Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
Loel
Jun 4, 2012

"For the Emperor."

There was a terrible noise.
There was a terrible silence.



You always make this a fascinating read :allears:

Adbot
ADBOT LOVES YOU

Doodmons
Jan 17, 2009
Yeah just chiming in to say that this is the best possible way to do a Factorio LP imo. I could never in a million years play Factorio with this unholy cocktail of mods as the base game makes my brain hurt enough as it is. I can live vicariously, though.

Strategic Sage
Jan 22, 2017

And that's the way it is...

Boksi posted:

Will we see the legendary braided double-wide sushi belt in this playthrough?

You never know, but highly unlikely. Most of those kinds of things don't look like they would play nice with Belt Overflow. More likely is a stupid-complicated train network at some point.

Loel posted:

You always make this a fascinating read

Doodmons posted:

Just chiming in to say that this is the best possible way to do a Factorio LP

Glad to know the approach is working -- let the incineration of brain cells proceed!

Pierzak
Oct 30, 2010
So... what happens if you dump stuff from a belt onto a loop that's already full? Does it end up on the ground, or does it stop on the belt?

Alkydere
Jun 7, 2010
Capitol: A building or complex of buildings in which any legislature meets.
Capital: A city designated as a legislative seat by the government or some other authority, often the city in which the government is located; otherwise the most important city within a country or a subdivision of it.



Woo! Factorio Madness LP! Skimmed through this in a few hours and am...impressed with the, um, masochistic mix you selected. I've never gone that deep myself.

Something I wanted to bring up from pages (and likely weeks/months ago) is everyone was talking about how all fluids more or less act the same and some silliness about various systems (I gave this game to my dad for father's day and his eye started twitching at turning petroleum into sulfur, going on about how you're supposed to be removing it during the refining/distillation process not alchemically transmuting hydrocarbons to sulfur).

But yeah, as people have mentioned with steam is that you can have backup steam. This game doesn't really do anything with actual thermodynamics besides heat pipes for nuclear so steam will sit in a pipe/tank forever, never losing heat/energy. This brings up the fact that if you don't want a bunch of power poles between your base and outpost, what you can totally do is bring power via train-hauled steam batteries to your mining outposts! Just have some steam engines/turbines and a few tanks on site to hold the trainload.

Speaking of trains and fluid dynamics, I noticed you setting your oil stop like this:


This is an ENTIRELY REASONABLE way to do such things for vanilla/small factories. However you have the entire Arch Angel's library which includes Petrochem and as someone who's had experience with that, you're gonna want something a wee bit faster due to the sheer amount of fluids you're gonna be bringing in (even if it's just to pump into your refineries). The issue is that pipes have a maximum throughput of how much fluid they can transfer per second. Tanks, and tanker cars, do NOT have this limit. So for something a bit faster you can place a tank to store the oil (or other liquid/gas) directly next to the train stop and just connect tank to pump to train car without any pipes. You'll be limited by only the incredibly high fluid transfer rate of the pump, which will let you fill or drain a tanker car in about 5 seconds.

Strategic Sage
Jan 22, 2017

And that's the way it is...
Retracing My Steps, Part I

Video

With a basically secure situation in place for now, it was time to do some things I'd already done before in the previous save. After thinking a bit, I wanted to get radar up first, stone walls next, and steel third. None of those were hugely involved projects; a bit of research and getting things in place but nothing overly complex. I'm still trying to be reasonably efficient in general, but not obsessing too much about it without the rampant threat hanging over my head.




I wanted this to be fairly centralized, yet out of the way. Here, on the undeveloped side of the coal line, seemed a reasonable location. As happened last time, our coal usage will now basically double to provide the power for it, but we have plenty coming in. Sometimes, as is the case now, I have the saphirite mining totally shut down due to oversupply. This of course is a good problem to have.

Pierzak posted:

So... what happens if you dump stuff from a belt onto a loop that's already full? Does it end up on the ground, or does it stop on the belt?

This image also provides a useful answer to this question. Short version is that it eventually ends up on the ground. All the coal on the loop there should by rights be on the same side, because the drills all empty onto that side. When it backs up enough, the 'merging' belts will force the coal on the loop over to the other side. If it gets too bad, as it did here for a bit, you can see where they fell off the side by the burner inserters, or the one that got shoved off the far side of the loop. Basically there's a small backup that you can sustain, but get more than a couple extra items on the merging belt and you start getting a mess.




