|
Majere posted:Production science is fun because it just has a ton of "basic" ingredients. I like your linear layouts, nice and tidy. That's more like a 'gigabase'. Eeep... Very impressive. I know for a fact that I would stop playing long before my UPS/FPS dropped to 10!
|
#
?
Jul 19, 2021 16:28
|
|
|
|
| # ? May 25, 2024 12:10 |
|
|
Majere posted:Production science is fun because it just has a ton of "basic" ingredients. I like your linear layouts, nice and tidy. My 'layout' is really done pretty much by accident. For the first 4 sciences I just started by making the unloading stops (except I had to redo it with purple to add more iron unloading) then the smelter arrays and then each other 'module' as needed working from basic materials to finished without any real plan at all. Each module put together on it's own and then figured out how to route the materials into it. In the Purple science setup I ran into a couple issues getting materials where they needed to go due to this lack of planning, so for Yellow science I'm working backwards. I started with the final yellow science assembler build, then the blue circuits and green circuits. Next I'll probably build the robo-frame production and work backwards making all the intermediate builds for that, and then the other component. Next, I'll build the oil and associated products outputs (I need lubricant for this section) and then finally I'll finish with the smelting arrays, and then the train unloading area. Finally, I'll drive the rail line from the unloading area south until I find all the needed resources and set up the mining stations and put in the trains. We shall see if it works better or worse this way! I'm not good at planning as until I actually 'build' a thing I have a hard time understanding in my head just how much space it will need.
|
|
#
?
Jul 19, 2021 16:38
|
|
|
The Locator posted:That's more like a 'gigabase'. Eeep... Very impressive. When you start building this stupid big it's all about UPS optimization. Which is a challenge in and of itself, testing the limits of the game engine. My 18k Factorio Extended base ran at about 45-50 UPS, so I figured I could do better in vanilla with what I learned from that. I just cant get into bobs/angels/seablock stuff with so many products & byproducts to manage. Then you unlock a new tech and that entire process you just built is useless. I've posted some of the other builds over on reddit https://www.reddit.com/r/factorio/comments/l34t57/all_you_need_for_21000_space_science_per_minute/ https://www.reddit.com/r/factorio/comments/l437yw/hotdog_rocket_fuel_20kmin_vanilla/ https://www.reddit.com/r/factorio/comments/l5uiyu/clocked_plastic_satisfaction/ Kinda burned out at yellow science right now. So on a break till the urge comes back.
|
|
#
?
Jul 19, 2021 18:58
|
|
|
Majere posted:I just cant get into bobs/angels/seablock stuff with so many products & byproducts to manage. Then you unlock a new tech and that entire process you just built is useless. I think you posted some of your builds here before, I know I've seen them, maybe just the links. I agree about the angels/bobs/seablock stuff.. tried multiple times and just sputtered out really early on. As far as yellow science goes... Yay!  I *think* it will work, but still have to go set up and connect coal and oil before I know for sure. The iron and copper are all set now with outposts done and trains running (well, currently parked as the input buffers are full). Watching the iron and copper trains filling up the buffers makes me kind of worried about train congestion coming into and leaving the unloading station, but I guess I'll just have to wait until I finish space science and actually have it running at speed for a while to see.
|
|
#
?
Jul 20, 2021 06:57
|
|
|
The oil/petroleum section of the Space Science build is larger than any I've made for entire bases before. I need to feed almost 60 fully beaconed machines making solid fuel in order to feed 93 assemblers making rocket fuel, and still need to have enough petroleum to makes a ridiculous amount of plastic. I'm guessing that this one module will be taking almost as much, or maybe more 'stuff' than the other 5 sciences combined.
|
|
#
?
Jul 21, 2021 22:15
|
|
|
I split oil processing into separate rocket fuel, plastic and “the rest” builds for my planned 2700 build. Each of those was way easier to plan out than one giant one for all of it. I think I’m also using basic oil for plastic, which costs more refineries but removes the entire cracking ordeal.
|
|
#
?
Jul 21, 2021 22:37
|
|
|
necrotic posted:I split oil processing into separate rocket fuel, plastic and “the rest” builds for my planned 2700 build. Each of those was way easier to plan out than one giant one for all of it. Yeah, in my earlier builds that only needed plastic I just did basic refining. I only moved to advanced and cracking when I needed lubricant. At this point I think I've got the build mostly done so I'll keep it as one gigantic refinery setup for now and see how it goes. Breaking it down like you did would have probably been a good idea, and I may do that if I have to rebuild this if it's not working.
|
|
#
?
