|
DreadCthulhu posted:How much water could one possibly need in a regular playthrough? Is it fair to fill in most water you encounter as your base grows with exception for a few squares that you'll need to pump from? Yes. You'll need a decent amount of water for oil processing and nuclear if you use it, but here a "decent amount" is probably fewer than a dozen pumps until you get to megabase stage.
|
#
?
Aug 23, 2020 19:08
|
|
|
|
| # ? Apr 27, 2024 22:53 |
|
|
Also there's nothing that says you have to pump water locally, we just tend to because it's easy. Water is just as transportable as any other resource, and mass-mining it is incredibly cheap. It's just the question people above you pointed out about how much you want to focus on building around the environment, and how much you want to just pave the world and focus on scale or optimization.
|
|
#
?
Aug 23, 2020 19:12
|
|
|
Regarding bringing water from far away, is there a limit on how far I can pump it from? I know that regular liquids apparently need a pump added every few stretches of pipe to keep the flow strong, but does that not apply to offshore water pumps? Is the only limitation the 1200 units per second?
|
|
#
?
Aug 23, 2020 19:42
|
|
|
LonsomeSon posted:Pretty sure a narrow neck of land with water on two long sides is an isthmus, a narrow neck of water with land on two long sides is a channel. You are correct. I should have said turning channels into isthmuses. You pedant.
|
|
#
?
Aug 23, 2020 20:13
|
|
|
DreadCthulhu posted:Regarding bringing water from far away, is there a limit on how far I can pump it from? I know that regular liquids apparently need a pump added every few stretches of pipe to keep the flow strong, but does that not apply to offshore water pumps? Is the only limitation the 1200 units per second? All liquids behave the same. You need pumps pretty often over any real distance to keep throughput up. Either train the water or build factories that need water near a body so you don't have to pipe it far. If you lay out pipes first you can hover over the pipes and see the change in levels over distance. It is quite drastic, something like every 20 segments you need a pump to keep throughput up if you need a lot of whatever fluid. For really low volume requirements you can easily get away with few pumps over a longer distance, as long as the source itself is strong enough. necrotic fucked around with this message at 21:52 on Aug 23, 2020 |
|
#
?
Aug 23, 2020 21:49
|
|
|
Peep dat  I missed this game so much. 
|
|
#
?
Aug 23, 2020 21:52
|
|
|
Far too tidy.
|
|
#
?
Aug 23, 2020 21:56
|
|
|
Oxyclean posted:Far too tidy. Give it time. 
|
|
#
?
Aug 23, 2020 22:06
|
|
|
Thanks! Makes sense. Trains: is there any upside to having rails be one-directional in your train network, vs using one railway and building a loop at the ends? Is it mostly so that you can have multiple trains all going in the same direction on the track? I'm assuming trains don't crash and burn if they're all going in the same direction? I saw a bunch of timelapses of people building one-way tracks and I'm not fully getting it.
|
|
#
?
Aug 23, 2020 22:10
|
|
|
Two-way rail lines (either with loops at the ends or using double-headed trains) can't really support more than one train on the line at a time, so once you get to the point where you need more than to maintain throughput, running parallel one-directional tracks allows you to scale that up massively at very little cost and effort past the initial investment in laying the track. It takes some time to reach that point though, so running two-directional lines initially can also be useful for a while.
|
|
#
?
Aug 23, 2020 22:29
|
|
|
While you need a pump after every 17 pipes to hit 1200/s, you can get 1000/s with pumps every 200 pipes. There generally isn't much reason to put loops on bidirectional rail, generally people will just double-head the train to save space when they're doing that. To help hammer home the idea of scaling up and throughput, think about the rate stuff a is crossing a single tile of rail when a train crosses it at maximum speed, and then imagine a rail completely packed with trains moving nonstop - that's the theoretical limit for how much a single one-way rail could be moving at once if trains were able to perfectly exit and merge from their sidings. It's really crazy.
|
|
#
?