This was a quick-n-dirty adjustment, flawed and not particularly pleasing aesthetically. It does the job though. Stone walls have been researched, and are now slowly being built. All of the stone is now directed into a single location. There is about 2k raw stone, and almost a thousand crushed stone currently in the system. That's enough to make a few hundred wall sections and still have enough left over for other production needs. I estimate not quite enough for a full perimeter barrier, but close once we chew through it ... and more will come eventually from the mining work.




Steel was next on the list. I didn't really have room for a separate area for it ... so I won't use all the mines for now. I just took the last furnace here and replaced it with a metal-mixing one set to steel to get a modest start-up supply going. Right now I'm running two drills, and I can upgrade to three without any worry of ore getting through with the four remaining iron furnaces.

And then comes the more complicated part; all the stuff required to build circuit networks.




This somewhat over-engineered location is between the greenhouse in the west and the rest of the current operation. Resin and rubber are produced here. I thought about how to get coal over here ... and then just decided not to, burning the wood directly as fuel in the rubber-making furnace. For review, resin is used to make rubber and also for solder which we'll need for the basic electronic boards; rubber is required to insulate wires, which we need for any kind of circuit network. So that's where those commodities fit into things right now.

At this point I decided to break off and do a bunch of research. The first few were military in nature(heavy armor, bullet damage, bullet speed aka rate of fire). These are no longer as urgent for the purposes of needing everything I can get to survive, but I still think they are important in a resource conservation sense. When the biters do eventually return in reduced numbers, I want to be able to dispose of them as economically as possible and get back to other priorities. The ammunition upgrades will help with that. I've also now got several other things going on; radar scanning our surroundings, stone wall sections being produced, startup steel supply being forged/smelted, etc. So it seemed like a good time to do this while all that was happening.




Another retread setup, but again with a bit of a twist. This is the carbon-producing section, with carbon dioxide and coke(from crushed coal at the top) combining. The red long-handed inserter here uses the bob's mods adjustable inserters feature for the first time. It takes the carbon and turns at a right angle, reaching across the coal line to the west instead of dropping it opposite the liquifier to the north where it normally would. This whole carbon setup is on top of where there was coal a few hours ago, but as we wear away that deposit to the east, I'm just repurposing the space.




This little display configures where to grab from and drop the next inserter you want to place. I don't imagine I'll be using this a ton, but in this case it makes it a lot easier than doing some convoluted stuff with underground belts and moving more things around. So it's a nice QoL feature in the mod that was brought up in the thread before, and from time to time as I find it useful I may do it for things like this.




The last remaining bit of infrastructure setup is adding tin and lead, from bobmonium and rubyte, so that we can make the solder that's required for basic electronic boards. Tin is also needed for tinning copper wire, which is a component of basically everything I'm trying to get done right now. I've got a bunch of nanobot tubes built, some termite ones as well to clear the way for this extension of the coal line, and I thought it was worth it to send this mostly around the saphirite field here to ensure we don't have to fight with it as much in the future.




And here's that last bit of furnacing. One for lead, one for tin, the middle one to make solder out of the two. That's the one for which these metal-mixing furnaces are required, because that's literally what it does; you can't put tin and lead together to make solder in a standard stone furnace. The line to the north eventually takes all the crushed stone, along with any tin and solder output here, and joins the main artery from the greenhouse to the assembly area right by the whole rubber-resin setup.




This drill is on the rubyte field in the southwest, but it has a patch of bobmonium within it, so that we get both ores. This has the advantadge of simplicity, and we currently have no bobmonium patches within our borders. However we'd actually like to have quite a bit more tin than lead, and this is going to end up with things going the other way around, so it's going to be highly inefficient and it'll probably take some time to get enough tin to do anything. Like many things, this is very temporary -- it's just here until we can expand and do something better. For the relatively small amounts we're going to need before then, it should be good enough.

Strategic Sage
Jan 22, 2017

And that's the way it is...

