|
NoEyedSquareGuy posted:Guess I never noticed. It's tiny, you have to really look inside while they're working to notice it. The Mk. I only has two robo arms The Mk. II has three. The MK III has no arms and 3D prints/materialiezes a big ol' block floating in air before it ka-chunks down.
|
#
?
Oct 19, 2022 11:46
|
|
|
|
| # ? Jun 7, 2024 15:50 |
|
|
Alkydere posted:It's tiny, you have to really look inside while they're working to notice it. No? The roof is different.
|
|
#
?
Oct 19, 2022 12:20
|
|
|
How do y'all handle trains? I've never used them, always seem to bounce off the complexity jump right about this point. Do you train in raw ores to a single location or are you making end units on site and bringing those somewhere?
|
|
#
?
Oct 19, 2022 13:12
|
|
|
Sipher posted:How do y'all handle trains? I've never used them, always seem to bounce off the complexity jump right about this point. Do you train in raw ores to a single location or are you making end units on site and bringing those somewhere? There's a lot of different train strategies. In my experience: * If it's low tier and easily pulled from local sources (screws) then do that and don't involve trains * If it's a deep intermediate needed in large quantities, like heavy modular frames, that's what you want on a train * Liquid trains suck. Make oil/nitrogen goods on site and ship the later products * Any given train goes A->B and anything past that is a logistics nightmare that will bite you in the rear end eventually * If you need more than two stations per factory you need to do more locally. * If it's a final product you want drones, not trains
|
|
#
?
Oct 19, 2022 14:03
|
|
|
Mailer posted:* Any given train goes A->B and anything past that is a logistics nightmare that will bite you in the rear end eventually The more advanced signal and roundabouts makes the network thing so stupid simple. Also drones suck rear end for any decent throughput. I rarely bother with them.
|
|
#
?
Oct 19, 2022 18:34
|
|
|
CainFortea posted:The more advanced signal and roundabouts makes the network thing so stupid simple. I think Mailer is talking about a train moving products from one source to one destination. Not the train track network being A->B lines. Which is definitely good advice, for anyone who needs beginner advice on trains. The item filters for trains mean you can do more complex things, and way more easily than before the big train update. But there are many kinks that can come up doing it and I wouldn't recommend doing multi-stop multi-product trains. Like, try doing mixed-product sushi on regular belts first. You get the same type of problems, but they show up much faster. CainFortea posted:Also drones suck rear end for any decent throughput. I rarely bother with them. Final products means like ultra-high-end items that you aren't gonna be producing in numbers that require a lot of throughput. Though drones don't have that terrible of throughput, really. You get ~mk5 belt speed for anything that stacks 100 or more. But they have plenty of other annoyances (like supplying the batteries).
|
|
#
?
Oct 19, 2022 19:51
|
|
|
one whole entire mk5 belt? I'm going to get the vapors! Kidding aside, I suppose a more better way to say it is that the throughput:effort ratio for drones is rear end.
|
|
#
?
Oct 19, 2022 21:05
|
|
|
CainFortea posted:one whole entire mk5 belt? I'm going to get the vapors! If you have a decent supply of batteries and plently of overhead on your power grid, drones are the lowest effort way of transporting things. Trucks require you to set up routes, trains require you to build track infrastructure across the whole world, drone ports let you set down A to B logistics without any fuss. A drone can't hold as much as a train car or truck, but throughput issues can be easily solved by adding additional drone ports as necessary. It's generally better to only use them for finished products but I'm going to use them for everything down to transporting raw ores whenever Update 7 comes out.
|
|
#
?
Oct 19, 2022 21:28
|
|
|
I feel drones are probably ideal for things you produce in relatively small quantities and want to transport long distances, like say if you're producing Project parts in a plant that's way the gently caress off away from your space elevator, and you don't feel like making a dedicated rail line/belt across half the drat map. The sole real benefit of the drones is that they don't NEED infrastructure between A and B beyond setting up the drone stations. Think of them like trucks that don't need to worry about pathfinding.
|
|
#
?
