This is a thread for those who love micromanaging, for those who don't mind spending 5 minutes working something out just to gain one more beaker, for perfectionist and arguably very anal civ players! This is a place to share all the extreme micromanagement that we use in our games.
The spiritual trait in itself makes quite a few more possible, but here are some basic tricks that i use in every game with every civ. Please feel free to add your own to the list.
Binary science rate
I always keep my science rate either at 0% or at maximum to avoid wasting fractions with bonuses. Here's the reasoning.
Imagine you have 10 commerce in your city. The city has 25% bonus on science with a library and also 25% on gold with a market. Let's say you want a 70% research rate. These numbers are just for the sake of the example; you can use any other numbers you wish. The effect of doing this micromanagement is almost always positive and never negative no matter the numbers used.
By setting science at 70%, you're getting 7 beakers and 3 gold. You get 25% more beakers, and 25% of 7 (rounded down) is 1. You get no extra gold, since 25% of 3 (rounded down) is 0. You're getting a total of 8 beakers and 3 gold per turn. After 10 turns, you'll have 80 beakers and 30 gold.
What you should do instead is run at 100% science 70% of the time, and at 0% science 30% of the time. Then when running science (for 7 out of 10 turns), you'll get 10 beakers. You'll also get 2 extra beakers (25% rounded down). This for a total of 12 beakers. On the turns you run 0% research, you'll get 10 + 2 = 12 gold. After 10 turns, you'll have 12 * 7 = 84 beakers, and 12 * 3 = 36 gold. This is 4 more beakers and 6 more gold than by doing the easy but inefficient 70% science rate all the time.
There's another advantage to doing this. By making sure you run at 0% science for enough turns to be able to afford 100% science for the duration of your entire research for a given tech, you won't get the tech later, but you will give the chance to the AI of getting the tech before you really start researching. The more AIs have the tech, the better bonus you get on your research.
Also, doing this on more than one tech, when the intermediate techs aren't immediately needed (and making sure you have enough accumulated gold for researching all the techs in the chain), you can get the intermediate techs later, which means that the AIs will get less research bonus on them since on less player knows those techs. This can be the difference between being the first civ to reach a critical tech at the end of the chain or coming in second and missing the one-time bonus.
Of course, by doing this, you'll often have large sums of money at hand. This comes with its own set of advantages and disadvantages. Mainly the AI is more likely to demand tribute when it sees you have lots of money in treasury (bad) but you have the ability to rush buy units in a pinch when in dire need (good).
Another advantage of this method is the added flexibility in choosing which tech to research. After running on 0% research for a few turns, you can change your mind about what tech to spend the extra gold on, if your priorities have changed in the mean time. Then you won't have lost any time researching the wrong tech.
Notice that specialists hurt this strategy, especially if you get one free specialist per city and he produces science or gold. This is because the specialist can make gold when running at 100% science, or can make science when running at 0% science. This causes potential wasted fractions for both science and gold in that city, instead of just one kind. However, the strategy always remains useful, even though the gain can be smaller.
There is an important note that i need to add after it was mentioned further down on this thread (thanks guys). On the last turn of researching a tech, you can exploit a bug in how carryover is calculated to your next tech. The gain from exploiting this will usually be larger than the losses from wasted fractions by going to a non-binary research rate on this one turn. Information on the bug can be found here : Technology research explained. Strategy to exploit it can be found here : Tech jumping. This bug has been corrected in patch 1.61.
Before you get your first building that gives a bonus to science or commerce (usually a library, but could be a monastery, a market or even an academy), binary science is of course useless in terms of fractional losses. It's still useful for having more money in the bank for emergencies, and more bonuses for other civs knowing the tech you research. However, these benefits are of limited value. Until you meet other civs (which on maps like archipelago could take a while), here's what you should do instead. Play around with the science slider, and for each setting check your financial advisor to see what the total research is. Pick a setting that gives a multiple of five. If you can't get a multiple of five, then 5n+1 is your best shot, followed by 5n+2. This way, you'll maximize the benefit of prerequisite techs, which give 20% bonus each towards research of the current tech. By picking a multiple of 5, you minimize the fractional loss on this bonus. This doesn't apply to the starting techs, which have no prerequisites. This doesn't apply either when you start meeting civs who might have the tech you're researching, because they add their own bonus, so the total bonus you get may not be a multiple of 20% anymore, and therefore aiming for a multiple of 5 becomes useless.
