Sacrificing the Depressed

[vale]

I think the problem stems from the fact that you are thinking of slavery as transforming into when a more accurate metric is that you are trading population-turns into production. Production in this case refers to both food and hammers and the production gained by a whip is shown in the following table where rows are whip size and columns are original city size:

Whipping something that does not increase happiness when you are at the happy cap always costs at least 10 population turns. There is no cheating the system here. That extra unhappy will always be relevant.

With a happy cap of 5 there are realistically 3 options that have a chance to be best for production:
1. Stagnate at 5: Only relevant in an extremely food poor city.
2. Whip 2 pop every 10 turns from pop 4 - this is sort of the "default mode" and is very easy to sustain as long as you still need workers and settlers.
3. Whip 3 pop every 10 turns from pop 6 - this is harder to sustain as the only early repeatable 3 pop whip is settlers and there usually is a limit to how many settlers are useful.

 

[Heathcliff]

People need to whip more, stop using the carrot guys!

 

[ben-jammin]

Wow, I honestly never knew this. It also works for workers it seems too. I don't think you actually gain anything out of this but it is definitely counter-intuitive.

Maybe the guy whos unhappy due to overcrowding is the one who gets to leave and settle a new, less crowded one

 

[Um the Muse]

Maybe the guy whos unhappy due to overcrowding is the one who gets to leave and settle a new, less crowded one

That sounds like it's as good a reason as any. I mean, this has got to be deliberate, right?

Anyway, I think that I finally realized that I was wrong (again). MyOtherName was right all along about not actually gaining anything by paying the unhappiness penalty.

I still can't post the screenshot and save yet, but I should be able to later tonight.

Edit: Ok, here you go. As you can see, while the capital doesn't have super strong tiles, they should be stronger than slavery. At least, I thought that they were. I don't know. The second city, on the other hand, had three really powerful tiles and then a floodplain.

I tried this map exactly twice (trying to transition between Marathon over to Normal speed) and pretended that I didn't know the map the second time through. End result was an extra city and one really close to being built.

The Capital:

Spoiler :

The second city:

Spoiler :

I didn't take a screenshot of the third city, but it didn't have production either (it had double gold, which I probably should've picked up later).

While I'm writing this anyway, I thought I'd ask about this:

Spoiler :

How come I'm losing food per turn and yet I've got a one food surplus going to the settler?

Note: To look at the savegames below, you'll need BUG/ BULL installed as a mod.

 

[Um the Muse]

Vale, can you explain your chart some more? In particular, what do the numbers in the cells represent?

 

[VoiceOfUnreason]

How come I'm losing food per turn and yet I've got a one food surplus going to the settler?

That's weird - I think it's a side effect of how the "food + angry citizen + production" works.

That is, you have 7 pop, and 13 food income, so you are running a 1 F/turn deficit. Your production into the settler is 34 hammers (overflow from a previous build?) + negative 1 food (because of the deficit) + 2 food (because you get the food that the unhappy citizen is eating back as production).

In short, it's "food eaten by angry citizen also contributes to the training of the worker/settler", as opposed to "angry citizen doesn't eat food when training a worker/settler".

 

[VoiceOfUnreason]

As you can see, while the capital doesn't have super strong tiles, they should be stronger than slavery. At least, I thought that they were. I don't know. The second city, on the other hand, had three really powerful tiles and then a floodplain.

Hmm, yeah, they might be. Financial floodplains are a lot better than non-Financial grassland, which is what I consider when running whip models.

 

[TheWilltoAct]

Haha, way to present a bizarro interesting case for the whipping brigade.

I could definitely see growing into anger as a compelling strategy here.

 

[vale]

Vale, can you explain your chart some more? In particular, what do the numbers in the cells represent?

Sure.

As others have pointed out in various places (not in this thread I don't think) when you are stagnating (either via net food = 0 or via building a worker or settler), the total production of a city is the sum of + - 2 for each worked non-center tile plus the + value of the center tile.

The same holds true if you are doing a whipping cycle that starts and ends at the same food level after N turns (you usually want N to be 10 or 5 with sacrificial altars if whipping is gaining anything). So when you whip, and convert a bunch of to what is listed in that table is the net production gain (i.e. gained - lost).

The cost for this production boost is the loss of population-turns. When you are dealing with whips from or near the happy cap, you are always losing at least 10 population-turns. So if the tiles that you could have worked over the lost population-turns would have resulted in more overall production, then the whip will actually cost you production.

