Score Points per Tile

Svar

King
Joined
Jun 10, 2003
Messages
615
Location
China Lake, Ca
As I have entered the real milking phase of my present game, I decided I needed a 'tool' to evaluate the worth of any city location.

The tool is really only a method to do this evaluation. I start from a 21 tile all grassland city. Railed and irrigated, this city will produce 83 food. That will feed 41 citizens. If I count 2 times the number of happy citizens (20) plus the number of specialists (21) and add the number of tiles all divided by the number of tiles, I get 3.90 points per tile. That will be my baseline for any evaluation. Any time I can place a city and get at least that value I place the city.

As a check, what will an all plains city give me? Railed and irrigated, the city will produce 63 food. Running the same method , I get 3.43 points per city. Any city giving lower than that number will be the first city removed when I want to place a new city.

This method has made me take a very close look at city sites that look really good at first glance. In my present game I have a city in a new prime location that has 43 citizens and close by there is another location that looks almost just as good. On closer inspection the 43 population city has 11 FP tiles a wheat and an oasis. The new location only has 7 FP tiles. With 7 FP tiles and all the rest either plains or desert this location has less value than an all grassland location.

Ok, what about smaller cities? I only considered 9 tile all grassland cities. The points per tile for these cities 3.78 and since these are so easy to place many of these can replace all those all plains cities or any city with mountains in the border.

I will have to see how this works in practice but it will have a profound effect on my next game because there will probably be many more cities that will be consideded as tempory until the milking phase. Right Now I have to consider removing old core cities because their contribution is low to the total score.
 
Interesting line of thinking. I would like to point out however that the 21 tile venue of thinking should can be questionable if you're playing an agricultural civ however. Since ag civ cities get +1 food in the city center, under those circumstances I would think it better to have more cities of lower average size. This would not only increase your number of +1 tiles (thus increasing your overall pop and score), smaller quantities of citizens are easier to keep happy. All luxuries connected and a marketplace in each city and you should be good to go.

It's also easier to avoid having wasted tiles within your borders grid/layout wise.
 
The 9 tile cities are much easier to place thats for sure but have a slightly lower point per tile count.

Here is my biggest city and it is half away around the world from the palace but as you can see there is no trouble keeping every citizen happy.

Super_City2.JPG



BTW the points per tile count for this city is 4.0.


Here is a city closer to the capital that actually is much smaller but has a point per tile count of 4.3 because there are only 13 tiles that count toward the domination limit.

Liverpool2.JPG


During the early phases of a game I have much closer city spacing than 21 tile cities but most of these are desiginated tempory and are lifted in the milking phase to free up the city center tile where there is no happy citizen.

Edited to reduce the size of the screenshots.
 
Why are you mentioning how far these cities are from your capitol? That should only affect corruption, not happiness or population...
 
Distance from the capital does effect corruption but when you are using the luxury slider to keep citizens happy you also need low corruption to make the slider more effective. The key to this method is all citizens are either happy or specialists. Remember this is primarily for the milking stage.

I will include another illustration of the use of the method when there is overlap as would be the case for 9 tile cities. In New Quirigua, 2 cities overlap, New 95 and Uaxactun. For this discussion we will ignore Uaxactun. Without New 95, New Quirigua would have 15 tiles 17 happy citizens and 7 specialists for a point per tile (PPT) count of 3.73. When I added New 95, New Quirigua dropped to 14 tiles 16 happy citizens and 6 specialists for a PPT count of 3.71. But if you look at both cities taken as a system the combined totals would be 17 tiles, 21 happy citizens and 7 specialists for PPT count of 3.88.

New_Quirigua.JPG


New 95 by itself has 3 tiles, 5 happy citizens, and 1 specialists for a PPT count of 4.67.

New_95.JPG


But the real reason for placing New 95 was to work 2 coast tiles that were already included in the domination limit and not being worked by anyone. If the 2 unworked tiles were included in the New Quirigua w/o New 95 calculation its PPT count would have been 3.41. So adding New 95 increased the PPT count for New Quirigua from 3.41 to 3.88.
 
Nice Svar!

What you're doing here is similar to the rules of thumb I use when reorganizing my holdings during milking. But you are being much more methodical and precise.

The rule of thumb I've used is to settle wherever a city's worked tiles will average close to 4 food/tile, i.e. close to irrigated+railed grassland. And I abandon cities where the average is much less than 3 food/tile. I add in food from sea tiles without adding them to my tile count when checking cities. Near the end of the process I start examining the worst remaining cities (in terms of average food/tile) and seeing whether there's still available land which is better. That final phase seems to usually be around the 3 food/tile mark. I.e. the worst land which is worth keeping (because there's nothing better) is usually roughly "plains" quality.