Alkydere posted:

as people have mentioned with steam is that you can have backup steam. This game doesn't really do anything with actual thermodynamics besides heat pipes for nuclear so steam will sit in a pipe/tank forever, never losing heat/energy. This brings up the fact that if you don't want a bunch of power poles between your base and outpost, what you can totally do is bring power via train-hauled steam batteries to your mining outposts! Just have some steam engines/turbines and a few tanks on site to hold the trainload.

That's in the plans. I haven't gone into what I intend to do in the mid-late game as it's sort of like talking about dinner when you don't know if there's anything for breakfast yet. But shipping around steam for power is something I'd like to do at various points.

Alkydere posted:

The issue is that pipes have a maximum throughput of how much fluid they can transfer per second. Tanks, and tanker cars, do NOT have this limit. So for something a bit faster you can place a tank to store the oil (or other liquid/gas) directly next to the train stop and just connect tank to pump to train car without any pipes. You'll be limited by only the incredibly high fluid transfer rate of the pump, which will let you fill or drain a tanker car in about 5 seconds.

This I might have thought of … eventually. Petrochem is the part that, from the bit I've read, is probably going to make my brain explode the worst. So it's good to have an idea that might make things just a bit easier, thanks!

Bremen
Jul 20, 2006

Our God..... is an awesome God

Thotimx posted:

This I might have thought of … eventually. Petrochem is the part that, from the bit I've read, is probably going to make my brain explode the worst. So it's good to have an idea that might make things just a bit easier, thanks!

The big issue I had with petrochem is pretty much every reaction makes at least two things, one that you need and one that is probably useful for something but not what you need right now, and if it is something you need then if you need less of it then the other output then eventually you'll fill up on one and the other will stop producing, etc. So basically you end up having to choose between voiding half the results and losing like half your efficiency each step (my inner perfectionist!) or having a petrochem layout that makes a CPU circuit diagram look simple.

Alkydere
Jun 7, 2010
Capitol: A building or complex of buildings in which any legislature meets.
Capital: A city designated as a legislative seat by the government or some other authority, often the city in which the government is located; otherwise the most important city within a country or a subdivision of it.



Bremen posted:

The big issue I had with petrochem is pretty much every reaction makes at least two things, one that you need and one that is probably useful for something but not what you need right now, and if it is something you need then if you need less of it then the other output then eventually you'll fill up on one and the other will stop producing, etc. So basically you end up having to choose between voiding half the results and losing like half your efficiency each step (my inner perfectionist!) or having a petrochem layout that makes a CPU circuit diagram look simple.

The cheapest way to make plastic is take naptha, steam crack it with red catalysts (copper+iron ore) to make...methanol I think? Basically whatever comes out of that is only 1-2 easy steps from liquid plastic. Any other method of early plastic means that as you mentioned you'll lose so much from each step that you get trivial final results. In one game I would bring in a 3 tanker-car train of Natural Gas straight from the pump (and this was back when tanker cars held 75K units instead of 25K) to make about half a box car load of plastic because I needed blue science to research the stuff that all the waste products could be used for.

Thotimx, I highly reccommend adding FNEI (Factorio Not Enough Items, inspired by the Minecraft mod Not Enough Items, it will scan the recipe list to find what all recipes create a product and what recipes use a product) or as you research petrochem stuff start creating a list/spreadsheet/flowchart of what items come from/are used in what processes.

Strategic Sage
Jan 22, 2017

And that's the way it is...
Retracing My Steps, Part II

Video


Checking in on the power situation, we're peaking at less than 350kw, or about a sixth of what our turbine can provide. So that's going fine at the moment.




Here's what the radar has revealed in the approximately hour and a half that it's been up. You can barely see it, but in the very west-southwest there are a couple of red pips indicating we're getting to where the biters are. Another stiratite vein in the south, and there's coal due east that is starting to be revealed. No sign of saphirite. When we expand again, I want to get that bobmonium patch just north of us, and probably go all the way over to the east where that coal is. Then we'll have lots of room inside the perimeter for setting up a minibus etc. That's actually more room than I want, but I need to expand some anyway and rather than build a second area for the coal and risk stuff in between getting destroyed by biter attacks I think I'll just enclose it all. As for our coal, we have 41k of the original 58k remaining. I'll want to keep a sharp eye on that as time passes.

