ZERO trade

[shade]

Now I really am confused,I knew that foodcaravans give sometimes 0 trade arrows for a traderoute(I learend to live with that)
But in the game I'm playing now I sometimes get those 0-trade arrow-traderoutes for real trade-stuff????
(How Is this possible???)
->the trade always is between cities of my own(since I can't trust the AI)

Shade

[Gramphos]

The cities are too close.

[shade]

cities that are closer gave me arrow(s).

so guess not

Shade

[Eyn]

if u export silk to a nearby city who also trade silk don't be surprise if u don't earn money !

[La Fayette]

shade
The commonly admitted formula for the value of trade routes has been given long ago by Robert Lancaster (you can find it in the GL, read 'Scrolls of Wisdom').
This formula is:
(THC + TDC +4)/8
where THC = Trade in the Home City
and TDC = Trade in the Destination City
This means that if your city has 2 trade arrows and the foreign city where you deliver your caravan also has 2 trade arrows, the value of the trade route is (2+2+4)/8 = 1
But this value is halved if both cities are yours.
Therefore, if you trade between your own cities in the early game, it is not uncommon to get zero trade routes (but the value will grow when your cities grow, later on).

[La Fayette]

BTW shade
as you can see when looking at the formula, the distance between the 2 cities has no influence at all on the value of the trade route (contrary to what gramphos told you).
But the distance has an influence on the value of the delivery bonus (the gold and BEAKERS that you receive on delivery; the 'scrolls of wisdom' in the GL also tell you that).

[shade]

thx La Fayette that could be a reason(don't know how many arrows there were)
btw how does that GL actually work???

Eyn: Duhhhh,I'm not entirely retarded,the shipments went to cities that DEMANDED them.

Shade

[La Fayette]

Originally posted by shade
btw how does that GL actually work???
Shade

Second thread on the strategy forum (the first thread is a discussion about improving the GL).
Click: anything is explained.

IMO anyone willing to post on the strategy forum should HAVE A LOOK AT THE GL FIRST (since most of the questions that might be asked have already an answer, easy to find in the GL)

[shade]

well that was the part I knew,the only problem is that it's quite Chaotic to find something through those threads/posts.
(I wasn't sure anymore why I never try to search it,so I went back to take a look,now I know again.)
All info might be included there but it's kind of a nightmare going through those threads.
IMHO there must be a better way to save all that info.Maybe someone should take the time to go through all those threads and condence all real info out of them.
(ie less chaotic and easier to search)
Also I found the trade formula in the scrolls of wisdom:

Trade = ( trade of home city + trade of destination city + 4 ) / 8

Note that it does not directly depend on city size or distance. The following (cumulative) modifiers apply:

Both cities are yours -50%
Freight instead of caravan +50%
Cities connected by road +50%
Cities connected by rail +50%
Cities on different continents +100%
Airports in both cities +50%
Superhighways in home city +50%

=> now lets put in my info
1)all cities are conected with roads=>+50%
2)It were all my cities=>-50%
3)Some where on different Continents=>+100%
the only thing there is not talked about is if those numbers are rounded up or down
If rounded up even 2 cities without trade arrows would get 1 arrow=>((0+0+4)/8)-50%+50%(these 2 cancel oneanother out)
=1/2 rounded up=1

So there still seems to be a problem

Shade

[La Fayette]

1) About the GL being easy to read
When I came to Apolyton, there was no GL yet and I read ALL the archive before posting my first post (took me 3 weeks!).
IMO spending a few hours to discover how the GL works is really a rewarding investment (and some people have done exactly what you wish, namely test almost everything and write down the main results; read 'diplos and spies' or 'the lone settler at work', for example).
2) About rounding up or down
AFAIK there is no rounding at all.
For example: (5 + 5 + 4) = 14
14 is less than 16
Hence a trade route between 2 of your own cities with 5 trade arrows each has a value of 0 (if there are no other modifiers at work)

[shade]

LF:I think you are a brave person,going through those archives.
Maybe some day (when I have time,ie not so soon) I will go through some of those threads and get the info out and compile this to something quick to use.

about the rounding, I'm quite sure there must be.
I had quite some trade early-on and all of them gave 1 arrow in each city(except for those trading with that one city)
and that was only for 1 of the 3 traderoutes,each new traderoute made the previous one become one??

I really need to take a better look at that.

thx for the help LF

Shade

[SlowThinker]

Originally posted by shade
All info might be included there but it's kind of a nightmare going through those threads.
IMHO there must be a better way to save all that info.Maybe someone should take the time to go through all those threads and condence all real info out of them.
(ie less chaotic and easier to search)
Shade

Go on! Build your own "condensed" thread. There are only three now: Combat, Dips/spies and Settlers/Engineers.
"The Great Library: a hierarchical structure"

[Marquis de Sodaq]

Don't always assume rounding of numbers in programming! Often, decimals will get truncated - even 0.9 gets passed on as 0. This is usually just "display" value, the real value is retained for any calculations.

