Belts, how far is too far?

[count_zero99uk]

Just really starting gettign to grips with the game and working on getting busses sorted and such, mainly thanks to the KoS series. But how far is too far to belt? Is there a point where its better to do it by train? Likewise whats the same for liquids?

I have some deposits that are 3 maybe 4 radars away from where they need to be. Would you belt those?

[Loewchen]

That is essentially personal preference, but once you transport any of the ore in question by train you have the unloading station already and hooking up even close deposits with trains is then easier than having a belt route. Performance wise, trains are always the way to go.
Pipes additionally need pumps in regular intervals to restore pressure, so they are even less practical over longer distance than belts.

[count_zero99uk]

Thanks for that

Any numbers on pipes loosing pressure?

[Loewchen]

Any numbers on pipes loosing pressure?

Sure: 19851

[count_zero99uk]

OK thanks

Bit too much info atm im sure ill end up needing it Not even made blue science yet

[Elok]

Personally, I never use train for a simple reason.

For a mining station to be completly independant, it need :

1- Power
2- Bots
3- Stock to replace what will be destroyed by the Biters

The last time I created a mining station with train, I tried to install solar pannel and Accumulators. But it took so much that I ended up plugging the electricity to my main base. Then, bitter attacks became stronger and, sometime, they will destroy a laser tower and I would have to travel to make the repairs. I finally decided to add drone station to the electric pole. Problem with this, bitters started to attack them so I had to add laser station along the way. So, in the end, why not replace the train, which create a lot of irregularity in the supply chain, by a bunch of conveyor that are both stable and easily scalable. And when you start to have a big base, conveyor production is not a problem.

[innesm]

Elok, how do the solvable problems that you hit with your train-based outpost get solved by using belts?

[ledow]

Always blueprint train tracks with the cheapest wooden pylons in - they are so cheap its ridiculous and your blueprints will take out a load of trees along the way to pay for them anyway.

That way, your power network becomes as route-redundant as your train network too, and you have power to all remote stations. Trying to be entirely independent is a bit silly, I think. You can do it. It's a fun exercise. But the costs are enormous and you end up having "spider" points, where you run a train to your central location and then belt from there to any nearby resources too (which takes MORE power, especially if you're doing a lot of speed-module-miners!).

So power has to be global, really, but there's no need at all to think that belts would be easier or cheaper.

Plus, trains can collect from whole rafts of places, drop off at intermediate stations, don't need constant balancing, don't get interrupted if one tile is attacked (you run redundant routes, right?). One train with 30 rocket fuel can go for AGES too, so you're saving masses and masses of base resources compared to making literally thousands of belts.

Additionally, belts will back up and criss-cross making your life difficult. Trains will sort themselves out and - if you design right - they make it so that you don't bother to mine more than your base can take anyway (enable/disable stations, let the local mining back up rather than a HUMONGOUS LONG belt.

You still need lasers, and bots, and power, but it doesn't mean that trains don't have advantages. Additionally when it comes time to move on, you can leave all the infrastructure in place, it makes no difference and just provides alternative routes for the trains which they will find for themselves.

When you start doing fluids, you can make one fluid train and a few tiles of track, or you can make literally THOUSANDS of barrels and belts trying to get it all back to base. They aren't "recoverable" because for as long as that oil field is in operation you need sufficient stock of barrels to keep it occupied and ship them back along your miles of belt. Or you could just load up one train with one train's worth of barrels only. It delivers empties, takes away fulls, and doesn't need more than the transfer rate of the oil (not the transfer rate + fill up the entire belt).

Also, trains are SO MUCH FASTER than belts that it isn't even worth considering. A handful of trains can swamp a 6-belt bus easily, and critically they can do it at almost any distance, so they can actually "teleport" your 6-lane bus 1000 roboports away to be a 6-lane bus next to your base way quicker than a belt ever could. And without having to fill up miles of belt first.

[Koub]

Yeah, but belts are a brainless solution. I built my first train I was well past my 15th map, and just built it so that "I have to try at least once". And I didn't like it, because I've spent more time reading tutorials and watching youtube videos of tutorials than actually playing with my trains.

All that signalling and stuff is necessary, but I just can't get my brain the right way to do it. I'd like to just plonk down the rails and the stops, and let the game figure itself what to do with it, but that's not possible.

So I lay a pair of belt tracks a few thousand tiles long, and that's all. No hours long debugging your signals. No hours trying to do an unloading station design to balance your output. Exceptionnally, I'll do a 2 stations standalone rail to connect two things, without crossings, and that's about it.

[MeduSalem]

Yeah, but belts are a brainless solution. I built my first train I was well past my 15th map, and just built it so that "I have to try at least once". And I didn't like it, because I've spent more time reading tutorials and watching youtube videos of tutorials than actually playing with my trains.

All that signalling and stuff is necessary, but I just can't get my brain the right way to do it. I'd like to just plonk down the rails and the stops, and let the game figure itself what to do with it, but that's not possible.

