RoRo vs Terminus (was Loop vs 2-headed train network)

[Avezo]

I renamed the topic a bit, because of this good points: viewtopic.php?f=5&t=35806&p=224629#p224542 next page

RoRo : Roll-On, Roll-Off; trains enter the bottom end of station and exit the other end


Terminus: These stations are the end of a railway line

See https://wiki.openttd.org/Railway_station#Basic_stations
-- ßilk

I did search about this, bit I didn't find what I really wanted - a real big single thread about this 'issue'. I think it is important enough to have a 'big' thread about it - which means I don't really have any point to make myself here, I just want to hear from people who prefer either of those options (and in result it will make others learn a lot or so I believe).

[Acarin]

For me, loop wins every time. It does depend on how big you plan to build your rail network (bi-directional is ok for smaller setups), but I always plan for high traffic. I also find that, while there is the disadvantage of having to lay twice as much track, I prefer not having to drag an extra locomotive backwards (which does slow your trains down).

[Mehve]

I'm firmly in the "Loop" camp, but neither approach is a game-breaker for most situations. I just can't think of any benefit to the double-headed method EXCEPT for when you don't have enough room to loop. Meanwhile, half the locomotives are deadweight at any given point, you need to get coal/wood/fuel to both ends of the train, and for a given power/cargo capacity setup, the double-header will actually need more room at the train stop.

[impetus maximus]

i started Factorio with single engine, loop stops,

and single two way tracks with bypasses.

having played on servers with two headed trains and terminal stations, i've switched to two headed, with terminal stops.
it makes building stops, and stations SO much easier, and you need less track to do it. makes signalling less of a headache as well.

[theRustyKnife]

I'm wondering, am I the only one here who uses both as I see fit?

For real tho, isn't that the superior solution? You can enjoy all the advantages of big single headed trains with loop stations while still being able to build tiny utility-like terminal stations for L-C-L trains and such.
Same goes for single bidirectional vs dual one-way track (considering you can upgrade to dual one-way tracks quite easily if you need more throughput).

I haven't had a problem with these types of setups since chain signals were introduced. Idk, maybe I'm just not building big enough factories to see why one should be considered the better option...

[DaveMcW]

single bidirectional
... I haven't had a problem with these types of setups

[theRustyKnife]

single bidirectional
... I haven't had a problem with these types of setups

What's wrong? Just surprised I had no problems with it?
In that case it's because I don't have the entire network like that, only the places where it's appropriate - without much traffic going through

[Mehve]

What's wrong? Just surprised I had no problems with it?
In that case it's because I don't have the entire network like that, only the places where it's appropriate - without much traffic going through

He might have just been pointing out the possible oxymoron there, you never know.

[ssilk]

I haven't had a problem with these types of setups since chain signals were introduced. Idk, maybe I'm just not building big enough factories to see why one should be considered the better option...

Hm. After playing >500 hours with a two-way train map of about 20 square kilometers, 300 trains and 100 stops I can say, that the chain signals needs a bit more tweaking to enable the player to understand and see them a bit better.

1. See the signals, that influence this chain signal (mark the covered block and the signals? The current arrows are useful, but a underlaying color-highlight would help a lot).
2. Easy switching between normal and chain signal.
3. Enhance the "first" normal signal in a way that you see, that it influences chain signals.
4. If chain signal selected, show the player "better places" to put a chain signal on (For example: show all possible places to put a signal on, but enhance places before crossings when a normal signal follows).

[theRustyKnife]

What's wrong? Just surprised I had no problems with it?
In that case it's because I don't have the entire network like that, only the places where it's appropriate - without much traffic going through

He might have just been pointing out the possible oxymoron there, you never know.

Oh well I meant one track with bidirectional traffic on it. What would the proper wording be tho?
Sorry, English is not my first language.

I have trouble with Czech too sometimes

I haven't had a problem with these types of setups since chain signals were introduced. Idk, maybe I'm just not building big enough factories to see why one should be considered the better option...

