How to plan for future traffic.

[startrecker 0]

Hello everyone.  So I apologize if this was answered elsewhere but I can't seem to find it.  I just recently got into skylines and I am confused about something.  Most of the ideas I have seen for dealing with traffic are great but they are expensive.  So my question is how do you guys plan for future traffic issues when starting a new city?  Do you just build cheap at first and then bulldoze to put in the needed infrastructure like DCMI's later or do you have some way of laying the groundwork at the start without bankrupting your city?

[BlakeMW 84]

Unfortunately because the traffic AI is pretty stupid highways are not really a good way to solve traffic problems. I am a big fan of "ramp highways"; because of AI inadequacies 3 lanes is not terribly better than 1 so you may as well go with the much cheaper and easier to use ramp. Also because cims aren't good at utilizing extra lanes it tends to be best to stick with 2 lane roads and use mass transit or bypasses to remove traffic when the road becomes overloaded (roads never become overloaded due to local traffic, it's always traffic driving across the city which overloads a road, so it can always by sent via a more direct & higher speed bypass, ramps make ideal bypasses).

As for planning, well that's really simple. Just use tunnels to make buried bypasses and highways, it's not terribly interesting but it's very easy to do and surprisingly economical. 

What we need to consider here is opportunity cost. When we build a highway on the surface we miss out on the taxes we could have earned from zoning that land. High Density Residential at lvl5 brings in about $4 per square. To be fair we should dock $1 for education & utility costs (which are per-cim or building and not area based coverage) and say $3 per square (as a note: Commercial with BBF Policy can be at least as profitable as Residential with virtually no per-building expenses and can definitely earn $3-4 a square per week)

Now consider the 2 square wide ramp which costs 0.32 per unit, the ramp is 10 tiles long. 

As a terrestrial road it covers 20 squares which carries an opportunity cost in lost taxes of $60/week and it costs $3.2/week in upkeep, making a total real cost of $63.2

As a tunnel it covers 0 squares and because the maintenance is 6x higher, it costs $19.2/week. With a tunnel you are winning big-time in terms of real costs.

Now consider the 4 square wide highway which costs 0.96 per unit, the highway is 10 tiles long.

As a terrestrial road it covers 40 squares which carries an opportunity cost in lost taxes of $120/week and it costs $9.6 in upkeep, for a total real cost of $129.6/week.

As a tunnel it covers 0 squares and the maintenance is $57.6/week, again with a tunnel you coming out ahead.

What this means is that tunnels come out as the preferable option for the budget-conscious Mayor, in fact we can be rather less generous with our assumptions about how much a square of residential (or other zone) can earn and still have tunnels the preferable option. The break-even point is $0.8 per square for the ramp and $1.2 per square for the highway. These numbers are comparable to the tax income from industry/office which means in those zones tunnels are still breaking even on the budget balance.

But when we consider that tunnels will be shorter and more direct, that they require minimal demolitions, that their on/off ramps and interchanges require a lot less space and that they are far easier to modify (especially with Moledozer mod) we are forced to conclude that tunnels under the city are the way to go. As previously noted, tunnels are not the most interesting option, but in terms of gameplay they are very very optimal, so good it's almost cheating.

 

[Linoa06 1,374]

Unfortunately because the traffic AI is pretty stupid highways are not really a good way to solve traffic problems. I am a big fan of "ramp highways"; because of AI inadequacies 3 lanes is not terribly better than 1 so you may as well go with the much cheaper and easier to use ramp. Also because cims aren't good at utilizing extra lanes it tends to be best to stick with 2 lane roads and use mass transit or bypasses to remove traffic when the road becomes overloaded (roads never become overloaded due to local traffic, it's always traffic driving across the city which overloads a road, so it can always by sent via a more direct & higher speed bypass, ramps make ideal bypasses).

As for planning, well that's really simple. Just use tunnels to make buried bypasses and highways, it's not terribly interesting but it's very easy to do and surprisingly economical. 

What we need to consider here is opportunity cost. When we build a highway on the surface we miss out on the taxes we could have earned from zoning that land. High Density Residential at lvl5 brings in about $4 per square. To be fair we should dock $1 for education & utility costs (which are per-cim or building and not area based coverage) and say $3 per square (as a note: Commercial with BBF Policy can be at least as profitable as Residential with virtually no per-building expenses and can definitely earn $3-4 a square per week)

Now consider the 2 square wide ramp which costs 0.32 per unit, the ramp is 10 tiles long. 

