Add logistics support to trains

[factoriouzr]

Add ability to set a stop as "default", "logisitcs requester", "passive provider", or "active provider". Then when a train stops at this stop, the train and all it's cargo wagons will be treated as the associated chest type and robots can interact with them.

The current system of putting requesters and inserters, or passive providers and inserters etc beside each cargo wagon is not very good.

The issues with it are as follows:
+limits you to only be able to carry x unique items per cargo wagon, where x is the number of inserters you can fit on a cargo wagon on both sides (i think 12).
+requires too much space for loading and unloading
+is not expandable because your base and each outpost requires the cargo wagons to be in the exact same order or the inserters won't match up to the right cargo wagon (the inserters need filters to prevent incorrect loading/unloading, but this prevents them from working if the wrong cargo wagon lines up with them). Thing supply trains that carry multiple goods, drills, lasers, walls, robots etc. to outposts
+is not expandable because as soon as you need more goods to be transported and you run out of room in your current cargo wagons, you have to add another cargo wagon. This then limits you to only add it at the end and you can't reogranize them without completely rebuilding all your outposts. If you just add it to the end and you didn't leave enough room at the train stops, you have to redo your loading areas and all your outposts.
+trains can only go in one direction if they have more then one cargo wagon because otherwise cargo wagons will end up lining up with the wrong inserters if the train turns around somewhere and nothing works
+if you use single cargo wagon trains then you can have double headed trains, but the outposts become too large when you need 5 supply trains (1 for garbage), player access stops, and the loading stops to bring back the ores etc. It gets so large, you can't even zoom out far enough to place the blueprint, so you have to break it up into multiple blueprints
+takes up lots of space in general for all the extra chests and inserters, making outposts with only 3 trains (ore train, supply train, and player access train), surrounded by walls, lasers and with roboport coverage.
+there are other issues as well

With my above solution, this would all be solved. It also would fit perfectly with the game because it could be locked behind another research after automated construction. You could still use the early mechanism for loading and unloading, but after more research you could get logistics support for trains. We already have a natural logistics upgrade to belts for our base factory. I think the trains should have the same logical progression. Right now trains literally fulfill the same role as belts, just with larger throughput and over larger distances. The natural upgrade to belts is logistics robots, and I think the natural upgrade to trains should be logistics trains.

Another great thing that would go well with my suggestion (though not required for it) is this suggestion: viewtopic.php?f=6&t=44238

[Engimage]

While this could really simplify the logistics I am against this.
The reasoning is pretty simple.
We can recall the initial idea of the game - make complex things out of simple parts. So there are already items fulfilling this role. So you are free to use them.
Logistic chests are already overpowered as they allow near to unlimited throughput. And now you suggest to remove the only throughput limitation - inserters - so that unloading will take 2 seconds or whatever.
As for supplying remote defence perimeters (the task that is mentioned the most for this change) you will need train stops anyways to make it automated. And if you make stops nothing can prevent you from using chests.

So again, I do not see how this change will benefit the game (other from making people even more lazy).

[urza99814]

Well, the first issue is that you really shouldn't be emptying every single train using bots. Any efficiency you gained by using a train you just lost by using the bots. Slow, expensive, and power-hungry.

Secondly, learn to use the circuit network. Almost all of the problems you described could be solved by configuring your filter inserters to set their filters based on network conditions. Most of my trains carry only one item, but my wall repair/supply train uses a system like that. I've got six boxes with stack inserters beside the train feeding an exact quantity of turrets, wall segments, all three types of power poles, both kinds of bots, concrete, gates, rails, roboports... at least a dozen different items. The circuit network puts a specific amount of each into the six boxes, when the train arrives it fills, and when the train leaves it resets and refills. That's all in one train car and using only three filter inserters (the rest are regular stack) -- and it could be done with less.