All of this analysis assumes that commerce is a non-entity. The only thing I am caring about here is maximizing production over time. I can't open your save because I installed BUFFY not BUG, but I'll look at the city sites in a second and see what could have been done for best early game production.

Question about those cities as you were developing:

How early were the flood plains cottaged? Were they ever farmed for early game growth/production? What did your capital look like before your happy cap lifted? Same questions for York?

 

[vicawoo]

So what's the question now? Should you grow into unhappiness and whip?

In the short run, you should grow because the riverside grassland mine is a better tile than being whipped away.

In this case, particularly if you're in a whip cycle you should grow into unhappiness for a 1 or 2 pop whip because your tile yield (4 without imperialistic bonus) beats the 2 food you lose from growing to a larger size then whipping.

Finally, this brings up another point, whipping doesn't help your commerce. Since you're running scientists, I'm assuming you really care about research, and whipping will hurt you.

 

[vale]

I don't think that is his question. I believed it was during the early game with a more constrained happy cap that he was wondering.

I installed bug 4.4 and FWIW I was able to get my third city out about 200 years earlier. What that means in the grand context of things is hard to tell because the endpoint only lets me see city found dates and not what they looked like at a particular moment in time. But I can talk about concrete production cycles out of your initial city.

To maximize production out of this city, there are two cases, imperialistic leader or not imperialistic leader.

Case 1: Non-imperialistic leader:

You need:
Granary
4 of your possible 5 4 food tiles actually producing 4 food (too early for lighthouses or civil service early game)
The plains hill mined

With this setup you can produce optimal production at the original happy cap of 5 by 3 pop whipping settlers every ten turns and carefully micromanaging your queue and tile usage. If you do so you will produce an average of 152 every ten turns. 100 of those will always go to a settler with occasional investments into workers to help avoid growth into unhappiness early since there is high food and only one good food deficit tile.

This is the least flexible method because of the requirement of making a settler every ten turns and being forced to invest into workers as well. You can make your hammers slightly more flexible by working worse production tiles (ideally a flood plains cottage but more likely to be a plains forest). Every population-turn you work a flood plains cottage, a forest, or a mined desert hill, you lower that 152 number by 1.

152 every 10 turns is vastly superior to what this city can do in stagnation at pop 5 which is 90 every 10 turns.

The compromise out of this city is 2 pop whips every 10 turns from size 4. In an ideal situation this generates an average of 139 hammers every 10 turns. Two pop whips are much more flexible so you could easily have a large portion of the hammer output being pumped into axemen if you so choose. There will still be a need for some investment in workers or settlers because of the prevalence of food, but not nearly to the degree experience when doing 3 pop whips.

Now to go back to the idea of growing well above the happy cap to perform a 2 pop whip. From a production standpoint you are screwing yourself in two ways:
1. Every turn you support an unhappy citizen while growing you are costing yourself 2 production.
2. The eventual whip converts to at a less efficient rate than a lower pop whip would.

You are gaining the possibility of fewer population-turns lost, but the drawbacks absolutely outweigh benefits.

The imperialistic leader case is sort of interesting because in order to set up a 3 pop settler whip it has to work grossly inefficient tiles for a turn every cycle. Essentially, the turn it preps a settler in the queue it must end up contributing less than 10 total production to the settler. The bonus is you can use this inefficient production time as a chance to cash in some commerce
clams
cottaged flood plains
lake
farmed flood plains

Now this cycle only produces an average of 149 over 10 turns but the difference is that only ~67 of those are being spent on a settler each cycle. So you are getting more free hammers to play with although some will still have to be dumped into workers because of happiness constraints.

 

[vicawoo]

Happy cap 5, continually 3 pop whipping is the best? That's a little hard to believe, since at some point you're growing 5 to 6 with an unhappy citizen. I guess ideally, you're at size 5 for only one turn. So you lose 2 tile turns with one tile turn from being at size 3 for at least 1 turn.
You need 13+14+15=42 food, gain 90 hammers, lose 2 tile turns, and lose 10 more tile turns from whipping.

Stagnating should provide 13 food/hammers per turn with 5 tiles.

So 130 + 90 - 42 - 24 = 130 + 24 = 154. I'm guessing that it's taking 2 turns to grow from size 3 to 4, so that should be the other -2, for 152.

2 pop, I guess you should go size 5 to size 3. If you grow from 3 to 4 in one turn, you lose 1 tile turn from growing.
So 130+60-27-22=141.

