Infrastructure: The Master's Key To Victory

[FriedrichPsitalon]

Greatness comes from a lot of work(ers)
Infrastructure. One word, containing in its meaning the most vital concept of civilization. Infrastructure. The engine which drives war machines, or the funding which drives your research. Infrastructure – the greatest warmonger is hopeless without it, and the best builder builds nothing without it. Underlying everything, the one core skill – the one that is so rarely talked about, but implicitly understood to be what separates the best from the next tier and so on – the skill that brings you those troops, that gains you your technologies, spawns your cities: your development of your infrastructure – your use of workers. The greatest strategist who fails to understand how best to use his workers will come to nothing, and the most brilliant warmonger who does not know that infrastructure is the engine of his war machine will never comes to greatness. Too often and too soon do many players relegate their workers to the AI (if they ever bother to guide them by hand at all!) and here lies the ultimate secret of the masters – the greatest skill of the best players: what they do with their workers and towns that you do not!

General Rules of Thumb
In order to become truly skilled in the use of workers, it is important to understand one of the fundamental rules of “Ancient Start” play, and Despotic governments in general: whenever you gain three or more of a type of income from a tile, one is removed. Thus, if you gain two shields from a square, you will see and use both. If you gain three, however, the third shield is removed – you do not see it even existing! Many players do not realize this because they do not see it; it does not show up as corruption – it simply does not show up at all! A mined hill in despotism will show 1 food, 2 shields if you were to examine it under a Despotic government, but doing nothing else except changing your government will cause that same hill to show 1 food, 3 shields. This is a fundamental concept that many players in Ancient Start games fail to grasp. The two places this shows most readily in the ancient start game are in grassland tiles that receive irrigation, and hill/mountain tiles that are mined. For example, one of the most common mistakes that I have seen made in the ancient era is the irrigation of a normal (or bonus – the kind that produce 2 food 1 shield) grassland tile. This raises the food produced on the tile from 2 to 3, but because of Despotic government, it drops back down to 2 – meaning the irrigation HAS NO EFFECT! Let’s examine, case by case, each kind of land with and without bonuses, and note what could and should be done with each.

Grassland

Regular grassland provides two food – or, if a bonus grassland, 2 food and 1 shield. In a Despotic government, grassland without a bonus such as wheat, game, or cows should never be irrigated – it has no effect other than to waste the time of the worker in question. Mining these squares grants the square 1 extra shield per turn – making it 2/1 in the case of regular grassland, or 2/2 in the case of bonus grassland. The largest portion of the time, grassland should be mined – certainly when in Despotism.

Grassland with game (only possible after a forest chop) or wheat normally produces 4 food, since wheat gives 2 extra food, but Despotism reduces this 4-food tile to 3 food. Irrigating it will allow you – in Despotism – to get back up to 4 food, but many players prefer to mine that tile. The reason for this is that during a Golden Age, a tile that produces 1 shield will produce 2 – but a tile that produces 0 will continue to produce 0, wasting the Golden Age’s effect. Whenever possible, you want every square to be producing at least 1 shield. Grassland with a cow would normally produce 5 food, but is reduced in Despotism to 4. Once again, see above – mining is probably a better option. (There are rare exceptions to the “mine grass-wheat and grass-cow” rules; special cases are addressed below under “Other uses for workers.”)

Disclaimer: There is a school of thought which argues that you should irrigate the above examples of grassland/game/wheat/cows. The idea is that you should irrigate these to get the maximum food incoming to your city, get it up to maximum size (6 or 12) and by growing more quickly, you end up netting more shields by sacrificing some early on for more growth, but benefiting with more citizens to give you more worked tiles later. Typically, this method also involves changing the irrigation to mining at a later date, but I feel that in MP, you can’t afford to waste time by hitting the same square twice, and early production is important to protect against the Dagger-type players.
Plains
A plains square on its own gives 1 food and 1 shield, offering a player a few different options, depending on what the other tiles around the city look like. Cities that have a great deal of food coming in ( multiple tiles of floodplain/wheat, grassland/cow, for example) should probably mine plains, but under most circumstances, you should irrigate plains. The reason for this is that in order for a city to keep growing, you need to have at least 2 food coming in from every tile – a city will stop growing after a certain point if it begins working 1-food tiles. (This is why a city in desert doesn’t cease to exist – the city square still produces 2 food.) If you have a square producing 3 food, you can afford to use a tile producing 1 food, like a mined plains, or a forest. If you have a square producing 4 food, you can afford to use a tile producing 0 food, or two tiles producing 1 food, and so on. In most circumstances, you will have grassland tiles, which produce 2 food – requiring you to irrigate your plains. Still, this is not an unhappy condition; irrigation is a fast worker action, and doing so gives you a square that produces 2 food and 1 shield – ideal for a Golden Age, where it will become a 2/2 tile. (Mined bonus grasslands, producing 2 food/2 shields normally, still will only produce 2 shields in a Golden Age, since Despotism steals the bonus shield.)