[factoriouzr]

While this could really simplify the logistics I am against this.
The reasoning is pretty simple.
We can recall the initial idea of the game - make complex things out of simple parts. So there are already items fulfilling this role. So you are free to use them.
Logistic chests are already overpowered as they allow near to unlimited throughput. And now you suggest to remove the only throughput limitation - inserters - so that unloading will take 2 seconds or whatever.
As for supplying remote defence perimeters (the task that is mentioned the most for this change) you will need train stops anyways to make it automated. And if you make stops nothing can prevent you from using chests.

So again, I do not see how this change will benefit the game (other from making people even more lazy).

Actually, you don't need to use chests, you could supply the stop directly from the train when it's there. Also did you look at the other suggestion I referenced with the link?

If you don't want to use this mechanism, then you can stick to your chests, nothing is preventing you from playing the game that way when this feature is implemented. However without this feature, players that want to play this way, don't have any options at all. So it's a win+win. This isn't a competitive online game.

[factoriouzr]

Well, the first issue is that you really shouldn't be emptying every single train using bots. Any efficiency you gained by using a train you just lost by using the bots. Slow, expensive, and power-hungry.

Secondly, learn to use the circuit network. Almost all of the problems you described could be solved by configuring your filter inserters to set their filters based on network conditions. Most of my trains carry only one item, but my wall repair/supply train uses a system like that. I've got six boxes with stack inserters beside the train feeding an exact quantity of turrets, wall segments, all three types of power poles, both kinds of bots, concrete, gates, rails, roboports... at least a dozen different items. The circuit network puts a specific amount of each into the six boxes, when the train arrives it fills, and when the train leaves it resets and refills. That's all in one train car and using only three filter inserters (the rest are regular stack) -- and it could be done with less.

You are making assumptions that bots are slow and that power matters.

In late game bots are fast (you know there are speed upgrades right? there are also carrying capacity upgrades), bots also don't have to go a long distance for this setup. When 0.15 is out, with infinite research, perhaps we will have unlimited bot upgrades. Even if we don't there are always mods that improve the speed.

Also regarding power, I don't see it as an issue at all. It's easy to get power, and an outpost doesn't use much power. I only use 1 roboport with 50 bots per outpost or less. All these are assumptions you are making based on your play style. It all depends how each player plays. It depends how often trains come by, how quickly you want to load etc. Also with nuclear power in 0.15 power will be even less of an issue.

How complex is your circuit network to do all this? How much space does it take in combinators etc? How much work is it to set up? Also keep in mind many players (especially new ones) don't like to play with the combinator stuff as it's really complex and intimidating for them. Robots are made easy, why not make trains easy.

Also it's very inconsistent to have a logical upgrade for belts, which is logistics bots, yet no upgrade from trains to logistics trains. Trains are literally an upgrade to belts, so I think they should have a logistics version too. It would just make sense. This is the same reason you don't use burner miners as soon as you get electrical miners. The same reason you upgrade belts and the same reason you use trains instead of belts later on.

[ssilk]

Well, I think this suggestion is one step too far. See at the logistic train network mod, which goes (in my eyes) into the other extreme (a bit too complex etc...). I think the right way is somewhere between theses extremes.

[Optera]

In 0.12 we had a mod named logistic rails turning any wagon parked on them into corresponding chests.
Currently you could try using Logistic Wagon. Configuration scares me a bit though.

Well, I think this suggestion is one step too far. See at the logistic train network mod, which goes (in my eyes) into the other extreme (a bit too complex etc...). I think the right way is somewhere between theses extremes.

You are among those most vocally requesting additional complexity through control signals or even splitting it into multiple networks

[ssilk]

You are among those most vocally requesting additional complexity through control signals or even splitting it into multiple networks

Hmmmmm....

When I say that I think to the future of Factorio. ( viewtopic.php?f=80&t=24090 Collection of End-Game-Ideas / Enhancing Endgame )
Adding more and more complexity to the current game will lead to gameplay without target. Playing Factorio due to mastering the complexity is not a good game. It should go into a "direction". It should tell a story.

The end-game-idea of the devs does that.

And what we need is something, that makes the Factorio life just comfortable enough, just to keep the game going and increasing for the next big steps.

