Interesting observations on the effects of roads on trade

[Caesar the Great]

I was playing around with this for a little while today and I made some interesting observations on how the cities must be connected in order to attain the 50% road/RR bonus for trade routes. This is what I have observed...

The cities must be connected by the most direct route or else there is no bonus.

Two cities connected on the same horizontal or vertical axis

- Both get the bonus

Two cities connected on the same diagonal

- Only opponents city gets the bonus, the road/RR must leave your city on a vertical or horizontal axis and enter opponents city on a vertial or horizontal axis or else only he will get the bonus. But in furthar testing, I ran into a problem, it didn't always work out perfectly, some cities got the bonus one way but not the other way, if your city is above the city you're trading with on the map, the road/RR must leave on the vertical axis and enter on the horizontal axis, if it is below the opponents city on the map, it must leave your city on the horizontal axis and enter on the vertical axis in order for your city to have the trade route bonus. If it is built both ways both cities get the bonus.

Two cities on different axis and diagonals.

- If the road/RR leaves your city on a diagonal and enters the opponents city on a horizontal/vertical axis only your city will attain the bonus, if the opposite is true only your opponents city will have the bonus, if both are true then both cities attain the bonus.

The effect of rivers

- Rivers seem to have an interesting affect on trade routes, if two cities are on a river, both cities will receive a trade bonus. Furthar testing will be done later....


This provides some interesting conclusions, if you know which path the trade is going along, you will know how to limit your opponents trade, maximize your own and not waste time building unneccessary roads/RRs over harsh terain.

[samson]

Caesar the Great,

The "most direct" route is not always true.
I have seen one case where NO route provided a bonus and others where the one that does provide it seems far from direct.

The river effect on trade is news to me. What exactly do you mean by:

if two cities are on a river, both cities will receive a trade bonus

The same river? Any river?
Are you sure you're not simply seeing the extra trade caused by working bridged river squares?

I have often thought this should be true and have looked for it but have not seen it.
How would you even know if it was true, since you can't eliminate the river?

samson

P.S.

This is definitely a topic that does need to be clarified, so go for it!

[Gastrifitis]

Yes, once this clears up I'll add it to the Road and Railroad section of the new Settlers file and link back to the topic.

[Caesar the Great]

What I meant by most direct is, the fewest number spaces possible, given the circumstances.... some of the routes I observed were in fact quite obscure, but they fulfilled the criterea and did provide a bonus.

What I meant by the statement on the river, I only did one test on this, but two cities were on the same diagonal, a river ran down that same diagonal, a railroad was built, and under these conditions the other city, as all other observations had shown, should have yielded a bonus, but didn't. Building roads/RRs in paths off of the river spaces between the two cities provided normal results, horizontal and vertical river spaces provided normal results. Your own cities/units won't interupt the trade route, enemy cities/units will. If there are water spaces in the way of the trade route no road/RR path can be established.

[Caesar the Great]

Presumably the trade bonus is due to the extra trade from the river itself, but I guess there is no real way to test it out.... everything else has stood up to the tests I've put it through....

[Caesar the Great]

River results were inconclusive, in other words, rivers seem to have no effect, and since they can't be the path the roads/RRs must follow to attain the bonus, they probably have no effect besides the trade/defense/movement bonus already known....

But the rest of the data has already stood up to every test I have subjected it to, in old saves, in scenarios etc.

[Gastrifitis]

Can we assume that, since city and airbase squares function as road/railroad squares, they fulfill the requirements of the connection?

[Caesar the Great]

When playing with the cheat menu for a little while, I found that for open terrain the above work perfectly fine, but water complicates the situation, creating some seemingly obscure trade road/RR routes, but when looking at what happened when terrain was changed to water, there definitely seems to be a pattern to it. With water squares, the route goes normally until it hits the water square, then it goes around the water sqaure, always in the direction of the city, but when it has gone around the water square, it doesn't continue with the normal route, but instead it calculates a new route from that square to the destination city. If it runs into more water, this happens again. With multiple squares it continue until it finds another direct route to the city.

[Caesar the Great]

Originally posted by Gastrifitis
Can we assume that, since city and airbase squares function as road/railroad squares, they fulfill the requirements of the connection?

