Grid system?

[L. Spooner Inc]

I'm wondering whether there's any drawback to dividing your entire play area into a grid system of tracks made up of identical, interlocked square roundabouts, and then allowing trains to find their own way from pick-up to drop-off.

[_dp_]

To call anything a drawback you need to state what is your goal first. For traditional goals like money or citybuilding grids are pretty much a waste of time. Mb for a long game with Busybee GS they'll be usefull by I haven't tried.

[jfs]

This question reminds me of an ancient challenge: TwentyTwoTowns challenge, for TTDPatch without vehicle/station/orders count limits (fully?) lifted. (Maximum 240 trains allowed, no other vehicles allowed.)

[Pyoro]

Due to the somewhat negligible effect of terrain and (possibly) the grid-based nature of TTD my networks naturally end up to be large grids. Not to the extend to what you describe, but it's a bit of an annoyance I usually try to avoid since real ones ... ... ... well, they kinda sorta sometimes also are but not to that extend

For example
download/file.php?id=189134

If you look at this:
https://upload.wikimedia.org/wikipedia/ ... 6-2010.png
It's more like major hubs are centers of wheel-like networks, with lines going in multiple different directions (still forming grids of sorts, though). But of course in TTD there's a bit of a limit to how many different directions you can take and I suspect most tend to avoid "diagonal" tracks, so that's even less directions, meaning you automatically get something that's more explicitly like a grid ...

Anyway, what you suggest is just doing it in a more fine grained manner. Of course that can be done.

[odisseus]

I'm wondering whether there's any drawback

The proof of the pudding is in the eating. Go ahead and try it out, then report here. I would be interested to look at the end result.

That said, there's one obvious disadvantage of a pre-designed grid compared to an organically grown network: the upfront costs of building and maintenance. A junction capable of handling 100 trains will be a huge money sink if you have only one train yet.

Then again, there is a whole community engaged in building pre-designed grids that cover the whole map.

[L. Spooner Inc]

I've been experimenting with a grid system for a while now, using very simple designs and allowing trains to find their own way from station to station through networks of grids which I slowly grow over time as congestion builds. It's not as efficient as purpose-built lines, but it's super-simple and scales well. I'm uploading the save file of my most recent game in case anyone is curious about what it looks like and how it plays.

[Kevo00]

It would probably pay off in the long run, but have the disadvantage that some lines towards the edge of the grid might not see as much traffic, presumably.

[L. Spooner Inc]

It would probably pay off in the long run, but have the disadvantage that some lines towards the edge of the grid might not see as much traffic, presumably.

One of the effects of using a grid is that those less-used areas act as a kind of buffer for overflow. When congestion gets higher, those areas effectively become sidings where trains re-route themselves to get around congestion in more populated areas of the grid. In fact, it acts as an early warning to expand the grid because volume of traffic is making it less efficient. Eventually, towards the end of the game, you get total saturation and the possibility of gridlock due to bottlenecks caused by things like city placement, but by then you'll have billions and be able to start soaking up excess production with ships and canals. The use of grids means canals can be easily weaved through the grid without disruption just by adding a bridge between each node.

[L. Spooner Inc]

Another sample game for people to check. This is a desert map with high water level and low resources, meaning everything has to travel long distances on circuitous detours using very slow trains, and this is trains-only. This really shows the strength of using a grid system, especially since there's so much empty space available for use. The one weakness which has emerged is the path-finding; trains often get lost trying to find their way from one end of the map to the other. I suspect it's because of bottlenecks caused by bridges over long bodies of water not allowing trains to pre-reserve rails and other routes already at full capacity. Still, it only happens with really long routes, and only later in the game when congestion levels go up, which means you'll have enough money by that point to build secondary industries closer to where you need it.

[Alberth]

If you watch an area for some time, you do see trains making weird moves every now and then, with eg 90 degrees turns.
I think it is due to the large areas that are fully connected where trains claim long paths, blocking the path for other trains.

Things may be better if you force a direction with stations, especially the large ones, so trains all move in the same direction around it rather than having 2-way traffic everywhere.
Smaller fully connected areas likely work also better. (Given a train more choices than you need means you it can pick bad choices as well.)

[L. Spooner Inc]

If you watch an area for some time, you do see trains making weird moves every now and then, with eg 90 degrees turns.
I think it is due to the large areas that are fully connected where trains claim long paths, blocking the path for other trains.

Things may be better if you force a direction with stations, especially the large ones, so trains all move in the same direction around it rather than having 2-way traffic everywhere.
Smaller fully connected areas likely work also better. (Given a train more choices than you need means you it can pick bad choices as well.)

Oh, absolutely, you'll get more efficiency by creating purpose-built systems, but my goal was to see whether a grid system was feasible and it turns out it is. It's not as efficient as it possibly could be, but it's efficient enough to make it viable. One other weakness of this system is that local authorities, which are frustrating and rage-inducing at the best of times, become absolutely intolerable when they're obstinately preventing you from destroying roads or houses to create part of a grid, resulting in bottlenecks which can strangle large portions of the system. Once you've turned the entire area into old growth forest, the only way to improve relations is to bribe them, and it can cost upwards of $50 million+ to bribe even a small town to the point where you can lay track through them. I think that's probably the biggest weakness of using grids. It ends up being cheaper to raise the bloody ocean bed for miles than bribing a small town to let you lay track through them, but that's probably more down to the utterly broken local authorities system.

[L. Spooner Inc]

I'm curious, has anyone else tried this strategy now? What have your experiences and observations been?

[odisseus]

I'm impressed that you have made the "grid system" work. I expected that the trains would queue up on the busiest sections, while the rest of the grid would be underused. However, it seems that the pathfinder does a good job at finding alternate routes.

Nevertheless, it still isn't a viable strategy. There's just too much rail and too many intersections. With enabled infrastructure maintenance, the cost of maintaining your grid is larger than your entire income.

[L. Spooner Inc]

Nevertheless, it still isn't a viable strategy. There's just too much rail and too many intersections. With enabled infrastructure maintenance, the cost of maintaining your grid is larger than your entire income.

Well sure, but if I was playing with infrastructure costs, I'd be using an entirely different strategy anyway. I've now played a half-dozen games start to finish using a grid system, and I've really been enjoying the change in thinking it requires. It's almost an entirely different game. My next experiment will be seeing if altering the scale of each node changes anything. Currently it's set up to make 6x6 squares to accomodate 5-length trains; I'm going to see whether 11x11 squares (which would reduce the infrastructure costs dramatically) will make any difference.

For anyone who hasn't tried creating a grid yet, I encourage you to give it a try.