[factoriouzr]

You are among those most vocally requesting additional complexity through control signals or even splitting it into multiple networks

Hmmmmm....

When I say that I think to the future of Factorio. ( viewtopic.php?f=80&t=24090 Collection of End-Game-Ideas / Enhancing Endgame )
Adding more and more complexity to the current game will lead to gameplay without target. Playing Factorio due to mastering the complexity is not a good game. It should go into a "direction". It should tell a story.

The end-game-idea of the devs does that.

And what we need is something, that makes the Factorio life just comfortable enough, just to keep the game going and increasing for the next big steps.

Everything I said supports this game.

By your logic we shouldn't have many things that are already in the game now. We shouldn't have personal roboports, (likely you consider that too comfortable), we shouldn't have regular roboports, no construction bots, no logistics bots. Based on your argument, why even have solar? Isn't that too convenient too? It doesn't generate pollution and doesn't need any resource to be feeding it. What about belts? Aren't upgraded belts too convenient as well? You can lay multiple slower belts and change your setup to get more throughput. There are tons more examples of conveniences such as universal barreling and fluid wagons that are in 0.15. The list keeps going. This game is about automation and researching upgraded trains to allow for train logistics makes perfect sense. Just like every other research upgrade in the game. I think having logistic trains would be another one of those great game changer researches like going from belts to robots, then from robots to personal roboport. There aren't many of these in game now, I think logistics trains would be an amazing addition. I love these game changer researches and I am quite sure the devs mentioned that they like them too. It makes the game more interesting. You want to play until you can get the next cool game changer upgrade, then play around with that tool once you get it.

I'm not trying to offend, but I don't see anything logical about your argument. You just picked on one thing that you like doing a certain way and ignored the conveniences that are already in the game. Many of these conveniences and upgrades have been requested by the community, such as personal roboports. These are all in game now, and the game is still fun to play. Just look at how much people are playing factorio, it's more popular then ever. If you don't like convenience then don't use it. Let everyone play how they like. I feel this would add a lot to the game. It would let people who want to use it, have that option, and you can still use the existing mechanisms to unload/load trains. One of the factorio devs also mentioned that he is thinking of making base deployment easier by launching a rocket that deploys a roboport and robots to a remote location to start off an outpost. I personally love this idea as well. It's all about progression. If you reach late game and are still loading and unloading trains in the same way as early-mid game, you haven't achieved anything meaningful, it's just repetition.

[ssilk]

Everything I said supports this game.

Yes. That is not the question.

By your logic we shouldn't have many things that are already in the game now. We shouldn't have personal roboports, (likely you consider that too comfortable),

No, I supported that idea a lot.

we shouldn't have regular roboports, no construction bots, no logistics bots.

I did never said that.

Due to that I cannot follow your argumentation. It makes no sense to me.

The list keeps going.

I repeat: I didn't argue into that direction. Ever.

upgraded trains to allow for train logistics makes perfect sense.

No question. I play around with logistic in the trains since 3 years and tested every mod and developed some very personal. But I repeat me, that there is a broad way between automate everything and automate nothing. The right way is not clear. I tend to find it somewhere between the extremes, but I cannot be sure. It's a hard struggle, it needs a lot of testing and playing around, cause what feels useful one day is a game-fuckup at the next.

personal roboports.

Well that is one good point that you are not right with your agumentation about me: The personal roboports look nice and it brings a lot of fun into the game, but they are still quite incomplete. I would call them alpha; the personal roboports is just not thought through to the end and currently it's sometimes a mess. For example: Ever tried to deconstruct something big, when your inventory is nearly full. I never feel so helpless - it's just not thought to the end. Or the personal request slots vs. auto trash slots. I can continue the list.

There are SO MANY things, that can be improved before introducing new stuff. As moderator I just try to look objectively to the suggestions and it is just so, that game-value for small improvements is just higher, than for a complete overhaul, or introducing new stuff.