Oct 19, 2022 21:29
|
|
|
NoEyedSquareGuy posted:If you have a decent supply of batteries and plently of overhead on your power grid, drones are the lowest effort way of transporting things. Trucks require you to set up routes, trains require you to build track infrastructure across the whole world, drone ports let you set down A to B logistics without any fuss. A drone can't hold as much as a train car or truck, but throughput issues can be easily solved by adding additional drone ports as necessary. It's generally better to only use them for finished products but I'm going to use them for everything down to transporting raw ores whenever Update 7 comes out. Making a decent supply of batteries is more of a pain in the rear end than just putting rail down I find. And there's also the drone network to set up to distribute the batteries.
|
|
#
?
Oct 19, 2022 22:03
|
|
|
I really wish that hypertube entrances could be used as speed boosters. I know you can build rail guns, but something that's more organic and feels more like intended behavior would be nice.
|
|
#
?
Oct 20, 2022 00:07
|
|
|
Lemony posted:I really wish that hypertube entrances could be used as speed boosters. I know you can build rail guns, but something that's more organic and feels more like intended behavior would be nice. what do you mean? they do give you a speed boost the booster I use (can be made at an entrance/exit or inline an existing tube network): line up stackable or normal hypertube supports, with 2 spaces between them. the more of these the faster you'll go then connect the last support to the exit hypertube, and put hypertube entrances on both sides of the rest of the supports. you'll get a boost going either direction and it's super easy to set up
|
|
#
?
Oct 20, 2022 00:28
|
|
|
tak posted:what do you mean? they do give you a speed boost To be fair, it never occurred to me to put entrances in both sides of the support, that's probably going to come in handy for inline boosters. I more meant that, as far as I am aware, for that kind of booster to work you have bare supports without much in the way of tube between them. It'd be nice to just be able to slap down some boosters onto existing tubes. That being said, I've been making some assumptions I think. Should probably do some actual testing. Anyone know if tube booster setups can only work horizontally, or can you run them vertically as well?
|
|
#
?
Oct 20, 2022 01:20
|
|
|
Klyith posted:I think Mailer is talking about a train moving products from one source to one destination. Not the train track network being A->B lines. This. Trains are limited to tracks but have the benefit of large amounts of cargo, so utilizing that fact bidirectionally makes sense... except that just sucks in Satisfactory. You'll always wind up with an odd amount of stuff in the train, which causes its return run to not arrive in time, and then the whole system breaks down. So you only ever want Train 1 moving screws from Screw Source 1 to Screw Dropoff 1 and that's it. Also one item dropoff per station. The station part sucks, because stations take up a truly stupid amount of space, so you really only can use trains for expensive intermediates. Even if you can produce a million screws a minute at some far-off factory you'll have to build an extra station for them at every single factory that requires screws. As an example, computers: you import the plastic (requires oil so do it at the oil site), circuit boards (mid-tier intermediate), then mine/smelt/create cable and screws in a separate series of factories piping into the computer factory. With a train station to output the computers you'll have three stations and two dedicated inbound trains. You've just added two gigantic stations and freight platforms to your world, a huge footprint of factories for all the smelting and intermediate stuff, and two more trains zipping around consuming blocks on your rail network. You've also consumed X iron and Y copper veins, so you really need to build the factory close-ish to them.
|
|
#
?
Oct 20, 2022 01:33
|
|
|
Lemony posted:To be fair, it never occurred to me to put entrances in both sides of the support, that's probably going to come in handy for inline boosters. I more meant that, as far as I am aware, for that kind of booster to work you have bare supports without much in the way of tube between them. It'd be nice to just be able to slap down some boosters onto existing tubes. they work vertically too
|
|
#
?
