 |
|  |
 |
|
UnityScoutChopper
|
 |
Brno, Czech Republic
Oct 2000 time: 06:31
|
|
Subtle inside joke for those who played the 1980's Star Wars videogame: "We've lost our shields!!"
In the beginning of the game, there is ample room for micromanagement without any real boredom, and in fact, it can be quite an interesting "game alongside the game." I'd like to take a look in this thread at some of the most common early-game food/shield "best counts" (counts achieved when a city is using the terrain that provides the most FOPs*, and problems or "problems" related to them. Only shields will be considered, as just this subject alone is already complex enough.
Cases where shield counts rise in mid-build/mid-growth will be mostly ignored here, as simple cases are complex enough. Only the early game will be considered, and only common food/shield counts.
An implicit guideline in the below is the "10-turn guideline": where no strategic consideration is involved (of course, one generally is), a city should try to get something built in 10 turns.
SHIELDS
1 shield - An uncommon shield count in the very beginning, this becomes more common once (if) you start REXing beyond the range of your workers. Horribly low, but no shield-count micromanagement problem here, except for the issue of a city that reaches 11 shields on turn 10 (due to growth) and thus must choose between a more expensive build and a waste of 1 shield (10%, which is low compared to many cases below). Especially if the city also has one food or if the next square will not increase the shield count, the best builds (all else being equal) for such cities are warriors or workers (and considering that such cities are often unconnected border cities, both are actually fairly appropriate builds even without micromanagement). For religious civs, a rushed temple also works. Basically, you want to get at least SOMETHING out of this city in reasonable time.
2 shields - will go into any multiple of ten, though your shield count will generally rise, possibly causing shield wastage, before you build anything more expensive than 10 shields.
3 shields - the first really interesting case. 10-shield items get built with 20% wastage in a 3-shield city. Often, even when you really need a 10-shielder, you can micromanage the city (if you can remember...) to trade one or both of those 2 extra shields for food or gold. Otherwise it's on to the next rank, 20-shield items, which get built with 5% wastage, not really all that bad. If you're obsessed with avoiding shield wastage and the city will be at 3 shields for a whole 30 shields (unlikely; more common is locking into a multiple-of-3 groove and 3-shields production at the same time), you can build 30-shield items, but this generally means overobsessing with avoiding wastage, as the only such items are horsemen (you should be building chariots instead if you're still in the micromanagement stage), swordsmen (you should be building warriors), religious temples (if they'll bring an advantage, fine, and at 30 shields they often will, but building them just to avoid wastage is silly), and settlers (you should be building them from your granary cities unless a non-granary city is outgrowing your tile development, and if you're outgrowing your tile development, a new settler will often only worsen the problem unless you can build a worker elsewhere to compensate).
A special case here is the ultra-early 3-shield capitol: do you build an archer/spearman to avoid that painful 20% wastage, or a warrior/scout because they are more useful in this phase? I agonize over this probably more than I should.
4 shields - generally the result of shield waste or not having access to two 2-shield squares, yet still a lovely count. Goes perfectly into 20-shield items (generally even before the count rises) and fairly well into 30-shield items (6.7% wastage), though the shield count may rise inbetween. Goes perfectly into 40-shield items (though these are rare in the beginning, and the shield count will probably rise inbetween).
5 shields - the king of early counts: a high count that goes perfectly into everything, including the hard-to-fit 10 shields needed for the highly useful warrior, scout, and worker units.
6 shields - a painful fit for 10 shields. A hideous fit for 20 shields (same percentagewise as for 10 shields, but 20 shields is so easy to fit otherwise that it hurts more emotionally). A perfect fits for 30 shields. Not counting variations in shield count and settler builds, 6 shields is a level that is best saved for after you've got HBR or hooked-up iron.
7 shields - terribly wasteful for everything but 20-shield items, 40-shield items, and really expensive items.
8 shields - like 6 shields, but more so, with the difference that it is viable for building really expensive items in reasonable time.
9 shields - painful wastage for everything less than about 60 shields -- this count is best left avoided, raised, lowered, or used for expensive builds.
10 shields - no comment.
---------
Since an average granaryless city (counting high-food cities) will grow in 8-9 turns, since workers will often raise shield counts in mid-build, and since a city may change squares worked to avoid food wastage when reaching the next size in that city or a neighboring city, most builds longer than 4 turns will see changed shield counts in mid-build, but still, there are enough constant-shield cases that the above considerations are often important.
Because lower counts are generally harder to fit well (each wasted shield is a higher % of wastage, at the very least), one guideline -- all else being equal! -- is to set each city to building a 10-shielder, raise it to 20 if there'll be considerable wastage, and later also raise it to 30 if there'll be considerable wastage.
