|
M_Gargantua posted:Ftfy, Multiply won’t average. You gotta divide. Another good reason to use warehouses too. And you gotta be careful what you set the inserters condition as because you can inadvertently make it lock if every chest is perfectly balanced (all 0, all full etc.) I normally use “Anything < 4 to prevent this. It allows the chests to continue being fed if its a few items above the average rather than locking, while still maintaining each chest balanced to within a dozen. Whoops, yeah, I meant divide. Thanks for the catch. Another way to avoid it locking is to add in a constant combinator that adds a couple extra items in, throwing the average off just a little bit.
|
#
?
Feb 14, 2018 15:10
|
|
|
|
| # ? May 23, 2024 08:25 |
|
|
You can't use use <= 0 to prevent a lock?
|
|
#
?
Feb 14, 2018 15:39
|
|
|
Foehammer posted:Ask and ye shall receive. 
|
|
#
?
Feb 14, 2018 18:17
|
|
|
Foehammer posted:You can't use use <= 0 to prevent a lock? The advantage of using a constant combinator for things like this is that if you want to change the offset, you can just change one combinator rather than dozens of inserter conditions.
|
|
#
?
Feb 14, 2018 22:33
|
|
|
Is there any way to put the power measurement back to MW from GW? I would like to have more accurate numbers while tweaking plants. Today I've been trying to make an easy hookup 12 reactor design. The point is that it only needs small coast exposure for pumps, compared to so many designs out there that pretty much require you to landfill out the middle of a lake. It took some serious piping to get it to where it is now.  One useful thing I discovered was a pattern you can use to get three pipes worth of liquid through a space two squares wide. I could have just made the plant a couple tiles wider, but what fun would that be?  The plant is showing up as producing 1.7 GW when hooked up to a void, which it should be, but I am worried that the furthest exchangers from the coast might be very slightly water deprived. It is tough to gauge how effective changes are without more accurate power metering. I've think periodic fluid pumps can help speed up liquid? Or would it be better to hook up a couple extra offshore water pumps? A 12 reactor plant does have the ratios of 176 heat exchangers, 303 turbines, 16 pumps. I split it into sixteen equal-sized blocks of 11 exchangers, 19 turbines, and 1 pump. Foehammer posted:Ask and ye shall receive.
|
|
#
?
Feb 15, 2018 03:05
|
|
|
How do you even sequence what is effectively a MIDI? Just a buttload of counters?
|
|
#
?
Feb 15, 2018 03:16
|
|
|
XkyRauh posted:How do you even sequence what is effectively a MIDI? Just a buttload of counters? Yes, but god forbid you try to do it by hand. https://www.miditorio.com
|
|
#
?
Feb 15, 2018 03:50
|
|
|
Never seen that before. Neat. now I want to see the 1812 overture synced up with factorio artillery. Can circuits turn artillery on/off? Put it close enough to an enemy base and it should autofire.
|
|
#
?
Feb 15, 2018 03:54
|
|
|
Filthy Monkey posted:Never seen that before. Neat. now I want to see the 1812 overture synced up with factorio artillery. Can circuits turn artillery on/off? Put it close enough to an enemy base and it should autofire. Maybe control an inserter to load one shell at a time?
|
|
#
?
Feb 15, 2018 04:00
|
|
|
Foehammer posted:True, true. My circuit network fun has been turning on an American flag made out of ~300 lamps and playing the national anthem (converted from MIDI) every time a rocket launches loving hell. That owns.
|
|
#
?
Feb 15, 2018 12:00
|
|
|
Foehammer posted:Ask and ye shall receive. That's fantastic
|
|
#
?
Feb 15, 2018 12:39
|
|
|
The hardest part was determining the "rocket on pad, ready to launch" condition. Counting parts is kind of tricky due to the productivity modules (You use 5000 parts every 7 rockets, I guess I could try to modulo). My workaround to not being able to connect the silo to the network is: If all 3 part belts have > 0 items on them, start a timer. If that timer reaches 2 seconds, a rocket is ready. Any activity from the 3 inserters putting parts into the silo resets the timer, any of the belts being empty resets and disables the timer. It has an integer overflow bug that requires you to launch a rocket once every ~5000 hours of gameplay to circumvent. I fixed it by making a looping 2-second timer that sets an S-R latch instead.
|
|
#
?
Feb 15, 2018 16:55
|
|
|
Is having 2 parallel side-by-side stretches of heat pipes coming out of your reactors beneficial? Or does one single heat pipe work just as well?
|
|
#
?