[shade]

Originally posted by SlowThinker

Go on! Build your own "condensed" thread.

It's on my to do list,but it low priority and the other items are BIG.
(I think by the time civ IV is out i'll get to it )

(display values are most of the time rounded values,but now you say,If they just use integer-math the solution is truncated )


Shade

[La Fayette]

Originally posted by Marquis de Sodaq
Often, decimals will get truncated - even 0.9 gets passed on as 0. This is usually just "display" value, the real value is retained for any calculations.

Thank you Marquis.
This is exactly what I tried to explain, but your english is much better than mine (even with that kind of slightly strange accent...Minnesota?)
If you truncate civ 2.99, guess what you get.

[SlowThinker]

Originally posted by La Fayette If you truncate civ 2.99, guess what you get.

La Fayette,
do you want to say that the Civ 2 was Civ 2.99 in fact and so there will not be a difference between both versions?
BTW, did you read my message (MP)?
Edit: I mean the PM about the multiplayer.

[shade]

Now there still is the question how the modifiers work.
eg modifiers m1,m2,m3
trade in city T1,T2

is it:
((((T1+T2+4)/8)xm1)xm2)xm3

or is it

((T1+T2+4)/8)+((T1+T2+4)/8)xm1+((T1+T2+4)/8)xm2+((T1+T2+4)/8)xm3

?????
(probably the second one I guess)

Shade

[La Fayette]

Originally posted by shade

?????
(probably the second one I guess)
Shade

If you wish to make friends with Slow Thinker, you should never guess, you should test .
Contrary to what you guess, the multipliers I have tested (namely 'both cities yours','road' and 'railroad') follow your formula #1.

(La Fayette, ready to dream about SlowThinker after that little bit of testing)

[La Fayette]

ST
I meant that civ2 was such fun that it might just as well be civ3 or 4 or more than that (but I strongly hope that civ3 will be even more fun than civ2).
I got your message and replied. Let me know if there was a problem with my reply.

[shade]

Originally posted by La Fayette
If you wish to make friends with Slow Thinker, you should never guess, you should test .
Contrary to what you guess, the multipliers I have tested (namely 'both cities yours','road' and 'railroad') follow your formula #1.

I'll do some testing in the morning but I am sure it will be Nr 2,when I guess I have my reasons
->easier to program
->easier math for the programmer
->les problems with values dissipathing

(little sidenote : ppl have always been complaning about the food model=> did some calculationson them lately=> SID ACTUALLY USED A REALLIFE POPULATIONGROWTHALGORITHM(ie: dN=aN(1-N/K) ) =>9.9/10 for realism)

Shade

[SlowThinker]

La Fayette,
the last PM obtained from you is 28-10-2001. Then I posted you a PM on 29-10 and then a copy tomorrow.
If you don't have them you can read
A game for rookies (for SP players).

Originally posted by shade
SID ACTUALLY USED A REALLIFE POPULATIONGROWTHALGORITHM(ie: dN=aN(1-N/K) ) =>9.9/10 for realism

Can you explain?
BTW, did you see this: http://www.steffengerlach.de/cz/

[shade]

Can you explain?

Ok here we go:
The most simpel population model you can imagine is one where growth is linear equivalent with the population(dN=growth of population,N=population,a=the constant)
then this would give you this formula
dN=a*N
when we differentiate this we get an exponential growth
N=e^(a*t)
When we now add a restriction=>"there is only a limited amount of food that can be produced",we get a new (constant)Variable K wich represents the "support power"(don't know a better translation) of your food.The new formula will look like this:
dN=a*N*(1-N/K).
(this shows that if you heve a bigger population then you can support N/K>1 and dN becomes negative ie. you have ppl starving.When your population N When you now plot a curve that shows how your population grows over time you get something that looks like a tangent put on his side.
When you plot out the population growth in Civ2 vs the number of turns it takes you get a curve that looks exactly the same,and you'll recognise terms in your calculation from the second formula.