Which doesn't mean, that a train logistic (and other things) should not come. But that step is just a very high one and a "Let us make some train logistic where we take chests and robots"-suggestion, is just not clear, not detailed enough, not thought to the end.

It is not that simple (I can say that, cause - as said - I played around a lot with the different stuff.) And what I try as moderator is pointing into a "direction".

...Maybe it's hard to see and can be misunderstanding...

[factoriouzr]

Everything I said supports this game.

Yes. That is not the question.

By your logic we shouldn't have many things that are already in the game now. We shouldn't have personal roboports, (likely you consider that too comfortable),

No, I supported that idea a lot.

we shouldn't have regular roboports, no construction bots, no logistics bots.

I did never said that.

Due to that I cannot follow your argumentation. It makes no sense to me.

The list keeps going.

I repeat: I didn't argue into that direction. Ever.

upgraded trains to allow for train logistics makes perfect sense.

No question. I play around with logistic in the trains since 3 years and tested every mod and developed some very personal. But I repeat me, that there is a broad way between automate everything and automate nothing. The right way is not clear. I tend to find it somewhere between the extremes, but I cannot be sure. It's a hard struggle, it needs a lot of testing and playing around, cause what feels useful one day is a game-fuckup at the next.

personal roboports.

Well that is one good point that you are not right with your agumentation about me: The personal roboports look nice and it brings a lot of fun into the game, but they are still quite incomplete. I would call them alpha; the personal roboports is just not thought through to the end and currently it's sometimes a mess. For example: Ever tried to deconstruct something big, when your inventory is nearly full. I never feel so helpless - it's just not thought to the end. Or the personal request slots vs. auto trash slots. I can continue the list.

There are SO MANY things, that can be improved before introducing new stuff. As moderator I just try to look objectively to the suggestions and it is just so, that game-value for small improvements is just higher, than for a complete overhaul, or introducing new stuff.

Which doesn't mean, that a train logistic (and other things) should not come. But that step is just a very high one and a "Let us make some train logistic where we take chests and robots"-suggestion, is just not clear, not detailed enough, not thought to the end.

It is not that simple (I can say that, cause - as said - I played around a lot with the different stuff.) And what I try as moderator is pointing into a "direction".

...Maybe it's hard to see and can be misunderstanding...

I was not in any way suggesting that you were against roboports etc. Sorry for any misunderstanding. I was just saying, that since you argued train logistics would be too "convenient", it's very arbitrary where you draw the line. I gave many examples of things that can be considered convenient, and you even said you supported some. This is exactly my point. It's arbitrary what can be considered convenient and that's why your argument doesn't make sense to me. To me if something improves the gameplay or gives more options to players without forcing players to use it, it should be added to the game if enough people want it (which train logistics and train improvements have been requested a lot).

But I repeat me, that there is a broad way between automate everything and automate nothing

I don't see your point here. We already have roboports which would do the exact same thing for pre-roboport stuff like belts as logistics trains would do for trains.

The personal roboports look nice and it brings a lot of fun into the game, but they are still quite incomplete. I would call them alpha; the personal roboports is just not thought through to the end and currently it's sometimes a mess. For example: Ever tried to deconstruct something big, when your inventory is nearly full. I never feel so helpless - it's just not thought to the end. Or the personal request slots vs. auto trash slots. I can continue the list.

Like I said, I agree that existing systems can use improvements, see my example below about the music making added to 0.15. However personal roboports are great right now (not perfect but I always use them as soon as i get them). I really don't like how your robots always take priority over network robots. It would be nice to have a toggle. However with logistics trash slots this doesn't bother me a whole lot anymore because I can just auto trash the stuff I don't want. However the implementation isn't as good as it should be right now (I suggested combining requester and trash slots). As for the inventory full, I agree. I posted a bug about it and many other people are tired of the constant beeping etc.

Which doesn't mean, that a train logistic (and other things) should not come. But that step is just a very high one and a "Let us make some train logistic where we take chests and robots"-suggestion, is just not clear, not detailed enough, not thought to the end.