Feb 15, 2018 17:10
|
|
|
I usually do it for symmetry's sake, since I mostly run 1x2, 2x2, and 2x4 designs.
|
|
#
?
Feb 15, 2018 19:27
|
|
|
Foehammer posted:The hardest part was determining the "rocket on pad, ready to launch" condition. Counting parts is kind of tricky due to the productivity modules (You use 5000 parts every 7 rockets, I guess I could try to modulo).
|
|
#
?
Feb 15, 2018 19:51
|
|
|
ArchWizard posted:My "ready to launch" alert goes off when an inserter picks up the satellite to put it into the rocket. That's actually "A rocket is currently igniting" as opposed to "A rocket sits at 100%, and you are free to insert a satellite when you reach the appropriate time in the national anthem" e- unless there's a way to tell an inserter "put that thing you just picked up back where you got it" Foehammer fucked around with this message at 20:16 on Feb 15, 2018 |
|
#
?
Feb 15, 2018 20:14
|
|
|
I picked this game up thanks to the LP of it, and it's scratching an itch I never knew I had. I literally cannot stop playing it. A game hasn't had a hold on me this strong in a while. Anyway, I've a question about nuclear power: I've used the tutorial in the wiki to set up a rudimentary setup. One reactor, four heat exchanges, eight steam turbines. This is supposed to output a ton of power, way more than what I would ever need in my little baby factory. It is, but not what's advertised. Checking it out, it looks like each turbine isn't working to capacity. They're processing around 20 steam/second instead of their max of 60. Everything is just connected via pipes. Could utilizing pumps increase the throughput here?
|
|
#
?
Feb 15, 2018 20:16
|
|
|
SettingSun posted:I picked this game up thanks to the LP of it, and it's scratching an itch I never knew I had. I literally cannot stop playing it. A game hasn't had a hold on me this strong in a while. You shouldn't need pumps at all with a small reactor like that, can you post your setup?
|
|
#
?
Feb 15, 2018 20:17
|
|
|
SettingSun posted:I picked this game up thanks to the LP of it, and it's scratching an itch I never knew I had. I literally cannot stop playing it. A game hasn't had a hold on me this strong in a while. Do you still have your legacy power running? The grid won’t draw from nuclear steam turbines exclusively, so you might have to shut down your old power generators first before the nukes take up the load. Also if you generate excess power it will just go away unless you’re trapping the steam. Even my huge base only needs intermittent powering from my nuclear reactors, most of the time they’re idle while the base runs off stored steam.
|
|
#
?
Feb 15, 2018 20:21
|
|
|
DarkHorse posted:Do you still have your legacy power running? The grid won’t draw from nuclear steam turbines exclusively, so you might have to shut down your old power generators first before the nukes take up the load. I'm at work (scheming about my factory instead of working) else I would post a picture but you know, it's probably this. I left my old steam engines running so I'll try turning them off. My peak power consumption hovers around 20 MW (I think that's the correct unit) so even a single reactor is overkill, but I'm just concerned with the experiment of getting one running for now.
|
|
#
?
Feb 15, 2018 20:26
|
|
|
It's not gonna output with no place for the juice to go.
|
|
#
?
Feb 15, 2018 21:15
|
|
|
How does the read stopped train setting on train stops work? I would have assumed that if a train is stopped there, the value of the output signal is the train ID and if there IS no stopped train, that value just becomes 0? Using that setting right now I connect the wire to a power pole on the same network to check what signals the network is seeing and it's not registering the output signal from the train stop at all no matter what train is stopped there. I'm trying to set up a basic circuit that'll check both that there's enough cargo to fill a train and that there isn't already a train parked at a stop to turn the stop on with a basic AND gate, but I can't get a network to register a signal from the stop at all. e: Fixed it. Seems like it's a bug with newly placed trains. I pulled the train out of the stop and back in and it's showing up now. Toadsmash fucked around with this message at 00:34 on Feb 16, 2018 |
|
#
?
Feb 16, 2018 00:32
|
|
|
Toadsmash posted:How does the read stopped train setting on train stops work? I would have assumed that if a train is stopped there, the value of the output signal is the train ID and if there IS no stopped train, that value just becomes 0? Using that setting right now I connect the wire to a power pole on the same network to check what signals the network is seeing and it's not registering the output signal from the train stop at all no matter what train is stopped there. I'm trying to set up a basic circuit that'll check both that there's enough cargo to fill a train and that there isn't already a train parked at a stop to turn the stop on with a basic AND gate, but I can't get a network to register a signal from the stop at all. yeah, the train must be running on automatic mode and has to stop there to generate an ID. Most people just smooth the signal out to tell if a train is stopped for various signals, but considering each train has a unique ID, I kinda wonder if there's a better way to use it, like telling stations that a train with X ID gets Y items or something.
|
|
#
?