Oct 20, 2022 02:09
|
|
|
Lemony posted:To be fair, it never occurred to me to put entrances in both sides of the support, that's probably going to come in handy for inline boosters. I more meant that, as far as I am aware, for that kind of booster to work you have bare supports without much in the way of tube between them. It'd be nice to just be able to slap down some boosters onto existing tubes. They do work vertically, as famously seen in the Lets Game It Out video that invented cannons, where his first experiment was with a vertical tube. But there are secondary effects that make them not work as well as a speed-boost in the middle of a vertical run. Number one, the velocity boost of entrances is proportional to your speed going in. That's why a cannon with 10 entrances throws you a few 100 meters, 12 throws you across the dune desert, 14 is most of the way across the map, and 16 is very likely to kill you. But when going upwards in a tube you slow down, which works directly against the speed boost. A bunch of entrances spaced out on a long vertical run do nothing. Two, when you're going back down you get acceleration from the entrances, and acceleration from going down. So your entrance in the other direction *is* super effective. Also, the short tubes between the entrances is not strictly needed. The velocity boost is purely from the entrance itself. There are reasons to have short tubes, the obvious one being that less time in tubes between entrances means faster acceleration. But there's a separate thing about game physics that can make a cannon fail with long tubes. But it's not the source of the speed boost.
|
|
#
?
Oct 20, 2022 02:23
|
|
|
Mailer posted:This. Trains are limited to tracks but have the benefit of large amounts of cargo, so utilizing that fact bidirectionally makes sense... except that just sucks in Satisfactory. You'll always wind up with an odd amount of stuff in the train, which causes its return run to not arrive in time, and then the whole system breaks down. Well, you can't unload and load the same train car in a single station anyways. But you totally can do more complex things with trains... you just have to figure out where/when those bottlenecks will happen, and have overflow into a sink to prevent jams. Probably this requires flowcharting things on paper and such. So the question is, how much extra time do you want to spend solving a complex problem on paper rather than building extra train stations? Here's a thing I had after U5 added item filters, in my abandoned pre-U6 save. Steel plant: supplies 2k/m steel ingots and 1k/m petcoke Aluminum plant: needs petcoke, supplies 500 aluminum ingots Main factory: needs steel and aluminum ingots Using one train, with two cars, and a single station in each location, I made this work. It's possible. (Though I didn't finish the main factory before U6 hit, but it was tested by sinking ingots at the main factory.)
|
|
#
?
Oct 20, 2022 03:04
|
|
|
Klyith posted:Well, you can't unload and load the same train car in a single station anyways. But you totally can do more complex things with trains... you just have to figure out where/when those bottlenecks will happen, and have overflow into a sink to prevent jams. Probably this requires flowcharting things on paper and such. So the question is, how much extra time do you want to spend solving a complex problem on paper rather than building extra train stations? Overflows only solve it if the overflow amount is so low (or the initial production so insanely overproduced) that you can afford to dump a cartload of screws into a sink. It also assumes that your cartload of screws sinks faster than the return trip. Since they stack differently to the plates (or whatever) you're also running in there you'll eventually hit a situation where you've sunk all your screws and your plates come in, then your screws come in, then your plates come in again and now you're sinking plates which sink much faster. Your whole network is just feeding to the sink most of the time with a small amount of production during times when the stars are right. I believe, with enough planning and timing of trains and production, it would be possible to math out multi-component trains via cart placement. This would also allow for bidirectional cargo. The logistics involved in this, and the time spent, would both be incredibly unfun. Like bending a track slightly off to maintain a .01s time difference so that wire will unload at a station in 103 seconds into the trip instead of 103.01... NOPE. quote:Here's a thing I had after U5 added item filters, in my abandoned pre-U6 save. I'm talking about distributed production on a two-way train network, not using push-pull trains as a A->B conveyor bridge. If you move the above to a distributed model then it falls over if I need aluminum ingots delivered to my heatsink factory, which does not produce petcoke to feed the aluminum factory. It also mandates X-car trains and marks car Y as aluminum-only which now mandates aluminum-consuming stations be bigger to support that car numbering even if they just need the ingots as their only input. It's totally a solvable problem, with Advanced Train Logistics Mk1 or whatever, and it's something I expect will be solved in future updates when they ask us to scale up again for phase 5. In the meantime, though... one platform, one train, one item.
|
|
#
?