I think it is thought out very well. It's very simple, have an option in each train stop that makes the train act as a requester or provider chest (as I explained). That's it. All the logic is already in the game with roboports, it just needs to be applied to trains. I don't see how you can say it's not thought out. Also you are saying that my suggestion of train logistics can come, but in your original post it sounded like you were completely against the idea. If you think the priority should be different, you should say that instead of "Well, I think this suggestion is one step too far.". When you say things like this, it sounds like you are completely against the idea of train logistics.

There are SO MANY things, that can be improved before introducing new stuff. As moderator I just try to look objectively to the suggestions and it is just so, that game-value for small improvements is just higher, than for a complete overhaul, or introducing new stuff.

This is not a complete overhaul, see above. All the logic is already in the game. It should be relatively simple to apply it to trains.

I also think (and have posted many times about this topic and gave many suggestions for improvement) that there are many things that need to be improved in the existing implementations of things, blueprints, combining logistics request and trash slots etc. For eg. the new addition of making music in 0.15 is cool, but I think the time should have been spent improving the existing features of the game, especially where we all made many suggestions and want improvements (I don't recall anybody requesting being able to make music). These features are great but there are so many often requested features the community wants, I think those should be higher priority then something nobody requested. Then once these are under control, add the new things. However right now in the game, I feel trains have gotten the least amount of love and are the weakest point of the game (they are an existing system already in the game). They stick out to me like crazy. Train schedule and outpost managements is a pain. The GUI sucks, it's too small. There is no easy schedule management and copy paste. There is no way to aggregate train stops into a train station and tell trains to go to a station and have them figure out what stop they need to go to in that station (eg by my suggestion of labeling stations and stops viewtopic.php?f=6&t=38496). The existing train GUI doesn't even display the name of the train, so once you have a few, it's almost impossible to figure out (at a glance) which train is what.

[factoriouzr]

Just a few examples of discussions around train loading and unloading and how complex it is. The generally accepted solution seems to be the one inserter per item solution with 12 inserters per cargo wagon. It would be better to implement train logistics support to alleviate these problems with train loading, unloading, and train schedule management.

https://www.reddit.com/r/factorio/comme ... _properly/
https://www.reddit.com/r/factorio/comme ... _outposts/

[Optera]

Just a few examples of discussions around train loading and unloading and how complex it is. The generally accepted solution seems to be the one inserter per item solution with 12 inserters per cargo wagon. It would be better to implement train logistics support to alleviate these problems with train loading, unloading, and train schedule management.

https://www.reddit.com/r/factorio/comme ... _properly/
https://www.reddit.com/r/factorio/comme ... _outposts/

Those are not the peak of complexity. Take a look at some of the smart loaders in this thread.
Thankfully 0.15 will make train loading a lot simpler.

[factoriouzr]

Just a few examples of discussions around train loading and unloading and how complex it is. The generally accepted solution seems to be the one inserter per item solution with 12 inserters per cargo wagon. It would be better to implement train logistics support to alleviate these problems with train loading, unloading, and train schedule management.

https://www.reddit.com/r/factorio/comme ... _properly/
https://www.reddit.com/r/factorio/comme ... _outposts/

Those are not the peak of complexity. Take a look at some of the smart loaders in this thread.
Thankfully 0.15 will make train loading a lot simpler.

I wasn't posting those as examples of maximum complexity. In fact quite the opposite. I was posting them as examples of how many people are having issues with train loading and unloading without even trying to do anything complicated.

How will 0.15 make train loading and unloading easier? The only thing I can think of off the top of my head is that you can set the inserter to only pick up one item, but the inserter can still get stuck if the chest is full or the train is full of the item being loaded.

[folk]

Shameless plug incoming. Take heed my warning, fellow lost souls, the mod is not for the faint of heart.

https://mods.factorio.com/mods/folk/folk-logistic-wagon

[ssilk]

Sorry for the long time between the posts, but I was a bit ill - as some might have seen due to my activity on this forum.

This is not a complete overhaul, see above. All the logic is already in the game. It should be relatively simple to apply it to trains.