Jul 22, 2021 00:01
|
|
|
I’ve started Factorio multiple times over the years and the farthest I’ve ever got was making a train one time. I just started again telling myself “ok let’s actually do this” and here I am already stuffed in a tiny space trying to make enough gears for green science 
|
|
#
?
Jul 22, 2021 00:49
|
|
|
Spread out as your building. Space is effectively infinite and, depending on biter settings, free.
|
|
#
?
Jul 22, 2021 01:12
|
|
|
Just beat it for the first time, 25 hours or so, after 100 hours total played (never got past the 2/3rd mark before, would compare sciences but those have changed a bit since I first played.) I never used logistics. They came so late and my megabus was good enough. My technique was megabus horizontally with a ton of space between sections to break off horizontally. Blue belts into copper wire and basic circuits was all the bandwidth I needed. Wished there were achievements or win conditions for throughput. I kinda crawled to the 100 rocket parts, took maybe 45 mins, so that felt cheap like when I beat DSP.
|
|
#
?
Jul 22, 2021 08:35
|
|
|
androo posted:Wished there were achievements or win conditions for throughput. I kinda crawled to the 100 rocket parts, took maybe 45 mins, so that felt cheap like when I beat DSP. I know there are some achievements for production milestones, like one for 20 million green circuits produced etc., but I'm not sure what they all are because I don't pay any attention to achievements.
|
|
#
?
Jul 22, 2021 08:44
|
|
|
If I remember correctly, there is only one total production milestone, and it is the green circuits. They are based on lifetime total, cumulative throughout every save file you play on. There are throughput ones though where you have make an X amount of items over Y time. In other words, you have to be able to make it very quickly over a duration. Logistics are really cool and powerful, but unless you know how to properly get it up and running it is reasonable to just not touch it to launch your first rocket. But if you know how to do it right, you'll get to first rocket launch in a factory faster by going out of your way to do logistics.
|
|
#
?
Jul 22, 2021 18:08
|
|
|
Are speed runs using logistics now? On paper it always looked hard to beat just more yellow belts if all you wanted was a early rocket launch. Logistics come more into the picture with a lot of logistics tech to make them move fast enough and beacons where the same beacon layout means doing annoying belt weaving or else greatly simplifying getting stuff off a train. I generally just get player logistics going or maybe just patch a connection I'm really not feeling like laying in belt in my normal games.
|
|
#
?
Jul 22, 2021 18:21
|
|
|
Speed runs use construction robots to speed up scaling, something that you very much can't do by hand. You are correct in that for your first rocket logistics bots are unnecessary (or even for your 100th) But they do P2P and autoload balance so they are much more efficient in megabases. Cheaper UPS than belts as well.
|
|
#
?
Jul 22, 2021 18:41
|
|
|
Yeah, I was pretty inpressed by the SGDQ run, and how much the estimate had gone down since I last looked at it a few years ago. They use bots primarily to build copies of stuff they’ve already built once they get close to the rocket to try and balance out the inputs needed for the rocket. Just stamp out three more groups of LDS producers and split the belt to hook everything up properly & you’re on your way.
|
|
#
?
Jul 22, 2021 19:48
|
|
|
Ahh yeah that makes sense. I think it's also partially my own problem for the mental block of construction logibots being set aside in my head as something else than earnest loginet use. Construction bots are a huge enabler of growing exponentially bigger and getting things done fast.
|
|
#
?
Jul 22, 2021 20:14
|
|
|
It also depends on the category. If the rules say no blueprints you will do as above, making some design once then copying it with bots. In 100% you use the bots way way more since you are plopping down a precreated base by BP rather than building a precreated base manually. But in 100% you can't use logistic chests only personal delivery so logistics is is minimal until you launch the rocket.
|
|
#
?
Jul 22, 2021 20:19
|
|
|
I'm really hating the look and design of a main bus setup. It's not so bad when I take the time to try and really make splitting off lanes tight and as compact as possible, but it still always looks like a mess at the end of the day. I'm definitely not going to go through the effort to try and plan out a spaghetti style base that can grow, so that's out of the question. I've only ever used trains in a dedicated line to mining outposts and even then I only have one or two outposts going at at time, so I've never had to really think about anything complicated. Is it feasible to use trains to ship goods between base modules, kind of like a city grid style setup just using trains as your backbone? Or does that basically going to look exactly like a bus, only with train track instead of belts?
|
|
#
?
