Post patch Detailed Trade Route Discussion

External trade routes, on the other hand, benefit the less productive city, so if you trade with a science powerhouse city you get a lot of science back, while they don't get much.

Yes, but it isn't a viable enough compensation for the internal trade route fail imo. I hardly encounter AI players that are ahead of me and whose cities I could actually profit from. :(

Did they get a variable the wrong way round on internal routes?

I hope so because I like it the other way around.
 
Yes, but it isn't a viable enough compensation for the internal trade route fail imo. I hardly encounter AI players that are ahead of me and whose cities I could actually profit from. :(

Play Apollo.
 
Did a few tests with food TR yield. One may (or may not) find it useful.
Spoiler :


Format is [total food generated/food yield]

Experiment #1

12/6 +2 -> 25/9
12/6 +1 -> 21/13
12/6 +5 -> 14/0
12/6 +5 -> 15/1
12/6 +1 -> 40/16 +1

25/9 -> 21/13 +2
25/9 -> 40/16 +3
25/9 -> 12/6 +2

First one sounds reasonable. Second - you get 1 food less from bigger difference in surpluses than in first line.
Third and Fourth - you get +5 food from a city, that has much less food surplus and not really different amount of food total.
Fifth line is the weirdest case.
So, a city with lesser amount of food causes a surplus in a sending city, and city with greater amount causes a surplus too.
But before saying 'Hey! So food only goes to the sending city!' check the second block, especially the last line. It's the inverse one of the first.

So basically two conclusions here
1) Trade routes were designed to be 'symmetrical' (no difference if you swap the sender and the target)
2) Either something else is taken into account, or the opposite - some part of yield is not taken into account. Otherwise it's not possible to explain why both high-food and low-food cities provide a surplus in a medium-food city.

Experiment #2
28/10 -> 22/14 +1
27/9 -> 22/14 +2
26/8 -> 22/14 +3
25/7 -> 22/14 +4
24/6 -> 22/14 +4
21/3 -> 22/14 +4
19/1 -> 22/14 +4
16/-2 -> 22/14 +4 (?!)
14/-4 -> 22/14 +4 (?!)

Makes little sense.
Changing the food surplus affected every TR available in the same way, although numbers deviated a bit. Two other cities were 40/16 and 13/5. If the 40/16 city is almost the same as 22/14 for the purpose of experiment, in the case of 13/5 city I've at least passed the point where my surplus became less than 5, so logically speaking TR should've inverted food direction, but that's not happened.

Experiment #3

28/10 -> 22/14 +1
28/10 +1 -> 21/13 +1 (!!)
28/10 +1 -> 20/12 (!!!)
28/10 +2 -> 19/11
28/10 +4 -> 18/10 (?!)
28/10 +4 -> 15/7

Inversion happened here.
Actually, that's kinda makes sense. Probably if I could've improved my food over the 28/10 in an experiment #2, I believe I would've seen this behavior.
Weird +1/+1 case seems like a roundup effect. No other possible explanation.

Experiment #4
28/10 +2 -> 19/11
27/9 +1 -> 19/11
26/8 -> 19/11 (!!!)
25/7 -> 19/11 +2 (!!!)

So yes, an inversion of direction happens whether you change yield in one city or in the other one. Note that +1/+1 TR magically changed into +0/+0, and the next step became +2 instead of +1. Really smells like a roundup.

Experiment #5.
Whether TR yields are counted into calculation.
Let's take this piece:
28/10 -> 13/5 +1
28/10 +1 -> 12/4 +1
28/10 +1 -> 11/3
28/10 +2 -> 10/2

So I sent an auxilliary TR from other city, which pushed a city on the right side from 10/2 to 12/4. But the effect of test TR was still +2 to the sending city. Therefore, TR yields are not taken into account. This explains why both sending a TR to city with high food yield and low food yield can induce a similar result

Experiment #6.
Similar to #5, but about buildings that give food. Buildings are counted into TR yield calculation.

Experiment #7.
Review of #2 and #3 results
City on the left side had +6 from TR's, and city on the right side had +9.

