Some suggestions regarding the rail system

[Glockshna]

Been playing Factorio for far more hours a day than I care to admit for the past week or two and am having a lot of fun creating what are essentially towns of buildings all interconnected by an automated rail network. The replay value by simply regenerating a map is insane so good on you guys for a job well done, keep it up! but on to the topic at hand: The rail system is a great tool and mechanic but it needs quite a few tweaks in my opinion. I've fleshed out my ideas and hope you take some of them into consideration.



Anyway please let me know what you think!

The numbers in the post are just place holders there to give you guys an idea of the relative balance between the tiers.
Regards

Here is the PDF in text form it isn't as readable here but...


A quick feature I’d like to see before I get into the more specific things is the ability to mark a segment of track as Mainline, Siding or Factory/loading/unloading. This way trains won’t try to route themselves through factory loading docks in effort to take the shortest route between waypoints and will instead prefer the mainline wherever possible.
1. The products and materials we use rails to transport are diverse and so should be the rolling stock.
There should be different types of rail cars that are better at shipping different materials.
• Flat Cars:
o Best at shipping Steel Plates, Iron plates, Copper Plates, Copper Wire, Stone Bricks, Wood, Power Poles, Rails

• Box Cars:
o Best for shipping assembled components like Gear wheels, Circuits, Science Packs, Speed Modules etc

• Tank Cars:
o Best for shipping liquids (especially with Oil coming out in the next update)

• Hoppers:
o Best for raw extracted materials like Ore, Coal and Raw Stone.

• Different Locomotives for different jobs:
o There should be different locomotives at different 'prices' that can be used for different jobs
o Stronger locomotives should have much higher torque to pull large trains but at the expense of top speed while
o express locomotives and pull small loads of specialty equipment at much faster speeds
o but have trouble with loads much larger than that.







2. Better Physical Realism
• Trains should be limited by weight
o Each locomotive should have a maximum consist weight it is rated for. Above that weight it should have greatly reduced top speed and acceleration
o Each locomotive and train car should have its own weight which contributes to the total weight of the consist.
o Larger (Higher capacity) train cars should be heavier than smaller ones obviously
• Acceleration should be a formula of Horsepower, and Consist Limit. Top speed and consist limit can be made up in a way that makes each locomotive an upgrade or unique to others.

3. The research path to max out railway functionality is way too easy.
Once you get Automated Rail Transport you are basically set to automate EVERYTHING within closed loops and it's not hard to get signaling afterwards to connect everything to a mainline. (Early alpha, not a big deal)
With that, and the new Locomotives and Rolling Stock in mind I would suggest a tech tree resembling something like this:













• Railroad:
o Requires:
 Steel
 Logistics
o Unlocks:
 Straight Rail
 Curved Rail
 Basic Steam Locomotive - Powered by coal;
• Good for ferrying freight between nearby factories in small quantities. Top speed of around 30Kph at load limit.
 Basic Box Car
• 8 Slots for any type of cargo
o Research Unlocks:
 Railway
 Freight

• Freight
o Requires
 Railroad
 Logistics 2
o Unlocks
 Medium Box Car
• 16 slots for any type of cargo
 Small Hopper
• 16 Slots
o Only holds Ore, Coal and raw stone
o Weighs less than a medium boxcar so it’s better to use a hopper when transporting the appropriate material
 Small Flat Car
• 16 Slots
o Only holds Flat Car materials (See first section)
o Weighs less than a medium box car
o Research Unlocks
 FreightWay






• RailWay
o Requires
 Railroad
 Diesel Fuel Production
 Electric Energy Accumulators
o Unlocks
 Basic Diesel Locomotive
• 50 Kph Top speed at Consist Limit
 Tanker Car
• Not sure how you guys intend to balance liquid volumes so I’ll leave this one to you guys
o Holds only liquids obviously
 Tanker Car Extractor
• Extracts liquid from a tank car.
o Research Unlocks
 ExpressWay
 Stations
 Long Range Transport
 FreightWay

• Expressway
o Requires
 Railway
o Unlocks
 Express Diesel Locomotive
• Lower Consist Limit than Basic Diesel Locomotive
• 105Kph Top Speed at Consist Limit