I usually don't explicitly factor in city size. I just figure that bigger is better where possible. (One exception I seem to get in most games is a city on a coastal point where all citizens can't be made happy, not enough specialists.)

Your approach to all this is more precise. I wonder where it will take you next :) - will it make a significant gain being that precise in placement? Will a quicker way to evaluate sites become apparent?
 
I don't know whether there will be a significant gain, there sure is a significant slow down in the amount of time it takes to play 1 turn. The calculation of the PPT count doesn't take very long, there are just lots of locations to look at. I'm sure by the end of this game I will develop some method to cataloguing the various city sites. I still am learning just how the cities play together when their cultural borders merge.

In the screenshot below, New 106 is almost fully developed. New 118 was planned to be a 9 tile city and only needs a couple of improvements to be fully developed. New 121 was just placed and my orginal intent was to make it a 9 tile city as well but moved it 1 tile SW to see what would happen. The border filled the 3 tiles between the cities. Now new 118 and New 121 can share the extra 3 tiles. BTW New 121 is located on a hill surrounded by grassland, that is a trick I just learned. I get 3 food for that tile instead of 1 if the city were located at the other MI.

City_repositioning.JPG
 
OK, I just want to make sure I understand the score components. So far I've got:
  • Territory : 1 point per Land/Coast tile
  • Happy Citizens : 2 points per happy citizen
  • Content Citizens : 1 point per content citizen
  • Specialists : 1 point per specialist

Are there any other components?
 
so... umm.. using this you can find the max end score of your current city placement in the game by summing up all city per tile scores multiplied by the number of domination limit tiles they use.. multiplying this value with the difficulty modifier.. dividing the value with 540 and multiplying with turns left in game.. then add your current score * turns so far / 540...

or did i miss something?
 
Gyathaar said:
so... umm.. using this you can find the max end score of your current city placement in the game by summing up all city per tile scores multiplied by the number of domination limit tiles they use.. multiplying this value with the difficulty modifier.. dividing the value with 540 and multiplying with turns left in game.. then add your current score * turns so far / 540...

or did i miss something?
Sounds like you've got the right idea, but I would write it differently:

(CurrentScore*TurnsPlayed)+( (2.HappyCitizens + Specialists + LandCoastTiles) * DifficultyModifier * TurnsLeft ) / 540
 
Yup.. that would be the same.. :p

Edit: Correction.. it would be almost the same.. your would be with current number of citizens and specialists.. mine would be with max pop the land would support with all irrigated tiles
 
You can also use the formula I used in the spreadsheet at http://forums.civfanatics.com/showthread.php?t=17550

The approach I took there is to use the difference between the prior turn's game score and the current term's game score, and based on it calculate what the score for the single current turn is. Then use that to project what the final score will be if nothing changes from the current turn through to 2050. I.e. if territory, population, and the level of happiness of the citizens all remain unchanged for the rest of the game.
 
I know.. the line I said would be the max score possible to sqeeze out of currentl territory thou.. assuming you immediately got all tiles irrigated, all citizens either happy or specialists.. and never got pollution :p

The other formula plus SirPlebs calculator assumes current population
 
punkbass2000 said:
Don't forget content citizens. They're worth one point each as well.
Thanks punkbass2000. I just checked to make absolutely certain, and you were right (of course) :).
 
2 thoughts/questions:

1. Wouldn't it be best to look at the relation between the happy folxs and the non-happy ones, and try to max it out?
Taking this course leads me to think large cities are best - with 20 happy people, all tiles are taken, and the rest are specialists.
This is true for any placement, but with larger cities this relation seems highest (from my scetches..).

2. I can't figure out why for 2 ocean tiles off the domination limit, you would soak up so much coast/sea tiles which don't help food as much as land tiles..
Why don't you just go total-land?? :confused:
 
boogaboo said:
2. I can't figure out why for 2 ocean tiles off the domination limit, you would soak up so much coast/sea tiles which don't help food as much as land tiles..
Why don't you just go total-land?? :confused:
Sea tiles dont count in domination limit, so if you can grab equal amount of coast and sea tiles, they give more points than irrigated grassland (2 happy people and 1 point for area vs 1 point for area and 1 happy and 1 specialist)
 
Gyathaar said:
Sea tiles dont count in domination limit, so if you can grab equal amount of coast and sea tiles, they give more points than irrigated grassland (2 happy people and 1 point for area vs 1 point for area and 1 happy and 1 specialist)

Good point! :blush:
So.. large cities, and covering sea tiles is my first method..
 
Back
Top Bottom