Chest-on-a-belt
Moderator: ickputzdirwech
-
- Long Handed Inserter
- Posts: 86
- Joined: Fri Jan 31, 2014 8:47 pm
- Contact:
Chest-on-a-belt
Abstract. A chest able to travel via conveyor belts. It distributes resources from chests to assembly machines to chests.
What is wrong with conveyor belts?
• If input is greater than output then the belt is full, and you have all the resources on it just laying around doing nothing. It also feels like your production is halted because resources seem static.
• If input is lower than output then the belt is empty, and only first factories are getting resources leaving nothing to the back end. The production freezes, even if you have enough resources to produce everything in a slower pace.
• You either produce at full rate, or produce nothing.
• If the belt cannot handle your factory’s needs, that is it. You need to upgrade the belt.
• Resources are placed and taken one at a time. If belt’s input inserters combined cannot feed your assemblers combined, your factory is starving.
• They take a lot of space. You need a separate belt for each kind of resource your factory needs. It makes ugly looking squiggly belt lines with tons of intersections.
• Belt based factories grow to maintenance hell very quickly.
The answer to the belts is logistics bots.
What is wrong with logistics bots?
Disclaimer: I have only played with 7.5 bots.
There are two major gameplay points in the game: designing and maintaining. Logistics bots take them both and turn into a joke. Here is how a bot based factory looks like
Requester |-> assembly |-> … |-> assembly |-> provider
(The number of assemblies in between is low: 1-3)
If you need to produce anything, you follow this pattern. This is a no-brainer, and no-brainers are bad.
A-chest-on-a-belt.
The design is similar to trains.
You place a (special) chest on a (special) belt.
A new piece of belt: station belt. It is not moving by default. Whenever a chest enters the belt, the countdown initiates. After player defined time passes, station belt propagates the chest to the next belt (probably a normal one). During this time stations inserters can take from and put into the chest. A special instruction to tell the station belt to wait until inserters fill the chest.
The chest is a requester chest. While the chest is waiting on a loading station, smart inserters fill the chest with requested items. The inserters can be programmed to fill the chest only if the network is short of some items, so the chest will wait at the station. While the chest is waiting on an unloading station, smart inserters unload some of its contents (defined at the station). The player defines the items to unload and their quantity.
Chest routing is similar to train routing. A chest has the list of the stations to visit and a splitter routes the chest to the defined station.
Vs belts:
• A player knows what resources he needs to build an item, so he can configure the chest to take the needed amount in the loading station. The chest will then go through unloading stations near the assembly machines and split the resources in such a manner that no overflow or starvation happens.
• A player knows the production time, so he can set chest interval such that the factory will get as much chests as it needs.
• Any factory will need exactly one belt to function
• If you have not enough resources per minute for constant production, the factory will produce at lower rates, but will not starve.
• The belt bandwidth is measured in chests per tile.
• The design is concentrated around the loading stations as you can have different routes feeding from one station.
Vs bots:
• Player needs to design harder when using chests-on-a-belt.
• Bots distribute resources better and easier.
• Bots are no-brainers.
Vs trains:
• Chests take less space.
• Chests can be programmed.
• Chests require no fuel.
• Trains feel like an overkill.
Overall.The chest are less of a maintenance hell than belts and more fun than bots. The bots are objectively better than the chests in terms of optimal play because they can distribute the resources among many requesters with ease.
Resources go through factories in chests, and with a proper design a single chest going through a factory guarantees exactly 1 (2, 3, ... depending on the design) end product.
What is wrong with conveyor belts?
• If input is greater than output then the belt is full, and you have all the resources on it just laying around doing nothing. It also feels like your production is halted because resources seem static.
• If input is lower than output then the belt is empty, and only first factories are getting resources leaving nothing to the back end. The production freezes, even if you have enough resources to produce everything in a slower pace.
• You either produce at full rate, or produce nothing.
• If the belt cannot handle your factory’s needs, that is it. You need to upgrade the belt.
• Resources are placed and taken one at a time. If belt’s input inserters combined cannot feed your assemblers combined, your factory is starving.
• They take a lot of space. You need a separate belt for each kind of resource your factory needs. It makes ugly looking squiggly belt lines with tons of intersections.