Now that I think about it, 4 yield tiles are pretty neutral to 2 pop whips, since if you grow a size you gain 4-2=2 yield from an extra tile turn, but lose 2 food due to growing taking an extra food at each size.

 

[vanatteveldt]

One thing missing in the 'mathematical' approach is that whipping allows you to work better tiles.

Often, a capital or good city will have two good food sources (ie 5 or 6 food), and these will be among the best tiles. Problem is, with a low happy cap there is no (good) way to 'spend' that food working mines and specialists. So, if you choose to remain at happy cap you will be forced to work tiles that are inferior to the food specials.

In these cases, the opportunity "cost" can even be negative as you can 'produce' more after the whip than before by switching from inferior hammer tiles to superior food tiles (counting food as production, as seems to be the norm in this thread ).

This, to me, is why the FP capital above does not scream whipping: there is no large food source and the surplus food of working rice plus 3FP (2+2+3) can be efficiently spent on the mines and/or librarians.

 

[Um the Muse]

Well, that's a lot to take in, that's for sure. Vale, you asked what I did with those floodplains. I farmed one of them and then cottaged the other when I had the relevant technology and I had spare worker turns. The farmed FP got converted a little later, I think it was around when I had sailing.

So, I found that chart pretty fascinating, though I still feel I don't completely understand. So, as long as you take exactly 10 turns to grow from the happy cap, those numbers above will tell me how much production (in the hammers sense) I get out of the food I put into it. Is that what you meant?

How does the happy cap play into those numbers? Was the chart specifically meant for happy cap 5, or was this a general purpose chart? If I divide your numbers by ten, do I arrive at the average effective production yield I have to beat in order for slavery to come out ahead of stagnating?

 

[dr_s]

Yeah, maybe I'm dense, but I don't understand those tables either. vale, can you explain in detail how one of the numbers is calculated, e.g. the 35 in the second table at pop 4/whip 2? Is that 35 just the 60 hammers you get from the whip minus the food required to grow back to size 4? (There must be some assumption here about the status of the food bar when you whip, no?)

 

[vale]

2 pop, I guess you should go size 5 to size 3. If you grow from 3 to 4 in one turn, you lose 1 tile turn from growing.
So 130+60-27-22=141.

This isn't sustainable over time. When you are regrowing you either run a net food deficit over the course of your cycle or you will be forced to support an unhappy citizen at population 5 for one turn. The sustained cycle will therefore only produce 139 over ten turns, the same as 2 pop whips from pop 4 coincidentally.

In the 152 3 pop whip cycle, you either spend an extra turn at pop 3 or you support an extra turn of unhappy citizens. That is why it is 152 not 154.

So, I found that chart pretty fascinating, though I still feel I don't completely understand. So, as long as you take exactly 10 turns to grow from the happy cap, those numbers above will tell me how much production (in the hammers sense) I get out of the food I put into it. Is that what you meant?

How does the happy cap play into those numbers? Was the chart specifically meant for happy cap 5, or was this a general purpose chart? If I divide your numbers by ten, do I arrive at the average effective production yield I have to beat in order for slavery to come out ahead of stagnating?

Basically the table is showing how much production is generated by the whip in a food neutral cycle. The city size (columns) is assumed to be near the happy cap. Whipping in cities well below the happy cap is its own optimization problem and usually more an issue of getting things more immediately regardless of efficiency.

In terms of evaluating the efficiency of slavery vs stagnation, it really is a case by case basis. The general indicators that would be used in evaluating a city would be:
1. Presence or absence of granary/sacrificial altar
2. Size of city: larger decreases production created by whipping.
3. Size of whip: larger increases production created by whipping.
4. Many food surplus tiles: decreases population-turns lost in larger whip cases.
5. Good food surplus tiles: increases population-turn efficiency during regrowth.
6. Sufficient numbers of good food deficit tiles: increases population-turn efficiency during stagnation.

Yeah, maybe I'm dense, but I don't understand those tables either. vale, can you explain in detail how one of the numbers is calculated, e.g. the 35 in the second table at pop 4/whip 2? Is that 35 just the 60 hammers you get from the whip minus the food required to grow back to size 4? (There must be some assumption here about the status of the food bar when you whip, no?)

Yes the table is simply showing gained - cost.

The only assumption made about the status of the food bar in that chart is that your granary has been in place long enough to be holding 100% of its fair share of the food. Other than that the status of the food bar does not have any bearing on that chart. What it will change is the number of population-turns you lose as a result of slavery.