Jul 22, 2021 20:27
|
|
|
nessin posted:Is it feasible to use trains to ship goods between base modules, kind of like a city grid style setup just using trains as your backbone? Or does that basically going to look exactly like a bus, only with train track instead of belts? Yes, absolutely. This got a lot easier to do in 1.1 with train limits on stations, too. That said managing a large train network that's moving a huge variety of items is its own serious logistical challenge. You should still use them for transporting large volumes between different base modules, but it sounds like what you really want is just logistics bots to solve your belt problem.
|
|
#
?
Jul 22, 2021 20:30
|
|
|
nessin posted:I'm really hating the look and design of a main bus setup. It's not so bad when I take the time to try and really make splitting off lanes tight and as compact as possible, but it still always looks like a mess at the end of the day. I'm definitely not going to go through the effort to try and plan out a spaghetti style base that can grow, so that's out of the question. I've only ever used trains in a dedicated line to mining outposts and even then I only have one or two outposts going at at time, so I've never had to really think about anything complicated. Is it feasible to use trains to ship goods between base modules, kind of like a city grid style setup just using trains as your backbone? Or does that basically going to look exactly like a bus, only with train track instead of belts? Yeah this is not just fully possible but a preferred enough technique among enough players that there are train-dispatching mods to help support it, if you want to really scale things. I’ve never used ‘em personally, the one whose title I remember is called Logistic Train Network. It doesn’t visually resemble anything like a main bus, since the production installations can be arbitrary distances apart, receiving only the materials they need, and a bus keeps everything fairly close together by delivering all materials down a central route. You’re probably good to go with spaghetti-fied outposts, too, since if they become non-extensible you can just drop another receiving station nearby to feed in from another direction, or another new outpost.
|
|
#
?
Jul 22, 2021 20:42
|
|
|
FWIW, while LTN is definitely a boost to ease of use, I had a pretty solid train-based base going without any mods whatsoever. Just be smart about train limits and it'll work.
|
|
#
?
Jul 22, 2021 20:48
|
|
|
nessin posted:I'm really hating the look and design of a main bus setup. It's not so bad when I take the time to try and really make splitting off lanes tight and as compact as possible, but it still always looks like a mess at the end of the day. I'm definitely not going to go through the effort to try and plan out a spaghetti style base that can grow, so that's out of the question. I've only ever used trains in a dedicated line to mining outposts and even then I only have one or two outposts going at at time, so I've never had to really think about anything complicated. Is it feasible to use trains to ship goods between base modules, kind of like a city grid style setup just using trains as your backbone? Or does that basically going to look exactly like a bus, only with train track instead of belts? I normally just make a spagetti main base, and then expand with various small factories connected with trains. Sometimes making just raw materials or finished stuff like circuits. Then I use the main base for rocket launching and science. Satisfying to finally remove old belts and assemblers and get a lot of space in the old main base. feels like house cleaning
|
|
#
?
Jul 22, 2021 20:49
|
|
|
Can someone explain train limits? Are you just using them to enable stations when a shipment is ready? When do you set the train limit to 0?
|
|
#
?
Jul 22, 2021 21:58
|
|
|
If you set it to 1 when a shipment is ready, and a train drives to it, it will prevent all the other trains from also driving to it at the same time. Train limit can be set to 0 when you don't have a full load yet. If you're fancy you can use some multiplying magic to have train limits above 1 if the station has a lot of materials avaliable.
|
|
#
?
Jul 22, 2021 22:24
|
|
|
The simplest way to do is to set the train limit to 1 with no circuits, then it will only ever send one train there, and it'll sit there until it has a full load (presuming you set up the schedule this way), so you won't ever have trains backing up behind it trying to pick up loads that don't exist. The disadvantage of this approach is that you need exactly one train for every pickup station and if it ever falls behind it probably won't catch up, but if you don't want to mess with circuits it'll probably work fine. If you want multiple trains at a station but still won't want to mess with circuits you can put a stacker instead, and set the train limit to 1+N, where N is the number of stacker slots. The same approach works for dropoff stations as well, with the only real difference being the train schedule itself saying "leave when empty" instead of "leave when full".
|
|
#
?