Oct 20, 2022 05:35
|
|
|
Mailer posted:This. Trains are limited to tracks but have the benefit of large amounts of cargo, so utilizing that fact bidirectionally makes sense... except that just sucks in Satisfactory. You'll always wind up with an odd amount of stuff in the train, which causes its return run to not arrive in time, and then the whole system breaks down. So you only ever want Train 1 moving screws from Screw Source 1 to Screw Dropoff 1 and that's it. Also one item dropoff per station. I do not understand any of this because this doesn't match my in game experience. Like, almost at all. The amount of stuff in the train doesn't meaningfully impact how fast it runs unless you specifically tell it to wait until it's full or empty. But there is never a reason to do that. And having 1 engine moving 1 thing is just so wasteful. Since the stations unload all the boxcars at the same time you want to have more than 1 car per engine. My main hub will have 1 stop with 4 freight platforms to unload with. I will make sure the trains are picking up in stations that are on the same spar to reduce load times. And like 80% of all the stuff i'm freighting around with trains doesn't matter if it gets stopped up. So I don't even bother sinking it.
|
|
#
?
Oct 20, 2022 21:44
|
|
|
There are two main things to avoid when designing your train networks. One is having the same train car carrying different loads at different points on its route. Say, the front car of your train takes steel pipe to a factory, drops them off, picks up computers then drops those off at another factory. Or at least, if you're going to do this, make absolutely sure your train is unloading everything faster than it's loading them. Otherwise some of those steel pipes won't unload, they'll get mixed in with the computers, and unloaded at the computer stop. The other is having a car drop off at two different places without re-loading in between. Again, you'll need some really careful belt balancing if you want that second station to get supplied evenly. In this case you'd want to make sure the first one doesn't take out more than it needs. And even then you have to wait for the first station's inventory to fill up before the second one will start receiving anything at all. Something you don't necessarily have to avoid but does need to be considered is having multiple products coming to the same station, either in the same car or from multiple different trains as a unified dropoff route. You'll need to sort everything that comes out, of course, and dump any overflow into a crusher, else any one product backing up will clog the entire station and make your train network unusable. The only real trap here is that you can have a maximum of two mk5 belts coming out of the station, so the combined loads feeding in have to total no more than that much. All of these worries are per car/per station unit, not per train. If a train makes four stops, loads steel pipes into car 1 at stop 1 and unloads at stop 4, then as long as nothing else touches that car at stops 2 or 3 it's still only really making a two-stop journey. Just use empty platforms to space out the cars you're not interacting with. This does mean wasting space at stations, of course, but if you're using trains this way it means fewer trains on the lines - which means less waiting for each other at signals and stations.
|
|
#
?
Oct 20, 2022 21:59
|
|
|
nielsm posted:No? The roof is different. Yes. I was saying there's additional differences in the animation between levels if you zoom in and look. In addition to the roof having a slightly different model.
|
|
#
?
Oct 20, 2022 22:17
|
|
|
Yesterday I accidentally discovered that you can use Ctrl+C and Ctrl+V to copy and paste sign settings without actually going into the sign interface. All you need to do is look at the sign so that it's highlighted. I have played the game for several hundred hours.
|
|
#
?
Oct 21, 2022 03:39
|
|
|
Why do I have the urge to completely rip and rebuild my primary home base whenever I hit a new phase?
|
|
#
?
Oct 21, 2022 04:11
|
|
|
neato burrito posted:Why do I have the urge to completely rip and rebuild my primary home base whenever I hit a new phase? Brain worms. I definitely haven't rebuilt my main base multiple times.
|
|
#
?
