My experiments with farming...

[RiverCocytus 0]

Gack! double post. Read below, please.

[RiverCocytus 0]

No, I do understand. (you're actually being relevant!) This is what Bones is saying...

pumpkin acres = 2 jobs for 1 tile or 2 jobs for 2 tiles (1 job per tile)

farmland = 1 job for 1 tile

so in the case that you have all pumpkin acres, you will be getting at LEAST 1 job per tile, if not 17 jobs per 16 tiles for my small 4x4 arrangment (or even more if you do fff.)

other farm types offer more jobs, but with a greater number of tiles.

For instance, jez knight acres (the building) could take up 6x5 tiles. it still offers 10 jobs, so that's... 1/3 of a job per tile. Not terrific.

So if pedriana gives 13 jobs and takes up, um, 4x5 minimum space, then you are seeing 13 jobs over 20 tiles, which is less than 2 jobs over 1 or 2 tiles. In the case of 16 tile farms, all pumpkin acres take up 1 tile.

So if we ignore the freight time related desirability (so we have a base factor) we would get

1 pedriana = 20 tiles = 13 jobs

1 pumkins = 1 tile = 2 jobs

if we do a lil' gaussian work...

20 pumpkins = 1 pedriana (in tiles)

6.5 pumpkins = 1 pedriana (in jobs)

so 20 pumpkins would be 40 jobs, whereas the 1 pedriana (in the same tilespace) is only 13 jobs.

OK-- so lets make this a real-life situation.

Say I need to zone a 5x4 minimum to get a pedriana. BUT-- since a pedriana's plants offers fewer than 1 job a tile (as farm field does) it is of advantage for pedriana to be on as a LARGE as possible lot.

Pumpkin acres offers MORE jobs per tile on average than farmland, so it is of advantage for pumpkin acres to be on as SMALL a farm lot as possible.

In other words, if I zone one 8x8 and I get a pumpkin acres, if it is a 1 tile pumpkin acres I'm getting 8x8 -1 = 63 jobs from farmland and 2 jobs from pumpkin = 65 jobs, and with 4 4x4's I'm getting 8x8 - 4 60 jobs from farmland and 2x4 (assuming 1 tilers again) 8 jobs from pumpkin = 68 jobs (marginally more.)

Now if that 8x8 has a pedriana on it instead, we get

(pedriana area is assumed to be 5x4)

(8x8 = 64) - (5x4 = 20) = 44 tiles of farmland = 44 jobs

1 pedriana = 13 jobs

max jobs from pedriana on 8x8 lot = 57

vs.

max jobs from pumpkin (1x1 size) on 8x8 lot = 65 (again, marginally more.)

This is, I'm assuming you're correct Sam in that the buildings do actually effect the number of jobs available (or rather, the capacity satisfied, which is altered by desirability levels.)

So if with smaller farms more farmland is closer to freight transit and thus has lower freight times, over time small farms would create a significant different in number of jobs.

I think maybe there should be a re-configuration of farm building exemplars, (though it is just me, most likely) as the higher stage farms actually offer LESS jobs than the lower stage farms. It totally doesn't make sense. This is only because of farm land, really, since it always offers 1 job per tile. (though I think the BSC farms alter this)

No matter how you do the math, per acre the pumpkin acres is a better bet. phew.. sad, really.

[infoscott 0]

I play a lot with rural cities, and use an entirely different methodology. Given that I grew up a poor hillbilly child, I don't bother much with the theoretical stuff. Just start counting things and do the 'rithmetic.

If you want to measure jobs on a farm at some particular time, count how many human legs show up to work and divide by two. The Route Query tool will tell you that. Sometimes desirability has more to do with how much moonshine the sharecropper drank the night before. I find that field hands hire on at a rate of about 1 every 15 tiles to 1 every 30 tiles, depending on the type of crop. We all know there are way more people working in the "Big House" than what is needed; too many chiefs.

My guess is freight out correlates to taxes, but I can't prove it. Freight truck count seems to be proportional to the farm's acreage. You'll get maybe 2-4 times more R$ tax than I-Ag tax for a given population. So basically you provide farms at a low carrying cost in order to profit from taxing local families. The aggy tax is enough to cover all the electricity used and a good chunk of the monthly transportation cost. Rail is .03 per month, Street is .05, and Road is .10, so I send out most of my aggy freight by direct connect to the rails or by a nearby (less than 100 tiles away) freight station.

You'll need three times the residential occupancy of nearby R$ to provide enough field hands. I try to build enough residences adjacent to the farms or in nearby hamlets so that that field hands have short commutes. Likewise, when first setting up a farming city I lay down the road and rail network so as to minimize the monthly maintenance cost and freight commute time. If the rural transportation network is griddy enough you can get by with rural streets and use few if any roads. The occaisonal crossroads Dairy Queen will also keep field hands from going into villages to spend their CS$ money.

[VDub4Life 0]

I also was puzzled as to why the job stats listed for a farm lot (i.e. querying a barn) differed from the Jobs I-AG number in graph view.

After reading this thread and a few others on the topic, I decided to test infoscott's (post above) theory.

Originally posted by: infoscott If you want to measure jobs on a farm at some particular time, count how many human legs show up to work and divide by two. The Route Query tool will tell you that. Sometimes desirability has more to do with how much moonshine the sharecropper drank the night before. I find that field hands hire on at a rate of about 1 every 15 tiles to 1 every 30 tiles, depending on the type of crop.quote>

I set out building a test farming community with rectangular lots of varying large size, and let it run for a few years without touching anything.  I paused the game, selected the route tool, hovered over each farm, added the cars & freight truck numbers together and divided by 2, finally I selected the query tool and added the numbers of all current jobs to the previous number.