Aug 23, 2020 23:35
|
|
|
You ever go back to a map and have absolutely no idea how some of your own combinator contraptions work? I had a really nice system for logistics train network that forced it to request items in nice batches once one item from a particular supplier passed a threshold. But going back to the map.... taking me forever to figure out how it worked. Also, if you've never tried it, consider Ribbon Maze. I had a blast making rail systems & LTN stations that would work efficiently with minimal rail traffic. EDIT: Ah-ha! I figured out how I made LTN not dumb. The main issue I had with it was if I had say a mall or something that requested wildly different amounts of things, it would generate lots of small loads, and when a supply depot could provide a variety of items, there was no way to tell LTN 'by the way, while you're going there, stock up on xyz as well'. So the logic is something like this: Set the lowest necessary request threshold on LTN. Use a decider w/ feedback outputting everything < 0 as your accumulator of current requests. Pulses from output inserters or differentiators on receive tanks will be used to add to that accumulator and zero it out as we receive things. Invert the output of the memory cell and combine it with the local logi port supply or chests/tank supplies, this is our 'available or in route' signal. From this supply, filter out the things we want to request into the groups we want to request them together (ie, while getting iron, get copper too) eg, if our mall requests iron, copper, and steel from our smelter, lets group those together using 3 filters to give a single signal. Sum that filtered supply with a constant combinator into a decider checking for <0 and outputting some flag. Decider is set to negative trigger threshold for each material. (say -500 each) When any of those go below their respective threshold, we'll produce a request. Feed that trigger along with the filtered supply and a new set of negative values (say -2000 each), our desired buffer into a filter checking for the flag. Now when any item trips their trigger, all the items will request just enough to get up to this desired buffer. Combine the output of all the batch circuits, filter for negative values, then drive through a one-shot so we produce a single tick into the main decider accumulator and send the request. To keep LTN from making duplicate requests, whenever anything is coming out of the train stop itself OR the train signal for the stop is red, set an inhibit signal and use that to block the negative value coming from our accumulator to LTN. This way our inventory can stabilize. This fixes all the derpyness of LTN that I found, I've been really happy with it. Rescue Toaster fucked around with this message at 01:58 on Aug 24, 2020 |
|
#
?
Aug 24, 2020 01:04
|
|
|
I use chunk aligned dual rails instead of 1 way rails, I got bored of troubleshooting the rails when they hosed up. I'm transitioning my base to megabase and I've found the best way to do it is to clear a new area of biters, wall it off and start building a second base. I've got 4-5k logistics/construction bots running around now, have nuclear power and have begun filling in water to square my base up. I'm going to set the biters to death world settings soon with a mod so they harass me constantly, but as I have a 3 layer deep laser turret grid with flamers and a maze of walls I don't expect them to get far.
|
|
#
?
Aug 24, 2020 02:14
|
|
|
How do I tell if a pumpjack is about to be done? Seems like about 5 old clusters of pumps are == 1? Do you all send a train through all of your old pumps once in a while just to squeeze a little juice out of them, or is not worth it?
|
|
#
?
Aug 24, 2020 05:25
|
|
|
DreadCthulhu posted:How do I tell if a pumpjack is about to be done? Seems like about 5 old clusters of pumps are == 1? Do you all send a train through all of your old pumps once in a while just to squeeze a little juice out of them, or is not worth it? I have all my remote oil stations named Remote Oil, and hook the tanks up to the station, and only turn the station on when there is an oil pickup ready. Boom, trains cycle in and out of all stations, favoring the "live" ones and only going to "dead" ones when it's effective to do so.
|
|
#
?
Aug 24, 2020 06:21
|
|
|
Evilreaver posted:I have all my remote oil stations named Remote Oil, and hook the tanks up to the station, and only turn the station on when there is an oil pickup ready. Boom, trains cycle in and out of all stations, favoring the "live" ones and only going to "dead" ones when it's effective to do so. Can you show how you wire that up? Sounds like a great solution.
|
|
#
?