Commerce and production in packs of 4
I keep my commerce and production per city at multiples of 4 whenever possible to get the most out of 25% and 75% bonuses. If i can't have a multiple of 4, at 75% bonus i aim for 4n+3, if not then 4n+2, and i avoid at all cost 4n+1. At 25% bonus 4n (ideal) > 4n+1 > 4n+2 > 4n+3 (avoid at all cost). If the bonus is 50% or 150%, i aim for a multiple of two (i.e. an even number) in basic commerce/production. If, because of monasteries, the bonus is 45% or something else that's weird to calculate, i try different combinations and choose the one that seems to give the best ratio of total commerce/production to basic commerce/production. At 35%, i just try to get a multiple of 3, minus one for every 21 (i.e. i aim for 3, 6, 9, 12, 15, 18, 20, 23, 26, 29, 32, 35, 38, 40, 43, etc.), the absolute best being multiples of 20, and the best after that being the numbers right above those, not below.
This implies switching worked tiles whenever there is an overflow that's not a multiple of 4 (for production), or whenever a cottage/etc. upgrades to another level. This can mean switching between cottages and mines, or cottages and forested grasslands, or cottages and hamlets, or river cottages and non-river cottages, etc.
This tile switching can be combined with the binary science rate by using some other tricks to improve results even more. For example, in my science cities i try to use less advanced cottages at 0% science, so that i can grow them without any penalty to my science output. On the other hand, in my cities without as many science bonuses (e.g. no academy), or with more gold bonuses (e.g. market in a holy city), i'll do the opposite and try to use the most advanced cottages while at 0% and grow them while at 100%.
I'll admit to not always doing this perfectly for commerce as it gets a bit tedious and can't always be switched efficiently, but i always check production carefully when i have any kind of bonus.
Keep on working while moving
When needing workers to move their full movement or more, this is what i do to avoid wasting worker turns. I move workers 3 tiles on roads, give them an order and then cancel the order. Next turn, i keep on moving to where i actually need to be. Here's why this is helpful.
Let's say that you need your worker to move 8 tiles through roads for his next task. You can take 2 full turns to move the 8 tiles. Then on his 3rd turn your can make him do his improvement. On the first two turns, he won't have done any improvements at all. A worker turn where all he does is move, if it can be avoided, i consider a waste. Exceptions include moving to forests or hills with no roads, where you can't avoid it.
I would micromanage it this way. On the first turn, i will make him move 3 tiles. He still has 1/2 movement left. Then i ask him to make an improvement there. It could be anything i know i'll need there in the future. Then i cancel the order. Part of the improvement will already be done for later when i come back to it. For example, if i ask for a cottage (4 turns) and cancel, when i come back to it, there will only be 3 turns left to do. Therefore the turn hasn't been wasted.
I'll do the same on turn 2. On turn 3, i'll move the remaining 2 tiles and do the actual improvement that i wanted there. It starts on turn 3 either way, so nothing has been wasted, but improvements on the 2 intermediate tiles have already begun, so i've gained 2 worker turns doing this.
An alternative method is moving to a tile where another worker is making an improvement, then start making the same improvement. Take the worker who was there initially and move him further than the other worker could have reached from his starting position.
Chop, prechop, pre-improve
I don't think i need to explain settler chops anymore, this has been done to death already. This allows for making settlers while still letting your city grow, by producing settlers entirely out of trees. Of course, i use the variant where you switch production in the middle of a turn for maximum efficiency.