22/4 -> 13/5 +1
21/3 -> 13/5 +2
20/2 -> 13/5 +3
19/1 -> 13/5 +4
18/0 -> 13/5 +4
15/-3 -> 13/5 +4
13/-5 -> 13/5 +4
10/-8 -> 13/5 +4
8/-10 -> 13/5 +4

22/4 -> 13/5 +1
22/4 +1 -> 12/4 +1
22/4 +1 -> 11/3
22/4 +2 -> 10/2
22/4 +4 -> 9/1
22/4 +4 -> 6/-2

Eureka.
1) Food TR yield depends on difference in yields
2) Other TR's are not taken into account (snowballing protection)
3) TR yield is given to the city with bigger yield
4) Negative yields are truncated.

Experiment #8.
There is still unclear stuff. Food TR yield doesn't seem to be linear function of difference of yields.
e.g.
22/4 +2 -> 10/2
22/4 +4 -> 9/1

1 point of yield difference is converted into 2 points of TR yield.
Another issue is the nature of those roundup effects.
All of that implies that TR gain may be non-linear.

Consider this:
37/13 +7 -> 18/0 13
37/13 +7 -> 19/1 12
37/13 +6 -> 20/2 11
37/13 +5 -> 21/3 10
37/13 +4 -> 22/4 9
37/13 +4 -> 23/5 8
37/13 +3 -> 24/6 7
37/13 +3 -> 25/7 6
37/13 +3 -> 26/8 5

Weird, right?
So, the more potential yield difference is, the less steep the curve becomes.
Why it's like that and how it works I'm not sure yet. I'm afraid I have to start a new game to check that. The current one had a terrible hill terrain and was pretty much ruined by my tests.

Good work
 
This is what i would recommend they change it to.
1 international trade routes.
Sender gets 0,5 gold/science per pop. reciever gets 0,25 gold/science per pop.
trade route value is allways set by the smaller city.

internal trade is set to +1 Production and Food per improved bonus tile and stategic tile that is different from the 2 cities. also adds 0,20 Food and Production per pop.
Exsample:
City 1 is size 5 and has improved tubers and fungus and oil.
city 2 is size 30 and has improved titanium and oil

city 2 would then get 2 Food and 2 Production from resouse difference and 1 Food and Production from pop for a total of 3 Food and Production.
city 1 would get 1 Food and Production from resourse difference and 5 from pop for a total yield of 6.

total trade value would then be 9/9 which is fairly decent for such a cheap unit.
 
I still feel the food yield directions are backward.

Since no food is ever taken from either city the most reasonable explanation for a city gaining food from a TR is that the TR food is food that is not usable by other city but is usable by this city. This makes sense given that variety in food is essential for a population; having massive amounts of one particular type would be a waste in terms of growing the population in a city - so some of this excess could be given away without having any negative impact on that city.

Why wouldn't the city optimise the variety of food it makes? Well, this may not be possible for several reasons such as lack of appropriate conditions, lack of experience in all the different varieties, etc. Futhermore, specialisation of a trade-able product can have benefits if you can trade for what you need in return (mass production via optimising conditions for one type of product, greater experience with that product, minimising of different equipment and space requirements, etc.). So, it would likely be quite common for a city to produce more of one particular product than it needs while lacking on other products - especially if they know they can make a trade.

Even if you take food to mean something more abstract it still makes far more sense that a city with a greater amount of whatever food is would provide a greater amount of TR food to the other city. It would take something particularly perverse to make sense in the other direction, I feel.

I feel, therefore, that food yields of the TR should be a percentage of the food each city produces being transferred (without being removed from the city) to the other. The issue here is granularity; with a small range to work from it is difficult to have percentages that result in meaningful yield differences; instead you get sudden jumps (thresholds).

For example, lets say we have a city, Bobcity, with 17 food and another city, Billcity, with 24 food; using a 10% yield and rounding we would get a TR yield of 2 food for Billcity and 2 food for Bobcity... even though Billcity produces 1.4 times the food that Bobcity does the TR yield is exactly the same! However, if Billcity produced one extra food the respective food yield would go from 2 to 3. This is caused by the coarse granularity involved.

