Hurry Production.

[Kramsib]

Hi everyone:

First of all I want to congratulate you for all your efforts to keep Civ II alive.

I have some questions for you :

I have always believed that hurrying production in Civ II costs 2 gold per shield, except when the production box is totally empty, then it costs 4 gold per shield.

But I was playing Civ II this morning and I realised that this ratio varies, depending on how much I have completed my production.

Do you know how much hurrying production costs exactly in every case?

Another question:

Do you think that knowing what a shield costs in gold is important in your PBEM games? Someone has a theory about shields value?

Thank you very much.

[N35t0r]

I know that this value is different for units, city improvements, and WoW's.



quick test: Units

total shields : shileds before rb : rb cost

20 with 10: 25
30 with 20: 25 (the last row, unless it is the first (10-shield units) always takes 25 gold

30 with 10: 60
40 with 20: 60 (so the next-to last row is 35 shields)

expected that aditional rows are +10 from the previous

20 with 1: 56 25+35-4
30 with 1: 100 25+35+45-5
40 with 1: 154 25+35+45+55-6

I expect that the first shield in each row costs 3+1*row#

(row# is 1 for last row, and goes up till first row)

40 with 19: 64 25+35+4
40 with 29: 28 25+3
40 with 39: 2

so the last shield in each row is 1+1*row#

hmmm... it seems that the cost for each shield is a base (which is either 2 or 1, makes up 15 g tot, and depends on the position of the shiled on the row) plus 1 times it's row #. (so each consecutive row is 10 more expensive than the previous one)

Shield# in row: 1 2 3 4 5 6 7 8 9 10
base Cost: 2 2 2 2 1 2 1 1 1 1

so the first row would be:
3 3 3 3 2 3 2 2 2 2 (=25)
the second
4 4 4 4 3 4 3 3 3 3 (=25+10=35)
and so on

which, added up, would be:
60 56 52 48 44 41 37 34 31 28
25 22 19 16 13 11 8 6 4 2

(ranging from buying 20 shields to buying 1 (if at any time the box is empty, cost is x2 (so a 20 shielder is 60*2=120) if the box was empty) to an almost full last row)



20 with 0: 120 (25+35)*2
30 with 0: 210 (25+35+45)*2
40 with 0: 320 (25+35+45+55)*2

so having no shields doubles cost

[Bloody Monk]

Incremental Rush-buying for Units

If there are more than one row of shields that need buying, and you have the build order slots for that row (row 3 = Diplomat), buy them one row at a time and save. This way the cost is 25g per row.

This does not have the same effect for non-unit buys.

Monk

[germanos]

Kramsib:

N35t0r showed perfectly how the cost of shields is handeled by the program.

Buying units 'per row' is often called "incremental rushbuying" and it is an old technique that is used extensively in PBEM's (and in any SP i play as well). There might be house-rules against it in some games.

Some scenario designers increase the amount of shields per row (for example Kobyashi's Battle for alpha-quadrant) which makes rushbuying much more expensive.

[DaveV]

In a nutshell, where N is the number of shields left:

Wonders and spaceship parts cost 4N to complete.
Other buildings cost 2N to complete.
Units cost N*N/20 + 2N.

In all cases, it costs twice as much if the box is completely empty.

Edit: added spaceship parts.

[debeest]

Wow. Cool. I knew the numbers, but I'd never tried to derive a formula for unit rushbuying cost.

As is so often the case, DaveV has spoken, and there's really nothing left to say.

[Kramsib]

Good analysis Nestor
and thank you very much DaveV for the summary.

Now, the reason why I am so interested in this topic is the fact that I think I have a rule to know how much gold a shield costs.

The game has no clear relation between gold and shields, except rushbuying or capitalization.

Capitalization turns each shield into gold in a 1:1 relation, one shield costs one gold.

Rushbuying, as you have explained, means different values for shields depending on what we are building and how much we have built.

My theory about shield value is based on some economic concepts, and I think I have an interesting explanation.

Under my point of view, neither rushbuying nor capitalization reveals the real shield value:

1. Rushbuying means a penalization in shield cost, so shield value is altered over its real value.

2. Capitalization considers that one shield costs one gold, but the whole quantity of gold and shields is different in a civilization (this is because the different terrain types and their different production), consequently their values must be different according to the fact that their relative abundance is different.

So, let me explain what is my solution to this problem.

Let's consider that citizens are our "labour factor". We employ our citizens to get resources, food, shields and gold. So, all the production in our Civilization is gathered with the same quantity of citizens each turn, or better said, the same quantity of "labour force".

It is logical to think that, as the whole production has been obtained with the same "labour force", the total production of food has the same value as the total production of shields, and both have the same value as the total production of gold.

Using a mathematical form

In turn #:

Food Value = Shield Value = Gold Value.

So we can translate, food and shields into gold, just like a real monetary system (with gold pattern )

How much gold does a shield cost?

Now, answer this question is very easy if we know where to get some information and what some demographics mean.