Oct 21, 2022 05:05
|
|
|
Lemony posted:Yesterday I accidentally discovered that you can use Ctrl+C and Ctrl+V to copy and paste sign settings without actually going into the sign interface. All you need to do is look at the sign so that it's highlighted. I have played the game for several hundred hours. You can do the same with machines. They added that recently. Makes the extended reach of the hover pack better.
|
|
#
?
Oct 21, 2022 06:58
|
|
|
One thing I would love for them to add, which would be amazing for train usage, is the ability to set filters for storage like in factorio. Setting slots to only be used for a specific item let's you do things like have a train go to station A to pick up plastic and then go to station B for your aluminium or whatever. Even with stations that provide multiple items, it would let you add them into a single cargo wagon with predefined ratios without having to add a wagon for each item. Plus, you'd be able to only use a few slots for more high end items and then set your train schedule to wait for full instead of only picking up what's there. Is there a mod that does this? I haven't really looked at the mods in forever but I don't remember this one being a thing the last time I looked.
|
|
#
?
Oct 21, 2022 13:11
|
|
|
Preview for update 7 this year. Not much content but we get automatic wall and ceiling mounts for belts. They also have a first implementation of game modes with peaceful mode. It doesn't outright disable monsters but at least they no longer attack you. ...Just realized they can't disable monsters as we need monster parts for some research and special items. https://www.youtube.com/watch?v=rX7OpLCg0l8
|
|
#
?
Oct 21, 2022 16:30
|
|
|
lagidnam posted:Preview for update 7 this year. CEILING MOUNTS FOR CONVEYORS 
|
|
#
?
Oct 21, 2022 16:32
|
|
|
That other guy been fired or something?
|
|
#
?
Oct 21, 2022 16:33
|
|
|
Combat Pretzel posted:That other guy been fired or something? They just have different community management roles. Also iirc they're still wfh so it's not easy to get both of them in a quick video like this.
|
|
#
?
Oct 21, 2022 16:59
|
|
|
They're also a Swedish company so they take some sizeable vacations from time to time.
|
|
#
?
Oct 21, 2022 17:14
|
|
|
Dunno-Lars posted:You can do the same with machines. They added that recently. Makes the extended reach of the hover pack better. .....gently caress. That's gonna save me some time going forward.
|
|
#
?
Oct 21, 2022 17:42
|
|
|
CainFortea posted:I do not understand any of this because this doesn't match my in game experience. Like, almost at all. The amount of stuff in the train doesn't meaningfully impact how fast it runs unless you specifically tell it to wait until it's full or empty. But there is never a reason to do that. And having 1 engine moving 1 thing is just so wasteful. Since the stations unload all the boxcars at the same time you want to have more than 1 car per engine. Stack sizes and the almightly 780 limit are the kick in the pants, here. You can't deliver Screws and Plates uniformly because they'll drain at different rates. If you drop two cargo wagons off with two different things (but it's still a dedicated train) then you're fine... except the pickup point (or points) have to ensure the same spacing. If circuit boards are on your second wagon you need an empty platform at the circuit board factory. This means all your other places producing/receiving circuit boards need to maintain that spacing or you break the system. To save space at one factory you're consuming space on all the other producers/receivers. To get back to the original comment - this is why bidirectional travel (pick up X, drop X, pick up Y, drop Y, repeat) is so hosed. You have to be able to pick up a full load of stuff AND drop off a full load of stuff and any time this isn't perfectly handled the whole operation falls over. There's no logistics to say "pick up 100 circuit boards" or "dump all circuit boards but it's ok if computers are left over" so you're dealing with a full wagon of parts every time. You can stuff your wagons full of undeliverables to reduce throughput but the effort involved in that is way more than the effort of powering another train to just grab the stuff. That's why the distributed production model falls over. Trains are fast and carry lots of stuff but their logic is limited so you have to solve it all yourself with more trains and more stations. The current design really wants a fps-killing megabase just to ensure you never have to deal with one-to-many relationships.
|
|
#
?