Enough for now!
USC
* Factors of Production. For the purposes of this thread, we will count 1 post-waste shield as one FOP, 1 food as one FOP, and one post-corruption gold as 1/4 FOP. Gold is a bit fuzzy here, as it could end up wasted on 40-turn science or, when using the luxury slider, wasted on luxuries; meanwhile, it may take a long time (depending on level, map parameters, etc.) to reach a gold-rushing government, and until you do, gold's value will vary from worthless (no spending options) to 1/2 FOP (upgrades) to nearly-unquantifiable (ROPs, maps, techs, etc.)
|
|
|  |
 |
|
Arrian
|
 |
Kneel before Grog!
Jul 2001 time: 00:31
|
|
I will try and adjust production so as not to waste too many shields (a 9-shield/turn city building warriors, for instance... eek). But no matter how hard you try, there is gonna be waste.
If you're willing to really get into micromanagement, you have to take city growth (and corresponding rise in shield output) and worker jobs into account. And that means figuring out exactly what happens when cities grow and new citizens are allocated:
-A city with high food intake will often autoallocate a new citizen to a low-food/high-shield tile (forest)
-When the city grows, the new citizen's shield production counts toward THAT turn. So, with a little luck, your high-food cities will get more shields than you would normally think, and still keep up their growth rate, since you can then move the citizen back to a high-food tile.
I don't often get into that level of detail in SP, and even if I do, it's just with the capitol. But in the demogames, for instance... well, you've got the time, you know?
-Arrian
|
|
|  |
 |
|
UnityScoutChopper
|
 |
Brno, Czech Republic
Oct 2000 time: 06:31
|
|
quote: But no matter how hard you try, there is gonna be waste. |
Trying to eliminate wastage would itself inevitably be wasteful even if it were possible -- for example building a barracks in an 8-shield town when what you really need there is a settler isn't too smart, and of course with no 90-shield buildings, you can't avoid waste with a 9-shield town. The idea is just to reduce it when reasonable.
quote: If you're willing to really get into micromanagement, you have to take city growth (and corresponding rise in shield output) and worker jobs into account. And that means figuring out exactly what happens when cities grow and new citizens are allocated:
-A city with high food intake will often autoallocate a new citizen to a low-food/high-shield tile (forest)
-When the city grows, the new citizen's shield production counts toward THAT turn. So, with a little luck, your high-food cities will get more shields than you would normally think, and still keep up their growth rate, since you can then move the citizen back to a high-food tile. |
There are too many special cases like this one involved once you move beyond the basic "x turns at a constant shields/turn", which is why I tried not to go into that much in what I wrote. But yeah, this stuff is important -- now what REALLY always gets me is the "grow, expand, entertain" bug, where AFAICT an auto-slapped entertainer gobbles that extra production you mentioned when you don't have a "spare" point of happiness in a town and its growth coincides with border growth (a nearly nonexistent phenomenon outside of each game's 10th turn, but then that's a very important turn).
USC
|
|
|  |
 |
|
Dominae
|
|
Interesting topic.
The main thing to keep in mind when counting Shields is that your cities grow, and this growth is better long-term than a "perfect fit" for Shields and production items. It's better to waste 6 Shields building a Horsemen in a 9spt city once, than to slow down its growth on the way to 10spt.
With this in mind, my heuristic is: grow cities as quickly as possible and as large as they will go under the constraints of happiness (and sometimes available tiles). While cities are growing, frequent micromanagement helps avoid waste by splitting, for example, a 30 Shield item into 2*7+2*8. This is the main reason why I like overlap between city radii (3-tile spacing or closer). Once the cities grow to their (temporary) maximum size, I then plan to slow their growth by working Plains with Mines, Hills, or Mountains, or simply by creating a Specialist (usually Scientists for 40-turn research).
Remember that, although a 5spt city does not waste Shields on a 20 Shield build, a 6spt city is usually better because you can transfer those extra Shields to some other city if your city spacing is close enough. So in general I do not agree with artifically halting growth just to avoid Shield waste.
Concerning the 3spt capital: yes, I obsess over this a lot too! One trick I've come to use is creating a "Commerce tile" that is only used when there is 1 Shield remaining in a build. If I've got a lake nearby, great, if not I'll put a Road on a Bonus Grassland by a River and then move on. This way, you can finish a 10-Shield unit wihtout wasting Shields, and get an extra couple Commerce in the process. I'll even use Coast (without Harbors if my Food output is high enough) to this end.