• Stations
o Requires
 Railway
 Robotics
o Unlocks
 Train Stop
 Rail Signal


• Long Range Transport
o Requires
 RailWay
 FreightWay
o Unlocks
 High Torque Diesel Locomotive
• 60 Kph top speed at Consist Limit
o Much larger consist limit than Basic Diesel Locomotive

• FreightWay
o Requires
 Railway
 Freight
o Unlocks
 Large Box Car
• 24 Slots
 Medium Hopper
• 24 Slots
o Weighs less than a Large Box Car
 Medium Flat Car
• 24 Slots
o Weighs less than a Large Box Car
o Research Unlocks
 Long Range Transport

[Balinor]

Some quite interesting suggestions there. I'd like to note also that if an expansion of the transport system is likely to take place then improving the control of the trains would also be a welcome addition. Having a single interface that shows all active trains and stations and allows micro management from that single window would make the job much easier in addition to being able to set conditions on transport, ie, wait until full, wait until empty, leave station when another arrives, etc.

[ssilk]

I must say that this suggestion is good, but in the wrong format: It can't be discussed, because it can't be cited; it is questionable, if the document is already online, when it is time to think about that. I prefer just text.

But to be not so negative :

• Different cargo wagons: Yes and no. If you mean with "best for" that it is really usable for other stuff: Yes. I would like to have it like in reality: The total maximum cargo depends on the form of wagon. For example there are the all-round box cargo wagon, which can transport nearly everything, but the loading/unloading is difficult (takes long) and the capacity is low.
• Better physical realism: I don't see the need to have a limit by weight. I mean it could simulate that a full train can slower accelerate and break, but the limits in real trains are only because the hills, because they cannot climb it. But of course if you put a small loco at 10 wagons loaded full of steel it will slipper on the tracks and takes ages, until it is at full speed. In other words: No limitations, but a correct physical simulation will be limit enough. And maybe we have sometimes hills in Factorio, then this is of course also a limit for trains.
• Research path: Hm. Well, sounds quite logical.

Think this is a bit oversized/too concrete. I cannot think to deep into that, because it will sooner or later change.

I like your suggestion and for the future you might limit the options. It's enough to say "We could have some other wagons, which are more specialized to types of items, like a flat car for steel, wood or a tank for fluids... etc."

[Glockshna]

I'll clarify some things.

I must say that this suggestion is good, but in the wrong format: It can't be discussed, because it can't be cited; it is questionable, if the document is already online, when it is time to think about that. I prefer just text.

It is text. It's a PDF, you can just copy and paste from it. But in case for whatever reason you cant, I've posted it in the OP but it's not very easy to read.

But to be not so negative :

• Different cargo wagons: Yes and no. If you mean with "best for" that it is really usable for other stuff: Yes. I would like to have it like in reality: The total maximum cargo depends on the form of wagon. For example there are the all-round box cargo wagon, which can transport nearly everything, but the loading/unloading is difficult (takes long) and the capacity is low.

Box cars would be able to transport anything but at lower capacity than equivalent weight specialty wagons.
As an example a Box car with 16 slots might weigh 20 Tons but a hopper with 16 slots might weigh 10 tons or something like that, but it will only accept raw ore, coal and stone making it better to use specialty wagons wherever possible. same type of deal for flat wagons.

[*] Better physical realism: I don't see the need to have a limit by weight. I mean it could simulate that a full train can slower accelerate and break, but the limits in real trains are only because the hills, because they cannot climb it. But of course if you put a small loco at 10 wagons loaded full of steel it will slipper on the tracks and takes ages, until it is at full speed. In other words: No limitations, but a correct physical simulation will be limit enough. And maybe we have sometimes hills in Factorio, then this is of course also a limit for trains.

Maybe you missed it but:

o Each locomotive should have a maximum consist weight it is rated for. Above that weight it should have greatly reduced top speed and acceleration
• Acceleration should be a formula of Horsepower, and Consist Limit. Top speed and consist limit can be made up in a way that makes each locomotive an upgrade or unique to others.

Taken directly from my OP. To clarify, Consist limit is the maximum total consist weight at which the train will be able to reach its top speed. Above that weight the train will have a reduced top speed and noticeably slower acceleration it's not meant to be a hard cap on the length of the train. Stopping distance is also something you have to keep in mind. If it takes 'ages' to get to full speed, it also takes 'ages' to stop.