When reading how many people have problems with simple rail networks then probably some parts of Optera's Logistic Train Network Mod should be a part of the basegame to alleviate the torture of setting up a simple rail network (doesn't need to have all of those fiddly options)... Couple of Provider Stations, couple of Requester Stations... add some trains and you are done.

But probably it wouldn't really solve the signaling problem which is what most people hate about rail networks.



Actually I only use trains anymore to get the ore from the outposts to my main factory... Everything else within the factory is done by robots. And I wouldn't consider doing it otherwise anymore because of how setting up train networks is a chore (even with blueprints and a modular base) and only pays off if the distances are huge... greater than one can efficiently use bots. I've not been using Belts other than during early stages of a map since the devil knows when... In late game they are just chokepoints... and no matter how wide you make the bus it always turns out it's just not enough. As a matter of fact... Belts don't scale well.

[Elok]

Elok, how do the solvable problems that you hit with your train-based outpost get solved by using belts?

If you have a defensive perimeter that protect your whole supply line to send power and drone to your mining station, then you can either use train or conveyor (since your conveyor are protected).

Not only conveyor are easy to build with drone remotely but the supply chain is a lot more stable with conveyor while you need to build a buffer with train.

In other words, if your base can supply a lot of blue conveyor fast, then using train doesn't worth it.

Here's my base : https://drive.google.com/file/d/0B45f44 ... sp=sharing

[PunkSkeleton]

Not only conveyor are easy to build with drone remotely but the supply chain is a lot more stable with conveyor while you need to build a buffer with train.

Not true. Mining outpost gradually output less and less ore as the drills in the center have a lot more ore than the ones at the edges. Moreover drills at the edges often catch only a few patches and run dry fast. With conveyors I'd either waste mining or run production at less than full capacity or... use buffer as with trains. With large buffers and trains I only have to add a new station when the buffers are running low.

In other words, if your base can supply a lot of blue conveyor fast, then using train doesn't worth it.

It depends on scale, really. If you're using only a few blue belts of each ore it is manageable. I'm using 12+...

Here's my base : https://drive.google.com/file/d/0B45f44 ... sp=sharing

[/quote]
And here's mine:
https://ibb.co/dvf5ab
https://ibb.co/idXU1G
https://ibb.co/ic4WFb
If you take a look at laser turret ranges you'll understand the difference in scale.

[Selvek]

I think there are a couple of big advantages to trains:

1) You'll probably want them eventually, as your close resource patches start to run out. Any time invested in a rail network early will help you out later, but time invested in belts really wont.
2) You can run multiple resources on one line. If you build a belt from an iron patch that's 4 radars away, and then discover copper that's slightly beyond it, you have to lay another parallel line for copper. When you need to hook up a uranium deposit, you'll need to add pipes for the sulfuric acid in addition to the uranium belt, whereas if you are using trains you can move the acid along the same tracks.
3) Throughput of a single two-lane train setup is, well, ludicrously large. You can easily saturate several blue belts of every resource with a single train line. If your total throughput increases later with belts, you have to go back manually to add more belts.

Against this, trains have a couple big disadvantages:
1) More research - in many games you are forced to expand out a ways with belts before the train research is done.
2) Tougher learning curve. This is, in my mind, the only real disadvantage of trains. Signals are pretty non-intuitive, and I imagine pretty intimidating to new players. I certainly belted my first outpost in my first game because I was too lazy to learn signals There are other concepts of station building that take some figuring out too - e.g. waiting bays.
3) Planning required. If you don't leave the massive space required for train stations, it's pretty hard to shoehorn them in later. In my first train station, I had weird tracks every which way to try and add more resource drop offs, and as my throughput grew, the deadlocks got pretty incredible.

So, to answer the original question - I would say if you're willing to put in the time to learn how signals work, and have a lot of space that you can reserve for a station, go ahead and try it with trains - you'll be glad you did later. On the other hand, if you need that iron ore NOW... go with the belts.

[PunkSkeleton]

Signals seem intuitive to me, what's non-intuitive about that? Green go - segment free, red stop - segment busy. Yellow = segment reserved but this is the same as busy. Much more intuitive than the railways in Poland. There's no real yellow, there's no green+yellow, there's no "red" that means stop and then continue at 20 kph unless you're a heavy cargo train on the hill that can continue without stopping behind that signal... Ahh, yes there's blue on the chain signal (green or red depending on the further path) in Factorio but I haven't really noticed that until recently and that knowledge does not help in building rail network.

But you really do have to reserve a lot of space and invest a lot of resources at the beginning if you want a good rail network.

[HurkWurk]

unintuitive, as i need to place a chain signal first, and not a signal.
to me a chain signal, should be, well, chained to another, not the other way around.

also right of way is not immediately grasped by many people. we tend to think of tracks as pass/no pass, and not as "oncoming traffic, leaving traffic"

to me its a backwards system. a signal should be two way. and a chain signal should chain off a signal.

[Elok]

Not only conveyor are easy to build with drone remotely but the supply chain is a lot more stable with conveyor while you need to build a buffer with train.