Shade

[shade]

As LF suggested I did some testing,and here are the results:

-Civ2.42
(small map/2 civs/no barbs)

First the formula
(THB+TRC+4)/8

-THB:Trade Home Base
-TRC:Trade Recieving City

This seems to be correct but there is one little thing that should be added: all modifiers seem to work on the truncated value of this formula.
==>Trunc((THB+TRC+4)/8)
(this means 2,001=2 and 2,999=2)

There were 5 cities
Washington(W)(1),Boston(B)(1),New York(NY)(2)=>1 counrty
Kyoto(K)(1),Osaka(O),(2)=>second counrty
(1) and (2) indicate the island they're located on
(letter)=shortcut for the name used lateron

Trade=route direction
wanted=if the tradecommodity is demanded by the reciever
Formula=basic calculated value
Arrows=tradearrows in the game
Mod's=modifiers that are in use

Trade |Wanted |Formula | arrows | Mod's
----------------------------------------------------------------
W->K | Y | 4,375 | 4 | none
W->B |N | 5,25 | 2 | My Cities
W->NY | Y | 6 | 3 * | diff continent/My city
W->O | Y | 4,75 | 4 * | diff continent
-----------------------------------------------------------------------------
W->K | Y | 4,875 +| 6 | roads
W->B | Y | 7,5 | 5 | roads/My Cities
W->NY | N | 6,75 | 3 * | diff continent/My city/roads
W->O | Y | 5 | 5 * | diff continent/roads
=> roads on diff islands mean that both cities are surrounded by roads,but there is no connection
-----------------------------------------------------------------------------
(using different amounts of starting trade from both cities)
W->B | Y | 7,5 | 7 | roads/RR/My Cities
W->B25,25| Y | 6,75 | 6 | roads/RR/My Cities
W->B19,16| Y | 4,875 | 4 | roads/RR/My Cities
W->B22,19| Y | 5,625 | 5 | roads/RR/My Cities
W->B44,42| Y | 11,25 | 11 (12)| roads/RR/My Cities
W->K | Y | 5,375 | 10(11)| roads/RR
W->K | Y | 6 | 15 | roads/RR/Super Highway

conclusions:
*=> different Islands DONT give any bonus
+=> shows that the value from the formula is truncated before further calculations(4,875x1,5=7,3 <>6!!!!)
=>since a RR square is always roaded(check it uot yourself using the cheat mode) I would say RR gives a bonus of 100% instead of having to take in account the fact that there are roads underneath them=>ie 2x+50%

Also am I very sorry to say that the modulators are used like this:
(((Formula)xM1)xM2)xM3
(and in those calculations there are no truncations until the final solution)

Also there still are some small discrepancies,look at the last tabel there the game values are 10 and 11 and the calculated values say 11 and 12(and I can't figure out where they are lost,it's not truncation,I'm sure=>calculated them)

Also an other little sidenote: 2 cities connected through roads does not mean you get +50% there seems to be a favored road(probably the one the AI would use in a go order )

Shade

[shade]

this is the test'game' I used

(couldn't add it while editing previous post, my white-spaces in my tabels are gone )

Shade

[SlowThinker]

Shade,
Welcome to the club of testers

I agree there is no different continent bonus and no freight bonus.

Two cities at a straight distance 3 get a road/railroad bonus only if there is a road/railroad with a bend.

And road/railroad routes are one-sided! Only one city gets a bonus. The (opposite ) route needed for the bonus of the second city is different.

I started with Scrolls of Ancient windows (Is there some better summary about the permanent trade bonus anywhere?).
The text about superhighways (Superhighways in home city...+50% ) is a little bit confusing. It is not the home city of the caravan that gets a bonus. It expresses that the superhiways needn't be in both cities (as airport) - only the city that has superhighways gets the bonus.

---

Originally posted by shade
SID ACTUALLY USED A REALLIFE POPULATIONGROWTHALGORITHM

I agree that the formula is for a limited food.

But:
1. I don't think it is natural real life: More food is usually available, but for higher expenses.
2. I think Sid didn't choose that formula. The civ city growths slower and slower, but don't stop growing.

[shade]

didn't test for freights and airports

1. I don't think it is natural real life: More food is usually available, but for higher expenses.

the formula I gave was a simplification used for simpel pop models(as is the exponential growth)and they simulate real population models.(but in a simplified way)

2. I think Sid didn't choose that formula. The civ city growths slower and slower, but don't stop growing.

1) the actual model I describe only reaches the limit at infinit(ie it never stops growing)it grows asymptotically.
2)It will not be exactly the same but it will be very close(eg the last time your pop grows 1 point is like 200 turns or so)
I have some graphs of it at school,I'll post it here(exel),it really shows the curve the described formula has.

Shade

[SlowThinker]

Originally posted by shade
the actual model I describe only reaches the limit at infinit(ie it never stops growing)it grows asymptotically.

do you speak about this:
dN=a*N*(1-N/K)

If you suppose you have always 2 food per square then
dN=K /(1+N)
(1/1+N represents the size of foodbox
K is a constant)

[shade]

do you speak about this:
dN=a*N*(1-N/K)

Yes.

btw I took a city with only grassland around and farmland=>4food/sq
Here you have to keep inmind that K isn't always constant(it grows with the population (for a while).
i'll add the exel-thingy I created.
(when you take a look at the graph you see pop/time.
When you plot that differential equation you'll get something similar)

Shade
PS:I forgot to take a look at it at school,and on this computer there is no exel so I'm not sure what the last graph was I saved