I'm sure about cities, but not 100% sure about airbases, I tested both of those, but I can't remember what result I got with airbases.....

[Caesar the Great]

In the first image, the red line shows the path the normal route takes, both cities have a bonus.

In the second image, the white line shows the path of the trade route taken from the bottom city to the top city.

In the third image, the yellow line shows the path of the trade route taken from the top city to the bottom city.

[samson]

A few clarifications.

Rivers do NOT create a connection between cities or provide a trade route bonus. Neither do they form a barrier to trade routes. Both of those statements appear to be misinterpretations of other effects.

Intervening cities can have a variety of effects. They can form part of the road/rail connection, they can prevent the connection from ever being achieved, or they can create the bonus in one city (to which they are connected) even though there is NO connection at all to the destination city.

Airports (in both cities) act as a 50% road bonus. But if either a road or rail connection exists, that bonus (50% for road, 100% for rail) replaces the airport bonus.

Non-allied foreign units on the road/rail connection (even non-military units) disrupt the trade bonus. Allied foreign units do not.

Also, a terrain square changed to Ocean via the CHEAT MENU may not function the same as an original Ocean square as it effects road/rail connections. So tests conducted using that technique, may be suspect.

[Caesar the Great]

Yes I know everything that you told me above, but all of my results using the cheat menu I confirmed and reconfirmed in regular games, scenarios and old saves and it has always confirmed the results which I had used with the cheat menu, I just used that example involving the cheat menu because it was the easiest to show the results.

[SlowThinker]

{caravan}{map}
{}{SlowThinker}

------------------------------
This is a post with keywords. See The Great Library: a hierarchical structure" thread.

[arii]

I posted about this about 6-8 months ago. Apparently the computer recognizes only 1 path between 2 cities. The path recognized between city A and B is different from the 1 from B and A. From my personal experience it is the same path for RR. A trick to know the recognized paths between 2 cities if you don't already have roads or no river between the 2 cities is to put a warrior in each city and order them to go to the other city.

[Gastrifitis]

Arii, do you have a link to that topic so we can see what was discussed? Finding stuff in the archives is always so much fun.

[Juggler_Bob]

Originally posted by arii
I posted about this about 6-8 months ago. Apparently the computer recognizes only 1 path between 2 cities. The path recognized between city A and B is different from the 1 from B and A. From my personal experience it is the same path for RR. A trick to know the recognized paths between 2 cities if you don't already have roads or no river between the 2 cities is to put a warrior in each city and order them to go to the other city.

And I was curious as to whether this applied to other calculations, like corruption. If so, it could be used as a sabotage technique against an enemy city. Cut the optimal path (pillage) to the capitol... Would the bribery cost be reduced due to increased corruption?

Just a thought...

-Bob

PS - This post just bumped me to chieftain!

PPS - I know, I'm not supposed to care about such things...

[Scouse Gits]

I clearly recall arii's post, but have just spent half an hour in fruitless searches of the archives - could it be that this thread did not make it to the archive? Ming? Any ideas?

[Paul]

The archives have not yet been imported into the new format. But the thread you're looking for is still active. http://apolyton.net/forums/showthrea...&threadid=2449

[Scouse Gits]

Thnks, Paul - it never occurred to me to search the active fora ... duh!

[Gastrifitis]

Definite that airbases, by themselves, do not contribute to trade connections. They don't hurt, either, you just have to have roads and railroads on the square too. The Trans-Siberian Railway doesn't generally help, either. The key seems to be, as arii mentioned, the Goto command. If there is so much water in the way that the Goto command fails, the computer will never recognize a completed route.

Sometimes, when testing for the shortest route with the Goto command, I will have a railroad going in the right general direction without following the rules for the shortest route. Often the test unit will take the railroad in that occasion. (Go figure. When I want them to take the railroad, they don't.) Just after running the route, if I check back in the source city, the trade route value has increased! It's only a phantom increase, however. If I look back in the city 2 seconds later, it's gone. But it does show how closely the Goto command is tied in.

