Railroads--worth it?

[basta]

The title pretty much says it all--is building railroads worth it, and how do you optimize them?

By the time the tech rolls around--usually an RA "gift" as I never tech for it (should I?)--I've disbanded most of my workers to save the high late-era maintenance costs and have a continent pretty much to myself.

I do build RRs but it's usually 1 E-W and 1 N-S route, one at least going through the capital, for quicker unit movement. Does the RR production bonus make it worth redoing your entire road network, and linking up minor cities?

 

[EasyGoin]

Depends largely on your economic state when you get RR, but overall they are absolutely worth it.

- Any city connected to the capital by RR gets 50% production bonus
- if you connect one city with RR by land and your capital has a harbor, any other cities connected by harbor get RR (as I understand it). In the alternative, if you RR to your harbor city, any cities connected by harbor get RR bonus. (again as i understand it).

Is it worth it to connect all your cities? probably not. But definitely to connect your top production cities.

you can just put the RR over your roads, unless your roads meander needlessly and take extra hexes. At 2gpt per tile you want to take the shortest route possible that hits all your primary production cities. Assuming my understanding above is correct, coastal cities would preferably be connected by harbor (3gpt upkeep if i recall correctly, equivalent to only 1.5 tiles of RR).

 

[diz2]

If you build them to an already strong production city they can bring a huge amount of production for a relatively small amount of gold.

You should generally try to set up your production cities so they can be railed/harbored together with reasonably few tiles.

One thing I've wondered about that maybe someone can answer: Is the cost of the railroad 2 gpt per tile, or 2 gpt per turn above the cost of the road you replace? I have always assumed it was the former. So, if you lay a 3 tile railroad over an existing road your cost goes from 3gpt to 6gpt. For that extra 3 gpt, you might get 20 or 30 extra production. That's sure worth doing.

 

[alpaca]

If you build them to an already strong production city they can bring a huge amount of production for a relatively small amount of gold.

You should generally try to set up your production cities so they can be railed/harbored together with reasonably few tiles.

One thing I've wondered about that maybe someone can answer: Is the cost of the railroad 2 gpt per tile, or 2 gpt per turn above the cost of the road you replace? I have always assumed it was the former. So, if you lay a 3 tile railroad over an existing road your cost goes from 3gpt to 6gpt. For that extra 3 gpt, you might get 20 or 30 extra production. That's sure worth doing.

Yeah it's 2 and replaces the road. The cheapest way to hook things up is harbors. In continent games I often find myself removing all my roads once I took control of the continent and replace them with harbors. They immediately connect your empire the turn you research them, too, so you don't have to wait for dozens of turns to build those railroads.

On Pangaea I'll typically have a few railroad connections but only to my most productive cities. It's worth a couple of tiles even in a mediocre city, though (2 is about 1 in most games). If you play Iroquois, you get the bonus but not the extra movement for all cities connected to your forest through your culture area - in my zero-culture ICS game I connected more than 20 cities with one railroad tile.

 

[Bibor]

Yes, the railroad connection (the 50% bonus) is awesome.

As for railroads, I have all my production cities connected to (possible) frontlines.
Other cities either get a harbor, stay without, or I grudgingly connect them via railroad.

 

[diz2]

It's worth a couple of tiles even in a mediocre city, though (2 is about 1 in most games).

I don't tend to connect my non-production cities because they generally aren't building things of significant value by the time railroads are available.

I guess one question would be, how much base production would you have to have to grow the production of Wealth enough to justify a railroad of X tiles?

What is the wealth conversion rate, 25% gpt/hammer?

So, if you start with 8 hammers, you get 4 more from railroads = 1 gpt if you produce wealth = not worth building even 2 tiles of railroad to connect. And I think most of my non-production cities have less than 8 hammers most of the time.

 

[Supr49er]

They definitely are worth it to connect your Spaceship part cities with your capital.

 

[ChevalierdeJstn]

I see a myopic lack of cost/benefit analysis in this discussion.

Let's say you have a city with production 20 per turn and it is 10 tiles from your capital.
To connect it with a railroad (without modifiers) would cost maintenance 10*2=20 per turn.
Post-railroad production would be 20*1.5=30 per turn.

For a 300 hammers-cost (just to make for round numbers), it now takes you 10 turns to build, during which you are paying 200 maintenance on the railroad. Without the railroad it would have taken you 5 extra turns to build. That's 100 hammers at the old production speed.

So you're spending 200 gold to save 100 hammers. Is there some reason that gold is only valued at half a hammer? Note this is not even including the extra maintenance from keeping workers around to lay rails.

If you're planning on building something like the Utopia Project which can't be purchased, in a city which is not your capital, then _maybe_ building railroads right before beginning said production is worthwhile. Otherwise, given the OODA-loop benefit of having the immediate purchase option...why would you make this trade?

Where could my example/analysis be wrong. Well, what if your average "production city" is not 10 turns away from your capital but only 5. Now RR maintenance is 10 per turn, so you're trading 100 gold for 100 hammers. That sounds more reasonable (though I strongly suggest that the immediate purchase option has a HUGE tactical benefit which is not being quantified here.)

So how often do you have a major production city which is 5 tiles or fewer from your capital? That's pretty much impossible in my games, simply because the capital is always one of the larger cities and is almost certainly working quite a few of the tiles which a nearby city would need to be a production powerhouse.

What's the other benefit from a railroad - 1/3 movement cost or thereabouts. Compare with a road which is 1/2 movement cost for half the maintenance. The tactical benefit of this increased movement depends on your specific situation, but...how long are you sitting around paying maintenance on railroads before using that movement benefit in a battle inside your own borders? Couldn't you have insta-purchased a unit (or more) where needed for what you paid in maintenance on the railroads?