• Belt based factories grow to maintenance hell very quickly.
The answer to the belts is logistics bots.
What is wrong with logistics bots?
Disclaimer: I have only played with 7.5 bots.
There are two major gameplay points in the game: designing and maintaining. Logistics bots take them both and turn into a joke. Here is how a bot based factory looks like
Requester |-> assembly |-> … |-> assembly |-> provider
(The number of assemblies in between is low: 1-3)
If you need to produce anything, you follow this pattern. This is a no-brainer, and no-brainers are bad.
A-chest-on-a-belt.
The design is similar to trains.
You place a (special) chest on a (special) belt.
A new piece of belt: station belt. It is not moving by default. Whenever a chest enters the belt, the countdown initiates. After player defined time passes, station belt propagates the chest to the next belt (probably a normal one). During this time stations inserters can take from and put into the chest. A special instruction to tell the station belt to wait until inserters fill the chest.
The chest is a requester chest. While the chest is waiting on a loading station, smart inserters fill the chest with requested items. The inserters can be programmed to fill the chest only if the network is short of some items, so the chest will wait at the station. While the chest is waiting on an unloading station, smart inserters unload some of its contents (defined at the station). The player defines the items to unload and their quantity.
Chest routing is similar to train routing. A chest has the list of the stations to visit and a splitter routes the chest to the defined station.
Vs belts:
• A player knows what resources he needs to build an item, so he can configure the chest to take the needed amount in the loading station. The chest will then go through unloading stations near the assembly machines and split the resources in such a manner that no overflow or starvation happens.
• A player knows the production time, so he can set chest interval such that the factory will get as much chests as it needs.
• Any factory will need exactly one belt to function
• If you have not enough resources per minute for constant production, the factory will produce at lower rates, but will not starve.
• The belt bandwidth is measured in chests per tile.
• The design is concentrated around the loading stations as you can have different routes feeding from one station.
Vs bots:
• Player needs to design harder when using chests-on-a-belt.
• Bots distribute resources better and easier.
• Bots are no-brainers.
Vs trains:
• Chests take less space.
• Chests can be programmed.
• Chests require no fuel.
• Trains feel like an overkill.
Overall.The chest are less of a maintenance hell than belts and more fun than bots. The bots are objectively better than the chests in terms of optimal play because they can distribute the resources among many requesters with ease.
Resources go through factories in chests, and with a proper design a single chest going through a factory guarantees exactly 1 (2, 3, ... depending on the design) end product.
Last edited by Plop and run on Tue Feb 11, 2014 4:03 pm, edited 1 time in total.
Black blood of the factory
Re: A-chest-on-a-belt
You may laugh, but you can already have "chests" on belts. Almost.
All that's missing is a way to pause the car.
All that's missing is a way to pause the car.
Re: A-chest-on-a-belt
Well there is also moving chest that pauses in stations. It's called a train.
-
- Long Handed Inserter
- Posts: 86
- Joined: Fri Jan 31, 2014 8:47 pm
- Contact:
Re: A-chest-on-a-belt
I did compare my idea with trains. See Vs train section.FrozenOne wrote:Well there is also moving chest that pauses in stations. It's called a train.
Black blood of the factory
Re: A-chest-on-a-belt
Id love a moving chest.
Maybe i can replace the 400 logistic bots i have on my current factory.
Maybe i can replace the 400 logistic bots i have on my current factory.
Re: A-chest-on-a-belt
I've experimented a lot with this. The best thing, what I've made was a car wash: A long belt, many inserters, I turned the speed to 0.1 and played this song: http://www.youtube.com/watch?v=3v8I5egzoMo
Sitting and laughing.
(Would be so cool: After a long run outside the field the car needs a wash: Dirty car in, clean car out)
Sitting and laughing.
(Would be so cool: After a long run outside the field the car needs a wash: Dirty car in, clean car out)
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Re: Chest-on-a-belt
How is your idea different from "make trains smaller and make them not use fuel"?
-
- Long Handed Inserter
- Posts: 86
- Joined: Fri Jan 31, 2014 8:47 pm
- Contact:
Re: Chest-on-a-belt
• You can specify what items (and their quantity!) the chest should take.immibis wrote:How is your idea different from "make trains smaller and make them not use fuel"?