Test #1
Data: (Cars+Freight)/2 + Current Jobs on farm
(300+79)/2 + 37 = 227
vs. graph showing 239 I-AG jobs
Basically the formula was still off by about 12 jobs from the graph

Calculating max capacity jobs instead of current/actual gives:
(300+79)/2 + 46 = 236
Only off by 3 jobs from what is displayed in the graph

Test #2
I than decided to bulldoze all the agriculture let it regrow, and see what the results were.
The city was simulated for a game year and the farms grew in a totally different configuration than before (same lot sizes, but different farms on each).
Results:
(270C+78F)/2 + 52Max Cap = 226
The graph showed 234 I-AG jobs were in the city (off by 8 this time)

Test #3
I did one last test.  This time I rezoned a lot of small farms (8x8 and 4x4), plus I kept a few of the larger ones.
Results:
(244C+83F)/2 + 85Max Cap = 249
The graph showed 252 I-AG jobs in the city (once again off by 3)

Conclusions (what this has taught me)
1.The formula could be off a little simply due the game rounding certain numbers up or down versus, what is shown in data views. 3 or 8 is certainly a lot less variance then querying every farm lot and adding the numbers together, and comparing it to the graph for I-AG jobs.
2. The jobs seem to be calculated citywide based on total zoned agricultural land, rather than individual farms and their associated desirability.
Case in point, during one of my tests I had 3 16x16 farms in a row surrounded on all 4 sides by street
Farm #1 was an Alvin's Acre
The farm lot occupied a 3x2 space and queried 5/6 current jobs (or 83% occupancy)
route query showed 55 cars going to and 13 freight leaving the farm

Farm #2 was a Barthelet Stables
Lot occupied a 3x2 space and queried 7/8 jobs (88% occupancy)
route query showed 44 cars and 13 freight

Farm #3 again an Alvin's Acre
However, it occupied a 10x8 space and queried full occupancy 6/6
rout query showed only 31 cars and 10 freight at this farm
If the game calculated I-AG jobs based on each zoned lot, I would say Farm #2 should have the most traffic and jobs generated, followed by Farm #1, and Farm #3 (farm #3 being least, because there was only 6x8 tiles of crops left after the farm lot was built).

3. This was not a test to show wheter zoning large farms or small farms was more efficient.  Rather it was only to show a way of calculating the correlation between actual data in your farming community and what the graph shows in agricultural jobs.

[Bones1 96]

It's not that complicated. The jobs-query simply shows the number of workers for that particular farm building. But that's not all the jobs there are. There are also jobs for each farm crop tile, which you can't query. So the graph is the total of the building jobs + the crop jobs. (I think it's 1 job per crop tile, IIRC.)

[VDub4Life 0]

Originally posted by: Bones1 It's not that complicated. The jobs-query simply shows the number of workers for that particular farm building. But that's not all the jobs there are. There are also jobs for each farm crop tile, which you can't query. So the graph is the total of the building jobs + the crop jobs. (I think it's 1 job per crop tile, IIRC.)quote>

 

I partially agree, but also disagree with you.  I believe the game calculates agricultural jobs based on the total citywide agricultural zoning(fully developed) you have which is similar to what you say, but it is certainly not calculated on a crop tile by crop tile + farm building basis.

I revisited the farming community from my tests above and started adding up the tiles.

- There was a total of 1728 developed tiles for agriculture

- 197 of those tiles were occupied by a farm lot/building (including attached vineyards)

- The remainder allotted to crops is 1531 tiles

- Even factoring in desirability of 77% in my farming community(I simply queried all farm buildings and divided the current occupancy with the max) still returns 1179 (1531*0.77=1178.9)

- That would total 1219 agricultural jobs, and the graph shows only 252.

Anyways, I feel my method is straight forward(certainly a lot easier then figuring out each lot size and adding them all together) and goes with what infoscott has posted.  Again I'll reiterate that you need to look at your city as a whole to figure out how many agricultural jobs you have, querying and calculating individual farms will return some quirky results.  Or trust what the graph says, it's about 98% accurate from my experience

Cheers

[Bones1 96]

1728 tiles for agriculture, and the graph only showed you 252 jobs? I believe the graph, so that must mean that there was some pretty big inefficiencies in the farming layout. How much of that farm land was dedicated to buildings? If the building lot is large, it can contain much less that 1 job per title. Also, is it possible that the desirability was less than 77%? You can't query desirability for the crop tiles, and perhaps many of them were abandoned.

[VDub4Life 0]

Inefficiencies, I'm not sure.  Again I wasn't trying at all to see what zone size produced the highest number of jobs or anything of the sort.  I merely discovered a formula that calculated within 8 jobs to what the graph showed no matter what size or shape of farms you had.

Anyways, not sure what else to say other than the formula works for my cities with farming.  Just thought I'd share it with the Simtrop community

[PuggyPoop 0]

Great idea....

[flame1396 9]

Bones1: relying on the graph is a bad idea... the simulater's raw data (shown by query, and tax money) is more reliable. The graphs need to interpret the raw data so errors could be possible... and even fundamental coding errors, after all SC4 was made by people just like us.

[dare0021 1]

I don't know about you people, but for me, farm lands are just eye candy. All my farms are as wide as possible, zoning again and again to prevent the "Zone too large" thing. Not very profitable, but looks good.

[Patricius Maximus 236]

I have the radical ordinance mod and it gives me $10000/month, and that takes care of all my city needs until I get a large population.

[IGGYMAN 101 1]

I will plop different kinds of crops.

[Craig-Abcvs 1,595]

This topic is now being unstickied, but a link has been put in the index thread.