[*] Research path: Hm. Well, sounds quite logical.[/list]
Think this is a bit oversized/too concrete. I cannot think to deep into that, because it will sooner or later change.
I like your suggestion and for the future you might limit the options. It's enough to say "We could have some other wagons, which are more specialized to types of items, like a flat car for steel, wood or a tank for fluids... etc."

None of this is supposed to be concrete it's laid out this way to give the devs an idea of the balance I had in mind with this idea especially with the research tree. I felt the game progressed too quickly past train research so I fleshed out the tech tree to give an idea of pacing. In the end it's up to them to decide how they want to implement it and what parts to implement, if they even like any of the idea in the first place.

[ssilk]

I prefer just text.

It is text. It's a PDF, you can just copy and paste from it. But in case for whatever reason you cant, I've posted it in the OP but it's not very easy to read.

Well, I saw it. In future: Recommend to write it not in a text-program. Or just format with underline-spaces.

Taken directly from my OP. To clarify, Consist limit is the maximum total consist weight at which the train will be able to reach its top speed. Above that weight the train will have a reduced top speed and noticeably slower acceleration it's not meant to be a hard cap on the length of the train. Stopping distance is also something you have to keep in mind. If it takes 'ages' to get to full speed, it also takes 'ages' to stop.

Hum. Not. A heavy train has not really a much longer stop distance, because every wagon has brakes. A little bit more, yes. This is mostly, because the brakes needs more air-pressure to brake the same amount of cargo. But once the pressure is reached, a heavy train breaks with the same deceleration. And the rise of the air-pressure in the breaking system is a combination of length of train plus the compressed air reservoir. Now it gets a little bit too deep into the details. And for the praxis and in the game, this doesn't make a big difference, because a train drives by the signals. When a train driver see that the next signal is red he knows, when to begin to brake (= begin to fill air in the braking system). The problem occurs, when a train has to halt in accident, but this doesn't currently happen in the game.

Much more interesting are: Is the track wet? Icy? How much sand can the locomotive put on the track to get grip etc. (Every train has a container of quartz-sand, which is used for that.)

Another thing is: How much is the train forced to reach the maximum speed in minimum time. Or in other words: The acceleration needs the most energy. When on speed a train takes only a minimum amount to keep the speed. Or in other words: It should take much more coal to run trains on a track which always stop and go as on tracks where they just stop at the train-stops.

[Glockshna]

I prefer just text.

It is text. It's a PDF, you can just copy and paste from it. But in case for whatever reason you cant, I've posted it in the OP but it's not very easy to read.

Well, I saw it. In future: Recommend to write it not in a text-program. Or just format with underline-spaces.

Taken directly from my OP. To clarify, Consist limit is the maximum total consist weight at which the train will be able to reach its top speed. Above that weight the train will have a reduced top speed and noticeably slower acceleration it's not meant to be a hard cap on the length of the train. Stopping distance is also something you have to keep in mind. If it takes 'ages' to get to full speed, it also takes 'ages' to stop.

Hum. Not. A heavy train has not really a much longer stop distance, because every wagon has brakes. A little bit more, yes. This is mostly, because the brakes needs more air-pressure to brake the same amount of cargo. But once the pressure is reached, a heavy train breaks with the same deceleration. And the rise of the air-pressure in the breaking system is a combination of length of train plus the compressed air reservoir. Now it gets a little bit too deep into the details. And for the praxis and in the game, this doesn't make a big difference, because a train drives by the signals. When a train driver see that the next signal is red he knows, when to begin to brake (= begin to fill air in the braking system). The problem occurs, when a train has to halt in accident, but this doesn't currently happen in the game.

Much more interesting are: Is the track wet? Icy? How much sand can the locomotive put on the track to get grip etc. (Every train has a container of quartz-sand, which is used for that.)

Another thing is: How much is the train forced to reach the maximum speed in minimum time. Or in other words: The acceleration needs the most energy. When on speed a train takes only a minimum amount to keep the speed. Or in other words: It should take much more coal to run trains on a track which always stop and go as on tracks where they just stop at the train-stops.