• You can configure your loading station such that the chest leaves only when it got it's items. Stations would send a new chest to the assemblies as soon as it has enough resources inside. It has advantages described in the Vs belt section.
• You can configure unloading stations such that inserters would take a specified amount of items, just enough to continue the production and leave the rest to the next assembly.
• Belts can have 90 degrees turns. Saves much space.
• Chests can not collide with each other. Especially useful when you have one chest waiting to be loaded.
• Trains are designed to carry tons of resources over large distances. Not 2 iron plates and 1 electronic circuit from a provider chest to an inserter factory.
• A wagon can be loaded/unloaded by 10 inserters simultaneously.
Black blood of the factory
Re: Chest-on-a-belt
• So basically a new kind of inserter that keeps a certain number of items in a container?Plop and run wrote:• You can specify what items (and their quantity!) the chest should take.immibis wrote:How is your idea different from "make trains smaller and make them not use fuel"?
• You can configure your loading station such that the chest leaves only when it got it's items. Stations would send a new chest to the assemblies as soon as it has enough resources inside. It has advantages described in the Vs belt section.
• You can configure unloading stations such that inserters would take a specified amount of items, just enough to continue the production and leave the rest to the next assembly.
• Belts can have 90 degrees turns. Saves much space.
• Chests can not collide with each other. Especially useful when you have one chest waiting to be loaded.
• Trains are designed to carry tons of resources over large distances. Not 2 iron plates and 1 electronic circuit from a provider chest to an inserter factory.
• A wagon can be loaded/unloaded by 10 inserters simultaneously.
• See above
• See above
• This comes under "smaller"
• What do they do instead - phase through each other?
• What things do matters much more than what they're designed to do.
• Side effect of "smaller"
Re: Chest-on-a-belt
This kind of discussion won't bring anything further. I can argue, that most things Plopandrun describes already work with trains. Or will work, because already discussed, or because it works like so in OpenTTD. The better argument is the funny fact, that I have also made the same suggestion, something like moving bots on tracks, like in real factories, which transports stacks and refill the assemblies directly. Or the containers idea...
I won't search it out (doityourself), because I think currently, that this is not the right way. And the reason is simple: Factories with that dense built up won't look good.
Why do they need to look good? The argument to save space is from a sight where we didn't see the possible task, that is to be done: bring down one million people. And because this game has so much space, which is currently not used. I found me often at points, where I come back from fighting or building in the outlands, thinking "woa, this is so small, why didn't I made more room?"
So, what I really like from Plopandrun's idea, is to close a gap between trains, belts and logistic bots, but it should take more room. There are many threads about "we need streets" or "we need ground plates" or "want to walk above the assemblies" and some others. And my current favorite is to have automated cars, like trains, but much easier and more flexible. A track for that needs about 3-5 tiles width in both directions. Of course, you can use it with the current car, to drive into town, all kind of cars are compatible to that system. The tracks are flexible, because they are just pairs of poles at the side of the track, very lite weighted, more or less like power poles, think to electrified busses. These tracks can be later advanced with plates, which makes driving faster.
And when I try to imagine, how that will look then, and what I need to change to my factories to include that, I think this could be very, very nice.
I won't search it out (doityourself), because I think currently, that this is not the right way. And the reason is simple: Factories with that dense built up won't look good.
Why do they need to look good? The argument to save space is from a sight where we didn't see the possible task, that is to be done: bring down one million people. And because this game has so much space, which is currently not used. I found me often at points, where I come back from fighting or building in the outlands, thinking "woa, this is so small, why didn't I made more room?"
So, what I really like from Plopandrun's idea, is to close a gap between trains, belts and logistic bots, but it should take more room. There are many threads about "we need streets" or "we need ground plates" or "want to walk above the assemblies" and some others. And my current favorite is to have automated cars, like trains, but much easier and more flexible. A track for that needs about 3-5 tiles width in both directions. Of course, you can use it with the current car, to drive into town, all kind of cars are compatible to that system. The tracks are flexible, because they are just pairs of poles at the side of the track, very lite weighted, more or less like power poles, think to electrified busses. These tracks can be later advanced with plates, which makes driving faster.