Feb 16, 2018 01:12
|
|
|
Foehammer posted:The hardest part was determining the "rocket on pad, ready to launch" condition. Counting parts is kind of tricky due to the productivity modules (You use 5000 parts every 7 rockets, I guess I could try to modulo). 
|
|
#
?
Feb 16, 2018 01:42
|
|
|
neogeo0823 posted:yeah, the train must be running on automatic mode and has to stop there to generate an ID. Most people just smooth the signal out to tell if a train is stopped for various signals, but considering each train has a unique ID, I kinda wonder if there's a better way to use it, like telling stations that a train with X ID gets Y items or something. But the train gets a unique ID visible in the tooltip immediately upon placement since .16. It's not that it doesn't have an ID.
|
|
#
?
Feb 16, 2018 02:26
|
|
|
I've been obsessing over nuclear design too much. I think I've figured out my new current favorite way of moderating a nuclear plant. It self syncs like the chest method, but doesn't require an extra set of chests. It also runs dry for one less inserter arm worth of time than the steel chest method. It does require more logic though. The method relies on three conditions. 1) A decider wired to all the steam tanks, set to output a signal if the steam is low. 2) The count of the used fuel cells taken out, modulo (num reactors). When the count is 0, I know that I have taken out a number of used fuel cells equal to the number of reactors. This part is responsible for syncing the reactors. 3) An SR latch that is set when a cell is pulled out, and reset when a cell is put in. The point of the latch is to stop repeat fuel cell insertion. If the steam is low, and the number of removed fuel cells modulo the number of reactors is 0, and the latch is set, then it is time to insert new fuel cells. The insertion resets the latch, stopping new fuel cells from getting inserted again. The used fuel cell count is read via hand pulse, so the new set of fuel cells goes in at the same time as the last fuel cell is coming out. That minimizes the amount of time the reactors will always run dry. I can't think of any way to do it faster without a timer, which introduces a whole host of other edge cases and brownout related slowdown issues. The six reactor plant I made with the moderation scheme runs at a sustained load of 795 MW, which I feel is pretty good given the theoretical maximum sustained load of 800 MW. I mostly attribute the difference to the one inserter arm rotation of dry time every 200s.  Also, it seriously takes more testing than you would expect to get the steam settings right. It is easy to have a plant that runs fine at 100% capacity, but fucky at 95% due to the dipping steam values and high consumption. I spent a lot of time at game speed 100 in creative mode, watching this plant do cycles at various power draws while I tweaked the storage. I can't promise this is perfect in 100% of cases, but it should be pretty solid. Edit: God drat it, I need to move a pump. Edit2: Pump Moved. https://pastebin.com/FHqngTkM Filthy Monkey fucked around with this message at 04:51 on Feb 16, 2018 |
|
#
?
Feb 16, 2018 04:14
|
|
|
Toadsmash posted:But the train gets a unique ID visible in the tooltip immediately upon placement since .16. It's not that it doesn't have an ID. Correct. It has the ID, but a station doesn't recognise a train until it has arrived at said station, which a newly-placed train has not done.
|
|
#
?
Feb 16, 2018 07:37
|
|
|
chairface posted:It's not gonna output with no place for the juice to go. For what it's worth this was the case. After turning off my steam engines output went up, and continued to go up when I added new things until they hit capacity. For some reason I thought all the power generators would just generate all their possible power at all times and what wasn't used would go off into the ether. It made sense in my head. Also I played for like 6 hours straight and managed to launch the rocket for my first win. I then immediately started a new game because I had some new layout designs I wanted to try out. So many ideas, so little time. This game is good.
|
|
#
?
Feb 16, 2018 20:51
|
|
|
SettingSun posted:For what it's worth this was the case. After turning off my steam engines output went up, and continued to go up when I added new things until they hit capacity. For some reason I thought all the power generators would just generate all their possible power at all times and what wasn't used would go off into the ether. It made sense in my head. Glad you're having fun! For what it's worth, it's usually a better idea to just get your base mostly self-sufficient, then go off to a new area to start a new base, that way you don't have to spend a bunch of time researching, bootstrapping up from nothing, and running around trying to hold everything together. That said, I did the exact same thing you did and it's not a huge deal, just slow to start up.
|
|
#
?