Hm. After playing >500 hours with a two-way train map of about 20 square kilometers, 300 trains and 100 stops I can say, that the chain signals needs a bit more tweaking to enable the player to understand and see them a bit better.

1. See the signals, that influence this chain signal (mark the covered block and the signals? The current arrows are useful, but a underlaying color-highlight would help a lot).
2. Easy switching between normal and chain signal.
3. Enhance the "first" normal signal in a way that you see, that it influences chain signals.
4. If chain signal selected, show the player "better places" to put a chain signal on (For example: show all possible places to put a signal on, but enhance places before crossings when a normal signal follows).

I think I agree with every single point here. Actually, I might try to make a mod for some of these things. At least switching between signal types with a hotkey (or something like that) should be easy

[Mehve]

I didn't think there was anything wrong, myself. Maybe "Single-lane, bidirectional", if you wanted to be specific?

[ChurchOrganist]

One thing I fail to understand about Factorio's rail implementation is why it doesn't reflect the ability of rail locomotives to contribute to power output whichever end of the train they are on.

As I may have said before, I came to Factorio via OpenTTD, and have always been mystified by this complete reversal of real life locomotive behaviour.

Certainly on British railways we have had control of multiple motor units from one locomotive at least since the Deltic was launched in the 1960's.

OK with steam locos you had to have crew in each locomotive, but Factorio's trains are supposed to be diesel!.

How about rectifying this strange situation devs?

Oh and to put this post back on topic - I prefer double-headed trains, but I'm pragmatic - if a situation calls for a loop-based solution I will build it.

[Mehve]

It's not that Factorio locomotives can't push, they simply have no reverse gear. But you can still stick them at the back, facing forwards, and get power from them.

I suppose it's just a gameplay thing. Similar to how the swinging inserter arms are limited strictly to 180 degree swings, with 45/90/135 degree movements impossible. Or how inserters will only place things on the far lane of a belt, despite clearly having the ability to grab stuff from either lane.

[impetus maximus]

what ChurchOrganist is saying is most diesel locomotives can go fast in reverse.
i have to agree. if a second locomotive is attached, the power output should double.

[ssilk]

what ChurchOrganist is saying is most diesel locomotives can go fast in reverse.

The train routing is not longer reliable then. In loop-based layouts you quickly run into situations, where the routing calculates shorter distances, if loop trains can change direction. Result is: Train changes direction which is in loop-based layout a catastrophe.

And I would say the way it is now is just fair: The afford to build loops is higher than with 2-headed designs... it would be unfair to have single-headed trains which can change direction; nobody would use then that and we had no reasons to discuss.

[impetus maximus]

i'm not saying a single engine train should be able to go full speed in reverse, or move in auto mode.
i'm saying if the head train is driving, the one in back should be able to double the torque.

[orzelek]

i'm not saying a single engine train should be able to go full speed in reverse, or move in auto mode.
i'm saying if the head train is driving, the one in back should be able to double the torque.

Currently data says that reverse power of factorio diesel is 0.6.
We can try to mod it and see if it behaves as expected (aka reverse engine will contribute to acceleration) or this attribute works in a different way.

[ChurchOrganist]

It's not that Factorio locomotives can't push, they simply have no reverse gear.

If that is the case, then how come you can drive a single locomotive train backwards in manual mode?

[Deadly-Bagel]

I don't really get two-way trains... You need two lanes of track anyway so trains can pass each other and you're saving the 50 or so iron of building a loop but spending 280 on the extra locomotive. To boot, having trains constantly switch direction means no consistency with the carriages' order.

[ratchetfreak]

I don't really get two-way trains... You need two lanes of track anyway so trains can pass each other and you're saving the 50 or so iron of building a loop but spending 280 on the extra locomotive. To boot, having trains constantly switch direction means no consistency with the carriages' order.

There is the option of single track with passing points/sidings however I very rarely see someone do that and the one time I did the signals were wrong (should be chain when entering the 2way track and normal when exiting). Not to mention it's a major bottle neck in a busy system.