Also, whenever you chop (doesn't have to be settlers, this applies to everything), there are some things to keep in mind. First, make sure that your total production for the turn, including chops, doesn't exceed the needs of the current build plus another identical build. For example, when making a swordsman (cost 40 hammers), and he's already at 30/40, you don't want to produce more than 10 (the remaining production needed for the build) + 40 (the production needed for another swordsman). This is because overflow is capped at the current build's cost. For example, when producing something that costs 40 hammers, your total overflow can't exceed 40. To avoid this, you can either switch what the city produces at the start of your turn, or you can stop the chop and finish it later.
In fact, prechopping is a good idea a lot of the time. The best example is when you know you'll get the technology for a critical wonder in a few turns, or a technology that lets you make superior military units that you want to use for an offensive, or any other technology that opens up a new option that requires a lot of production quickly. In such a case, you may want to chop each forest for two turns, then stop and build a road. The forest is still there but almost completely chopped. Whenever the technology arrives, send your workers back to the forests and complete the chops instantly. This trick sometimes allows you to completely build a wonder on the first turn it's available, without using a Great Engineer!
Prechopping is also good if you ever find yourself with workers that have nothing to do anymore. In this case, rather than simply chopping, it's often better to order the building of an improvement over a forest. This gives the idle worker more turns of work to keep himself busy. For example, a forest takes 3 turns to chop, while building a cottage requires 4 turns. Ordering a cottage over a forest thus requires 7 turns, while leaving the forest up even after 6 turns of work. Another alternative is to pre-build a cottage over a farm that you know you'll want transformed later, or any other improvement switch where you still want the current improvement for now but know that you'll want a different one later. This way, whenever you're ready for the other improvement, you can have it right away. Just remember that on hills and in forests, when prebuilding, you should always put down a road so that it's easier to come back later to finish the job.
I prebuild an axeman, then before i finish it, i prebuild a swordsman, then before i finish it, i prebuild a spearman. Then i finish them all in the order i started them. Because i finish the first two units later, i'll save on maintenance during the turns where they sit in the build queue, since units in the queue don't cost maintenance, but finished units do. This of course assumes that i don't need the first two units immediately. It also has the convenience of being able to finish a unit or two in one turn each in times of crisis because they're already almost finished. Just be careful not to leave a unit in the queue for more than 10 turns or you'll start losing production on it.
Alternately, you can prebuild a unit for this city's defense, then make a building that takes 10 or less turns to finish while the unit is in queue. This is just as good as actually having the unit defending the city because you can switch back and have it in one turn at the first sight of trouble, but saves you the maintenance cost while you're making your building.
Whip 'til your hands bleed
Pop rushing is way overpowered in this game, and should be abused to the max if you intend to master this game. Part of the reason why it's so overpowered is because of a bad calculation that they haven't even bothered to fix in patch 1.61.
With pop rushing at normal speed, you're supposed to get 30 hammers for every pop spent. That's already pretty good, considering that with a granary, a city only needs about 11 (at level 1) to 31 (at level 21) food to grow back the pop you've spent. Spending 11 food to get 30 hammers is already mighty good, but because of the exploit, it gets even better.
Assuming the building or unit is already started (so you don't get a penalty), what the game does is it checks to find how many base hammers you need to complete the build, and charges you an amount of pop based on this. Let's say there is a forge in your city, or you're using organized religion. This means you have a 25% bonus on production. With 30 base hammers, you then get 38 total hammers (small rounding error, should have been 37). If what you need to complete the build is 38 hammers, the game will only charge you one pop.
However, the game always gives you an amount of hammers that's the smallest multiple of 30 needed to complete the build. This is total hammers, not base hammers. So in the case above, you'll get 60 hammers because 30 wouldn't be enough to complete it. Congratulations, you've just received 60 hammers by expending only 11 food (assuming a size 2 city going down to size 1). That's a grand total of 5.45 hammers per food spent! Tell me that isn't overpowered. Even taking into account that normal hammers would receive a normal 25% bonus, you still need 4.36 normal hammers to equal every food used. No improvement anywhere in the game can provide 4.36 hammers to compete with the 1 food a farm provides.