There's a lot to talk about here, but it's all theory and therefore rather bland. What would be great is a few examples from real games (sort of what we've tried to do with AUSG101).
Dominae
|
|
|  |
 |
|
Arrian
|
 |
Kneel before Grog!
Jul 2001 time: 00:31
|
|
I've got a great example, except it's "state secrets" from the PTW Demogame II. Or, for that matter, PTW Demogame I... IIRC, there was some pretty serious micromanagement of Eye of the Storm (not to mention other cities) for, oh, like most of the game. To the point where it was beyond me.
Dominae makes a good point re: city spacing... yet another reason why the spacing I hate works so well - at least early on when it counts the most. *sigh* I've gotten used to 4-tile, but though I've used it, I really don't like 3-tile.
Another thing I've been working on lately is trying to avoid wasting worker turns whenever possible. In the past, I wouldn't have batted an eyelash at ordering 2 non-industrious workers to both road a tile of flatland (1 such worker needs 3 turns, the two together do it in 2... but it's inefficient). I've been trying to avoid stuff like that, particularly early on. I haven't been playing industrious civs much, so I need to use my slowass workers the best way I can. 
-Arrian
|
|
|  |
 |
|
Dominae
|
|
quote: Originally posted by Arrian
Another thing I've been working on lately is trying to avoid wasting worker turns whenever possible. In the past, I wouldn't have batted an eyelash at ordering 2 non-industrious workers to both road a tile of flatland. |
Unlike counting Shields, I think that Worker actions is a bit of micromanagement that becomes second nature if you do it enough. I'm not much bothered by ordering every single Worker until well into the Medieval era, not only because I know it's important but because I spend very little time thinking about it.
Unfortunately, this will change with Conquests, because Industrious Workers are suddenly slower. This throws off a lot of math that I've come to rely on.
Dominae
|
|
|  |
 |
|
Dominae
|
|
quote: Originally posted by Jaybe
When they become available, teams of 1 Worker + 1 Captured Worker work real well for mines and roads on grass/plains. |
Industrious civ putting a Road on flat land (Plains, Grassland, etc.), from least efficient to most efficient:
1 slave Worker
1 native Worker
1 slave Worker, 1 native Worker
2 native Workers
But notice that this list is also from most time-consuming to least time-consuming. So it's a judgement call whether or not you really need to send those extra Workers to Road that tile. If you're keeping up or ahead on tile improvements, it's always better to just send one slave Worker to do the job.
Dominae
|
|
|  |
 |
|  |
 |
|  |
 |
|
UnityScoutChopper
|
 |
Brno, Czech Republic
Oct 2000 time: 06:31
|
|
Arrian wrote:
quote: I play with the happiness governor off, so I don't get autoplaced entertainers. |
I play with him off, too, but this always happens to me anyway in cases where growth coincides with expansion of cultural borders. Maybe this was fixed in PTW?
quote: I often forget, and find myself loading up the autosave (unlike reloading for better RNG results, I have no problem with fixing something I genuinely meant to do, but forgot) - at least in the crucial early game. |
This is a bit OT, but... I personally feel guilty doing that. In fact, I kept having this problem where I would load from autosave to negate that type of disorder, and then I would sooner or later have such a strong feeling that the game was "invalidated" that I started up another. I restart a lot anyway due to the "lure of the start", even when my games are going well, and thus it gradually became ridiculously uncommon for my games (huge/16) to last longer than, say, 1000 BC. I thus turned off AutoSave and, while I find myself gritting my teeth a lot, I'm moving farther into the game and, as a bonus, am learning a la sink or swim to avoid disorder... which, considering I have ADD, is a bit of a feat.
Oh, the joys of neurotic gaming! :-)
Dominae wrote:
quote: The main thing to keep in mind when counting Shields is that your cities grow, and this growth is better long-term than a "perfect fit" for Shields and production items. It's better to waste 6 Shields building a Horsemen in a 9spt city once, than to slow down its growth on the way to 10spt. |
The growth issue is "unfortunately" complicated too. Very large towns in the ancient age are often a liability, because they often outgrow the empire's overall luxury-slider needs or their palette of developed tiles, or encroach upon other towns' (especially camps') developed-tile needs, or have trouble reaching 10 shields without an odd tile allocation, etc. But yes, food micromanagement is a consideration equal to, if not greater than, shield micromanagement.
quote: While cities are growing, frequent micromanagement helps avoid waste by splitting, for example, a 30 Shield item into 2*7+2*8. This is the main reason why I like overlap between city radii (3-tile spacing or closer). |
Amen brutha! This goes for food micromanagement as well, of course.