For now, there would be a wait for us to get some tin and solder plate going. While waiting for that, I worked on some miscellaneous tasks:

** Got myself a steel axe
** Building more nanobot tubes, always needed in higher quantity than I expect.
** Built up a reserve of basics(copper wire, iron gear wheels, etc).
** Added a 5th coal mining drill to help fill up the longer belt for that.
** Researched circuit network

Essentially I just wanted to be ready to move forward as soon as the resources were available to do so.




I soon realized my setup had a flaw. Rubyte was making it's way further into the system than it was supposed to, because the furnace had a smaller internal buffer for it than I thought and within about ten minutes I already had a backlog of lead. So I just put this chest here to store any extra; I can let out a bit at time this way. It's just one more spot where I'll have to deal with re-routing the surplus at some point, but them's the breaks.




After about 25 minutes we had a starter supply of tin and solder. And also this, which is good news and bad news. The coal field at the extent of the radar's scanning to the east holds almost 2 million. It's also got biters by it. How many, we don't know, but the enemy is most definitely here.




Then it was time for this time-sink. Just like the first run, here I'm moving everything in the main storage chest to a new one, so I can put a larger steel chest in its place. The multiplying of the number of different items currently being used makes this necessary. It also gave the system more time to produce more tin while getting something useful done. I only needed two basic electronic boards for the fast inserter, the first two we've produced. There will be many more.




I also used some of the time to put up walls in the west and the south, the areas where I don't intend to expand further for a while at least. A little over 100 sections went up, so we now have a partial barrier in those directions. It was about a 6-minute task each way with the fast inserter, emptying the chest and then putting the stuff back into the new, steel replacement. Just seems longer than that because much of it is spent waiting for it to be finished so I can start making things again.




Next, with storage back in operation, the constant combinator goes up for the CCC(Central Control Center) because Dept. of Redundancy Department. Here once more I'll store values to control how much of everything I want to keep on hand, so that instead of running around turning inserters more and more often, the network will handle it for me. At first I only have enough wiring to hook it up to some nearby things; carbon, saphirite mining drills, and steel furnace. I want a few more but that's a start on making the factory work 'smart' and synchronize it's activities to current needs. While doing that, I also build heavy armor; I'd waited on that to get the fast inserter because of the fact that it takes a while to move all of the ingredients(100 Copper plates, 50 steel plates).




I'm sure you recognize this from the previous save; a starting squad of chaingunners. Ammunition is produced on the opposite side so it'll be slightly inconvenient to reload, but I shouldn't need them as much. That's partly why I only built four of them instead of the six last round. Once the radar gives us a clearer picture and some more things around here are attended to, it'll be time to go a hunting.




This is an upgrade to my logic for the iron plate production. Basically each of these decider combinators compares our supply of transport belts, magazines, and stored iron plates to the targets I've set. If we're short of any of them, the inserters feeding fuel into three of the drills turn on. They switch off if all three conditions are met. I need another one for steel to add to the system, and a similar setup to control the greenhouse for wood/resin/rubber, but I'm out of wiring again. Good news is we've got a decent supply of solder so I've shut that down. All the tin we get will come up to central storage now.




I've also got this all wired up, so that I'll keep a reasonable amount of stone and bricks in reserve, and also throw them on the main supply belt if central storage needs them. The conversion to 'smart production' is mostly of the way there. While I wait for the tin, ammunition, and belts to stockpile, I do a quick estimation and figure we're going to need about ten more turrets for the next perimeter expansion to encompass the bobmonium and coal patches. I'll build a dozen just to be safe. Going to need a least a thousand belt sections I think and we only have about 300 right now, so there's a lot of work to do.

There's always a lot of work to do. The last half-hour-plus was spent making most of the things we'll need for those two projects(quite a bit more quickly now that we've got a star fast inserter to match our star yellow assembling machine), as well as getting started on the most expensive research to date, an advance that will have benefit now while being essential not too far down the road. The steel combinator was laid down as well to complete the iron-smelting logic.

To accomplish the next update I have to get past a wall in terms of my understanding of combinator logic. It's the only part in the near future that has me stymied - I may well end up seeking help from the gurus either here, at the factorio forums, or both in order to wrap my head around it.