Feb 16, 2018 21:55
|
|
|
That factory was held up with duct tape and a nightmarish web of belts and pipes. It was supposed to be a 'get my bearings' game that lasted 40 hours. Frankly I'm glad to start anew. I also turned on peaceful mode for this new game so I don't have to drive my tank out every 30 minutes to clear out the rabble that moves in.
|
|
#
?
Feb 16, 2018 22:00
|
|
|
SettingSun posted:For what it's worth this was the case. After turning off my steam engines output went up, and continued to go up when I added new things until they hit capacity. For some reason I thought all the power generators would just generate all their possible power at all times and what wasn't used would go off into the ether. It made sense in my head. The reactors will generate all their possible heat at all times, but the turbines will only consume steam as necessary to meet current needs.
|
|
#
?
Feb 16, 2018 22:24
|
|
|
GotLag posted:Performance was total rear end, which I assume is the reason why every other wireless mod has been abandoned as well. Aww. Oh, well. Well, considering I needed a new computer to actually run mid-game AngelBobs at more than 20 UPS, probably for the best. quote:Edit: As for channels, the problem is how to implement that UI-wise. The signal selection was just a constant combinator's GUI. I was thinking a new variable, like you had for Reactors, that would never be sent through an antenna but would instead set the antenna's channel number. Of course, if the performance of wireless signals was already bad, then adding the ability for UINT64 new channels would probably make it even worse. Psawhn fucked around with this message at 23:15 on Feb 16, 2018 |
|
#
?
Feb 16, 2018 23:00
|
|
|
Any idea how the wireless signals are slower than regular signals? Can you just make an infinitely long invisible wire link between two wireless points? Or are green and red the only usable connections since they’re hard coded? I’m imagining some mod that adds purple and blue and yellow wire too.
|
|
#
?
Feb 16, 2018 23:37
|
|
|
M_Gargantua posted:Any idea how the wireless signals are slower than regular signals? Can you just make an infinitely long invisible wire link between two wireless points? Or are green and red the only usable connections since they’re hard coded? I’m imagining some mod that adds purple and blue and yellow wire too. Or rather, an infinitely long invisible wire link between every wireless point and an imaginary power pole that every wireless point connects to. Otherwise, you get n^2 connections.
|
|
#
?
Feb 16, 2018 23:50
|
|
|
You can't mod in extra kinds of circuit wire, or make a given wire invisible. Wireless mods work by creating custom constant combinators then overriding their output settings every update (typically every tick). This involves a bunch of lua table creation and iteration, which is slow.
|
|
#
?
Feb 17, 2018 01:48
|
|
|
So what in the code is linking two entities linked by green or red wire? Could you just override that value and tell it two entities are on the same 'red' network even without a sprite between them? Something like Far reach but for wire connections.
|
|
#
?
Feb 17, 2018 01:55
|
|
|
http://lua-api.factorio.com/latest/LuaEntity.html#LuaEntity.connect_neighbour
|
|
#
?
Feb 17, 2018 02:08
|
|
|
Collateral Damage posted:Rampant - Improves the biter AI and makes it more interesting. They won't just throw themselves blindly into your wall of turrets any more, but pull back and try to find other routes and exploit vulnerabilities in your defense. Would anyone like to tell me more about Rampant? I installed it but haven't faced the biters yet. I'm unsure how to configure properly it so it won't be too much for me.
|
|
#
?
Feb 17, 2018 03:26
|
|
|
https://www.miditorio.com/ + http://www.midiworld.com/files/295/ + maybe a flamethrower wall or something and a recording of biters getting wrecked
|
|
#
?
Feb 17, 2018 05:12
|
|
|
|
| # ? May 23, 2024 08:25 |
|
|
I've heard of (but not seen in implementation) a theoretical "train stop wifi" which I can only assume involves train stops being turned on and off and somehow reading that remotely, but the only way I can think of involves having a rail connection (to avoid no-pathing) and reading (and then stopping, via a controlled red signal) a train that tries to leave the receiver. e: After a bit of googling maybe it's possible with multiple identically named stations as well, which would enable a true "wireless" connection. You'd still be limiting to sending one bit per train and the latency would be awful though. Xerol fucked around with this message at 05:56 on Feb 17, 2018 |
|
#
?
Feb 17, 2018 05:53
|
|