And when I try to imagine, how that will look then, and what I need to change to my factories to include that, I think this could be very, very nice.
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Re: Chest-on-a-belt
With all this discussion I realized that the big chest I always wanted to have is already in the gave, just a wagon with the minimun set of rails. Cool for modules productions where all of them need circuits but the module from before
Re: Chest-on-a-belt
And how do you limit the production?
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Re: Chest-on-a-belt
Haven't reach that part of my plan yetssilk wrote:And how do you limit the production?
-
- Long Handed Inserter
- Posts: 86
- Joined: Fri Jan 31, 2014 8:47 pm
- Contact:
Re: Chest-on-a-belt
Ok, it's been a long time, but can you tell me what you don't like? Thanks!ssilk wrote:what I really like from Plopandrun's idea
Black blood of the factory
Re: Chest-on-a-belt
If I say I like something, it doesn't mean, that I don't like the rest!?
Well. I liked the idea of having smaller units moving around your factory. Not the needed low size, not the other details, not that it is too much "rectangle", and not round, not natural. And in combination with the other things (imibis post directly behind yours), I had my own ideas. And when you read my posts since last week about cars I wrote many things about cars moving around.
{Currently:
The streets are simple: Just some "buttons", you place. Like the poles, but no wire (because e. g. by dragging you tell the direction of the connection to other buttons); a new street is a thing of some seconds and reminds a bit to electrified bus-lines, but the cable runs on the ground and is not electrified, it's only that the cars find the way (electrification maybe later).
You have car-stops and automatic-cars can drive in this network, like trains, but the streets are much more "flexible", a 180 turn within 4 tiles, and you need to take care when building the track, if the cars touch something and break it or it collides with other cars on the opposite track... (more interesting if we are able to add trailers). Might be a very interesting thing to plan routes through your factory... You need also to make one-way-streets or have traffic-lights. etc. Must all be researched and OpenTTD is the big prototype for that, but - as said - the streets should be much more free and the automatic cars should be definitely no concurrence to the train. They are something between expensive belts, cheap logistic bots and narrow-gauge/one-rail train, very flexible, very fast built up.
And the side-effect is is the needed size and that this makes a real-looking street, which I can use (if researched) also with my car to avoid crashing everywhere.}
So your idea was inspiring. And that the reason for writing them down here.
Well. I liked the idea of having smaller units moving around your factory. Not the needed low size, not the other details, not that it is too much "rectangle", and not round, not natural. And in combination with the other things (imibis post directly behind yours), I had my own ideas. And when you read my posts since last week about cars I wrote many things about cars moving around.
{Currently:
The streets are simple: Just some "buttons", you place. Like the poles, but no wire (because e. g. by dragging you tell the direction of the connection to other buttons); a new street is a thing of some seconds and reminds a bit to electrified bus-lines, but the cable runs on the ground and is not electrified, it's only that the cars find the way (electrification maybe later).
You have car-stops and automatic-cars can drive in this network, like trains, but the streets are much more "flexible", a 180 turn within 4 tiles, and you need to take care when building the track, if the cars touch something and break it or it collides with other cars on the opposite track... (more interesting if we are able to add trailers). Might be a very interesting thing to plan routes through your factory... You need also to make one-way-streets or have traffic-lights. etc. Must all be researched and OpenTTD is the big prototype for that, but - as said - the streets should be much more free and the automatic cars should be definitely no concurrence to the train. They are something between expensive belts, cheap logistic bots and narrow-gauge/one-rail train, very flexible, very fast built up.
And the side-effect is is the needed size and that this makes a real-looking street, which I can use (if researched) also with my car to avoid crashing everywhere.}
So your idea was inspiring. And that the reason for writing them down here.
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Re: Chest-on-a-belt
I like the idea!!! may I offer some opinions??ssilk wrote:If I say I like something, it doesn't mean, that I don't like the rest!?
Well. I liked the idea of having smaller units moving around your factory. Not the needed low size, not the other details, not that it is too much "rectangle", and not round, not natural. And in combination with the other things (imibis post directly behind yours), I had my own ideas. And when you read my posts since last week about cars I wrote many things about cars moving around.