Aug 24, 2020 06:40
|
|
|
Slung Blade posted:Can you show how you wire that up? Sounds like a great solution. Wire up the tank(s), hook them to the train station. Set the train station to activate when there is enough to hold a full train load, e.g. "oil > 100k". That's it. This is the main method of automating larger train setups in vanilla if you don't have any special train mods. You name stations the same thing, so in this case maybe you'll have "Oil Loading", which would be the name of all your stations that load oil. Then you have another set of stations named "Oil Unload". Your unloading stations will be wired to disable unless storage falls below a set amount. When it does, the station will turn on, and any full oil trains that are idle will instantly start heading towards it. Once they're unloaded, they'll loop right back to any active oil loading stations and will fill immediately. This is especially good with ore mines. Say you have 5 separate places that can grab iron ore - all of them are named "Iron Mine". Each is wired to be deactivated until there is enough storage to fill a train. Then you have your smelting location, named "Iron Unloading", wired to be deactivated unless storage is low. Then your iron train will have this setup: Iron mine - wait until full Iron ore unloading - wait until empty Your trains will automatically route themselves to any station that has is able to load/unload them. Note that this method will require you to have stackers at every station to avoid clogging your train system. This is especially nice though because it allows you to have multiple consumers that will automatically get filled whenever they need it, so e.g. you can have a plastic outpost that has coal and oil stations, but also have coal and oil stations at your main base for other products. That way you don't have to have dedicated loops for each input/outputs that may have not be balanced equally so they spend a lot of time idle. All you have to do is name the stations the same thing and wire them up to activate on whatever conditions you need. You can modify this with full/empty train depots to take some strain off per-station stackers, but that's a little more complex and is only relevant once you're getting above 2k spm. Taffer fucked around with this message at 07:16 on Aug 24, 2020 |
|
#
?
Aug 24, 2020 07:14
|
|
|
Do you have problems with multiple idle trains waiting for a pickup load all converging on the same station when it pings out an "I'm full!" signal? Or is the vanilla train 'scheduler' smart enough to only send one at a time?
|
|
#
?
Aug 24, 2020 15:00
|
|
|
Slung Blade posted:Do you have problems with multiple idle trains waiting for a pickup load all converging on the same station when it pings out an "I'm full!" signal? Or is the vanilla train 'scheduler' smart enough to only send one at a time? Generally when I use the method ascribed above, there's only a single train dedicated to that resource. Thinking through the problem scaling up, this is why you'd want waiting areas for trains. If you have multiple trains, you can add a third "overflow" station. If the train gets to the waiting area and the stop it was going to is turned off (say from another train taking all resources from the mine or filling your buffers) then the third stop which is always active would be selected when it stops at the signal before the now deactivated station. Bonus is that if you're expanding super far, you can have this waiting station split the difference between your offload and active mines. And the same applies to the waiting station: using the same name on multiple stations (even if they're all active) will allow the train to wait at it ready to go. Of course I usually launch the rocket before I get that far scaling up, get bored and start a new save....
|
|
#
?
Aug 24, 2020 15:14
|
|
|
Vanilla trains are stupid and will dogpile. The best fix to this I've come up with is to have all your loading stations have generic names i.e. Iron Plate Load but to name your unload stations specifically - Iron Plate Unload Green Circuits. Then you set up the specific number of trains you want to grab from Iron Plate Load and unload at Iron Plate Unload Green Circuits and that way you don't have to build a huge stacker which can accommodate every train moving iron plates.
|
|
#
?
Aug 24, 2020 15:37
|
|
|
Slung Blade posted:Do you have problems with multiple idle trains waiting for a pickup load all converging on the same station when it pings out an "I'm full!" signal? Or is the vanilla train 'scheduler' smart enough to only send one at a time? Yep, hence the stackers. It's not a perfect system, but it's decent enough. I was able to get to 2k spm using that method. I had some ideas to make it smarter but I ended up just moving to train mods in more recent runs to avoid mega circuit setups at every station.
|
|
#
?
Aug 24, 2020 18:51
|
|
|
Playing again now that 1.0 is here. Love the new (to me) basic oil processing recipe only producing gas. Determined to actually launch a rocket this time and get No Solar and No Laser achievements. The last time I actually launched was pre 1.15.
|
|
#
?
Aug 24, 2020 18:56
|
|
|
I started a new game for 1.0 too! I launched my first rocket a few months ago, despite playing since .12. How do y'all choose what to build first? I always get really overwhelmed in the beginning, major analysis paralysis once I get basic belts + red science automation going.
|
|
#
?