Right, I think that is as I see it currently:
- because it is just based on existing mechanics it doesn't bring in new - logically. (Yes, it's not a complete overhaul and I never said that, sorry for being unclear; my statement was more about general improvement out of the sight of a moderator). Which is a sign of low added game-value.
- There where similar mods about this. So this can eventually be also implemented as mod. Which leads me to this thought: If I can write one mod, I can write many (in theory and - as you might see with the number of mods counted in this thread - also in reality). And when there are many mods possible around one basic concept, then this is just one idea of many possible.

It doesn't mean, this suggestion is bad or so. It just has more meaning: There is no pressure to implement that exactly like so. And - that is for me the important idea - it points into a direction of "There is a possible game value, that just needs to be digged out". And "Just" is underlined 20 times as an ironic understatement.

These features are great but there are so many often requested features the community wants, I think those should be higher priority then something nobody requested.

You're completely wrong with this statement. All of that has been requested many times, but it was "before your time". About 3-4 years ago. I remember for example this one (because I made it myself):
viewtopic.php?f=6&t=921 Isle of tune mod
or this one:
viewtopic.php?f=6&t=902 Control voltage
And so it is with the others!

That ideas have been implemented NOW. And we as community need to learn to think in about this time-distances of more than 1-3 years, before something will be implemented into Factorio on a regular base. (Which doesn't mean, that all things will need so long, but most.)

Thinking into that direction it is clear, that the planning and evaluating becomes much, much more important. We - as community - need to think into that dimensions, because any change will make Factorio codebase more complex, integrated and entangled. It will become harder and harder to implement simple details. Code must be rewritten and that will change behavior. But this is not the subject of this discussion; I wanted to explain that developing an idea is had work.

Then once these are under control, add the new things. However right now in the game, I feel trains have gotten the least amount of love and are the weakest point of the game (they are an existing system already in the game). They stick out to me like crazy. Train schedule and outpost managements is a pain. The GUI sucks, it's too small. There is no easy schedule management and copy paste. There is no way to aggregate train stops into a train station and tell trains to go to a station and have them figure out what stop they need to go to in that station (eg by my suggestion of labeling stations and stops viewtopic.php?f=6&t=38496). The existing train GUI doesn't even display the name of the train, so once you have a few, it's almost impossible to figure out (at a glance) which train is what.

THAT are suggestions which are going much more into that direction. Cause they will bring much more (in my opinion) quick game-value and enable a change, which is currently unthinkable. Changing and tidying rails/trains up will open the clear view of "what's really needed" to be improved with the train-logistics. Cause currently it just "works". (Quite elegant in the beginning, but with advanced knowledge you miss more and more things.)

For example I pick one of the not-so-obvious things out and explain it:

Since nearly the implementation of trains it is possible to have train stops with the same name. And nearly anybody that hears that says "Cool trick". And it is, what it is: A nice and elegant programming trick to reduce game logic.

I explain it a bit better, so that everybody can follow: If you name two stops with the same name, they act like one "train station" (hm, about like so, the distance plays a big role here, but who should name two train stops equally, which have two kilometers distance? (Right: That is the nature of such "tricks": They are based around "obvious behavior".)).

This trick works - as said - nicely with the current train-system. Indeed it's the perfect solution!

But with the LTN-mod for example (and with any serious implementation of logistic for trains, which means: there are no longer static routes, the routes are calculated by the logistics part) it will not longer be possible to use that: If I request the same amount of items over two train stops (for example, I request 1000 iron ore at two train stops) it is not clear for the game (and the mod), what's really meant. Do I request 1000 iron at station A and 1000 at A'? (note the small ' !) Or 1000 at one of the two A-stations? Or does it mean: First 1000 at A and when delivered 1000 at A' and vice versa?

The correct answer of this question is not simple. They introduce other questions: Do the train-stops sit in the same logistic network? How distant are they? And many more questions.
Currently the LTN mod forbids this case by checking the stop-names, but for a Factorio version of LTN this needs to be much more integrated.
The point is: Even if the game would be able to answer those questions and decide how the two train stops should be handled, it could be wrong. There is no limitation of using this feature. And the right way to do that is not clear: There are many ways to do it "about right".