Strategic Sage
Jan 22, 2017

And that's the way it is...
Expansion

Video




Here's our lastest view of the surroundings. That orange stuff at the northern extremity of radar range is something called Thermal Water. I don't even wanna know at this point.




Gradually ramping up, but we're still under half a megawatt. The only concern here is having enough fuel.




The primacy of coal is clearly seen here. A lot of iron, but even more of the 'black gold'. Also, in the last hour I've produced 44 different items. That would have been on the high side for the vanilla 1.5 run during the mid-late game. That's what I'm making now, when I haven't even researched steam power, which vanilla starts with. Just ponder that fact for a bit.




Still ok for now, but soon it will get a bit convoluted cramming all the mining drills in. It seems 4k per hour is the standard usage right now. That's just over 8 hours worth. Given the necessary transition and priming time when setting up a new field(esp. one a long ways away), it's clear this needs to be dealt with soon or there's going to be major pain. I really don't want to build a bunch of greenhouses just to inefficiently fuel things with wood. The general rule of thumb that I found in the vanilla game is that anything lower than 10 hours supply is a significant issue that shouldn't be put off. By comparison, we have about 120 hours worth of saphirite, hundreds worth of stiratite, etc.

Resource-wise, coal is the whole ball-game right now.




75 red vials, which get processed a bit faster than I can produce them, bring us filter inserters. For the price of a fast inserter and a few more basic electronic boards we add the ability to choose what items to move.




Here's what I want to use it for immediately. The crushed stone will all bypass central storage from the iron direction now, and overall that's where the majority of it comes from. That'll mean less diversions to have the second assembling machine make regular stone, and more of building the stuff I want it to - I would add carbon to the filters a bit later. I also used to throw a few bits of saphirite ore that made their way up here some hours ago back on the belt to get gobbled up by the furnaces, just a small bit of maintenance to satisfy my OCD.




Then I decided to take the chaingunners for a stroll, across the desert, to clear out the area by the coal patch in the east. What happened here was hilarious overkill and a total waste of the coal they used to power themselves. There were two biters here. Not two spawners ... just two biters with no apparent home nearby. Rather amusing that, but they were eliminated and the path cleared. Back to base, fellas. It was time to start putting up the extended perimeter, with unexpectedly low resistance. That meant a lot of running and thankfully few termite bots due to the clear, sandy terrain. It was far too big to blueprint, and I had to do it in sections, returning to build more nanbots while the transport belts caught up.

It also soon became clear that I'd miscalculated the number of turrets I needed, by roughly half. Not sure why, but the why doesn't matter as much as fixing the problem. Further work on circuit network stuff was suspended until we could get that operational.




A little while longer, and that was accomplished. The entire thing has power, inserters, turrets -- and soon, ammunition. Controlling how much ammo to put on the supply belt is just a case of eyeballing things to start/stop and manually turning the inserter. I need a better way, but that's where things are for now. The focus is getting mining up at the new coal patch, now that it has basic protection.




Here's the completed line as I get my first few coal drills in place. It's so long now that there's going to be a gap and some problems back at the factory while the belt fills up - I'll turn off the saphirite and stiratite mining for a brief while to minimize the pain. Basic idea here though is that these drills add to the belt, it circles around the back of the resource patch so that I'm not crossing paths anywhere, and then the journey back west across the desert to the main production area.




An overview of the factory to see the general layout.

Next up, I do really need to deal with the thorny circuit network issue mentioned earlier. Both for ammunition for the turrets, and coal coming from the drills, the question is how to set things up to self-regulate the amount I have out on these massive supply belts. I have part of the answer, but not all of it yet. That'll be the focus of the next section as we proceed.

Alkydere
Jun 7, 2010
Capitol: A building or complex of buildings in which any legislature meets.
Capital: A city designated as a legislative seat by the government or some other authority, often the city in which the government is located; otherwise the most important city within a country or a subdivision of it.



Your dedication to using the belt overflow mod amazes and horrifies me.