{Currently:
The streets are simple: Just some "buttons", you place. Like the poles, but no wire (because e. g. by dragging you tell the direction of the connection to other buttons); a new street is a thing of some seconds and reminds a bit to electrified bus-lines, but the cable runs on the ground and is not electrified, it's only that the cars find the way (electrification maybe later).
You have car-stops and automatic-cars can drive in this network, like trains, but the streets are much more "flexible", a 180 turn within 4 tiles, and you need to take care when building the track, if the cars touch something and break it or it collides with other cars on the opposite track... (more interesting if we are able to add trailers). Might be a very interesting thing to plan routes through your factory... You need also to make one-way-streets or have traffic-lights. etc. Must all be researched and OpenTTD is the big prototype for that, but - as said - the streets should be much more free and the automatic cars should be definitely no concurrence to the train. They are something between expensive belts, cheap logistic bots and narrow-gauge/one-rail train, very flexible, very fast built up.
And the side-effect is is the needed size and that this makes a real-looking street, which I can use (if researched) also with my car to avoid crashing everywhere.}
So your idea was inspiring. And that the reason for writing them down here.
-For this to not replace compleatly the train, waggons should be powered up, maybe having the train carry up to (+-) 6 wagons without slowing it so much, or making the wagons carry twice as many items than now.
- The auto cars should be slow (if you're talking about they traveling through a line... MAYBE make them MONORAILS????)
- Have them speed up with research, but not too much
- Have "car stops" of 2 or 3 kinds: one that work as provider chest, one as requester (and maybe one as storage) so that the car knows if it has to load or unload there... also make them work NOT ONLY by time schedule, but also by "Stay here untill car is full/empty".
- MAYBE Make them come with a robotic arm (buit in inserter) so they can load-unload automatically in the stop (i don't think i'd like having inserters all around the street to load the car, but this is only IMO)
Love the ideas of the plated roads, they would look sooo nice!!!!
Re: Chest-on-a-belt
One of those big postings.
Well I see them as option: 3 locomotives, 10-12 wagons, a distance of 1500 tiles, double tracks, and 6 of those trains on this tracks. Transportation 10000-30000 items per minute are possible with that, no question: THAT is fun. But for the needs of gameplay I would say, small is beautiful.
But they will look really cool... I think I could sit before the screen and watching them driving around.
Hmmm.
Now, when I have some more experience with building big factories, I think this is so very usable to cover the "area".
What do I mean? We have currently the belts, the logistic bots and the trains.
- The belts can cover big areas, but they are not thought for it. It takes much resources and much time and after clearing all resources in the area, they are laying on the ground and doing nothing, they are not worth to be mined. What a waste!
- The roboports have a coverage of 50x50 tiles for the logistic area. Now when you want to cover a big area, it hurts. Alone the time to produce the roboports. And the placement! And the energy! (I think they use too less energy yet and the bots should also use more energy.)
- Then we have the trains. But they hurt also. They are not thought for such small distances under 200 tiles. That takes too much space and is just ugly gameplay.
My picture of that is like so: There are "islands". Eventually you know the game Siedler II, which has islands? An island has also requests and offers, and a boatman transported them between island and land. This here is about the same idea, but a bit more complicated.
So lets assume we have solved the problem, that the "islands" know, how much is available and how much is needed. This alone is a big task.
But then you can create real logistic between those islands.
Which might look so:
- A truck drives to a truck-station.
- Is loaded.
- If full or nothing else available he drives to the target and is unloaded.
Not simpler thoughtable!
(Well of course you could also use transport belts for that job of exchanging between the logistic areas. But as mentioned above the problem in my opinion is, that it takes much longer to built such long transport belts. With this truck system which needs no streets in the first tier you don't need not built anything, but the truck-stations. So less resources and less items, so it is faster built up and can be built by yourself, while a far away belt is an difficult task, you need hundreds of belts and cannot hold everything into the inventory.)
Well, not everything is thought through. The loading and unloading for example. But when trucks are available before logstic I don't see any other option as with inserters and chests. But this is not so important! The point is, that this works in some ways much more effective than the trains, the belts and the logistic bots! It's just a 4th option. I didn't liked the trucks/busses in OpenTTD not that much, because of their inefficiency and the need to so much planing and micromanagement, but I think this trucks I would love, because they work from in the form of how you place the truck stations and where and how many streets you built afterwards. In the beginning they drive criss-cross and then you see the most used paths and make more place for them and place poles. Then you see that you loose many trucks because the trucks and trains collide, because the trucks cross the tracks without looking left and right and you built some kind of train-crossing. Then you see, they find a way through a wood and are killed by some biters.