Plains with wheat are a little trickier. A plains square with wheat in Despotism APPEARS to be 2/1, but we know that it is actually (3-1)/1. The wheat, giving 2 food, actually makes the plains square a total of 3 food, but the cruelty of Despotism lowers it to 2. Irrigating it moves it up to 4 food – meaning you will see 3. Mining it moves it up to 2 shields, making it a 2/2 square. Here we have to make a decision – do we want a 3food/1shield square, or a 2food/2shield square? The answer is difficult and refers back to what we discussed above. With a 3 food 1 shield square (irrigated plains-wheat), you can then work another square that only gives 1 food – like a plains with a mine, or a hill with or without a mine. That totals, for the two tiles, 4 food (3+1) and 3 shields (1+2). With the 2 food/2 shield square (mined plains wheat), you need to work another 2 food square – meaning, most likely, a grassland with a mine or an irrigated planes. This totals 4 food (2+2) and 3 shields again (2+1) ! It would appear that the two are relatively equal in what you get for your work. The difference – and the answer to this puzzle – is in the time it takes to create each. Irrigation is quicker than mining, and with an irrigated plains-wheat and a mined plains or hill, you have your workers moving on to other tasks more quickly than if you mine the plains-wheat, and then mine another 2-food square. This may seem confusing, so to put it in simpler terms: It is always faster to get an irrigated tile than a mined tile, so when both combined with other tiles give you the same result for a city – irrigate! (Or the simplest possible rule: unless you have irrigated floodplains or irrigated wheat/cow grassland, always irrigate plains.)

Plains with cows are a no-brainer; these should always be irrigated in Despotism. This is because the cow gives 2 food and 1 production, making plains-cow a 3food 2 shield square (which, in Despotism, will appear 2/2.) Thus irrigating the square makes it 4 food (appears 3) while mining it would result in only 3 shields (appears 2) which makes no change at all in Despotism.

Hills, floodplains, and mountains are also no-brainers; there is only one choice as to how to improve these squares. Typically, mining hills and mountains should be a relatively low priority – it takes a very long time, and the benefit will still be dropped down to 2 shields by the Despotic penalty. Irrigating a floodplain, which makes this 3 food (appears to be 2) tile into a 4 food (appears to be 3) tile isn’t a bad choice, but again, should be lower priority – since 3 food and no production is not always desirable – except possibly for worker factories.

Tundra on its own is useless – 1 food, 0 shields. Mined, it becomes 1/1, and even with a railroad it is still only a 1 food 2 shield tile… so why bother mining it at all? Planting a forest makes it a 1 food 2 shield tile to begin with.

[FriedrichPsitalon]

Other general rules of thumb worth knowing….

When possible, the order in which you should improve the land around a particular city:
Mine or irrigate any square (as appropriate) with a “bonus resource” on it.
Mine bonus grassland/irrigate plains/irrigate deserts (if you are AGR).
Mine regular grassland.
Irrigate floodplains.
Chop down jungle around the town.
Drain marshland around the town.
Mine hills or irrigate deserts (if you are NOT AGR). (Usually, you should be working through this phase when a ladder game ends.)
Chop down forests and treat those squares as above. (If you get this far in a typical ladder game, something odd has occurred.)
Plant forests on tundra squares. (If you get this far in a typical ladder game, something pretty odd has occurred.)
Mine mountains. (If you get this far in a typical ladder game, something wildly odd has occurred.)

(It may help you to print that list out or try to memorize the first few sections – it can be tough to manage a lot of workers at first. Also, having a set pattern means not wasting a lot of time trying to decide what to do with any kind of square, which can make it easier to manage more workers.)