As a terrestrial road it covers 20 squares which carries an opportunity cost in lost taxes of $60/week and it costs $3.2/week in upkeep, making a total real cost of $63.2

As a tunnel it covers 0 squares and because the maintenance is 6x higher, it costs $19.2/week. With a tunnel you are winning big-time in terms of real costs.

Now consider the 4 square wide highway which costs 0.96 per unit, the highway is 10 tiles long.

As a terrestrial road it covers 40 squares which carries an opportunity cost in lost taxes of $120/week and it costs $9.6 in upkeep, for a total real cost of $129.6/week.

As a tunnel it covers 0 squares and the maintenance is $57.6/week, again with a tunnel you coming out ahead.

What this means is that tunnels come out as the preferable option for the budget-conscious Mayor, in fact we can be rather less generous with our assumptions about how much a square of residential (or other zone) can earn and still have tunnels the preferable option. The break-even point is $0.8 per square for the ramp and $1.2 per square for the highway. These numbers are comparable to the tax income from industry/office which means in those zones tunnels are still breaking even on the budget balance.

But when we consider that tunnels will be shorter and more direct, that they require minimal demolitions, that their on/off ramps and interchanges require a lot less space and that they are far easier to modify (especially with Moledozer mod) we are forced to conclude that tunnels under the city are the way to go. As previously noted, tunnels are not the most interesting option, but in terms of gameplay they are very very optimal, so good it's almost cheating.

 

More highways isn't always the solution anyway even if the AI is perfect - there's always some extra demand beyond the capacity you add to your network after all, though whether this is verified in the game is yet to be seen. (Perhaps, as upgrading roads may make some agents use it when they used a different route before)

Tunnels are pretty good though albeit slightly more expensive, as with metros I wish anything underground related was far more expensive because yes right now it's almost cheating

[startrecker 0]

So do you just take standard straight roads in from the highway at the start with normal intersections when your beginning.  I am having a hard time wrapping my head around the starting layout.  I am coming from sim city 4 and the roads here feel much more expensive relative to the starting money compared to that game that I don't quite know how to set up a layout with such difficulty laying out the road network.  The ways I have tried either uses standard intersections and get clogged up fast or they are very expensive with interchanges and I barely can stay afloat money wise.

[ethansprang 41]

The answers are

1. Yes, you'll probably bulldoze a bunch later in the game to upgrade capacity.  You will have TONS of money.

2. Your starting point will all be low-density, but the later additions will be higher density, so the traffic problems will move away from them, especially if you add a second interchange when you get about 10K population.

So, how?  Start with dirt roads everywhere.  Very cheap - and big enough for quite a while.  Make the streets leading from the offramp into basic 2-lane roads, but that's about it, until you get firmly in green and your pop is over 2K.

[JJogurt 18]

A single 4-lane road starting from the highway works well. Then I'll zone residential at the far end of the road, industry near the highway, and commercial in the middle. Everything else comes as your city develops and you get more money.

[BlakeMW 84]

So do you just take standard straight roads in from the highway at the start with normal intersections when your beginning.  I am having a hard time wrapping my head around the starting layout.  I am coming from sim city 4 and the roads here feel much more expensive relative to the starting money compared to that game that I don't quite know how to set up a layout with such difficulty laying out the road network.  The ways I have tried either uses standard intersections and get clogged up fast or they are very expensive with interchanges and I barely can stay afloat money wise.

I'm a very boring person and always use a square grid which looks like this:

If you look carefully you see the grid is a square of 4x4 blocks, I divide the square in two with a road which usually ends in a cul-de-sac, making it clearer this is a square grid and not a rectangular grid, also not joining the dividing road adds 6 more tiles which can be occupied by a tax-paying building.

