granaries and barracks: when?

[ramses II.]

Do you build them? When?
Since I play Deity (Single Player), I cannot afford building them in the beginning. I need the shields (and the money - I must pay for the support)for the caravans needed for wonder building.
I build granaries in the industrial age, after a temple, an aqueduct, a colosseum and a market place.
What's your strategy?

And with some exceptions (cities with direct contact to AI civs - then a must have) I won't build barracks until mobil warfare. I must sell them too often.
That results of my strategy: I want to reach AC, I'm don't interested in long wars. I prefer building libraries and caravans instead of dragoons and howiitzers ...Why shall I kill an AI that brings me so much money by caravans and freight?
In my last ganem for example, a transport with 8 freights brought me up to 10,000 Gold (I needed three turns to deliver them till I had my ship chain).
I can image that barracks are a must have in MP games - or am I wrong?

[Six Thousand Year Old Man]

I usually play SP on Emperor (ToT) or a mix of Emperor/Deity on 2.42.

I never build granaries anymore. They're costly and while not having them (or Pyramids), slows early growth, using WLT*D seems to make up for that once I get to representative government. Ideally I can get Pyramids, too, but I usually want Colossus/HG first, and can't always get the first 3 wonders.

I seldom build barracks unless I have a Super Production city - one with Shakespeare and a lot of shield specials, or KRC. Then I'll build one to get vets, but only one. I lose vet effects via Leos, so barracks don't make much sense to me. And once Leos expires, usually vet effects don't matter anymore

STYOM

[Scouse Gits]

I'm a Barracks fan - often the only improvement in a city - but I believe in crushing the AI with Crooks and 'Clads - if you go on a rampage it is absolutely imperative that your troops are vet -- that extra 50% makes all the difference

What's a Granary? "" ""

SG[1]

[DaveV]

I rarely build barracks against the AI, although Sun Tzu's is one of my favorite wonders. As to granaries, I'd much rather have three more cities than two granaries. Exception: I'll build a granary in a single city if I'm trying to play a trading game (WLY growth under monarchy: join settlers up to size 8, then one food caravan per year).

[rah]

I can probably count on my hand how many times I've built a granary in the last decade.

Barracks. Those I build in trouble spots, or when I'm prepping an offensive. It's silly to build an army to attack if a majority of them aren't vets.

RAH

[War4ever]

i personally like the pyramids as my first wonder....must be a carryover from civ1 when pyramids WAS the wonder to have....

barracks...for war only......granaries....hahaha....temples are better than granaries

in fact i will build harbours before granaries......let me rephrase...i will build almost anything before a granary

[Juggler]

If I'm not playing a scenario I usually bild barracks as soon as possible. They allow my units to have a certain advantage over other enemy units.
Granary is actually last on my building list (and I try to be the first to build the Pyramids anyway).

[atawa]

I'll only build baracks when the time comes to start mass-producing howitsers.

As for graneries: I usualy grow so big I have a hard time trying to keep everyone happy as it is, no use in ading to that by bulding graneries

[Six Thousand Year Old Man]

One of the main things I think about when Civving is whether it's worth my while to build a particular improvement, versus a caravan or a settler (or a military unit).

A caravan takes 50 shields and zero support.
A settler takes 40 shields, and at least 1 food support - usually a shield support as well.

I always think of these 2 things and whether I would rather have one more city, or be anywhere from 1/4 to 1/12 of the way to a WoW that will help my whole Civ, or have a science bonus and the cash and trade arrows which will last the rest of the game... versus spending a lot of shields to build, and paying cash to support an improvement.

If you build a granary, your city will grow faster, but in 2/3 of the time you could have built a settler. Yes, you lose 1 point of population with a settler, but you get that back if you use the settler to found another city. And a size 1 and size 2 city will get to size 2 and 3, faster than a size 3 will get to size 5. Usually. (This is the basis behind ICS - 5 size 1 cities will work 10 squares of territory, while 1 size 5 city will work 6 squares).

Add in the fact that I will usually WLT*D as soon as I get to Republic... and I never build granaries.

A barracks is no guarantee of a win in combat - it helps, no question, but bad luck can still happen. If I have a city that's threatened, I'd rather have an Elephant or Crusader for the same shield cost. Or a second rifleman. Also, in the later stages of the game, it becomes easy to quickly reinforce threatened cities along railroads. So my defenses are usually minimal in most individual cities, but with a centralized force of 2-or 3-move units that can quickly respond. But then, that's my SP style. Obviously it would be different if I was a pure ICSer, or played MP.

Anyway, those are easy choices for me. I also never build City Walls, SAMs, or Coastal Fortresses, except in very key cities (once again.. I play SP ), any of Power/Hydro/Nuke Plants, or Police Stations. I tend to only build Superhighways and Research Labs only in the SSC.

What are other people's 'never-build' lists like, and what are your reasons for not building?

STYOM

[Smash]

Only build a granary if you can rush in 1>3 turns and only in a high food settler spitting city.And only at 1x production.Or in your Commie cap/ssc ala DaveV for food caravan growth.

If you must build a granary,might as well throw 4 camels into the Pyramids.

At least 1 barracks in a centrally located high sheild city.

[Bloody Monk]

Why not have both?? But think outside the box; do it creatively.

It's 500b in the Campo Challenge. I usually follow the previous advise: why capture AI cities when you can trade w/ them for mucho gold, etc?? But this Challenge is a conquest game and it is easier to take a city before it gets walls. So I built a couple of barracks to get vet Crooks and extras Setts to build roads to pave the way for capturing the Aztec Cap just after it finishes the Pyramids.