MINE/IRRIGATE FIRST, road second, but always road! Always when improving a tile, mine or irrigate first – food and production are more important than commerce, 99% of the time. Do not fail to road the tile after improving it, however – you can never tell where you may need to move units quickly later, and if a worker needs to “double back” across territory at some point, having a road in place makes him get to his new target tile that much quicker – making YOU develop that much faster! This means in most situations, rather than using the “Road to” command, you should use the “Go to” command.

Too Much Of A Good Thing Is Almost As Bad As Too Little
Knowing exactly how many workers to have can be difficult, and often requires a good sense of judgment and the situation. A good general target is to have 1.5-2 workers per city (1.25-1.5 for IND civs) for fast growth. Conveniently enough, most cities can build 2 workers in the time it takes to build 1 settler. You may want to move this number up if you have a lot of poor terrain (if you see a lot of jungle chopping/marsh draining in your future, particularly), but it is rarely a good idea to move that number per city down – even the most wonderful cow, game, and grape-filled paradise will still benefit immensely from improvements! Generally, if you have one city building nothing but settlers from a granary, and one city building nothing but workers from a granary, the worker-town will keep pace just fine.

If for some reason you have more workers than you can effectively manage, then and ONLY THEN should you consider automating them – automated workers make BAD choices. Want proof? Play single player sometime, and let the AI run your workers while you just hint “enter” over and over. Note how many irrigated grasslands you’ll have in despotism, and how many mined plains you’ll see that a city won’t use because they don’t have surplus food from another source. If you run into maintenance costs, don’t be afraid to have workers join other cities – particularly those with very good land around them. Having a city that’s nothing but floodplain pump workers into another city that has cows and hills is a great idea! (In fact, the secret to getting massive production centers for wonders early on is exactly this.)

Other ways to skin a worker....er....cat

Although practical uses of the “other functions” of a worker are comparatively few and far between, there are cases where workers are handy for something other than irrigation, mining and roads. A brief list, then, along with the “good” uses of these functions.

Forest Chops: The problem with cutting down forests is that it cannot be used for one of the things it would be most useful for: wonders. Having said that, some expensive structures like marketplaces and aquaducts can benefit from the 10 shield bonus of a forest chop – usually saving between 1-2 turns. Whether this is worth the average of 4 worker turns needed to chop is up to you.

Colonies: This is probably the best possible use of a worker. If you see a resource that is considerably out of reach of your empire, but not in the direction of an enemy – send a worker or two that way with a “road to” command. (You may wish to send more or less workers depending on the terrain.) If barbarians are set to a troublesome difficulty, you may wish to send a warrior or spear along – they can both ensure the safety of the workers, and protect against annoying colony-overruns later on. Luxury or strategic resource regardless – it can be worth it to establish a colony on that resource. True, colonies are vulnerable to attack and having roads severed, but risking one or two workers is a lot less dangerous than risking putting a town in a hazardous, exposed, and isolated position. Learn to love colonies!

Outposts: The second most useful and unfortunately far less frequently used function of the worker is to create an outpost – in doing so, a worker “kills” himself and an outpost will appear wherever the worker was, with the same sight range as a worker. There are three advantages to this, however – first, outposts require no maintenance; a handy way to rid yourself of an overabundance of workers as the game progresses. Secondly, outposts on mountains see further than units can by 1 tile – an outpost on a mountain can see 4 tiles! (Try sneaking a sea attack by one of THOSE watchers.) Lastly, outposts cannot be “obstructed.” A normal unit, if watching from a mountain or hill, cannot see past another mountain or hill next to it (the second tile in that direction will be dark.) An outpost is not restricted by obstruction, which can be very useful when you have to watch for incoming units over a large mountain chain.

Fortresses and barricades, unfortunately, are not often useful, but there are situations where they can come in handy: bottlenecks, for example. A bottleneck on a plains square can be a pain to defend, but your workers can build a fortress there, and then barricade it, making it the equivalent of a mountain! A bottleneck mountain or hill, on the other hand, becomes almost unassailable with a barricaded fortress present. Barricades can also be good on critical squares – for example, a mountain square which is within double-move reach of two towns, etc. Unfortunately, due to the length of time needed to create barricades, one should be careful not to build them where they ARENT definitely needed, as workers often have better things to do!