Jul 23, 2021 00:11
|
|
|
The method I'm experimenting with is to use stackers and train limits to divvy up the train pool. So for example my yellow science build has 4 iron outposts and I have 4 iron ore unloading stations. All unloading stations have the same name, and all loading stations have the same name. The unloading stations all share the same stacker, while the loading stations all have a 6 train stacker. All loading stations have a train limit set to 6 while the unloading stations have no limit set since they all feed from the same 24 lane stacker. 24 trains are set up, all with the exact same orders: 1 - Iron ore loading station - Wait for condition, full cargo. 2 - Iron ore unloading station - Wait for condition, empty cargo. At the unloading stations I have a requester chest with 3 nuclear fuel set and a blue inserter to keep the trains gassed up. That's it.. super simple, but it's also a closed network where no other iron stations (loading or unloading) can get to this section of track, which is isolated to serving the input side of yellow science. The train that supplies the fuel to this setup is on the 'main line', while the outbound science is on yet another isolated train network that doesn't connect to the supply side or the mainline. The outbound science train is also fueled at the yellow science module area, and there will be 2 trains assigned to each science. If my math is correct, a full 1-4 science train will hold 32000 science and should make a trip with a full load after just under 12 minutes. I'm sure I it would work fine with a single science train for each color, but I'd rather have a train being loaded while the other one is unloaded just on general principle, and there's plenty of room at each end for the extra train to park and wait.
|
|
#
?
Jul 23, 2021 00:30
|
|
|
This is the first time I’ve made it to oil. Is there a fluid pressure system or any reason I shouldn’t just make a 5 mile long pipe tube to bring it where I want it?
|
|
#
?
Jul 26, 2021 00:08
|
|
|
Flow rate reduces over distance. It's nonlinear. Don't pipe things long distance. Trains are the way to go for long distance. K8.0 fucked around with this message at 00:27 on Jul 26, 2021 |
|
#
?
Jul 26, 2021 00:15
|
|
|
drunken officeparty posted:This is the first time I’ve made it to oil. Is there a fluid pressure system or any reason I shouldn’t just make a 5 mile long pipe tube to bring it where I want it? Fluid flow is kinda annoying to deal with since it's a pretty opaque system compared to belts. You can get more mileage out of it by using pumps or underground pipes (the squares in between aren't modeled, so they always act as two pipe segments no matter how long the gap is), but both are more expensive than ordinary pipes. For longer-distance stuff, you'll want to use trains witt fluid tanks. Barrels aren't worth it most of the time, but they do have some interesting uses.
|
|
#
?
Jul 26, 2021 01:25
|
|
|
You can fit a lot of fluid in one wagon. Nearly 600 plastic worth of crude with just the initial recipe before advanced + cracking
|
|
#
?
Jul 26, 2021 01:54
|
|
|
M_Gargantua posted:You can fit a lot of fluid in one wagon. Nearly 600 plastic worth of crude with just the initial recipe before advanced + cracking Or you can train coal to your oilfield and plastics back. One carload of coal brings back 4000 plastic (half a car) before productivity bonuses. It really helps to have a balanced oil blueprint with cracking to just plop down next to oilfields and train the gas back or pipe it a few blocks over and build a small plastic factory. For rocket fuel you can either siphon off some of the crude or just use another oilfield (replace the light->gas cracking with the rocket fuel builds, convert the extra gas into more solid fuel). Re: main buses, compact is not the way to go if you want it to look neat. You have infinite space, use it. I haven't done very many like this, it does take a bit of extra effort to not just plop down a temporary-eventually-permanent gear assembler next to the iron bus, but the best looking base I ever made had about 30 tiles either side of the bus before I started plopping down auxiliary builds. I actually overplanned and had too much space, so if I did this again I could probably make it a bit more compact, but again, you have infinite space.   The hardest part is probably making sure you don't over-customize the build to where your initial resource patches are; sooner or later you're going to be training it in anyway, so your starter base might as well only bother going up through green + military tech (just 1-2 assemblers for military, if playing non-peaceful) but largely focus on stockpiling resources to build the real base, and designing that from scratch with trains in mind. You can feed 8 lanes of red belt steel furnaces with a four-car train (unload from both sides) and still have about 95% uptime if it's all being consumed (which it won't be for a long time), and 8 furnace lanes is plenty of space between train stations, so I tend towards designs with that density in mind.
|
|
#
?
Jul 26, 2021 02:20
|
|
|
I feel like I’m spending way too much time traveling long distances to stop the loving bugs from breaking my poo poo. Or refilling turrets or oops missed a spot now they took down a power line. I also don’t understand trains and can’t do anything more than an engine on both ends, 2 stops at both ends, one track per resource.
|
|
#
?