Aug 24, 2020 19:33
|
|
|
Taffer posted:Wire up the tank(s), hook them to the train station. Set the train station to activate when there is enough to hold a full train load, e.g. "oil > 100k". That's it. I've had really bad experiences trying to wire Enable conditions to both the Mining and Delivery stops, if you're using more than one train for that resource. I've encountered a situation where, say, two Mining stops get just enough ore for a train to come pick it up, but the first full load being returned puts the Delivery stop over capacity, so it disables... now you've got a train full of ore literally stopping dead in the middle of the tracks until that Delivery station opens back up. That situation is the primary reason we switched to LTN! Edit: We usually ended up having Delivery stations never disable themselves, and only wired the Mining stops to flag Enabled if they had enough for a full train. 
|
|
#
?
Aug 24, 2020 19:36
|
|
|
Solumin posted:How do y'all choose what to build first? I always get really overwhelmed in the beginning, major analysis paralysis once I get basic belts + red science automation going. As long as biters are kept at bay (which is a pretty huge focus at the early game on default settings, to be fair) it's just a matter of seeing what you're being starved of and expanding the part that produces it while climbing the science trees, with speedbumps around oil and probably nuclear. I don't sweat too much about implementing new tech if it's expensive either, fast inserters and steel power poles are essential past an early point IMO but you can get away with yellow and red belts all the way to the rocket, and blue assemblers to 90% of the way there. You can fall into a trap of putting in loads of effort to produce and use something you won't get major benefits from until you've expanded every other part of your production. I usually regret not laying out some stone processing and military science early on since they have unusual ingredients, and i end up stockpiling hand grenades fairly early on just to deal with trees blocking any building plans.
|
|
#
?
Aug 24, 2020 20:52
|
|
|
Should I be building a ton of radars? Right now I have one per outpost, but I could easily have 100 of them in a cluster somewhere. Does that get me anything?
|
|
#
?
Aug 24, 2020 21:24
|
|
|
Nah. More radars means a particular area can get scanned up quickly but unless you are specifically making a scouting outpost that's not really a big concern, and one is plenty so you have the vision... Maybe a second is a backup in case one gets eaten by bugs.
|
|
#
?
Aug 24, 2020 21:27
|
|
|
Radars have a view area of 7x7 chunks (224x224 tiles), which let you zoom in the map as if you were there in person. And that easily allows placing ghosts and blueprints from afar once you get to the robot stage. They also have a much larger scan range of 29x29 chunks (928x928 tiles), which slowly circles around the unexplored terrain and reveals resources and biters, but doesn't affect the map fog-of-war Stacking multiple radars just increases the scan speed, but not the scan range or view range.
|
|
#
?
Aug 24, 2020 22:28
|
|
|
XkyRauh posted:I've had really bad experiences trying to wire Enable conditions to both the Mining and Delivery stops, if you're using more than one train for that resource. I've encountered a situation where, say, two Mining stops get just enough ore for a train to come pick it up, but the first full load being returned puts the Delivery stop over capacity, so it disables... now you've got a train full of ore literally stopping dead in the middle of the tracks until that Delivery station opens back up. Yeah, I graduated to TSM for similar reasons. Its UI is a little rocky but its functionality is great. Felt like a real natural progression from the way I did trains before.
|
|
#
?
Aug 25, 2020 00:09
|
|
|
XkyRauh posted:I've had really bad experiences trying to wire Enable conditions to both the Mining and Delivery stops, if you're using more than one train for that resource. I've encountered a situation where, say, two Mining stops get just enough ore for a train to come pick it up, but the first full load being returned puts the Delivery stop over capacity, so it disables... now you've got a train full of ore literally stopping dead in the middle of the tracks until that Delivery station opens back up. Yeah to overcome that you need sr latches and read when an arriving train is on the rails before turning off the station. In fact, you have to put a 'pre' station for this. It gets complicated.
|
|
#
?
Aug 25, 2020 00:51
|
|
|
Wallrod posted:That's the beauty of factorio, even if you're putting stuff in chests because you haven't built the next part of production, you're still making progress towards something, and you can automate a really messy fast solution while you figure out the Right Way for your next goal. Once red is automated i start to make a small parts mall to stockpile inserters, splitters, assemblers and belts and such just to streamline the expansion process, and green science is only a few steps more to automate over red. This helped push me over, just finished setting up a nice, ugly smelting area for copper, iron and steel. Stone bricks is next. I've never enjoyed having biters attack me early, but having them not attack at all has been boring. So I'm splitting the difference with a 200% starting area and Rail World (so no expansion (and also TRAINS!!)) in the hope that these settings will have me facing biters in the early-mid-game.
|
|
#
?