And the biggest advantage is in my eyes, that you are forced to leave some space through your factory. I guarantee, that this will make the factories look much nicer and simpler.
For the devs the big advantage of that: when it is clear, that this new type of transportation (transport by the need of the single logistic networks) works, it can be also introduced for the trains! Not too much differences. The advantage is then, that we don't need to handle all the routing stuff. I liked that routing options, the "how long should a train wait at each station" for OpenTTD very much for that, but I think Factorio is one or two numbers smaller than that (or bigger, depends from standpoint ) and I think making all this micromanagement doesn't bring this game very much forward.
Yes, of course this is a bit naive. It doesn't make sense to order a truck from about 2000 tiles away. Maybe they have a home-station. But I think routes are not so much needed. We should not think, that everything needs routes and all the micromanagement. This brings the game much more forward and the rest are solvable small problems.
Ok, there should be routes for the trains. But only the big routes, the very, very big amounts of resources. The long ranges, because that is more efficient. In OpenTTD we made up routes, which are as long as possible to get the biggest cash, but in factorio we made those routes, because it is more efficient to have some big transport over long distances and not many small for short; not for those 2-4 mining drills 350 tiles outside, which needs a connection to the rest. This eats just up time...
Exactly therefore are the trucks!
And in the end I would like to see trains, which run in network to every point the train can ride to, to fetch up the next important items, too. Some train depots, which I the train drive into, when currently not needed (or the trucks, to come back to the streets). So the route-handling isn't needed to be so deep as in OpenTTD; the signal handling must not be so perfect, the path-finding could have some small issues, which make the trucks/trains look stupid, but in the end the network works and does it's job, the players have a playable game and the devs can do more important stuff, instead of fixing the bugs for a perfect routing.
This are all advantages, which will bring the game very much forward into the wanted direction.
Additionally: what I currently think is, that we have trucks which works like a rolling roboport. You can fill in some robots and the needed items for some blueprints and drive them to some area. You place stuff via blueprints/ghost-building and robots swarm out and place it. You built up a small mining site within 1-3 minutes, instead of 10-20. Powered by solar. Without any other needed connection. And the resources are transported by the trucks to the next big train station, just because it is needed so. Just without very much further programming/micromanagement.
Well. I think 6 wagons is a bit too long. Because with 6 wagons you have about 6000 items you can transport per run. Two chests of resources. And the train stations get so big. You need to built up so much. Even if blueprints are a big help, but I mean this is not needed. And there are many other issues. I mean indeed, that the optimal size for wagons is about 3-4 per locomotive. I tested other combinations and my feeling here is the best. Train stations and tracks for 6 wagons are so ... big and you have so much afford with them.Sedado77 wrote:I like the idea!!! may I offer some opinions??
-For this to not replace compleatly the train, waggons should be powered up, maybe having the train carry up to (+-) 6 wagons without slowing it so much, or making the wagons carry twice as many items than now.
Well I see them as option: 3 locomotives, 10-12 wagons, a distance of 1500 tiles, double tracks, and 6 of those trains on this tracks. Transportation 10000-30000 items per minute are possible with that, no question: THAT is fun. But for the needs of gameplay I would say, small is beautiful.
Thought a bit about that. No monorails. Currently I want free streets, no rails. You need to built nothing but truck-stations and trucks! But YOU CAN built streets or parts of streets. To speed it up, to charge the trucks while driving (so it may look a bit like monorail), to tell them: Don't take the shortest way, take the way around... And you can tell the construction bots to built streets over the most used routes first. And yes, they are slow, compared to train. I think about as the bots, 0.8-8 tiles/sec. And curves will slow them very strong.- The auto cars should be slow (if you're talking about they traveling through a line... MAYBE make them MONORAILS????)