Pop rushing is so overpowered in this game that i find slavery to be by far the best civic, all branches included. I abuse it to such an extent that in a typical game, until gold rushing comes along, about 70% to 80% of all my production is obtained through this mean, and only about 20% to 30% from actual hammers obtained from tiles. It's so overpowered that a city full of farms becomes a better producer than one full of mines!
The best ways to abuse the system are using 1 pop for 60 hammers when needing 31 to 38 with a 25% bonus (getting 30 hammers for free because of the bug), or using 5 pop to get 210 hammers when needing 181 to 187 with a 25% bonus (getting 60 hammers for free). However, it's good at all levels, as long as you make sure to be at a point that will provide more hammers than you spend pop for.
Because of the unhappiness penalty involved, you don't want to whip much more than once every 10 turns (so the unhappiness has time to wear off). However, because it's so overpowered, you definitely want to use it as soon as the unhappiness penalty wares off, if not earlier. You don't want to waste a single turn that's not being spent on making unhappiness disappear. Yes, this means that you want to be in a constant state of unhappiness due to whipping!
For cities with low food, it will take them time to grow back the population, but little time to get rid of the unhappiness. For those, you ideally want to whip a single pop every 10 turns, and no more. This increases the bonus obtained from the bug, while minimizing your pop spent. Time your production and use your queue to make sure that you have a building that will have just the right amount of production done by the time unhappiness falls down to zero. With a 25% bonus, this means you should have something that needs 31 to 38 hammers to finish. With a 50% bonus, that's 31 to 45 hammers.
For cities that have high food outputs, growing is not a problem, and unhappiness quickly becomes huge. In this case, you want to whip as many pop points as possible every time you whip. This means you should have buildings barely started waiting to be whipped. Those buildings should have somewhere between 1 and 30 hammers done on them, depending on the total hammers required, to ensure the biggest gain possible as described above. For example, a bank requires 200 hammers. Therefore with a 25% bonus you want to whip 5 pop to gain 210 hammers. This can be done by having 181 to 187 hammers left to the build. Therefore you should let normal production of the bank reach 13 to 19 hammers, then put it in the queue until you can whip the rest. Actually, due to a rounding error this will work at 12 hammers as well.
Another thing you want to keep in mind is when to rush. The ideal time is usually not on the very turn where unhappiness disappears, but before that. Ideally, you want to be as close to that point as possible, to keep unhappiness at a minimum and be able to use as many productive tiles as possible for as long as possible. However, there is another factor to consider, one that becomes more significant the more pop you whip in one swoop.
You see, the cost of whipping depends on your city size. Assuming a granary, whipping one pop costs 1 more food for every extra pop your city has before the whip, because it will cost one more food, for example, to go from size 4 to size 5 than it would cost to go from size 3 to size 4. As you whip more pop at a time, however, the effect gets bigger. Whipping 2 pop will cost 2 more food per extra initial level, because each of the two levels you need to grow back will cost 1 extra food. Similarly, whipping 5 pop will cost 5 extra food per initial level, i.e. it will cost 5 more food at level 11 than at level 10, and 10 more food at level 12 than at level 10. This really adds up.
So you want to be at high levels as long as possible to use more tiles, but you want to whip at the lowest level possible. Solution : whip at the end of a level. Use you level to the max, wait until you're right about to jump to the next level, and whip right before this happens, rather than right after it does. Ideally, what you want to have is the following. The last turn before whipping, you want to make just enough food to reach the next level, and not one more. Set you governor to "avoid growth". Then the next turn your city will be at 30/30 food or something like that. Set the governor back to normal, whip and witness your city now at 30/26 or something like that (which is nice because now you know the first two increases will come quickly), and let it grow again.