Coarse (or low) granularity can be made less of an issue by increasing the range involved (making it finer, or higher); example, instead of one city making 17 food and another 24 food, they could respectively make 170 food and 240 food (i.e., increased by a factor of 10). Problem is that this makes things less easy to comprehend at a glance, as well as the problem of displaying the amount on the world tiles. The TR yield (at 10%) here would be 17 and 24, though - quite different from each other, unlike before.

Another way to make coarse granularity less of an issue is to have fluctuating yields based on probability, revolving around the (true) value. So, the before example with 10% yield as having a (true) value of 1.7 food and 2.4 food, both rounding to 2 for the yield. By fluctuating around the (true) values (say by random equally-distrubuted chance of adding a number between -0.5 and 0.5) we can get yields varying between 1 and 2 (averaging out as roughly 1.7) for the first yield of 1.7 and 2 and 3 (averaging out as roughly 2.4) for the second yield of 2.4. This could occur dynamically each turn - however, I am not sure how much turn-time overhead this might add (it is a pretty simple calculation and there wouldn't be that many TRs...). The true value could be displayed in the TR panel, so that you can properly compare TRs. Note, though, that this could lead to cities growing a turn earlier/later than would have been the case if there was no randomness involved, but this is not necessarily a bad thing.

P.S. The same thing could be done with production.
 
There has to be more to it than differences in yields.
I have a size 53 city making over 200+ surplus food, and trading with newly founded cities is no more beneficial that trading with size 6 or 7 cities.
Maybe potential dirt yields or a weird table that scales parabolic.
 
Outstanding work by those who tested. Bravo.

The amount of effort Firaxis has put into a terrible idea is remarkable, truly remarkable.
 
Thanks for the work on yields.

I think there's gotta be two different systems, one for Food, one for Production. Food is based on relative surpluses - lets call it "Food, Base Excluding TR" (FBETR). It only goes one way. Net food from a TR can vary from 0-10, at least.

But why does total production yield from a possible TR seem, except in extreme cases, to be constant, but the distribution changes? Note that the total nuts from a hypothetical route from Sender won't change even if the base nuts in the destination city vary widely between Destinations A, B, or C. Option A might give 100% of the production to Sender, Option B might give 80% to sender, Option C gives 90% to Destination - but the total nuts from the route is constant. The only real exception is when one of the cities is newly founded and has VERY low production.

Sender is not unique. A, B, and C exhibit the same behavior. Because the routes mirror, if all the routes into and out of Sender give the same net nuts (except the extreme, as noted), then it follows that all the routes into and out of A, B, and C need to give the same net nuts - and we end up with a situation where it seems that the vast majority of the potential internal routes available will yield the same net nuts. That value changes over time - it clearly rises - but it is fairly constant colony-wide, seeming to govern pairs of cities with high and low production, and high and low differentials.

So what is that Net Nut value based on? Faction-wide production? Very odd.
 
Historically speaking, water trade routes no longer receiving increased yield doesn't make sense. Sure they circumvent land borders and terrain obstacles, but the ease of travel over water and the capacity to carry more is why sea trade today is as important as it is today.

I guess that's why 'Beyond Earth' and not Civ 6 =\
 
Sea trade is exactly why coastal cities flourish, in modern day civilization.
I guess we're really beyond earth with this change in mechanics.

#unreal

Not talking historically but in terms of game mechanics.

If you wanted to model them historically, sea routes would not give more yield, but would cost less maintenance for certain distance.

Something like
Trade vessel= .05 energy per tiles distance
Trade convoy= .3 energy per tiles distance (.2 if over roads, .1 if over magrails)
Trade airlift = 1 energy per tile distance

all with the same yield
 
I am in shock. Please tell me I am not the only one who thinks these new trade route calculations are almost game breaking. What the point of even having trade routes if they're going to benefit your competitors 2 or 3 times more then you? I understand the argument of "well it makes your choices more important", and that'd be great, if it wasn't for the fact that I'm getting 2, 3 energy max from these routes.

Even worse, internal routes are completely broken now too. I'm playing a game where I'm trying to get my second city off the ground and my ITR to it is giving it literally nothing but for some reason bringing 7 food and 8 production back to my capital? What's the point? I don't need it for my capital, I need it for the new city. That was always sort of the purpose of ITRs to begin with, since in Civ5 and BE getting a new city off the ground took a ridiculous amount of time and instead fixing the problem they just added a new component to the game we had to take into consideration instead.