Valid for trains, too.- Have them speed up with research, but not too much
What I really would like is, that I don't need to program anything. Place a truck-station, place a truck on ground, put coal in it and he will begin to drive himself around and brings requested items from one station to the other, more or less like the logistic bots, but they have to look for the other traffic. For the devs they are surely more complicated to program then the logistic bots.- Have "car stops" of 2 or 3 kinds: one that work as provider chest, one as requester (and maybe one as storage) so that the car knows if it has to load or unload there... also make them work NOT ONLY by time schedule, but also by "Stay here untill car is full/empty".
But they will look really cool... I think I could sit before the screen and watching them driving around.
Hmmm.
Now, when I have some more experience with building big factories, I think this is so very usable to cover the "area".
What do I mean? We have currently the belts, the logistic bots and the trains.
- The belts can cover big areas, but they are not thought for it. It takes much resources and much time and after clearing all resources in the area, they are laying on the ground and doing nothing, they are not worth to be mined. What a waste!
- The roboports have a coverage of 50x50 tiles for the logistic area. Now when you want to cover a big area, it hurts. Alone the time to produce the roboports. And the placement! And the energy! (I think they use too less energy yet and the bots should also use more energy.)
- Then we have the trains. But they hurt also. They are not thought for such small distances under 200 tiles. That takes too much space and is just ugly gameplay.
My picture of that is like so: There are "islands". Eventually you know the game Siedler II, which has islands? An island has also requests and offers, and a boatman transported them between island and land. This here is about the same idea, but a bit more complicated.
So lets assume we have solved the problem, that the "islands" know, how much is available and how much is needed. This alone is a big task.
But then you can create real logistic between those islands.
Which might look so:
- A truck drives to a truck-station.
- Is loaded.
- If full or nothing else available he drives to the target and is unloaded.
Not simpler thoughtable!
(Well of course you could also use transport belts for that job of exchanging between the logistic areas. But as mentioned above the problem in my opinion is, that it takes much longer to built such long transport belts. With this truck system which needs no streets in the first tier you don't need not built anything, but the truck-stations. So less resources and less items, so it is faster built up and can be built by yourself, while a far away belt is an difficult task, you need hundreds of belts and cannot hold everything into the inventory.)
Well, not everything is thought through. The loading and unloading for example. But when trucks are available before logstic I don't see any other option as with inserters and chests. But this is not so important! The point is, that this works in some ways much more effective than the trains, the belts and the logistic bots! It's just a 4th option. I didn't liked the trucks/busses in OpenTTD not that much, because of their inefficiency and the need to so much planing and micromanagement, but I think this trucks I would love, because they work from in the form of how you place the truck stations and where and how many streets you built afterwards. In the beginning they drive criss-cross and then you see the most used paths and make more place for them and place poles. Then you see that you loose many trucks because the trucks and trains collide, because the trucks cross the tracks without looking left and right and you built some kind of train-crossing. Then you see, they find a way through a wood and are killed by some biters.
And the biggest advantage is in my eyes, that you are forced to leave some space through your factory. I guarantee, that this will make the factories look much nicer and simpler.
For the devs the big advantage of that: when it is clear, that this new type of transportation (transport by the need of the single logistic networks) works, it can be also introduced for the trains! Not too much differences. The advantage is then, that we don't need to handle all the routing stuff. I liked that routing options, the "how long should a train wait at each station" for OpenTTD very much for that, but I think Factorio is one or two numbers smaller than that (or bigger, depends from standpoint ) and I think making all this micromanagement doesn't bring this game very much forward.
Yes, of course this is a bit naive. It doesn't make sense to order a truck from about 2000 tiles away. Maybe they have a home-station. But I think routes are not so much needed. We should not think, that everything needs routes and all the micromanagement. This brings the game much more forward and the rest are solvable small problems.
Ok, there should be routes for the trains. But only the big routes, the very, very big amounts of resources. The long ranges, because that is more efficient. In OpenTTD we made up routes, which are as long as possible to get the biggest cash, but in factorio we made those routes, because it is more efficient to have some big transport over long distances and not many small for short; not for those 2-4 mining drills 350 tiles outside, which needs a connection to the rest. This eats just up time...
Exactly therefore are the trucks!
