Food trickle efficiency: Ice House vs Corral + Cows

[ffm2]

I was recently calculating whats a better investment for food trickle, a ice house or a farmstead and cows garrisoned.

 

ICE HOUSE (Limit: 5)

Base building:
  Cost:       100 stone, 100 wood
  Build time: 60 s
  Trickle:    1 food / 2 s   (0.5 food/s)

Aqueduct upgrade:
  Cost:        300 stone, 300 wood
  Research:    40 s
  Trickle:     2 food / 2 s  (1.0 food/s)


5 ICE HOUSES (max):

  5 × (100 stone + 100 wood) = 500 stone, 500 wood
  1 × (300 stone + 300 wood) = 300 stone, 300 wood

  Total resources:  1600
  Total trickle:    5 food/s

  Cost per 1 food/s:
  1600 / 5 = 320 resources


CORRAL + COWS (Limit: 50 cows)

Corral:
  Cost:        100 wood
  Build time:  50 s
  Capacity:    8 cows

Cows:
  Cost:        150 food
  Produce:     60 s
  Trickle:     3 food / 3 s  (1.0 food/s)


7 CORRALS + 50 COWS (max cows):

  7 × 100 wood         =   700 wood
  50 × 150 food        =  7500 food

  Total resources:       8200
  Total trickle:         50 food/s

  Cost per 1 food/s:
  8200 / 50 = 164 resources


Summary:
Ice House:      320 resources per 1 food/s
Corral + Cows:  164 resources per 1 food/s

 

Corrals win by far. I think the ice houses should be better considering they are a civ specialty.

Edited Friday at 20:52 by ffm2

[chrstgtr]

Interesting but it’s an incomplete picture. 
 

1. Corrals use food as a resource input, which is slower to gather. 
2. Corrals use of food to gather food also means that the food collection chart will have lots scarcity dips caused by capital investment costs. This is pretty dangerous and can bankrupt a player quickly. 
3. Corrals require more micro.

4. Corrals require continual investment/management whereas ice houses just need to be built once. 
5. Ice houses are unique to Persia, which has a higher pop cap/ eco potential, so more it’s easier to collect the input resources 

6. You’re missing the corral training tech

7. Cows can be batch trained 

8. You can build 7 (or more corrals) all at once  

 

all in all, I agree ice houses need a buff but I don’t think it’s as desperate as the above portrays 

Edited Sunday at 22:02 by chrstgtr

[Arup]

320 per 1 food? if you add the house tech into it,it probably become somewhere around 180 (guesstimate, not calculating)

that's not bad for a obscure piece of the biggest civ in game right?

the persians need a lot of balance changes (both buffs and nerfs) perhaps I'll make a thread on its own

[Dakara]

don't buff persians 

plus persians can cumulate theses 2 solutions 

btw corals

[wowgetoffyourcellphone]

On 15/02/2026 at 4:01 PM, chrstgtr said:

3. Corrals require more micro.

This is an important one and the first thing to come to my mind when reading the OP. Passive bonuses, even if not as "lucrative," are generally better because one can use the APM you'd use on a corral for something else. An Ice House is a "build it and forget it" feature. Not so with a Corral. Similar to the difference between a Global aura vs. a Ranged aura for heroes. Global auras should always be weaker than Ranged auras, and vice versa. 

[ffm2]

On 15/02/2026 at 10:01 PM, chrstgtr said:

1. Corrals use food as a resource input, which is slower to gather. 
2. Corrals use of food to gather food also means that the food collection chart will have lots scarcity dips caused by capital investment costs. This is pretty dangerous and can bankrupt a player quickly. 
3. Corrals require more micro.

4. Corrals require continual investment/management whereas ice houses just need to be built once. 
5. Ice houses are unique to Persia, which has a higher pop cap/ eco potential, so more it’s easier to collect the input resources 

6. You’re missing the corral training tech

7. Cows can be batch trained 

8. You can build 7 (or more corrals) all at once  

It's a good thing that it gives the thing it costs. If i trade, I start to set the trade on 100% metal to finance the training of traders until they are enough. So I see 1. more as a pro corral argument.

The corral training tech. It slows the investment down, so one can invest more of the gains the current cows are giving. This is not a important point though.

It is more micro, but not much. You can do it in a non-crusial time of the game. It is also a set and forget thing. Just place a coral at some time, Just activate auto-training, set a way point on the ground next to the corral, set a way point on the corral to garrison. So its something like ~5 clicks. The cow production stops automatically if there are 50 cows.

The value of food varies during the game very hard. Since populating up is a extreme high priority one make >10 fields. Then when you reach pop cap, the value suddenly drops, while you still have the farmers on the field. The best use is to be the first at the market then and converting it. But then when the game drags on it's best invested in the corrals to free up the pop of the woman on the fields.

Persians sure can do both but it's advisable to do the less efficient one later when the efficient method is fully exhausted.

Just for note there is also a little trick possible, build a house as extension of your territory, build the 5 ice houses as the extension. Delete the house. Let the ice houses go to gaia. Build 5 new ice houses. Capture the old 5 ice houses back. Interesting for the "survival of the fittest" game where the Persians are OP.