Oct 22, 2022 01:57
|
|
|
lagidnam posted:Preview for update 7 this year. RIP to Paradise Island. Was hoping they would extend the death barrier a bit and put a rare node over there or something, guess it's easier to just get rid of it entirely.
|
|
#
?
Oct 22, 2022 02:32
|
|
|
Mailer posted:Stack sizes and the almightly 780 limit are the kick in the pants, here. You can't deliver Screws and Plates uniformly because they'll drain at different rates. If you drop two cargo wagons off with two different things (but it's still a dedicated train) then you're fine... except the pickup point (or points) have to ensure the same spacing. If circuit boards are on your second wagon you need an empty platform at the circuit board factory. This means all your other places producing/receiving circuit boards need to maintain that spacing or you break the system. To save space at one factory you're consuming space on all the other producers/receivers. Distributed production just means production at many sites, not production distributed solely and inflexibly by train. Yes, delivering screws via train sucks. Don't do it. If you need screws and plates in one place, deliver iron ingots. Or better yet find the nearest iron node and mine it. You can't throw a rock without hitting an iron node. To do distributed factories you should be looking for locations with most or at least some of the resources you need locally, and specialized resources delivered. If you are building an electronics factory that is based around Silicon Circuit Boards, you should be building it somewhere with quartz and copper. Each factory needs to aim for fairly high-tier goods, or something that you need in a quantity that is hard to do in anything but a specialized setup (oil products, aluminum). And eventually yes, you need several train stations per factory. I have a plan for a build that needs 4 train stations, 2 truck stations for local ore collection, plus a drone port on the roof. Yes, all that takes some space. The rest of the factory will have like 200+ buildings, 37 of which are manufacturers. The stations are not the biggest thing here. If you're not building on that scale and train stations are bigger than your factories, you should just megabase -- your megabase won't be fps-destroying. Trust me, at one point I agreed with you that train stations were too big. Now I don't because my factories got much bigger. (Clipping freedom helps too.)
|
|
#
?
Oct 22, 2022 03:44
|
|
|
Didn't really have any good ideas for what to build when update 7 comes out but I've made an important discoveryquote:Trivia They have an inventory slot of 1, like an even more useless factory cart because you have to work in fuel logistics as well. After a bit of testing...    Going to build a nuclear plant with this bullshit somehow. The all-factory cart save was already a nightmare, this should be significantly worse.
|
|
#
?
Oct 22, 2022 04:36
|
|
|
NoEyedSquareGuy posted:RIP to Paradise Island. Was hoping they would extend the death barrier a bit and put a rare node over there or something, guess it's easier to just get rid of it entirely. I'm not sure how they went "the death barrier interferes with this really cool island we have" and came to the conclusion "we should remove the island". 
|
|
#
?
Oct 22, 2022 05:43
|
|
|
Mailer posted:Stack sizes and the almightly 780 limit are the kick in the pants, here. You can't deliver Screws and Plates uniformly because they'll drain at different rates. If you drop two cargo wagons off with two different things (but it's still a dedicated train) then you're fine... except the pickup point (or points) have to ensure the same spacing. If circuit boards are on your second wagon you need an empty platform at the circuit board factory. This means all your other places producing/receiving circuit boards need to maintain that spacing or you break the system. To save space at one factory you're consuming space on all the other producers/receivers. You don't need to deliver things uniformly if they're being drained at different rates. You just need to be able to deliver enough item to refill the station between runs. If that is 10 or 1,000 doesn't matter at all. It takes the same time to unload the train.
|
|
#
?