So, the only logical answer is: This question can only be answered by the player! The player must explain how to use two somehow connected stations. And that they have equal names is then just an option.

This feature is missing BEFORE anything serious can be done with logistic trains and such stuff. I mean serious enough to be implemented into the game. Cause I want to avoid misconceptions like with the personal roboport.
And I can list you some more of such "must haves" , before we can think about logistic trains, but I just don't want to dictate this board. Ehm.

So what doing is I'm searching for are solutions which are clean and simple and integrate perfect. This is not the case with this suggestion, cause it doesn't even mention this problem. Which doesn't mean it is bad or wrong or something! It just means: We (the community) needs to think and search deeper.

[Optera]

How will 0.15 make train loading and unloading easier? The only thing I can think of off the top of my head is that you can set the inserter to only pick up one item, but the inserter can still get stuck if the chest is full or the train is full of the item being loaded.

Train stops can output train inventory. It seems like they don't output fluids though

[factoriouzr]

Sorry for the long time between the posts, but I was a bit ill - as some might have seen due to my activity on this forum.

This is not a complete overhaul, see above. All the logic is already in the game. It should be relatively simple to apply it to trains.

Right, I think that is as I see it currently:
- because it is just based on existing mechanics it doesn't bring in new - logically. (Yes, it's not a complete overhaul and I never said that, sorry for being unclear; my statement was more about general improvement out of the sight of a moderator). Which is a sign of low added game-value.
- There where similar mods about this. So this can eventually be also implemented as mod. Which leads me to this thought: If I can write one mod, I can write many (in theory and - as you might see with the number of mods counted in this thread - also in reality). And when there are many mods possible around one basic concept, then this is just one idea of many possible.

It doesn't mean, this suggestion is bad or so. It just has more meaning: There is no pressure to implement that exactly like so. And - that is for me the important idea - it points into a direction of "There is a possible game value, that just needs to be digged out". And "Just" is underlined 20 times as an ironic understatement.

These features are great but there are so many often requested features the community wants, I think those should be higher priority then something nobody requested.

You're completely wrong with this statement. All of that has been requested many times, but it was "before your time". About 3-4 years ago. I remember for example this one (because I made it myself):
viewtopic.php?f=6&t=921 Isle of tune mod
or this one:
viewtopic.php?f=6&t=902 Control voltage
And so it is with the others!

That ideas have been implemented NOW. And we as community need to learn to think in about this time-distances of more than 1-3 years, before something will be implemented into Factorio on a regular base. (Which doesn't mean, that all things will need so long, but most.)

Thinking into that direction it is clear, that the planning and evaluating becomes much, much more important. We - as community - need to think into that dimensions, because any change will make Factorio codebase more complex, integrated and entangled. It will become harder and harder to implement simple details. Code must be rewritten and that will change behavior. But this is not the subject of this discussion; I wanted to explain that developing an idea is had work.

Then once these are under control, add the new things. However right now in the game, I feel trains have gotten the least amount of love and are the weakest point of the game (they are an existing system already in the game). They stick out to me like crazy. Train schedule and outpost managements is a pain. The GUI sucks, it's too small. There is no easy schedule management and copy paste. There is no way to aggregate train stops into a train station and tell trains to go to a station and have them figure out what stop they need to go to in that station (eg by my suggestion of labeling stations and stops viewtopic.php?f=6&t=38496). The existing train GUI doesn't even display the name of the train, so once you have a few, it's almost impossible to figure out (at a glance) which train is what.

THAT are suggestions which are going much more into that direction. Cause they will bring much more (in my opinion) quick game-value and enable a change, which is currently unthinkable. Changing and tidying rails/trains up will open the clear view of "what's really needed" to be improved with the train-logistics. Cause currently it just "works". (Quite elegant in the beginning, but with advanced knowledge you miss more and more things.)