Gold Value is equal to GNP in Demographics, as it reflects the total quantity of trade rows which can be turned into gold.

Shield Value is equal to Mfg. Goods in Demographics, as it reflects the total quantity of shields which are produced in your Civilization.

So the ratio:

GNP / Mfg. Goods

gives the real quantity of gold each shield costs.



As this ratio is different for each Civ, it reflects the different prices every nation has in real world, Inflaction and Deflaction can be also simulated. If the quantity of shields grows up (industrialization), the ratio goes down, the abundance of shields makes them cheaper (deflaction). If the quantity of gold raises (massive building of commercial facilities or roads), the ratio goes up and shields become less abundant in comparison with gold, so we have inflaction. (Just like in the real world ).

In a PBEM, we can establish a real market of items, every nation will has different prices, so they can especialise and compete like a real market

Another aplication for this ratio, is that it can be also used individually for each city to compare if rushbuying or capitalization is really profitable.

If a city produces 10 shields per turn and 30 gold, the ratio will be 3 gold per shield, so if rushbuying costs less than 3 gold per shield, let's say 2 gold, rushbuying is cheaper and a profitable option. If rushbuying costs 4 gold, you will lose money. In this specific case, you can realised that Capitalization is not a good option, it will give you one gold per shield when each shield is equal to 3 gold.

What do you think?

[debeest]

Your definition of the GNP in demographics is wrong.

In the demographics screen (F11), GNP = gold income plus 2*luxuries. (It's probably the only thing I ever discovered and posted here before anyone else, so I have a stake in making sure everyone gets it right. )

[Bloody Monk]

I am much in favor of extensive micro-management, but, this makes my hair hurt. "Food value = shield value"??...not in most of my games, where controlling growth/happiness is usually a consideration. And what is the value of those demographic screens anyway?? Even if you can determine what they are measuring (thanks, debeest!!), does this info really help one play the game better??

It is an interesting article and I stand ready to be corrected, but it seems like over-thinking to me. More game playing, I say.

Monk

[Kramsib]

Originally posted by debeest
Your definition of the GNP in demographics is wrong.

In the demographics screen (F11), GNP = gold income plus 2*luxuries. (It's probably the only thing I ever discovered and posted here before anyone else, so I have a stake in making sure everyone gets it right. )

I see, ...

I was working with Stefan's definitions

http://stefan.winkler.net/civ2.html

GNP is undefined and I thought that it could be the same as in Civ III where GNP is defined as the sum of all the gold gathered.

I was wrong obviously, thank you debeest.

But my idea is still there, the ratio:

Total Gold / Total Shields, give the price for each shield.

Bloody Monk says:

I am much in favor of extensive micro-management, but, this makes my hair hurt. "Food value = shield value"??...not in most of my games, where controlling growth/happiness is usually a consideration. And what is the value of those demographic screens anyway?? Even if you can determine what they are measuring (thanks, debeest!!), does this info really help one play the game better??

It is an interesting article and I stand ready to be corrected, but it seems like over-thinking to me. More game playing, I say.

Monk

Yes I know it can sound like overthinking but I think it can help to play better.

If you know how much gold is a shield you can know the real value of any unit, any building or any Wonder without the default punishment for rushbuilding.

When trading units in your PBEM's (if you use Yaroslav's CivDip you could trade even with cities - see the Spanish Civilization Site for further details, look at my sign for a link - ) you would be able to calculate a more accurate price.

This ratio can be also useful to programme a more efficient AI for new Civilization Games becouse not only it gives the shield value, but also indicates the relation between wealthy and industrialisation.

Of course you can live without all this knowledge, but for me is really amazing, discovering new things in Civ and how it seems to reality.

[Bloody Monk]

Please do not misunderstand my comments. I do not in any way disparage the search for more knowledge about this wonderful game. When I said it "makes my hair hurt" I was thinking how hard it is (for me) to determine whether this calculation is meaningful or whether the assumptions are unwarranted. I have seen many discussions over the years trying to nail down this idea (SlowThinker, IIRC), and it would be very nice if your idea is the accurate solution. It will fall to others to determine that, however.

Thanks for starting the thread. Maybe this time will see this answered.

Monk

[Kramsib]

Originally posted by Bloody Monk
Please do not misunderstand my comments. I do not in any way disparage the search for more knowledge about this wonderful game. When I said it "makes my hair hurt" I was thinking how hard it is (for me) to determine whether this calculation is meaningful or whether the assumptions are unwarranted. I have seen many discussions over the years trying to nail down this idea (SlowThinker, IIRC), and it would be very nice if your idea is the accurate solution. It will fall to others to determine that, however.

Thanks for starting the thread. Maybe this time will see this answered.

Monk

If this topic was dealt before by eminent Civ II players, perhaps I should send them a PM to know what they think.

[Ben Kenobi]

Shields are worth about 2.5 trade, all other things being equal.

Capitalisation gives you a poor return for your money, you will do better, just letting the city produce, and increasing your income elsewhere.

[SlowThinker]

Hi guys, I was invited by Kramsib...