quote: Once the cities grow to their (temporary) maximum size, I then plan to slow their growth by working Plains with Mines, Hills, or Mountains, or simply by creating a Specialist (usually Scientists for 40-turn research). |
"Temporary maximum size" is something I find impossible to keep track of in my games, as it requires a level of context that I can't see when looking at a city... and I find it too distracting to jump out and back in again just to make sure I have a good reason for limiting growth. In other words, I find maximum sizes quite volatile, as I am constantly building more garrisons, raising happiness or lowering pop enough to free up garrisons elsewhere, improving tiles to add food or enough gold to change the luxury situation, bringing new luxury tiles online, hooking border towns onto the grid, etc. etc. etc.
quote: in general I do not agree with artifically halting growth just to avoid Shield waste. |
Jeepers, where did I recommend that? My lines of defense against shield waste are:
-- starting builds at 10-shields, completing them if they're a fit, raising to 20 if not, raising again to 30 if not, then just biting the bullet unless there's a useful 40+-shield item to build
-- OCP-filling camps, 3-spaced cities (my spacing is an "opportunistic" mix of 3, 4, and 5, with camps to fill the worst gaps), and tile redistribution
-- grabbing the chance to get a little more food or gold. (Yes virginia, I sometime leave max food squares unclaimed, if they provide only a little more food and are otherwise far worse than the best available alternative tile, except in some cases where the town is in a high-corruption area.)
quote: There's a lot to talk about here, but it's all theory and therefore rather bland. |
Dominae, the great thing about Civ and 'Poly is that it's a very varied experience for a very varied crowd. And me, I LOVE theory!
quote: Another thing I've been working on lately is trying to avoid wasting worker turns whenever possible. In the past, I wouldn't have batted an eyelash at ordering 2 non-industrious workers to both road a tile of flatland (1 such worker needs 3 turns, the two together do it in 2... but it's inefficient). |
Funny... worker micromanagement, on the other hand, I really can't get into. I mean, the worker dogpile stuff. I "elide" this micromanagement by just following the rules:
- never send a worker to an unroaded tile more than once, unless there is a strategic gain involved or the gain in turns-to-first-benefit will be immense. (For this reason, I can't see why dogpiling roadbuilding on plains would be an interesting case -- it's something I'd normally never do, unless maybe there was a forest there at first and I had nothing but slaves for the job!)
- never pair slaves and natives, because otherwise the math becomes too confusing
- never pair up when there's an odd # of turns to completion, unless
- never end a worker-turn moving unless moving into an unroaded square
- "always spread outwards" -- all else being equal (which it never is), when choosing a direction to move, choose the one farthest away from your existing cities, and when choose a target space, choose the one providing the most balanced "spread" of worker locations.
- "avoid the capitol" (pre-PTW only) -- once the capitol's needs are met, never unnecessarily improve its area, because you will constantly be buying workers that will have no efficient choice but to be doing precisely that.
Enough for now, fer God's sake!! :whatevertheiconisforlaugheyface:
USC
|
|
|  |
 |
|
Dominae
|
|
quote: Originally posted by UnityScoutChopper
Jeepers, where did I recommend that?
|
Nowhere. I was just using your excellent points to make an additional one of my own.

quote:
-- starting builds at 10-shields, completing them if they're a fit, raising to 20 if not, raising again to 30 if not, then just biting the bullet unless there's a useful 40+-shield item to build
|
Careful not to keep raising the cost just to avoid Shield waste. Sometimes those cheaper items are more useful, and well worth the wasted 2-4 Shields. The best example of this is China with a 3spt capital: do you build an Archer first, wasting only 1 Shield, or 2 Warriors, wasting 4 Shields? I think the right answer is the two Warriors, because early scouting power is so important.
Dominae
|
|
|  |
 |
|
WackenOpenAir
|
|
hmm really looks like you go way too far in this.
-micro should not interfere with macro strategy. what has to be build, will be build.
-i don't see production as a constant thing, i can very well produce at 6 shields for 2 turns, then swich a worker to a flood plain for example and produce 2 more turns of 4 shields to finish a 20 shield thing. Much better than looking for something that costs 30 shields.
in other words, i will never alter my production to match the shields, i will try to alter the shields to match production. mostly that is changing it near the end of producing something.
only for a production of 5 or 10 shields per turn, i will pay extra effort to reach that.