We are getting a bit beyond the scope of the discussion I wanted to have here. I'm not looking for 100% physical accuracy otherwise this wouldn't be a game. I'm just looking for more in depth train game mechanics is all.

[ssilk]

Of course, this board is about ideas, not 100% physical simulation. Good ideas bring the game forward.

But have you ever heard from a game, where the train behavior was influenced by the weather?
If I have an idea, I write it down. Otherwise it's forgotten.

[Garm]

Am I the only one, that thinks trains need more profound changes first?

I mean the suggestions listed here are nice and interesting, but they are superficial: They are about the outside feel of the train system in game.

Yet we are still have so much missing from the core of train handling:

- better schedule system is needed
1. IF/THEN schedules (on demand routing)
2. FOR i=n, WHILE schedules (loops within loops)
3. more fluid player interruptions en-route
4. Logic system routing
- Train link modifications
- Bridges
- signalling with logic inputs

[3LollipopZ]

I would concur. there are pressing issues like bridges, advanced track layout, logic in signalling, etc.
Running power alongside of rail is a pig as well I'm just sick of doing it

[ssilk]

Yeah, but that's no new ideas.

I would priorize it like so:
- Because of oil special wagons and special stuff to load/unload, and eventually a second locomotive (just more ore less power)
- the automatic laying of tracks (drag new rails, like in OpenTTD), because I measured some times in my last game and laying tracks is not good.
- Helper grids for train stops (to see, where I can set up the inserters and how long the train is), help grid for signals (because length). Help grids could also be very useful for placing any stuff.
- When doing automatic laying of tracks, nice end-pieces of tracks should also be done.
- and "half s-curves" for one rail-width and other special rails.
- Because of many questions about that, the highlighting of blocks (see which segments a signal will watch).
- Because it is very easy, train stops without halt (run thru).
- Because it is a mess, a complete rework of station name handling (and names overall).
- And in a first version a signal type, which works like a combination of block signal and smart-inserter-logic (a smart block signal), so that the players are able to begin to play around with it.

The other stuff: well, nice, but when I think of playability and realistic time estimations, I think every estimation will end up somehow there.

[haunted_1]

Yesterday, I was thinking of to add more stuff to the rail system too. It's actually funny that some of your (ssilk) ideas are the same, e.g. the diffrent cargos for diffrent materials.

Also, i was thinking about three diffrent trains. Those are my thoughts:

Trains that use wood/coal:

- can only drag a small amount of cargos, lets say 3 or 4 cargos otherwise the train would be really slow or doesnt even start running
- acceleration and top speed is not very high (lower than the current train)
- the amount of coal/wood that this train uses must be much higher in my opinion compared to the current one


Trains that use oil (which is coming up):

- can drag a decent amount of cargos
- accerleration and top speed is about the same that the current train
- a special entity for inserting the oil into the train (fuel load station)


Trains that use electricity (which would add electric railways)

- can drag a huge amount of cargos
- sucking your electricity from your electric network
- you need to add electric poles to the railway or there is a new railway with electric poles
- accerleration and top speed is very high


More signals!

- Block signals/One-Way signals
- Signals for the maximum highspeed


Those are my thoughts from yesterday and i think its quite similar to ssilks in some way.

[Plop and run]

How would you use speed limit signals

[haunted_1]

How would you use speed limit signals

I'm a trainiac, i can use them

[ssilk]

Ever get killed by a train? In my last train system I got killed four times, even, when I watched carefully. Not that this is bad, but I see some sense in lower speed, when the train comes near to a place, where I need to cross often.

And ever watched trains in curves? That's ridiculous. I think they should automatically drive a bit slower in curves. Interesting game element, because it prefers straight tracks.


Edit: the block signals are by default one way. I reworked the whole part in the wiki, I think it's much more clear now. I would appreciate corrections.
https://forums.factorio.com/wiki/inde ... ork/Signal

[haunted_1]

Ever get killed by a train?

It's game ender number one in the late game. I got killed by my trains more then by the biters in overall i think.

[BurnHard]

In late game, with enough shields in my armour only my train gets destroyed when I am on the track at the wrong time

[ssilk]

That's what I mean: the slow train gets not so broken, when driving slower.