Win, win, win: granaries for everyone forever paid for by the AI, weaken an AI civ, barracks in place for more vets. Plus, my cities will grow w/o having to slow things down in Rep WLT*D, saving two revolts and keeping research focused on Conquest.

I had never done early conquest before, always playing AC. This was a new and interesting and fun alternative. I just love this game!!! Three years, scores of games, the thrill of discovering a whole new (for me) Strat. Hope this is helpful.

[Bloody Monk]

Originally posted by rah
I can probably count on my hand how many times I've built a granary in the last decade.

Barracks. Those I build in trouble spots, or when I'm prepping an offensive. It's silly to build an army to attack if a majority of them aren't vets.

RAH

I think this is "the" definitive and consise response to the orignal query, which I understood to apply to SP, 1X games, at Deity. Until you get Mike's Chapel, managing growth is a problem making granaries ( and Pyramid) wasteful, IMHO.

There could be special cases where they would be helpful, though...in cities w/ Gold,for example: the extra trade becomes more meaningful and the city can grow more easily. Also, to boost the effect of your SSC: more Pop=more trade arrows.

Barracks: It's not a bad idea to have a few Vet Killers to take care of Barbs. Warriors don't cut it against Barb Legions.

[Campo]

DaveV and Bloody Monk both touched on my preference. I like to build a granary in my SCC. If I build HG there it gets the happiness bonus, which generally controls the problem (along with temple, marketplace, etc.) for that city. With a granary it can grow faster and produce more science.

The problem is fitting in it. My SCC strives for Colossus, HG, Cope's, and Issacs, and of course temple, marketplace, library, bank, university, etc., plus caravans for trade routes. Can't always find the time (or money) to build a granary when most needed.

I use WL days for the rest of my cities and eventually capture Pyramids as well.

Barracks -- I didn't used to because of the cost and obsolescence, plus the loss of vet status due to Leo's upgrades. My challenge game convinced me of their worth for early conquest, but I'll probably skip them in favor of Sun's for more leisurely games.

[Rufus T. Firefly]

About 2/3 of the way through a game, I generally come to a point where a city can only build food caravans and already has all of the "basic" improvements (walls, temple, library, aquaduct, sewers if necessary, and all the economic improvements). At that point, if I have Adam Smith's, I'll build a granary.

I rarely play world conquest, so rarely build baracks. When I do play for conquest, I don't really begin to conquer until late in the game (cavalry and artillary are my preferred units), so I build them in high production cities then.

[Julius Brenzaida]

I have to agree with everybody. Granaries are not a priority. 60 shileds to bring you unhappiness faster is the beginning of a deity game, are you crazy. What's more the pyramids are so cheap, you have them for 3 granaries and some, and growing with WLTPD is always an option. So I almost never build granaries.
As for barracks, I usually set a few high producing cities to make war units for everyone and they will have barracks. But of course that it depends on your strategy : AC or conquest.
I am going to shamelessly use STYOM idea and start a thread with the least favored improvement.

[Marquis de Sodaq]

I only build granaries in cities with really slow growth (+1 wheatsheath), if at all.

Barracks usually come later in high production cities. I just use them for pumping out bad guys as soon as the city has any happiness issues resolved. I rehome units in cities that are busy building improvements or caravans. If I sack an AI city with one, I never sell it unless I have SunTzu's. Or if the city is unwalled and the war rages on nearby...

[geofelt]

Granaries are good, but I only build it in my SSC if I can fit it in early. The pyramids are a good buy. Early on, I might build a barracks in a high shield city and use it to produce vet defenders for everybody. Later on, every city producing units that are expected to be used will have a barracks.

[SlowThinker]

Originally posted by Marquis de Sodaq
I only build granaries in cities with really slow growth (+1 wheatsheath), if at all.

I disagree with this approach.

I think you should build granaries in cities that often grow (ideally cities that pump out settlers or cities that admit food caravans).

Imagine you have just 2 cities of size 1, 1st city has the food surplus 1, 2nd city has the food surplus 2, both will grow next turn, one of them has a granary.
1st city will build 1/30 of a new square each turn, with the granary 1/15, i.e the granary gives you an additional 1/30 of a new square each turn.
2nd city will build 1/15 of a new square each turn, with the granary 1/8, i.e the granary gives you almost 1/15 of a square each turn.

I often turn 1-surplus cities into 0-surplus cities (working a forest...).

---

Barracks:
Vets are almost 2times more powerfull (1.5 times higher attack and defense means the vet unit is about twice stronger against units of a similar strength and 1.5 times stronger against units of a very different strength (for example warrior against armor or armor against warior)), i. e. the use of shields must be more than 2times effective (the difference between survival/death).
(You can "train" units in battles (I mean send them to a defensive terrain near AI activity, fortify and let the stupid AI make them veterans). But this way is costly (shield support). )
Conclusion: I think barracks are a must in most cases.

Granaries:
Unfortunately, deity is extremely unbalanced, the shields are too important. In the beginning of the game you just need to build settlers and settlers.
This fact causes granary doesn't pay off. Maybe in chieftain, but I never calculated that.
Granary may be good for special purposes only (pumping out settlers or pumping in food caravans).

[TCO]

I very much like both Pyramids and Sun Tzu. I rarely build barracks or granaries.

[rah]

I'll also sometimes build it in that slow growth city just to get it to size three for celebrating faster (normally because it's built on gold). And then only when it's an outpost, because if it's close to your core, it's easier to send a settler or two to grow it that way.

RAH