Aug 25, 2020 01:19
|
|
|
I've automated the mario colored and green mario colored science in sea block and it feels satsifying. Like when I first started with factorio.
|
|
#
?
Aug 25, 2020 02:40
|
|
|
Now look at blue science and feel absolutely lost. Like when I first started with Factorio. (Red Circuits and how to produce Resin that doesn't gobble up all your free space and oh gently caress your tin and lead are backed up because you're producing gold/aluminum/cobalt/zinc/nickel without catalysts ugggggggh)
|
|
#
?
Aug 25, 2020 02:51
|
|
|
How do you all deal with constantly respawning bug bases? Do you keep running around the map taking them down, or is there a way to make it less of a nuisance? At some point you're big enough that there's always something under attack :/
|
|
#
?
Aug 25, 2020 06:34
|
|
|
Once you get sufficient artillery range research, you can clear your pollution cloud, then build artillery strongpoints that will wipe out any bug bases that move back in, and the (weak) counterattack from destroying those bases will hit your artillery base instead of somewhere undefended. Before then, or if you don't want to do that, you mostly just build a big wall to keep the biters out. Look for chokepoints between water and cliffs to make it a little cheaper.
|
|
#
?
Aug 25, 2020 06:44
|
|
|
DreadCthulhu posted:How do you all deal with constantly respawning bug bases? Do you keep running around the map taking them down, or is there a way to make it less of a nuisance? At some point you're big enough that there's always something under attack :/ Well, personally I turn off biter base respawning at worldgen. If not doing that you end up needing to set up a big perimeter wall. Enclose your base(s) in walls, turrets, flamers and an autosupplied ammo belt if using gun turrets (generally a good idea). You want to get roboports and construction robots set up pretty fast to automatically repair your defenses, getting them should be your first goal after doing initial oil processing and setting up blue science. The goal is that, yes, you'll be constantly under attack but you don't personally have to do anything about it since your defenses have enough firepower to kill the attacks before they destroy anything, and your bots automatically repair any damaged wall segments or turrets. In the late game you'll have artillery and a platoon of angry spider robots, but it doesn't sound like you're near launching rockets yet.
|
|
#
?
Aug 25, 2020 06:50
|
|
|
Enclosing your entire base in walls is unnecessary. Build firebases with a couple of walled-in turrets and a box of ammo at even distances. Stock biter attack waves will aggro on anything that attacks any member of the group, so they'll stop there anyway. Add a speaker that starts to beep when the ammo box is low and just stock it up manually.
|
|
#
?
Aug 25, 2020 08:25
|
|
|
Collateral Damage posted:Add a speaker that starts to beep when the ammo box is low and just stock it up manually. This way you can have all manner of alerting and monitoring systems for various resources that might run out
|
|
#
?
Aug 25, 2020 11:17
|
|
|
Eh, I find encasing my base easier than making bunkers once in mid game. You just draw a line of walls, then a line of turrets, then a line of belts. Using bunkers might save me some resources but require me to think about where they go and potentially block useful base areas. Once I'm in blue science ease-of-build becomes the bigger concern to me. It's not super important though. The main takeaway is that the solution to biter attacks in the mid game is the combination of more guns, either by belt-fed from your ammo assemblers or from a chest full of ammo, combined with roboports and construction drones with access to repair packs to keep the turrets and walls up without your intervention.
|
|
#
?
Aug 25, 2020 11:25
|
|
|
|
| # ? Apr 27, 2024 22:53 |
|
|
Xerophyte posted:Eh, I find encasing my base easier than making bunkers once in mid game. You just draw a line of walls, then a line of turrets, then a line of belts. Using bunkers might save me some resources but require me to think about where they go and potentially block useful base areas. Once I'm in blue science ease-of-build becomes the bigger concern to me. * or disco lasers
|
|
#
?
Aug 25, 2020 11:27
|
|