Some simple switch

[vezrilx]

https://steamcommunity.com/sharedfiles/filedetails/?id=2503294900

Do you guys use the same sort of things?

Or want to share any experiments?

https://steamcommunity.com/sharedfiles/filedetails/?id=2487665353

[coenvijge]

Good compact junctions.

I'm using seldom a connection like the first one because I prefer 4-track lines like this:

<------ slow --------<
<<<----fast-------<<<
>>>----fast------->>>
>------slow---------->

[Saulius]

No offense here
It looks nice in the picture, but ..... in this game You are asking for really big traffic problems

[coenvijge]

Yes, but depending on the traffic you have at the moment on that piece of your tracks.

[vezrilx]

Ah well what about this then?
https://steamcommunity.com/sharedfiles/filedetails/?id=2511673687
https://steamcommunity.com/sharedfiles/filedetails/?id=2511673701

And to be fair, the 8 wide is not worthwhile to make and use, just a good exercise to see what is possible. I do think more people should be aware of splitting 2 into 4 in parallel though.

[gussmed]

Generally, there's not a lot of need for anything more than 4 lanes on a single route. Trains are generally just 2 categories: freight (slow) and passengers (fast). A set of rails for each handles most situations.

I don't even use separate fast / slow lanes that often. Much more often, I'll just add an occasional passing lane. They're inexpensive compared to laying down 2 additional tracks, and generally once the fast train is ahead of the freight trains, it's clear for a long time afterward.

More to the point, branching a 4-lane track is a huge PITA. The track-laying engine doesn't really understand anything wider than 2 rails, so you have to make long, elaborate shunts, or get into large, expensive overpasses.

[chaney]

I'm with you gussmed. Further reason not to branch too many things is swap time. If incoming trains have to wait for more switches to be cleared when an exiting train leaves, it delays operations. This added delay really makes single switches near the Stations perform better than the high numbers of switches to more tracks shown above.

[vezrilx]

I honestly do not see at all where you guys are coming from. But mostly I am interested in seeing funky screenshots and experiments people try out. Perhaps this does not suit everybody but I do not think these setups lack efficiency if used correctly and monitored.

I do make use of "cheater, speeder" signals, to help along flow, where they may bind if there are too many trains or do not function properly if trains of the wrong length enter, but that is a process that comes with observation of my lines.

https://steamcommunity.com/sharedfiles/filedetails/?id=2487664714

TLDR: screenshots section of steam is full of still story images or old screenshots. I want to see some cool stuff but started with absolute basics of structure that I use.

A rural station with pass-through, and dedicated platforms for stops on each side, for instance..

https://steamcommunity.com/sharedfiles/filedetails/?id=2229723496

Which makes for a great reason to split from 2 into 4.

[genemead]

To me, this looks like solution looking for a problem.
A nice exercise for fun, not for any practicality.

[vezrilx]

Sidedness issues at cities may indeed cause other issues up the line that do demand solutions, none of these were solutions created without an apparent need however.

Seriously though, just trying to pique some interest in some screenshots feel free to post something with explanation if needed.

Maybe it backfired including the 2 into 8.. but the point of it was to be ridiculous in the first place.

[vezrilx]

I REALLY like to provide dedicated waiting rails at least enough to keep trains with another purpose from waiting. And it's not ineffective.

This is why I felt like making a 2-8 structure, the bridge I come off of is a little much.

https://steamcommunity.com/sharedfiles/filedetails/?id=2513313364

Sure it's not super high capacity but everything has a distinct purpose, and learning how useful that particular switch is will be taken into account later.

[chaney]

I like how you are riffing here. Sometimes a degree of sorting has some value.
Usually it is done to try and improve total throughput and priority throughput at a Station when I have done it. Ultimately if there are frequently trains waiting, some of those trains may as well not exist, so a lot of thought that would apply in other situations/games/real life doesn't translate so well here - I think that's where some of us are coming from.

Some of us have put up pictures in the past, as you are doing here, to show ideas we have had. There are some good screenshots out there.

[gussmed]

It’s true that long wait times effectively nullify the value of a train, but almost always, long wait times are due to platform shortages at the station. It doesn’t matter how many tracks you add leading to the station if trains have to wait for an open platform.

Big cities need more than the 4 platforms you get with one station, and the biggest cities need 12 platforms. I’ve gone as high as 16, 2 large stations and 2 large warehouses, when congestion is serious.

Also included in that is the need for a least one set of tracks bypassing the station. Though these days I tend to go for “cities as detours off a straight mainline track” rather than a city bypass.

[chaney]

Originally posted by gussmed:

It’s true that long wait times effectively nullify the value of a train, but almost always, long wait times are due to platform shortages at the station. It doesn’t matter how many tracks you add leading to the station if trains have to wait for an open platform.

...

Also included in that is the need for a least one set of tracks bypassing the station. Though these days I tend to go for “cities as detours off a straight mainline track” rather than a city bypass.

100% agreed with these thoughts.

[vezrilx]

Making use of even 2 platforms with the same 2 wide rails and reliance in a switch can lead to more waiting that needed with at least 3 trains, simply let the 2 trains sharing a platform run in succession, and the third waits behind that.

Signaling control stations have proven to me at least, to be plainly inadequate in overall throughput, so I am considering the other solutions.

How exactly do you plug into 12 platforms at a city? It could be all from 1 side but likely not... still should make for a nice picture.

[chaney]

I strongly believe Signal Control Stations are terrible for maximum throughput needs. They have those long switch networks that make for long swap times.

Point to point, isolated, dedicated track as a rule works incredibly well. Each Platform is connected to a SINGLE route which goes to only a SINGLE other stop in the most strict form.

With that setup, you don't end up with waiting because trains will space out quasi-automatically as each train can only be blocked by the one using the Platform directly before it and then the waiting train has un/load time that will give it some separation before the next stop could make for interference. This is true for symmetric loads, but if you dead-head one way (run empty) things will bunch a bit at the source end, but that's not the bottleneck anyway, and if it is, you can add a second Platform there as a concession to the strict P2P dogma.

Another modification can be made by allowing 2-3 *low demand* routes to share a Platform. Swapping which side is accessed reduces swap time if you do that. You can also set up trains to always enter one side and leave on the other for maximum throughput by minimizing swap time.

These are basic approaches that can be applied to get lots of deliveries with the minimum of infrastructure in terms of track, Platforms, and engines.