(extra effort being sacrificing multiple trade or units or even a food, to get that 10th production, or giving that city a very high priority when deciding what a worker should do)
|
|
|  |
 |
|
UnityScoutChopper
|
 |
Brno, Czech Republic
Oct 2000 time: 06:31
|
|
Dominae wrote:
quote: Careful not to keep raising the cost just to avoid Shield waste. Sometimes those cheaper items are more useful, and well worth the wasted 2-4 Shields. The best example of this is China with a 3spt capital: do you build an Archer first, wasting only 1 Shield, or 2 Warriors, wasting 4 Shields? I think the right answer is the two Warriors, because early scouting power is so important. |
Heh, I was waiting for that. The Persians (industrious, spearman vs. warrior) Zulus (non-industrious, true, but archer vs. scout) and Russians (non-industrious, but spearman vs. scout) are in a similar situation. Sometimes a rise in shield count in mid-build (e.g. 3+3+4 or 3+3+5) saves one the agony of the decision. Otherwise I too generally go the for warrior, but it all depends. Actually, for non-expansionists, the spearman is the most tempting of the warrior alternatives, as it can pop huts with relative impunity - nothing worse on a Huge map than dozens of turns of progress into the darkness because your luck ran out during a hut pop.
WackenOpenAir wrote:
quote: in other words, i will never alter my production to match the shields, i will try to alter the shields to match production. |
I do both, but I too usually do the latter.
quote: -micro should not interfere with macro strategy. what has to be build, will be build. |
I find that by setting builds to the cheapest useful item, then next cheapest, etc., what has to be built does get built anyway, and then you just use your road network and move it where it belongs.
quote: -i don't see production as a constant thing, i can very well produce at 6 shields for 2 turns, then swich a worker to a flood plain for example and produce 2 more turns of 4 shields to finish a 20 shield thing. Much better than looking for something that costs 30 shields. |
Discussing production as an inconstant thing would be too complex for the article. But there are many cases where you can predict in advance your next x turns of shield production -- which in extreme cases may be constant as many as 19 turns (for new towns in the midst of a peninsula, forest, and/or hills)
Incidentally, to tie into an earlier point of Dominae's and an old important food micromanagement oddity: when a city grows, it gets immediate new SHIELDS from the new tile, but not immediate new FOOD, and the auto-placed worker generally goes to the TILE CREATING THE MOST BALANCE. The upshot? If you're going to be micromanaging food for a town because its bonus-food tiles that don't add up to 5 (or 4 for an ungranaried town), it's best to "siphon" food from the town to achieve a good/perfect foodbox fit BEFORE the turn the town grows; if it grows and has plenty of food but not shields, it'll get a free 2-shield tile; if it grows and has plenty of shields but not food, it'll get a high-food tile, whose effect will be wasted.
USC
|
|
|  |
 |
|
Dominae
|
|
quote: Originally posted by UnityScoutChopper
Incidentally, to tie into an earlier point of Dominae's and an old important food micromanagement oddity: when a city grows, it gets immediate new SHIELDS from the new tile, but not immediate new FOOD, and the auto-placed worker generally goes to the TILE CREATING THE MOST BALANCE. The upshot? If you're going to be micromanaging food for a town because its bonus-food tiles that don't add up to 5 (or 4 for an ungranaried town), it's best to "siphon" food from the town to achieve a good/perfect foodbox fit BEFORE the turn the town grows; if it grows and has plenty of food but not shields, it'll get a free 2-shield tile; if it grows and has plenty of shields but not food, it'll get a high-food tile, whose effect will be wasted.
|
Yes, you always want to time things such that your highest Food output is the one that fills the Food box. The Governor will then choose a tile with high Shield output for the new laborer when the city grows, and these Shields are immediately counted toward that turn's production.
For this reason I like to keep an unchopped Forest around my 3+ Food surplus cities (non-River is better than River). Such a tile is the highest non-bonus Shield output you can get in Despotism without tile improvement, and so is perfect for boosting production when a city grows. It does not really matter that the tile does not produce good Food or any Commerce, because only the Shields are added when the city grows.
Dominae
Last edited by Dominae on 03-10-2003 at 18:50
|
|
|  |
 |
|  |
 |
|
Dominae
|
|
Severe micro is standard fare for some players. And, once you get used to it, does not really take that much more time. It's one of those things you have to pick up on the way to being a truly excellent player (or so I've heard!).
Dominae
|
|
|  |
 |
|
Dominae
|
|
I expect all current Emperor-level players here in the Strat forum will really enjoy the new Demigod level. It provides that nice challenge we all look for in our games, without being too in your face with bonuses like there are on Deity.
Dominae
|
|
|  |
All times are GMT. The time now is 05:31. Apolyton Time is 00:31. |
top of page
|
| archivepost |
|
Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is ON
vB code is ON
Smilies are ON
[IMG] code is ON
|
|
|
|
|
|