Originally posted by Kramsib
If you know how much gold is a shield you can know the real value of any unit, any building or any Wonder without the default punishment for rushbuilding.

I agree, if you can find a correct ratio between food/shield/arrow then you will be able to evaluate every production and expenses by one number. Then it will be much easier to find out how to play and which production you should prefer.

Let's consider that citizens are our "labour factor". We employ our citizens to get resources, food, shields and gold. So, all the production in our Civilization is gathered with the same quantity of citizens each turn, or better said, the same quantity of "labour force".

It is logical to think that, as the whole production has been obtained with the same "labour force", the total production of food has the same value as the total production of shields, and both have the same value as the total production of gold.

Using a mathematical form

In turn #:

Food Value = Shield Value = Gold Value

Bad idea. You forgot that labour forces are not equally effective.
In order to simplify situation imagine there are only 2 types of squares in Civ2: forest produces 2 shield only and grassland 3 food. Then you can exchange 1 shield for 1.5 food. (now we disregard the fact that shields may be traded for food also using food caravans)

You have always to consider rates you can 'trade' resources (replacing workers is also a kind of trade). Therefore rushbuy cost + capitalization cost (or improvement building and its subsequent sale) are possible methods how to compare gold and shield. Which one to use, rushbuy(RB) or capitalization? It depends on situation, shields may be used for city production, gold for bribing. If bribing is more effective then you will build improvements, sell them and bribe: you will consider 1 shield=1 gold. If city production is more effective then you will RB units and improvements: you will consider 1 shield = approx. 2 gold (a digression: gold has one big advantage: it is freely transferable over your empire. Therefore with large civ it may be 1 shield = 1 gold; but also in small empire the possibility of transfer has its value, because gold may be used immediately (in 1st city that is able to finish production with RB), in opposite cumulated shields must 'wait' 10 or 20 turns before they start to be effective).
The price of bribing is quite high in Civ2, therefore you usually prefer not to bribe but to rushbuy (not speaking about MP duel now ), and so the ratio should be about 1 shield = 2 gold (and less, because of free transfer).

the ratio:
GNP (total gold) / Mfg. Goods
gives the real quantity of gold each shield costs.

This is a good idea if you want an 'average' price of a resource. But in a real play you shouldn't think this way, because prices are different in different cities.
Again our simplified example: imagine we have also Plains that produces 2 food. One city has only plains+forest around, another one has only grassland+forest. Of course you won't want to use a general, 'average' price food/shield but you will consider each city separately.

BTW, Civ2 is not 'analog' but 'digital', production is completed at a stroke. So in the beginning you are going for Monarchy, you know it will cost say 200 beakers. If I compare the asset of the Monarchy with 200 beakers I switch to 60% taxes. But later, when I am 20 beakers from Monarchy, I not only switch to 60% science but also replace workers to Ocean.
Similarly when the foodbox is nearly full you should switch to food-productive squares.

If a city produces 10 shields per turn and 30 gold, the ratio will be 3 gold per shield, so if rushbuying costs less than 3 gold per shield, let's say 2 gold, rushbuying is cheaper and a profitable option. If rushbuying costs 4 gold, you will lose money. In this specific case, you can realised that Capitalization is not a good option, it will give you one gold per shield when each shield is equal to 3 gold.

Don't forget gold is transferable. But let's suppose you play OCC:
You are probably trying to think this way:

If a city can choose to produce 10 shileds OR 30 gold...

Remember my example with 2-shield forest and 3-food grassland...


_________________________________________

Originally posted by Bloody Monk
I have seen many discussions over the years trying to nail down this idea (SlowThinker, IIRC), and it would be very nice if your idea is the accurate solution.

I Campo's challenge thread I tried to open a debate how to mathematically evaluate different possible decisions of a player, i.e. ways of development (for example in the beginning of the game - when should you prefer shields and when food?). I indicated a possible method ... but nobody wanted to talk with me :weeping:. But I finished an algorithm which I use in play ( debeest give me a lesson of english if this is ridiculous): I compare a level of exponential growth for each possibility: for example by building Settlers you can double your output in 15-20 turns, with caravans and shipchaining in 5-10 turns etc.
This algorithm could be used to write a very good AI (from economic point of view, so under peace conditions) for a very simplified Civ2. For real Civ2 it would be extremely laborious, because each rule and each exception should be taken into consideration.

[Ben Kenobi]

I did a test of this once.

I built one city with all production squares, another with all trade, and another with all food.

The reason why food is so devalued has to do with WLtD. It is always prefereable to grow your cities through WLtD rather than through the accumulation of food.

Also, as time passes, your value of food drops, as the cities get larger. Food cannot be any greater in value than 1 food = 10 shields, but for a city size 10, food cannot be greater than 1 food = 1 shield. For a city size 20, this becomes 1 shield = 2 food. The value may be less, but it cannot be greater than these numbers.

You can do very well without ever having a large surplus of food, as you will still be able to grow your cities. Production is something that you cannot live without.