This design is rather boring but it is incredibly effective (by the way I rotate the dividing road at random because it makes the grid look much less boring). The spacing between intersections is ideal for reducing congestion and the 3-way intersections cause virtually no congestion because there is no through-traffic and a vehicle hardly ever turns at them. The use of square grids makes it easy to spam buslines (I have a bus line going up and down every single road both North/South and East/West, each line forms a loop up one road down the next), the 4x4 block square grid has another very attractive property - if you spam buslines so that each edge of the square has 2 bus stops  (one on the near side of the road, one on the far side) the 8 bus stops will be just enough to max out the transportation rating. As transportation is the most valuable service to provide for leveling up buildings and as bus stops are free to create and as buses provide you with income as fares this will dramatically improve the financial situation of the city.

Early in the cities life (as in the screenshot) I just attach the grid directly to the highway. Later on the demands of high density residential/commercial will make this stop working, so usually once I have a nest egg of about $200,000 and have purchased the block of land the highway comes from I bulldoze the existing highway attachment point and extend the highway (as a tunnel) under the city, extending at least 2-3 blocks, I make the required on and off ramps, attaching them all to different blocks to distribute traffic. As the city grows larger the highway tunnel can be extended further under the city with more on and off ramps added to further distribute traffic and it is a good idea to have direct connections to any industrial block as you really don't want trucks trundling along random city streets.

Chokepoints are the enemy of good traffic flow. The square grid is ideal for minimizing choke points and distributing traffic evenly, the roads will be well used but because the traffic is distributed evenly and because the spacing between intersections is excellent there will be no real congestion and traffic flow will be smooth. It is hard to improve upon this design for a few reasons. If you use wider roads they cost a lot more in upkeep and come with traffic lights (unless disabled by a mod). Because they take up twice as many tiles there is significant opportunity cost in terms of lost tax income. From a financial situation it's a no-brainer to use the cheap standard road (I also use the more expensive "road with trees" but only selectively for example around a metro station in a residential district because it cuts down noise a lot - but most the time there either isn't much noise or the zone type doesn't care about noise so the standard road or the road with bike lanes works just as well). It really doesn't work at all to mix standard roads and wide roads - because the wide roads have a higher speed limit they act as traffic magnets and because they are attracting traffic like crazy they then cause congestion either on the wide road or the connecting standard roads.

You can almost get away without using highways or ramps at all, the square grid of standard roads with bus line spam and metro support is very very powerful. By the time the city grows large enough for it to become congested the game tends to start reducing the number of agents it generates solving the problem for you. But I do like to use highways and ramps and it can certainly be done.

The best way to join a highway to the city streets is to use an approach something like a tree's roots, have lots of on and off ramps joined at lots of different points along the highway so it can collect traffic from a large area of the city and so that cims aren't all forced onto the same street to get onto the highway.

I like to build ramp tunnels under the city as high speed bypasses - especially between industry and commercial (car traffic is better off put onto buses and metro). One critical aspect to making this work well is to make sure they only connect distant parts of the city. If two nearby parts of the city are connected by a high speed bypass then it acts as a traffic magnet and causes congestion thanks to short-cutters. My normal high speed bypasses might have 3 on-ramps at one end which pulls traffic from 3 different blocks, it then might cross 10 or more blocks across the city and end in 3 off-ramps which distributes traffic over 3 blocks. Along the way there are no on or off ramps because that would permit short cutting and the bypass should only be used by traffic which has to travel long distances.

Following these kinds of principles I've created cities which generate in excess of $50,000 profit/week (at population about 80-90k), profit is a good indicator of how efficiently a city is running because a big chunk of profit comes from sales tax which requires the whole industrial->commercial->shoppers supply chain to be working as smoothly as possible.

[LivingInThePast 364]

 

Unfortunately because the traffic AI is pretty stupid highways are not really a good way to solve traffic problems. I am a big fan of "ramp highways"; because of AI inadequacies 3 lanes is not terribly better than 1 so you may as well go with the much cheaper and easier to use ramp. Also because cims aren't good at utilizing extra lanes it tends to be best to stick with 2 lane roads and use mass transit or bypasses to remove traffic when the road becomes overloaded (roads never become overloaded due to local traffic, it's always traffic driving across the city which overloads a road, so it can always by sent via a more direct & higher speed bypass, ramps make ideal bypasses).

As for planning, well that's really simple. Just use tunnels to make buried bypasses and highways, it's not terribly interesting but it's very easy to do and surprisingly economical. 