And in the end I would like to see trains, which run in network to every point the train can ride to, to fetch up the next important items, too. Some train depots, which I the train drive into, when currently not needed (or the trucks, to come back to the streets). So the route-handling isn't needed to be so deep as in OpenTTD; the signal handling must not be so perfect, the path-finding could have some small issues, which make the trucks/trains look stupid, but in the end the network works and does it's job, the players have a playable game and the devs can do more important stuff, instead of fixing the bugs for a perfect routing.
This are all advantages, which will bring the game very much forward into the wanted direction.
Additionally: what I currently think is, that we have trucks which works like a rolling roboport. You can fill in some robots and the needed items for some blueprints and drive them to some area. You place stuff via blueprints/ghost-building and robots swarm out and place it. You built up a small mining site within 1-3 minutes, instead of 10-20. Powered by solar. Without any other needed connection. And the resources are transported by the trucks to the next big train station, just because it is needed so. Just without very much further programming/micromanagement.
Yes. But - as you saw - I turned my mind a bit and see them currently only as an option, because it doesn't bring the game really forward to have streets, when we need transport instead.- MAYBE Make them come with a robotic arm (buit in inserter) so they can load-unload automatically in the stop (i don't think i'd like having inserters all around the street to load the car, but this is only IMO)
Love the ideas of the plated roads, they would look sooo nice!!!!
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...
Re: Chest-on-a-belt
TL:DR
see above (not quoting for obvious reasons)
see above (not quoting for obvious reasons)
Re: Chest-on-a-belt
This is my idea for Transport Carts.
Carts can only move in one direction determined by the monorails.
Monorails types:
Each monorail segment can be occupied by only one cart and cart can move only when segment before him is free.
Straight rails have higher priority, right the lowest.
When the cart stopped at the station, is treated like a logical chest. You can connect the pole at the station with the logical system.
Each cart need energy to move around and gets it out of the rails. Energy flows through the rails just as liquids through pipes.
First tier of cart have only 1 slot and move with normal speed. Each next tier have +1 slot and move little faster but consume more energy.
That system can by something between belts and trains.
Carts can only move in one direction determined by the monorails.
Monorails types:
Each monorail segment can be occupied by only one cart and cart can move only when segment before him is free.
Straight rails have higher priority, right the lowest.
When the cart stopped at the station, is treated like a logical chest. You can connect the pole at the station with the logical system.
Each cart need energy to move around and gets it out of the rails. Energy flows through the rails just as liquids through pipes.
First tier of cart have only 1 slot and move with normal speed. Each next tier have +1 slot and move little faster but consume more energy.
That system can by something between belts and trains.
Re: Chest-on-a-belt
Now i picture it better... But still i'd life some "Road" or "walkway" or something to make the islands you say. Something to mark for myself "leave this X amount of tiles for the trucks to go by".ssilk wrote:One of those big postings. [...]
It doesn't have to be for the robotrucks or have any utility, just aesthetics. filling the empty places and have it look nicer!
I love this idea. I think trains need buffing. In amaunt that can be carried AND in speed. I usually fill only 1 chest with 1 entire train (3 waggons). This makes it impossible to use 1-way rails, because you NEED to have 4 or 5 trains in the track for getting enough resources... if with 2 trais you were allowed to transport the same, you can have them in a 1way track, and since tracks are EXPENSIVE AS HELL for the amount you need for a long journey, 2ble rail is not allways an option from the start...ssilk wrote:Valid for trains, too.sedado77 wrote:- Have them speed up with research, but not too much
Talking about this, maybe make another rail type: LONG STRAIGHT RAIL, make it cost twice as much as the current straight rail, and have it be 3-5 times longer. This would make it more viable for trains!!!!
I Often see let's plays in wich they use LOTS of curved rails, where the train goes in zig-zag, and even if it is A LOT SLOWER, it ends up being more resource-efficient (because you would use 4 curved tracks for the space you'll need to place like 8 straight tracks) at early game than a lot of straight rails :S
CANT WAIT TO SEE THIS HAPPENNNNNNNNN. I swear I'll be standing next to an S shaped train track for hours just to see that happen XDssilk wrote:Then you see that you loose many trucks because the trucks and trains collide, because the trucks cross the tracks without looking left and right
Keep up the great work ssilk and the rest of devs. Your game Is getting better and better