Not true. Mining outpost gradually output less and less ore as the drills in the center have a lot more ore than the ones at the edges. Moreover drills at the edges often catch only a few patches and run dry fast. With conveyors I'd either waste mining or run production at less than full capacity or... use buffer as with trains. With large buffers and trains I only have to add a new station when the buffers are running low.

No matter how much you twist it, train add turbulence in the supply chain.

Not only your unloading must be way faster than your intake because of the downtime between train, but there's the logistic of train that don't always scale well (adding more train raise the odds of stopage and traffic) unless you have a dedictated rail for each of your mine.

When you use conveyor, a full conveyor 100 miles away will stay full until it reach it's destination. So it's only a matter of making sure you mine enough ore to fill them up.

In other words, if your base can supply a lot of blue conveyor fast, then using train doesn't worth it.

It depends on scale, really. If you're using only a few blue belts of each ore it is manageable. I'm using 12+...

Taking my base in example, I got a conveyor highway with my 9 line of blue conveyor (5 for iron, 4 for copper) at 100% capacity. And it scale well so I could raise it to 20, 50, 100 easilly (It's a matter of a few copy&paste).

The only, only real downside of conveyor compared to train is the increased cost of iron. But unless you play in a ressource depleded map on expensive mode it won't be an issue. I estimate that less than 5% of my raw ressource that are used to build those conveyor highway. But the time I gain on productivity (adding a new highway is a matter of taking my blueprint and paste it a few time, 1 min top) allow me to work on other things and expand faster, neutralising this downside.

Here's my base : https://drive.google.com/file/d/0B45f44 ... sp=sharing

And here's mine:
https://ibb.co/dvf5ab
https://ibb.co/idXU1G
https://ibb.co/ic4WFb
If you take a look at laser turret ranges you'll understand the difference in scale.

Nice base. You should make a video out of it. If you got Windows 10, there's a great video recorder included : https://gizmodo.com/windows-10-is-hidin ... 1719196149

[Bauer]

You will change your mind in the "Belt vs. Train" discussion if your base becomes bigger. For 9 belts of iron+copper (i.e. 9x40=360 per second) belts are the prefered choice (IMHO). If your base consumes more than about 1000 per second, which is 25 blue belts, it's a real pain with belts. Then you can merrily do your logistics with bots (and maybe a few belts to transport shit to and from you closer mining outposts).

At a level of 4000 per second, you are happy that you can use a few 1000 bots to unload your incoming trains and sort the stuff into the right boxes.
If you're still happy with belts only, it just means that you base is tiny.

Think big!

[Ringkeeper]

Not only your unloading must be way faster than your intake because of the downtime between train, but there's the logistic of train that don't always scale well (adding more train raise the odds of stopage and traffic) unless you have a dedictated rail for each of your mine.

Thats why you have chests at mining outpost/unloading. If you load/unload on each side of the train with max chests per waggon you have more then enough buffer. 1 Station for L-CCCC-L can support 8 blue belts. Need more? Build a second station next to it. Ore not comming fast enough? Add another train.
Even in vanilla, without any train mods, its really simple. If you use just 1 decider combinator per station you can even close them if they are above X amount so trains automatically take next station or wait.

Thats my base at the moment..... starting to build a bit bigger. Not much needed with trains to support this. I would need ALOT of belts for that^^ Specialy when i start to spread out to get everything to "bigger".

https://imgur.com/a/HhCGx
That are 6 iron plate and 9 copper plate smelting rows on top. Full beaconed for compressed blue belt each. Filled from 2 train stations for each ore and with only 2 trains per ore. Oh and one oil and coal train on the top left for the plastic.... red belt of blue circuits needs alot of stuff

*edit* If you play without mods (so no FARL) there are awsome 2/4 lane blueprint books (LHD and RHD)
https://www.reddit.com/r/factorio/comme ... int_books/
https://www.reddit.com/r/factorio/comme ... rints_rhd/

So also not much headache with signaling and you can learn from them, how to do it on your own if you want.

[Elok]

You will change your mind in the "Belt vs. Train" discussion if your base becomes bigger. For 9 belts of iron+copper (i.e. 9x40=360 per second) belts are the prefered choice (IMHO). If your base consumes more than about 1000 per second, which is 25 blue belts, it's a real pain with belts. Then you can merrily do your logistics with bots (and maybe a few belts to transport shit to and from you closer mining outposts).

At a level of 4000 per second, you are happy that you can use a few 1000 bots to unload your incoming trains and sort the stuff into the right boxes.
If you're still happy with belts only, it just means that you base is tiny.

Think big!

As I said, scalability isn't a issue with belt. In fact, it's even easier than train imo since you don't need to add new unloading station or increase train logistic. You only need to add new belt in parallel to the existing one. And with tens of thousand drone, it doesn't take long.

No matter what medium you'll use, in the end it's all conveyor.

[quyxkh]

I'm a little worried that the belt optimizations in 0.16 will make this the most UPS-efficient delivery method too. I don't want to see sixty blue belts replacing one train line.