For example I pick one of the not-so-obvious things out and explain it:

Since nearly the implementation of trains it is possible to have train stops with the same name. And nearly anybody that hears that says "Cool trick". And it is, what it is: A nice and elegant programming trick to reduce game logic.

I explain it a bit better, so that everybody can follow: If you name two stops with the same name, they act like one "train station" (hm, about like so, the distance plays a big role here, but who should name two train stops equally, which have two kilometers distance? (Right: That is the nature of such "tricks": They are based around "obvious behavior".)).

This trick works - as said - nicely with the current train-system. Indeed it's the perfect solution!

But with the LTN-mod for example (and with any serious implementation of logistic for trains, which means: there are no longer static routes, the routes are calculated by the logistics part) it will not longer be possible to use that: If I request the same amount of items over two train stops (for example, I request 1000 iron ore at two train stops) it is not clear for the game (and the mod), what's really meant. Do I request 1000 iron at station A and 1000 at A'? (note the small ' !) Or 1000 at one of the two A-stations? Or does it mean: First 1000 at A and when delivered 1000 at A' and vice versa?

The correct answer of this question is not simple. They introduce other questions: Do the train-stops sit in the same logistic network? How distant are they? And many more questions.
Currently the LTN mod forbids this case by checking the stop-names, but for a Factorio version of LTN this needs to be much more integrated.
The point is: Even if the game would be able to answer those questions and decide how the two train stops should be handled, it could be wrong. There is no limitation of using this feature. And the right way to do that is not clear: There are many ways to do it "about right".

So, the only logical answer is: This question can only be answered by the player! The player must explain how to use two somehow connected stations. And that they have equal names is then just an option.

This feature is missing BEFORE anything serious can be done with logistic trains and such stuff. I mean serious enough to be implemented into the game. Cause I want to avoid misconceptions like with the personal roboport.
And I can list you some more of such "must haves" , before we can think about logistic trains, but I just don't want to dictate this board. Ehm.

So what doing is I'm searching for are solutions which are clean and simple and integrate perfect. This is not the case with this suggestion, cause it doesn't even mention this problem. Which doesn't mean it is bad or wrong or something! It just means: We (the community) needs to think and search deeper.

So what doing is I'm searching for are solutions which are clean and simple and integrate perfect. This is not the case with this suggestion, cause it doesn't even mention this problem. Which doesn't mean it is bad or wrong or something! It just means: We (the community) needs to think and search deeper.

Your criticism has nothing to do with my suggestion. I think this "problem" you are mentioning has nothing to do with my suggested solution and doesn't need to be accounted for in my suggestion. It's up to the player what they decide to implement. When a train stops at a stop, it is treated as a normal train or a requester, or a passive provider based on the setting of the stop. If you have a separate roboport network, then it will behave as such in that network, if this network is part of your main factory, then it will still behave the same but in your main factory. This is exactly by design of why I suggested it this way. This gives the flexibility to the players and allows for things like making the train a requester when it stops at your main factory and the robots will bring the requested items, and we can make it a passive provider at each outpost (each outpost is it's own robotic network).

I think you are just confusing ideas.

Simply put, robot logistics are in game now, treating a train as a chest is completely consistent with the current design of the game and would fit nicely.

There are other features needed for trains, such as better GUIs, better schedule management etc. This feature that I'm suggesting would fit perfectly into the current game and only enhance it. I'm not sure why you are so adamant about arguing against it.

[factoriouzr]

How will 0.15 make train loading and unloading easier? The only thing I can think of off the top of my head is that you can set the inserter to only pick up one item, but the inserter can still get stuck if the chest is full or the train is full of the item being loaded.

Train stops can output train inventory. It seems like they don't output fluids though

Right, that's a cool feature. It doesn't replace my suggestion in any way, but it's definitely a good change that will help dealing with trains in different situations.

[ssilk]

I think you are just confusing ideas.

Indeed.

Because of the title I thought you meant with "logistics" that trains search their way dynamically.

But communication is always between two sides.