Oct 22, 2022 06:21
|
|
|
Klyith posted:Distributed production just means production at many sites, not production distributed solely and inflexibly by train. Yes, delivering screws via train sucks. Don't do it. If you need screws and plates in one place, deliver iron ingots. Or better yet find the nearest iron node and mine it. You can't throw a rock without hitting an iron node. Yup, I mentioned this before too. Only ship complex intermediates (ie, things that require multiple things... I don't have a good term for that) and that limits you to, in most cases, 2-4 stations. quote:And eventually yes, you need several train stations per factory. I started trains extremely late, like in phase 4, so my perspective on what is ok before you're producing mass quantities is probably skewed. Maybe it's cool at low volume to mix bins and whatnot but by the time I'd laid down the trackwork and sorted everything out I was in a state where I had crap like this:  That factory sucks and is the driving force behind what I'm saying. It's got mixed platforms, which led to logistic hell later. It's delivering basic goods (notably four cars worth of screws) which was always somehow too much and too little. It's also producing far too much for one location which leads to needing further trackwork to get trains in and out fast. It was interesting to build but ultimately just multiple lessons in how not to do things. This is all just general train stuff, though, and I only brought it up in relation to why bidirectional trains don't work once you scale up into distributed production. CainFortea posted:You don't need to deliver things uniformly if they're being drained at different rates. You just need to be able to deliver enough item to refill the station between runs. If that is 10 or 1,000 doesn't matter at all. It takes the same time to unload the train. This is true in my One Platform One Item and (unless you need more) One Train design that I'll be sticking with for the future. My point was that if you tried to do this bidirectionally, then that leftover stuff from your last delivery takes up the space for your other delivery on the return trip. This leads to shortages and overstocks at the various consumers/producers as your train rumbles along with a weirdly-distributed mishmash. As long as your circuit delivery is only circuit delivery, then having it back up just leaves the train parked at the destination until it's needed again and that's a perfectly acceptable situation unless you're running some very weird JIT delivery shenanigans. I'd toyed with the idea of just having all producers mandate four outbound platforms to always allow for bespoke multi-car delivery stations at consumers but in the end this seemed like way more space and work for goods that had increasing inbound requirements and decreasing outbound demand. I'd also have to drastically overproduce (I don't need 3120/m circuit boards) just to fill the platforms. In the end I just accepted I'd eat the space for inbound stations and run more trains. TL;DR - Trains are dumb. Having smarter trains would be cool, but if they're gonna be dumb I'm just going to let them be dumb and run a ton of them with a well-designed track network. Then just replace the final few tiers with drones unless I'm trying to set a record for tickets.
|
|
#
?
Oct 22, 2022 08:56
|
|
|
|
| # ? Jun 7, 2024 15:50 |
|
|
Mailer posted:I started trains extremely late, like in phase 4, so my perspective on what is ok before you're producing mass quantities is probably skewed. Maybe it's cool at low volume to mix bins and whatnot but by the time I'd laid down the trackwork and sorted everything out I was in a state where I had crap like this: Ok, I think I'm starting to see where you're having the problems that make something like overlapping delivery impossible: you also don't have balanced production and consumption. If you are producing 9000 screws/min and only using 5000/min at the destinations, of course there will be too many screws. And because they're screws you're gonna have trouble getting them out of the station fast enough to dispose of. Don't load 9000 screws onto the train in the first place. If you cut production of screws down so that the train was delivering just the right amount (or a fraction more) at each stop, you don't have a train that's full of screws. Then you can do things like unload the screws from car #1 while loading circuit boards into car #4 in the same station. And you don't have 10 extra stacks of screws to overflow into a sink. Mailer posted:This is all just general train stuff, though, and I only brought it up in relation to why bidirectional trains don't work once you scale up into distributed production. So I totally agreed with you when presented as general advice for people who aren't sure how to organize trains. But you are absolutely wrong that it doesn't work. It can be done. But you need a plan, and to have some control over item production at the high level. Overproduction is bad for logistics. (Also for anyone reading this: don't put loving screws on trains. Ship iron ingots or steal beams and turn them into screws later. Same with wire. Think about item compression.)
|
|
#
?
Oct 22, 2022 15:28
|
|