When to build your first granaries?

[georgjorge]

I also probably build Granaries later than most other people here. But the Settler-Granary comparison isn't entirely watertight on high difficulties.

The maintenance wall comes very early on Deity, at about 3 cities before Writing (without luxuries), maybe 4 or 5 before Currency. So there are not that many Settlers to build anyway. In contrast, growing one of your 3 or 4 cities faster so you can run early specialists or work cottages can be good. The only reason I don't build Granaries super-early is the low happiness cap in the BCs, so it might be better to chop, say, a Library first and run the specialists earlier than chop out a Granary and whip in a Library. Or work cottages early and slow-build a Granary instead of whipping it in.

Whether most hammers come from chopping or whipping depends entirely on the map. 2-pop-whipping something gives 60 hammers - that's 3 forests and 9 worker turns (pre-Maths). If you don't have lots of forests AND a good amount fo workers you will probably make more hammers by whipping in the medium term if you have good food surplus.

 

[lindsay40k]

Here's my flowchart, assuming not CRE or Stonehenge (Monuments irrelevant) or Inca (build a Granary everywhere immediately) or REX mode (prioritise high food /forest sites and whip & chop Settlers and Workers, building warriors or archers or work boats whilst waiting for cities to grow to use all their high food tiles):

- City location has a food resource in the first ring: Granary is first build
-- City location has important resource/s in second ring: Monument is second build; otherwise, Barracks or Library or Forge

- City location has a food resource only in the second ring: Monument is first build, then Granary, then Barracks or Library or Forge

- City location has mediocre food: eh, still want a Granary in case I have to whip to improve defenses

- City has awful food and exists to grab a couple of Tundra Silver or a Desert Iron: this rubbish outpost will be lucky to get anything

It's the ability for a city to whip in emergencies and recover that's good about a Granary in a city that's not routinely pumping units. Yes, your 5 river cottages site with not even a dry Rice tile might be capped at size five, but if Toku arrives with a half dozen Axemen and a Catapult then you want to whip two more Archers to stop him banzai charging whilst your own army travels to deal with him. Think of the opportunity cost as an insurance premium.

 

[Jivilov]

Playing Warlords on Emperor level didn't have a THIRD city settled until 1460 BC my last game. This is clearly unsatisfactory since a general rule is at least four by 2000 BC. Prioritizing granaries was part of the problem. So at what year or city number point should one generally start building them? Thanks.

 

[Lazteuq]

Thanks for the flowchart, Lindsay40k. It would be nice if this forum had some kind of upvote feature....
Jivilov, I agree that sounds like slow development, but I don't think there is really a year when you should build granaries, civ4 is too nuanced for that sort of easy guidelines. I think you should decide on a city by city basis, considering what that specific city needs to become more productive. Some cities might not ever need one, as said above.

 

[Jivilov]

Well said Lazteuq. This game is indeed a Byzantine web of complexity.

 

[Rusten]

Unit costs (hammers) don't scale properly to marathon speed IIRC, so you're coming from a different perspective in that regard. Settlers are going to be cheaper compared to a granary for you than for a normal speed player.

In simple terms granaries double your food surplus in the city. That means that their output usually rivals, or is better than, that of getting a new city even if it has an improved food resource right off the bat. Your capital for instance often has more than +6 food surplus. The granary's effectiveness can hardly be overstated.

But; early on the happy cap is a big issue. This means that you want to spread the food surplus out over more cities to have more tiles used. Thus it's usually better to settle a new city over getting a granary. You should rarely get a granary before you have your initial cities settled. I hesitate to give a number because it varies, but I think the 1500 BC - 1000 BC range is where I often get granaries. Calendar is certainly too late, unless you beeline it. Maybe you're not pursuing monarchy hard enough?

There's also the argument of grabbing city sites before the AI does for long-term gain, but that's beside the point here. I'm only referring to the efficiency.

 

[Jivilov]

Unit costs (hammers) don't scale properly to marathon speed IIRC, so you're coming from a different perspective in that regard. Settlers are going to be cheaper compared to a granary for you than for a normal speed player.

But; early on the happy cap is a big issue. This means that you want to spread the food surplus out over more cities to have more tiles used. Thus it's usually better to settle a new city over getting a granary. You should rarely get a granary before you have your initial cities settled. I hesitate to give a number because it varies, but I think the 1500 BC - 1000 BC range is where I often get granaries. Calendar is certainly too late, unless you beeline it. Maybe you're not pursuing monarchy hard enough?

There's also the argument of grabbing city sites before the AI does for long-term gain, but that's beside the point here. I'm only referring to the efficiency.

Thanks Rusten. Play normal speed myself so we're on the same wavelength. Your points about spreading out the happy cap over more cities and settling initial cities BEFORE building granaries was what I was looking for. That has been a consistent mistake which must be corrected. Armed with this insight perhaps Immortal level may not be unattainable. Never had a domination/elimination win at any level; it's high time this barrier was breached. Cheers and thanks again!

 

[Fippy]

In the other thread, you agreed that it's not good to build a granary before a settler and worker in your first city. OK. So we agree on that. What makes this true for your first city, but not true for every city?

It seems to me that its a general principle that a settler is a better long term investment than a granary, until you reach the point where you've already built every city you're planning to build.

Too general, bigger cities can be worth more than 3 small cities.
If you are in no hurry securing land from AIs, you may not want to use every city for expansion. Especially true for your Cap.

(and while most peoples are probably not playing Deity, it's even more true here. Try maximum settlers / expanding every game, ignore granaries, and lose most if not all your games)

 

[nate46]

Settlers aren't cheaper on marathon, but workers are.