Conclusion: MAYBE in specific cases where you plan WELL in advance to make the best use of the railroad's hammers-for-gold trade, railroads are worthwhile. It seems likely that this celestial alignment will be rare.

Final note: I am not at all a proponent of realism in Civ (it's a game, not a simulation) but it's interesting to note that this treatment of railroads is completely bass-ackwards. Historically, railroads cost a lot to build but result in improved commerce AND production once completed. Like roads, with judicious placement the trade route benefits of railroads should (realistically) at least pay for the maintenance cost.

 

[DaveMcW]

ChevalierdeJstn is wrong.

alpaca is right.

 

[Monthar]

I see a myopic lack of cost/benefit analysis in this discussion.

Let's say you have a city with production 20 per turn and it is 10 tiles from your capital.
To connect it with a railroad (without modifiers) would cost maintenance 10*2=20 per turn.
Post-railroad production would be 20*1.5=30 per turn.

For a 300 hammers-cost (just to make for round numbers), it now takes you 10 turns to build, during which you are paying 200 maintenance on the railroad. Without the railroad it would have taken you 5 extra turns to build. That's 100 hammers at the old production speed.

So you're spending 200 gold to save 100 hammers. Is there some reason that gold is only valued at half a hammer? Note this is not even including the extra maintenance from keeping workers around to lay rails.

First off, you most likely already have a road connecting the capital to this city, so you're already paying 10 gpt maintenance on that road. So you're really only increasing the cost by 10gpt or 100 gold for the 10 turns in your example. Thus it's 100 additional gold for 100 additional hammers.

You have to subtract the trade route income from cost of the railroad. If your trade route income is 100 gpt and maintenance cost of all your roads is only 50 gpt and this one 10 tile route is the only one you are going to upgrade to a railroad. You're only increasing the maintenance cost by 10 gpt. 100 TR income - 60 tile maintenance = net profit of 40 gpt. Which is more than enough to cover the cost of those workers you had building the railroad.

 

[JohnnyW]

Yes, you must account for the trade route income. We should also consider potential factories.

Let's say I have a city to have the great luck of a bunch of riverside hills. It's pop10 and making me 20 base While being 6 spaces from my capital. It also has a factory for a total of 30 /turn, and I currently have a road going there.

Road: -6 +12.5
Factory: -3 +10.
Subtotal: +3.5 +10
Rails: -6 +10 .
Total: -2.5 +20

Only rails: +0.5 +10.

edit: each hex closer/farther from capital is +2/-2. Each is +1.25. Each base is +0.5 for rails and factory.

 

[Bibor]

Let's say you have a city with production 20 per turn and it is 10 tiles from your capital. To connect it with a railroad (without modifiers) would cost maintenance 10*2=20 per turn. Post-railroad production would be 20*1.5=30 per turn.

No. Your calculation is way off. There must exist either a harbor or road already. And cities are rarely if ever more than 5 hexes apart. Except on another landmass, which again points to a harbor. Thus, realistically, as you already have a road connection, the railroad simply adds 2 PT to 5 PT to your existing road network costs. Usually its 4 PT, which makes is slightly more expensive than a factory but spends no coal and costs no hammers.

If your city is size 10, which it should be, it's already paying its (rail)road connection costs by trade routes alone, even if its 5 hexes away.

The railroad unit movement speed is so negligible that you only need to connect your inland cities via a railroad network; coastal cities might as well build a harbor to save on maintenance costs. I never play on large or huge maps though, so the movement bonus might be worth it there.

 

[Smote]

Its simple.

A railroad connection is worth building if you can spend less than 2x gpt to get x hammers.

In the limiting case (alpaca's 1 railroad with iroquois), he spent 2gpt to get probably about 100 hammers.

 

[Maestro_Innit]

Crikey, I couldn't imagine not using railroads! I make a good effort to research it as quick as possible since production is probably the most elusive thing int he game and a 50% boost in, say, your top five cities is amazing.

 

[ryantherenegade]

rails are defo essential for the science and conquest victories. Maybes not the culture or diplo.

 

[SeerTenedos]

+1 ryantherenegade. It all goes back to your victory type, and what you want. I think by the time you have RR tech you throw the 2 gold = 1 hammer ratio out the door to meet your specific needs for what remains in the game. For late game scenarios where I am behind in tech and relying on RA's and still trying to get more votes my ratio is almost completely inverted.

rails are defo essential for the science and conquest victories. Maybes not the culture or diplo.

 

[JimT]

rails are defo essential for the science and conquest victories. Maybes not the culture or diplo.

On culture, you need any boost you can get for the Utopia project. This is unless you're building it in the capital so a link isn't needed and you only need to research it.

 

[code9]

What I usually do is plan far ahead. 3 or 4 of the cities closest to my capital usually have some forest or hills around them so they become production centers. This is useful in the early game, as well as later because you only need small amounts of RR to get larger total bonuses in your cities.

 

[metatheory]

Maybe I'm blind or just haven't noticed but all my railroad connected cities have not shown a 50% production boost. I made a point to watch production numbers when I finished connecting two cities last time and didn't notice a change. Is there any where in the UI to verify this bonus?

 

[Bibor]

Maybe I'm blind or just haven't noticed but all my railroad connected cities have not shown a 50% production boost. I made a point to watch production numbers when I finished connecting two cities last time and didn't notice a change. Is there any where in the UI to verify this bonus?

Hover over the total yields display in the city screen. You'll get messages like "unit domain +25%" or "Railroad connection +50%" or "City modifier +65%" depending on what you are producing and what buildings you have.