Jul 26, 2021 22:25
|
|
|
drunken officeparty posted:I feel like I’m spending way too much time traveling long distances to stop the loving bugs from breaking my poo poo. Or refilling turrets or oops missed a spot now they took down a power line.  Do 1-way rails, engine on one end, make sure to have like "pit zones" so trains can pull off to a stop and not block traffic. Create separate pit zones/drop off/pickup points per resource. IIRC, the general idea of rail signals is basically breaking up the railway into sections that say "only one train can be in this section," but the above image should be a good rule of thumb of where you need to place signals.
|
|
#
?
Jul 26, 2021 22:35
|
|
|
The other rule of rails, which is simple enough that I think people don't say it, is to always always always use one-way rail. It's much harder to get two-way rail to work properly, so most players who love trains will forgo it unless they're engaged in a fun passion project of getting it to work.
|
|
#
?
Jul 26, 2021 22:42
|
|
|
I literally just said that! But yes, it bares repeating. Having one track for one direction massively simplifies stetting up rails, and makes it way easier to have multiple trains on the same network. Basically doubling the amount of rails you need to put down might feel costly when you first unlock the tech, but it's worth it to do it right.
|
|
#
?
Jul 26, 2021 22:59
|
|
|
drunken officeparty posted:I feel like I’m spending way too much time traveling long distances to stop the loving bugs from breaking my poo poo. Or refilling turrets or oops missed a spot now they took down a power line. In the long run, you'll want to use two one-way tracks rather than one two-way track. It requires more track, but it vastly increases throughput and is much simpler to actually use. The fundamental rule of trains is that only one train can be between any set of signals. The track is divided up into sections of track called "blocks" based on the presence of signals, and the signals will only allow one train to be in their block at a time. So if you've got three signals in a row on a straight track, then the first signal will turn red (blocking any further trains) as soon as your train's nose passes the signal. The first signal will remain red until the entirety of the train has passed the second signal. Meanwhile, the second signal turns red as soon as the front of the train passes it, and remains red until the entirety of the train has passed the third signal. And so on. Any time any part of a train is between two signals, the first signal will be red. It gets more complicated once you throw intersections into the mix, but the same basic principle applies - signals divide the track into "blocks", and only one train can be in a block at any time. Typically, you want to have signals on all the entrances and exits of an intersection. This ensures that only one train can pass through the intersection at a time, while also keeping the block compact so that it doesn't stop trains that aren't going through the intersection. Lastly, there's chain signals. They only turn green when both their block and the next block are open. They're useful for various edge cases that pop up when your train traffic gets congested. For example, what if traffic gets backed up shortly after one exit of a four-way intersection, turning that exit signal red? With normal signals, the entry signal will be green while the exit signal is red, so a train will drive into the intersection and stop right in the middle - blocking up the whole intersection in all directions and making your traffic problem much worse. On the other hand, a chain signal at the entrance will detect that the exit is blocked, and prevent trains from entering the intersection if they're bound for the blocked exit. That keeps the intersection open in the other direction.
|
|
#
?
Jul 26, 2021 23:10
|
|
|
Also turn off bugs. After one run with them on, you've seen everything they have to offer. The only thing you're missing out on is building a military science factory.
|
|
#
?
Jul 27, 2021 01:45
|
|
|
I personally kind of liked turning bugs to passive - gives you something to do with the military, but takes the pressure off. But if you just want to build unimpeded, I get turning them off.
|
|
#
?
Jul 27, 2021 01:54
|
|
|
bi-directional tracks can work great on the way to mining outposts that don't see trains more than once every ~30s. Just put a siding down every 2000 tiles. (remember though that a track is always a 2x2 not a single tile) Much simpler to lay down too. As there are no intermediate signals to place, just track in a straight line. As long as you have as many stations + sidings as you have trains on the line it will never lock up. Here is how you make sidings:  You'll note the red X's where there is no signal at all, even though you'd expect them if you were used to regular uni-directional intersections. This one is eastbound biased, meaning that the eastbound is a nonstop train and the westbound all jump ahead one siding at a time. You can technically alternate eastbound/westbound bias every other siding to make the system even more efficient.
|
|
#
?
Jul 27, 2021 03:19
|
|
|
|
| # ? May 25, 2024 12:10 |
|
|
So it's been a while since my first successful rocket playthrough, and I'm getting the itch again. Would the goon recommendation for a second playthrough be to go for the fast launch achievements or start in on something like Space Exploration? I tend to drop out when things get too complex but I wouldn't mind the game going longer after that first launch.
|
|
#
?
Jul 31, 2021 01:58
|
|