Strategic Sage
Jan 22, 2017

And that's the way it is...
I hate to think what your reaction to the machinations in the next update will be …

Strategic Sage
Jan 22, 2017

And that's the way it is...
'Fun' with Combinators

Video





We've found a bunch more coal; that field in the northwest is 4 million strong. I'm no longer concerned about looking for more of it. I do think I've reached the point though where it's good to review all of our resources at the start of each update, to see how things are changing and what needs attention.

Resource Overview

** Saphirite(361k) -- This is now the top concern. We're using between 3 and 3.5k per hour, so this is still more than a hundred hours worth, at current levels. Which will increase by quite a bit once the minibus gets going. The concern though is that I haven't found more of it. I've found more of everything else, and a good distance away a small patch of less than 200k for saphirite ... which would just be a stopgap really, only half the size of the area we're currently mining.

** Stiratite(383k) -- Using about 600 or so per hour, so this has a really long time on it. Also, I've found three more fields within radar range, one just over a half-million and two about a million each. It's going to be a really long time before I need to be concerned about this.

** Rubyte(222k) -- Only trace amounts of this needed right now.

** Bobmonium(nothing) -- The field in the northwest of our base is well over 300k, so once I get mining set up there this will not be an issue at all.

** Coal(1.8M) -- We've got 24k more at the starting field, and it's going fast. Up to about 4.5k used/hour, which would mean our new patch will get us to about 400 hours from now. So until we ramp up production a LOT, this is fine too.

** Action Plan -- No short-term concerns here at all at this point, in terms of resources available.

Supply Belt Solution

For both the coal and ammunition belts, in order to know when they need more, I have to be able to tell how much is on them. That part wasn't a problem:




Here is a basic 'clock', in factorio terms. There's a T signal in the CCC set to 1. This decider will output whatever comes in, and then 1 will be added to it, so that every tick T increases by 1 -- until it reaches the value of Y. In this case, I've set that to 900. There are 60 ticks/second, so a little math tells us that this will run for 15 seconds, then reset and run again. That part was pretty simple to understand, and those who are really good at combinators use this concept a lot, in more advanced application of course.




Here I can read the amount of material on the belt. All I need for this is a red or green circuit-network wire hooked up to the belt. The belt itself is in 'pulse' mode, meaning each item is only counted once, when it enters this belt section. The 'hold' option counts things as long as they are on the belt, which is not so good for keeping a running count over time like I'm doing, but can be used for other things like if I wanted something to run only when the belt was empty or full, etc.




This arithmetic combinator increments the coal count every time more enters that section of belt, storing it in the Z signal. Again, not difficult. But this is where I ran into problems. What I need to do here is that when the clock resets to zero:

** Store the value of Z so I can do something with that information, and;
** Reset Z to zero, so it will start counting from the beginning.

I was able to solve the second problem with a moderate amount of brain-melting, after happening on a suggestion that hadn't occured to me before. Three more combinators were added:




This will output A as 1 when the timer resets to zero; at all other times A will be 0. This basically connects the clock/timer to the rest of the logic -- when A = 1, do a bunch of stuff related to the end of the timer cycle being reached.




If A is zero it'll stay zero, but on the reset(when A = 1) it becomes -1 here.




This sends 0 out if A is zero, but will send out the inverse of Z when the timer resets and A=1. That outputs to the combinator that is adding up all the coal, and zeroes out the count at the same time as the timer resets. I tried various other convoluted ways as doing this before, all of which ended badly. The idea of multiplying by -1, aka the inverse, didn't occur to me for whatever reason but once I happened upon it after far too many searches, I realized it was the solution I needed. So after testing that this worked, I then needed to find a way to take the coal count just before it resets to zero, and do something intelligent with it.




This here is a basic memory cell, storing the results of the most recent 15-second count. There are two keys to making this work:

** The output is tied right back into the input, which is how the 'memory' part works. It gets the one 'pulse' every time the timer resets from the combinator below it, the one that is adding all the coal up and then zeroing out when the timer resets.

** By itself, this will just keep getting larger and larger every 15 seconds. If there's 50 coal on the belt in that time, then it'll read 50, then 100, then 150 ... the number's going to keep growing larger and that's not what we want. So this outputs to an arithmetic combinator to multiply the value by it's inverse(-1) like I did before, which then sends it to another decider, which only pumps that inverse value back into the memory cell(to zero it out) when the timer resets; once every 15 seconds.