Siege Roads :The “siege road” is a concept I first brought up in an older article, but rarely sees use. The idea embedded in this concept is that when you launch your attack force of, say, 35 units, you are still producing more units of some sort back home. Usually people wait until they have another 35 units and throw those in at a later time, but with a siege road, it becomes possible to have units race down this road and “meet up” with your main attack force. Typically, the siege road involves take 2-4 workers out (probably under guard) and building a road towards your enemy – usually, you start this just before your first attack force sets off. The first attack force will have to move at “normal costs” – meaning, a maximum of 2 tiles per turn. The siege road, however, if built well (leapfrogging workers and so on) can allow units catching up to move at triple-speed up to the main force. Instead of attacking with 35 and then 35 later, you could attack with 50 on your initial strike, even though your force left with only 35! It may take some practice to learn the timing of sending out your “siege road” and how long you can keep funneling reinforcements down it, but once mastered, siege roads can make your original attack – and later reinforcements and counterattacks! – much more successful. One warning, however – make sure to avoid opposing sentinel nets when building siege roads, because the disadvantage to a siege road is that your enemy can use it, too!

[AlisonD]

Great article , since I usually have difficulties to decide, what to do in what priority. Well yeah, I tend to irrigate normal grassland

There are two things, that are not clear to me:

"Tundra on its own is useless – 1 food, 0 shields. Mined, it becomes 1/1, and even with a railroad it is still only a 1 food 2 shield tile… so why bother mining it at all? Planting a forest makes it a 1 food 2 shield tile to begin with."

So, what does that mean? When I plant a forest on tundra first and chop it then (basically it looks the same like before), but is it that I have this tile permanently turned in a 1food 2 shield tile (with railroad) for the whole game?
And the other question is, that I think having read somewhere, when planting wood and chop it then, I get the shields bonus only the first time I chop a forest, but not anymore, when doing the same procedure again on that tile?

And last question: Can someone explain to me what is the movement cost (or movement penalty), that is refered to in many charts or articles. Not sure, if it only relies to troups or to workers too. I can't find anything to this subject.

Keep up writing these good articles. I have read a lot and learned a lot with them.

AlisonD

[Swissy]

The lack of workers is probably the biggest noobie mistake. I smile everytime I see somebody with a pop 2 city on turn one. Especially true now that a pop 2 city can't produce a Jag in one turn.

One thing I'd add is to clear marshes before jungles. It takes less turns and I've found a higher percentage of bonus grassland is revealed. Don't know if this is a design feature or my luck.

Also, don't forget to irrigate late in the game if you get to change govs. Many games decided by a few points go to the player with more population. And, join those workers to cities on the last turn.

[TheBrain]

I have to say thanks to Fried for all this articles.
His strategy articles helped me alot when i was a noob and im sure it helped a lot others too.
Also veteran players can get something new out of most of his strategy guides.

about this one i have to say that i prefer to build roads first in some cases:
At start i prefer to build 2 roads first so that i can build my 2nd city faster and so that my 2 citys are connectect. ( to prevent jag rush) I always build roads before i build a new city.

I also like to clean forest to build temples faster (it works the best with egypt fast worker and fast temples )

I like the siege road tactic but sometimes its also nice to build it make your enemy think u will attack there and then just attack somewhere else with fast horses or boats.
another nice effect of the siege road is that once you take an enemy city its allready connected with your citys and its more easy to defend it later ( if u need to)

[Ktulu]

You must teach me your tricks someday!

[BoNeHeD]

I love this read! ;D Fried you must do a book for the rest of the world someday.
I like cutting forests on river plains then irrigate.I try to time the 10 shields to build an improvement and not waste them for a spear. unless its on auto harvest)
draining marshland yields bonus grassland mostly(fertilized) mostly from what I've seen so far. Bad land can be overcome.Stacks(chain-gangs) of workers can clear/irrigate or mine/road fairly easy even jungles! Too bad harvesting jungles didn't yield some bonus(cutting down rain forest) ;D
8-)Seige roads are a must for my legions.(LEGIONS RULE!)

[Dr Shot]

Cut the trees down,takes time, lose the two shield production to one or none.