Note that sometimes it may be better to whip right after growth rather than right before. This depends on too many factors for me to provide precise guidelines, but here are a few pointers. By whipping after growth, you get at least one turn (the one where you click on the whip button) at one size higher. If you whipped more than one pop, you may also (but there's no guarantee) grow some of the levels back one turn earlier. Therefore, by doing this, you're trading away food in exchange for whatever the tile gives you (don't forget to count -2 food for what the tile gives you, because it needs to feed a citizen). If all you intend to get from the tile is 1 food (a grassland farm), then it's never worth it to whip after growth (you'll save more food by whipping before growth than you could gain by whipping after it). If it's production you want from that tile, again it's not worth it, because the food saved otherwise can be turned into lots of more through whipping later on, than what you'd get from the tile. If what you want is commerce, than it might be worth it, since there's no easy way otherwise to turn food into commerce, especially if you're already working all the cottages you can.
Whipping at the middle of a level is never a good thing. Usually it's better at the end of a level, occasionally it may be better at the beginning of a level if you want commerce, but there's never any point in doing it somewhere in the middle. Note that if whipping just after growth, the use of the city governor is still recommended if the growth will lead you into unhappiness. What you want to do then is grow your granary to the max with "avoid growth" turned on. Next turn, turn it off and grow as much as you can. This leads into unhappiness but it has no impact since the instant you get there, you'll whip away the malcontents.
When to do this and when to let your city grow instead? Well, the gist of it is that if in the time it takes for your whipping related unhappiness to drop down to zero, you could grow an extra size and fill that granary again (or at least come close to filling it), then you should do that; otherwise you should whip at this level and not wait for unhappiness to subside. By going down in pop, happiness won't be a problem for a while anyway. Another thing to keep in mind is that it's sometimes preferable to wait until you're at max happiness before whipping; this maximizes usage of tiles at the cost of less whipping (i.e. less abuse). There are also other factors to consider, like whether or not you've got extra 3-food or 2-food tiles you could use, but i'll let you find that out through your own experimentation.
One last thing. Whenever you've got a big overflow from whipping (around 30 hammers at normal speed), it's often a good idea to use that overflow on a worker or settler. This way, you're not wasting your time making them the normal way and getting only 1 hammer per food. Also, workers and settlers make good candidates for the whip for the same reason.
Just how efficient is this tactic? Further in this thread, for one particular example, it was calculated that doing this was 16% more efficient in terms of hammer and commerce than simply using hills and cottages without whipping. The particular example included the following :
- city of size 8 (bigger = less efficient whipping, smaller = more efficient)
- 5 excess food (more food = more efficient whipping, less food = less efficient)
- 2 grassland hills and 2 plains hills available (better tiles = less efficient whipping, worse tiles = more efficient)
- whipping 3 pop at a time (more at a time = less efficient whipping, less at a time = more efficient)
For a nice spreadsheet showing the points that provide more hammers than you paid for at each game speed, see this file graciously provided by Malekithe.
You can finish all worker actions and all other queued actions by pressing Ctrl-A.
If you don't mind re-issuing every worker's task every turn, you may want to use Alt-Escape on one worker at the end of every turn. This will cancel the current task of all workers so that you can re-issue a task next turn. This works well with chopping to ensure that you can switch production to whatever you want to rush before the chop comes in.
When changing from serfdom to something else, make sure to do it after all your workers have done their actions. When switching from organized religion to something else, make sure to wait until after all your chops for the turn have come in, so that they still get the bonus.
If you have units on a tile that need to move, and workers making a road on that tile, make sure to order the completion of the road before you move the units out. Of course, this only matters if they can be moved out to another tile with roads. Obviously, the same thing applies to the tiles they're moving to, not only from.
I won't even talk about assigning workers manually, assigning citizens manually and switching them around whenever needed, assigning specialists manually and switching them as needed, moving each military unit manually every turn, and so on. I consider these as basic micromanagement that everybody, micromanager or not, should use. These are not, at least in my book, cases of extreme micromanagement!
Anything else that i've forgotten or simply don't know about? Please add any extreme micromanagement trick that you use that's not listed here!