What we need to consider here is opportunity cost. When we build a highway on the surface we miss out on the taxes we could have earned from zoning that land. High Density Residential at lvl5 brings in about $4 per square. To be fair we should dock $1 for education & utility costs (which are per-cim or building and not area based coverage) and say $3 per square (as a note: Commercial with BBF Policy can be at least as profitable as Residential with virtually no per-building expenses and can definitely earn $3-4 a square per week)

Now consider the 2 square wide ramp which costs 0.32 per unit, the ramp is 10 tiles long. 

As a terrestrial road it covers 20 squares which carries an opportunity cost in lost taxes of $60/week and it costs $3.2/week in upkeep, making a total real cost of $63.2

As a tunnel it covers 0 squares and because the maintenance is 6x higher, it costs $19.2/week. With a tunnel you are winning big-time in terms of real costs.

Now consider the 4 square wide highway which costs 0.96 per unit, the highway is 10 tiles long.

As a terrestrial road it covers 40 squares which carries an opportunity cost in lost taxes of $120/week and it costs $9.6 in upkeep, for a total real cost of $129.6/week.

As a tunnel it covers 0 squares and the maintenance is $57.6/week, again with a tunnel you coming out ahead.

What this means is that tunnels come out as the preferable option for the budget-conscious Mayor, in fact we can be rather less generous with our assumptions about how much a square of residential (or other zone) can earn and still have tunnels the preferable option. The break-even point is $0.8 per square for the ramp and $1.2 per square for the highway. These numbers are comparable to the tax income from industry/office which means in those zones tunnels are still breaking even on the budget balance.

But when we consider that tunnels will be shorter and more direct, that they require minimal demolitions, that their on/off ramps and interchanges require a lot less space and that they are far easier to modify (especially with Moledozer mod) we are forced to conclude that tunnels under the city are the way to go. As previously noted, tunnels are not the most interesting option, but in terms of gameplay they are very very optimal, so good it's almost cheating.

 

More highways isn't always the solution anyway even if the AI is perfect - there's always some extra demand beyond the capacity you add to your network after all, though whether this is verified in the game is yet to be seen. (Perhaps, as upgrading roads may make some agents use it when they used a different route before)

I like what I'm reading about highways, especially about the lost revenue. The Katy Freeway in Houston comes to mind, carrying five lanes of traffic in each direction, plus two lanes of HOT/HOV lane traffic in each direction, and another 3 lanes of frontage lanes in each direction. The amount of clearances it took to build such a thing is another issue entirely, and nearly wiped out a small urban municipality's entire commercial tax base. Of course, that doesn't mean you have to be "railroaded" into adding rail-based transit to make things work, because if we were also going by reality, rail only works in high density situations/centralized job centers, otherwise you'll never reach critical mass and be pouring money into an ineffective system that barely keeps pressure off of highways. 

Hate to say it, but tunnels should be MORE expensive than elevated structures.

[Linoa06 1,374]

 

Unfortunately because the traffic AI is pretty stupid highways are not really a good way to solve traffic problems. I am a big fan of "ramp highways"; because of AI inadequacies 3 lanes is not terribly better than 1 so you may as well go with the much cheaper and easier to use ramp. Also because cims aren't good at utilizing extra lanes it tends to be best to stick with 2 lane roads and use mass transit or bypasses to remove traffic when the road becomes overloaded (roads never become overloaded due to local traffic, it's always traffic driving across the city which overloads a road, so it can always by sent via a more direct & higher speed bypass, ramps make ideal bypasses).

As for planning, well that's really simple. Just use tunnels to make buried bypasses and highways, it's not terribly interesting but it's very easy to do and surprisingly economical. 

What we need to consider here is opportunity cost. When we build a highway on the surface we miss out on the taxes we could have earned from zoning that land. High Density Residential at lvl5 brings in about $4 per square. To be fair we should dock $1 for education & utility costs (which are per-cim or building and not area based coverage) and say $3 per square (as a note: Commercial with BBF Policy can be at least as profitable as Residential with virtually no per-building expenses and can definitely earn $3-4 a square per week)

Now consider the 2 square wide ramp which costs 0.32 per unit, the ramp is 10 tiles long. 