[Quantum Satis]

Here is a way to find out if a trade route has the road or railroad bonus. Change workers in the city to Entertainers and look at how the value of the trade route change. These values are for foreign trade routes:

No con: 0,1,2,3,4,5,6
Road: 0,1,3,4,6,7,9
Rail: 0,2,4,6,8,10,12

With Superhighways the patterns are different:

No con: 0,1,3,4,6,7,9,10,12,13,15,16
Road: 0,2,4,6,8,10,12,14,16,18,20,22
Rail: 0,2,5,7,10,12,15,17,20,22,25,27

If the routes are domestic their values increases by one regardless of whether the cities are connected or not. The exception is when a city has the Superhighways and is connected to the other city by railroad:

Rail: 0,1,2,3,5,6,7,8,10,11,12,13,15,16,17,18,20...43
(4, 9, 14, 19...39 are missing)

If the distance between two trading cities exceeds 22 squares, or the path between the two cities is too complicated then you don't get the road and railroad bonuses. Diagonal distances are the same as others.

[Quantum Satis]

Deleted

[Quantum Satis]

(Trying again to attach a file)
Here is a picture showing where to build roads so that the trade routes receive the road bonus. The center square is the receiving city. The picture is made with the assumptions that there is no water blocking the way, and that the date line is not crossed. Only distances up to 10 squares are shown but the road bonus can be collected for trade routes up to 22 squares long.

[La Fayette]

Picasso? Miro? Chagall?
Those people did not play civ2, I suppose.
And that is why your image is somewhat nicer looking than anything they painted.
Congratulations,QS!

(this guy is La Fayette, willing to be able to do the same)

(La Fayette, knowing that this never happens)
( perhaps another day I have a try at checking whether this is really the way it works, but today is the french '4th of July' and my only wish is to go to bed with such a beautiful image in the back of my eyes)

[Adam Smith]

Quantum:

Very interesting. What was your methodology for developing this pattern??
I notice that the pattern is not symmetric on the right side of the southwest quadrant. This means that the pattern is not invariant with respect to rotation. (I.e., if you rotate the grid by 90 degrees the path for these squares will be different than it was.) Is this possibly an error?

[Rasputin]

hmm m completely lost me with that pic of the roads system ... what are you trying to see, build roads everywhere >?>>

[Quantum Satis]

La Fayette - Thank you, I'm glad someone liked the picture. Unfortunately I cannot claim any credit for it since all I did was to plot down the outcome of the GoTo-routine, so all credit goes to the person who wrote that routine.


Adam Smith - I made a map containing a continent with a size of 21 x 21 squares. A game was started and Rome was built on the center square. The Extra Land unit had been changed so that it was like an Explorer but with 16 movement points. One such unit was placed on each of the 80 outermost squares of the continent and sent to Rome, with their choice of route noted down . Some more of those units were sent from selected squares to find out the finer details of the pattern. I assume that the game checks the path of a trade route with the GoTo-routine and, as DaveV pointed out in another thread, it is likely that the programmer reused the GoTo-routine rather than writing a new for this function. So far this theory seems to be correct. The asymmetry in the pattern that you mentioned is intentional, that is, by me at least.

Rasputin - The picture shows where to build a road between two cities, the receiving city is the center square, the city sending out the caravan is any of the other squares and the road should be built following the line connecting those two squares. The picture shows only the forward route, if you want the bonus for the return route you have to place the other city in the center and see if more roads are needed. Did this make the picture any clearer?

There is another use for the picture. When a city is connected to the capital, waste (loss of shield production) in that city is reduced. A city that can produce 4 shields without waste in despotism, can then produce 6 shields without waste. This connection needs only to be built in one direction, towards the capital.

[Blaupanzer]

Quantam Satis, very impressive; thank you. Any thoughts, anyone, on why one quadrant would be assymmetric to the other three?

[SlowThinker]

bump.

Originally posted by Blaupanzer
Any thoughts, anyone, on why one quadrant would be assymmetric to the other three?

Is the down-left quadrant an error?

Did anybody verify the picture?

[DrSpike]

to SlowThinker for bumping this thread. That diagram above is the most useful thing I've seen in months.............all those occasions swearing due to no bonus..........I hope it turns out to be accurate.