(then)Build Mine or irragate,takes time. The 10 shields gained by KILLING the forest can be gained a few more times over by not killing the forest. Sure this depends on what other terrain is around and what goals you have;growth,shileds ,pop boost ,killing/armies/units/buildings
I would rather have forests than cut them, unless you are in a bind for good land. One must adapt to the surroundings....Hey, that really is nice land you have there, wish I had some like it....hmmmm ;D

[Armageddon1]

35+35=70, not 50

[MiloMind]

Nice work! Well thot out & presented.
I too luv those workers. Main reason i prefer the industrious civ's. This great game is as much about BUILDING as it is destroying.
Your "siege road" use is so important too. An army always needs some "combat engineers" to effectively & efficiently fight. Plus it speeds the return of damaged units to get repaired; especially those valuable retreated fast units that are down to their red damage zone.
(BTW: Arma... his math is a guessimate of how many units would be built & able to get in on the FIRST wave attack + it would cost a few combat units to build adequate # of workers)

Nice points on clearing marsh vs jungle from U other guys too. Never thot about it that way.

The colony thing IS a good idea that i'v not had the guts to try. But U raise a good case for it, especially if one is playing at the higher levels of difficulty or government, since your always unhappy citizens cry for some luxury items.

As for the road or improvement first: gotta go with improve 1st. Used to road, then improve. Since reading the "Master's" same thread over on Apolyton, i'v tried it & switched. Now the exception would be, as was cited, if U need to connect that new city asap for military &/or sharing of those wonderful resources (since 1 of each satisfies EVERY city... IF they are connected).

One reminder i mite add: Don't OVER do it. Nothing is a bigger waste than to more squares improved than a city can regularily use (ie, constantly shipping out settlers or workers).

Another hint for SP: During war, use workers as bait for the AI to come & capture them. Thereby wasting a combat unit's turn & getting itself out in the open where U can nail it & get your worker rite back. The AI is really stupid for capturing workers... REALLY stupid.

[rosco74]

Great thread,

Sofar a couple of things I didnt know about workers and despotism , the more I play the more I hate this form of government, the only thing its good for is killing population to hurry spear in a city under seige or hurry rush temples in fast growing citites or in cities to spread sphere of influence to gain access to a resource,
so my idea is there a possibility to design a scenario that has a monarchy start for all civs?
(just to reduce corruption), abit like a med start or QC but everything else is set to start from the beginning??

maybe we could call it a rosco..like everyone saying lets play a rosco game:LOL (i just want to be immortalised like cton;)

And something else im working on was I had a tendancy to build workers and then find them improving out lying cities only to have my first big 4 cities ( the first 4 that I usually build) slumping later in the game, so to combat this I have started making that I have always at least 1 worker 2(optimal) improving every city and I stack groups of 3 workers and have them as mega road builders so they build 1 road per turn meaning I can get all my cities linked quite quikly resources pouring in and i can defend sieged cities quicker and easier.

[Trip]

Most players definitely don't build as many Workers as they ought to. Between that and loose city spacing (less common in MP... for those who live anyway) it can cost a great amount of economic potential from most empires. Having a lot of tiles improved and close city spacing also allows the sharing of tiles (less viable in an MP game, but still possible to MM early on with only a few cities and not a huge number of units).

Economics! ;D

[RedPhoenix]

Trip said:

Most players definitely don't build as many Workers as they ought to. Between that and loose city spacing (less common in MP... for those who live anyway) it can cost a great amount of economic potential from most empires. Having a lot of tiles improved and close city spacing also allows the sharing of tiles (less viable in an MP game, but still possible to MM early on with only a few cities and not a huge number of units).

Economics! ;D

Isn't sharing tiles a cheat, or at the least bug abusing? It is clearly not how the game was intended either way.

[Trip]

RedPhoenix said:

Isn't sharing tiles a cheat, or at the least bug abusing? It is clearly not how the game was intended either way.

I don't mean sharing as in using the same tile twice (or more) in one turn. What I'm referring to is switching city labor placement between cities on different turns. For example, if you have 2 cities, both with Granaries, and a irrigated cow (4 food), you can give the cow to 1 city for 2 turns (8 food) and then switch off (final 2 food) and give it to the other city that turn, keeping wasted food to a minimum and concurrently helping other cities grow/increase production.

[RedPhoenix]

Trip said:

I don't mean sharing as in using the same tile twice (or more) in one turn. What I'm referring to is switching city labor placement between cities on different turns. For example, if you have 2 cities, both with Granaries, and a irrigated cow (4 food), you can give the cow to 1 city for 2 turns (8 food) and then switch off (final 2 food) and give it to the other city that turn, keeping wasted food to a minimum and concurrently helping other cities grow/increase production.

I do that sometimes, but really that kind of level of micromanagement in general is a bit too much for me gets way too boring