As a terrestrial road it covers 20 squares which carries an opportunity cost in lost taxes of $60/week and it costs $3.2/week in upkeep, making a total real cost of $63.2

As a tunnel it covers 0 squares and because the maintenance is 6x higher, it costs $19.2/week. With a tunnel you are winning big-time in terms of real costs.

Now consider the 4 square wide highway which costs 0.96 per unit, the highway is 10 tiles long.

As a terrestrial road it covers 40 squares which carries an opportunity cost in lost taxes of $120/week and it costs $9.6 in upkeep, for a total real cost of $129.6/week.

As a tunnel it covers 0 squares and the maintenance is $57.6/week, again with a tunnel you coming out ahead.

What this means is that tunnels come out as the preferable option for the budget-conscious Mayor, in fact we can be rather less generous with our assumptions about how much a square of residential (or other zone) can earn and still have tunnels the preferable option. The break-even point is $0.8 per square for the ramp and $1.2 per square for the highway. These numbers are comparable to the tax income from industry/office which means in those zones tunnels are still breaking even on the budget balance.

But when we consider that tunnels will be shorter and more direct, that they require minimal demolitions, that their on/off ramps and interchanges require a lot less space and that they are far easier to modify (especially with Moledozer mod) we are forced to conclude that tunnels under the city are the way to go. As previously noted, tunnels are not the most interesting option, but in terms of gameplay they are very very optimal, so good it's almost cheating.

 

More highways isn't always the solution anyway even if the AI is perfect - there's always some extra demand beyond the capacity you add to your network after all, though whether this is verified in the game is yet to be seen. (Perhaps, as upgrading roads may make some agents use it when they used a different route before)

I like what I'm reading about highways, especially about the lost revenue. The Katy Freeway in Houston comes to mind, carrying five lanes of traffic in each direction, plus two lanes of HOT/HOV lane traffic in each direction, and another 3 lanes of frontage lanes in each direction. The amount of clearances it took to build such a thing is another issue entirely, and nearly wiped out a small urban municipality's entire commercial tax base. Of course, that doesn't mean you have to be "railroaded" into adding rail-based transit to make things work, because if we were also going by reality, rail only works in high density situations/centralized job centers, otherwise you'll never reach critical mass and be pouring money into an ineffective system that barely keeps pressure off of highways. 

Hate to say it, but tunnels should be MORE expensive than elevated structures.

I kinda hope there's a mod out there that makes this part of the game more challenging :>

Metros and tunnels should be far more expensive. I see that rail tends to be more expensive and more disruptive to build, so even for underground trains I actually use rail, it feels a bit more like I have to watch my budget and think twice before linking an area.

  Edited November 24, 2015 by Linoa06  

[Ltw 401]

On 24-11-2015 at 10:08 AM, Linoa06 said:

I kinda hope there's a mod out there that makes this part of the game more challenging :>

Metros and tunnels should be far more expensive. I see that rail tends to be more expensive and more disruptive to build, so even for underground trains I actually use rail, it feels a bit more like I have to watch my budget and think twice before linking an area.

That anoys me too. My metro system is very extensive and I can easily pay for that. However, the expensive network is rail. In fact I like the fact that you can easily pay for the metro (even though it might be more realistic to have a more expensive metro system it is nice to go crazy with it). However, the expensiveness of the train system doesn't really make sense to me, mostly because it seems like a less expensive alternative to the metro system.

Nevertheless, metro is a great way to plan for future traffic. If you can plan lines in a way that they will service both commercial zones and residential zones they could be used by a lot of people. Furthermore I personally like to make the lines as long as possible and building some stations on which the citizens could switch lines (there are some nice 4-track and 6-track stations on Steam). I really like this option because you can extend the lines quite easily and it doesn't take that much real-estate. Next to that you can make money out of it.

I pay less attention to roads, mostly because of the extensive metro-system I'm running. I have some highways, mostly serving industrial zones outside of the city (but also the city itself) and I have one big avenue running through the middle (six lanes in both directions). Furthermore I use six-lane avenues a lot, or four-lane avenues if I don't expect a lot of traffic. You can also uprgrade four-lane avenues quite easily so that is nice too. At last I try to increase bike-traffic (more on this topic in this thread). Bike-paths are quite nice because they don't take a lot of real-estate and could be quite effective, for example my bike-paths are all constructed in the existing city without demolishing a lot of infrastructure.