you then run into inefficiency converting it into something else due to over-growth) - IE if you already have irrigated corn, a farm is probably a poor build - and also you can only work X tiles before a city becomes unhappy.

[couerdelion]

Originally posted by uberfish
The thing is the value of +1 food diminishes when you already have a good food surplus in a city (as you then run into inefficiency converting it into something else due to over-growth) - IE if you already have irrigated corn, a farm is probably a poor build - and also you can only work X tiles before a city becomes unhappy.

Agreed, the value of more food is lower when food surplus is already high.And where you are running up against happiness limits (or have exceeded them) you could even argue that food has a negative or zero value.

I think above all, I would advise people to consider other implications rather than following these sorts of rules blindly. It is very convenient for modelling purposes that we can make such grand assumptions such as the ability to appoint specialists, the unlimited happiness and health, ample free space for setting up cottages and the worker capacity to build them.

As is often the case in this game, you’ll need several things in place for a given tactic or strategy to work. Here I’m really trying to apply an overall framework to assess the value of workers (in particular the first one or two). But the values are very sensitive to assumptions

At 3% it works out to be about 400g
At 2% it works out to be over 1000g

Fortunately both of these give a strong justification for getting a worker out and also having the techs to use that worker. Perhaps there are stronger plays but these have to give you more than the worker option. On the assumptions I gave the worker can produce 6% per turn return to a civilisation so that’s a pretty tall order for anything to match.