I have no idea why anyone would defend the changes made to this system. It is nonsensical.
 
Yeah, the system creates some really odd scenarios.

I am currently experimenting with a strategy where I found one city in a desolate place, keep it at size 1 with minimum yield and then re-route all my trade routes to it to maximize trade yield for my other cities. So far it seems shockingly effective (I have gotten up to 16 total yield per trade route out of it with Industry tree), particulary when I already have a big empire (so that the science and culture penalty from the new city has less impact).

Tbh I'd prefer if we could just go back to a CIV5 style trade system. Limited number of trade routes, one way only, sortable by max yield and delta yield. The current system is just a PITA...
 
If the external route is bad then don't use it. You got internals for that situation. In my current game several cities have good choices for good external routes while others have none. I do agree though that more options doesn't mean more choices. The choice between ice-cream and frozen-yogurt isn't that different from a choice of, ice-cream, frozen-yogurt, or a punch in the face. Options ≠ choices.
 
I am in shock. Please tell me I am not the only one who thinks these new trade route calculations are almost game breaking. What the point of even having trade routes if they're going to benefit your competitors 2 or 3 times more then you? I understand the argument of "well it makes your choices more important", and that'd be great, if it wasn't for the fact that I'm getting 2, 3 energy max from these routes.

Even worse, internal routes are completely broken now too. I'm playing a game where I'm trying to get my second city off the ground and my ITR to it is giving it literally nothing but for some reason bringing 7 food and 8 production back to my capital? What's the point? I don't need it for my capital, I need it for the new city. That was always sort of the purpose of ITRs to begin with, since in Civ5 and BE getting a new city off the ground took a ridiculous amount of time and instead fixing the problem they just added a new component to the game we had to take into consideration instead.

I have no idea why anyone would defend the changes made to this system. It is nonsensical.

Externals make sense to me - the city producing more science/energy gets less from the route. So what you do with those is get your new tiny cities to sponge off the enemy's old huge cities.

Internals are the other way around - the bigger producer gets more out of the route. As you've correctly noted, this means you can't use trade routes to pump up your new cities any more.

I think they got a variable the wrong way around on the internal trade routes, but people have argued that it's working as intended, and who's to know either way? As far as I know Civ's designers have never said what the trade system is actually intended to achieve, so it's difficult to say if anything's broken or not.
 
Yeah, the system creates some really odd scenarios.

I am currently experimenting with a strategy where I found one city in a desolate place, keep it at size 1 with minimum yield and then re-route all my trade routes to it to maximize trade yield for my other cities. So far it seems shockingly effective (I have gotten up to 16 total yield per trade route out of it with Industry tree), particulary when I already have a big empire (so that the science and culture penalty from the new city has less impact).

Tbh I'd prefer if we could just go back to a CIV5 style trade system. Limited number of trade routes, one way only, sortable by max yield and delta yield. The current system is just a PITA...

you can use that very same strategy to grow your empire... settle the new city, send trade routes to it, but don't grow that new city until the next one is settled... this means settling a new city every 20 turns or so... rinse and repeat, always having a 1 pop city to send trade routes to. Done expanding, then that 1 pop city in the tundra is a great destination for your large cities.
 
that strategy is actually bad.

It is simple nature of trade route yield, you only win 3-4 production from sending to no production city compare to medium production city and medium city will bring you Mach more of everything . you only need 2 pop max to cover difference in trade route yield.
so growing your cities is always the best strategy.
 
that strategy is actually bad.

It is simple nature of trade route yield, you only win 3-4 production from sending to no production city compare to medium production city and medium city will bring you Mach more of everything . you only need 2 pop max to cover difference in trade route yield.
so growing your cities is always the best strategy.

well that is your opinion, but when I am getting 3 or 4 food, and 8 or 9 production per trade route to my 1 pop city, it grows my empire quite quickly. All trade route yields are based upon the differential of the two cities, with the largest producer getting the benefit.

And I never said, I don't grow my cities... I just always have a 1 pop city to send my trade routes to.
 
Top Bottom