Not confused enough yet? Good, because we're not finished.




There's two more combinators on the left here. An arithmetic one(second from the left, third from the right) takes the output from the memory cell and subtracts whatever the target value is. For this, I want it half full, which would be 75 coal. Then it feeds into this decider shown here, which can finally output 'X', the 'final decision variable', to the mining drills. If the count is 75 or higher, then X this condition will come out false; say 80 for example; 80-75 = 5 input to this combinator, which isn't less than zero so X will be null(or 0). If it's lower, the condition will evaluate as true.




Instead of putting the condition on the inserters, it's better to keep the drills full of coal and just have them operate themselves. It runs if X=1(i.e., there are less than 75 coal coming through the belt-reader in the previous 15-second period). If there's more than that, it shuts off. In either case the timer cycle keeps running, so four times a minute the status will be re-evaluated.

I need a lobotomy. So this is the test setup. It requires, not including the CCC:

** 2 decider combinators for the timer; one for the timer itself and another one to broadcast the 'A' signal which tells everything else that the timer reset, triggering all of their various activities which need to be syncrronized at that time.

** 5 arithmetic and 4 decider combinators for the coal-counting and various resetting & logical steps associated with that.

** Bunch of red and green wire, naturally, to hook everything up.

Having determined that all this works(it's likely possible to do it with a couple combinators fewer, but it works and that's good enough for me right now), I then needed to put it in the right place. This was just the test setup, because I want these drills by the production area to run non-stop. The ones I want to be 'smart' are the ones at that big patch in the east. Which means a bunch more wire, and poles, to connect the network all the way over there. If I don't do it that way, then I'm reading a totally different section of the belt than the one they are going to empty their product onto, which doesn't do a whole lot of good. Reading the part that's several seconds away from reaching the drills will have them emptying, or not emptying as the case may be, onto the part that was just counted.




I also ran into this, which really threw my whole belt-assembler for a loop. A mod update, so now basic transport belts require iron plates and motors - it was iron plates and iron gear wheels before, which was easier. That means I now need another assembler in that process, which I don't have room for. It also means they are almost twice as expensive(5 Iron Plates per, instead of 3). I basically dealt with this by deciding not to deal with it, since I wasn't going to need any more belts in the immediate future and had almost 50 of them on hand. But yeah ... that would have made the perimeter expansion a lot more painful, and the next time I need to build any significant production line the belts are going to set me back a lot more.

For now though I had more the more pressing issue of getting the combinators going.




Sometime in between all of the building of combinators and stringing all the wire out to the eastern coal patch, the radar finished extending our territory as far as it could. And look at what it found, among other things -- the biters have expanded, establishing a nest not far to the southwest. And still no more saphirite. I shudder to think of how far we'll have to go to get that when we need it. Probably far enough for a train, before we have trains to go get it with. The fun part about that is we'll need a lot of iron in order to make the belts to ship it in with -- enough of that tangent. That can of worms will, and must, wait.




It took me a rather absurd amount of time to actually get this working, but here's the eastern coal-mining operation with everything hooked up. There were a couple times when I was getting strange results and couldn't understand why. Values that should be stable or increasing slowly would skyrocket into the millions(plus or minus) within seconds. I don't understand why, but in some cases just changing the wire color would cause information to be transmitted properly. It's like I got things stuck in some sort of strange programming loop otherwhise, even though seemingly identical setups wouldn't have that problem. I also spend a stupid-long period figuring out that the decider which announces that the timer is resetting (the one that puts out the A signal) needed to be out here, not back by the clock itself in the production area. The reason for that is because it can't hook directly into the other combinators, but rather was going into the network, which doesn't work when precise timing is needed since you lose a tick in comparison. So nothing was working and I'm ashamed to say I spent hours trying different things until I understood what the problem was.

Grrr. But this is now operational. The four drills out here, and I need to add more before long, are now intelligently hooked up. They're still running full-blast because the belt is between a quarter and a third full in most spots and the targeted amount is half-full, but they will slow down once it reaches that. I want this field to be used only when needed to reach that point -- the drills back by the fading patch at the production area will run full-blast to clear that out first.




Here in a closer view, you can see the lone red wire connecting a couple of the combinators. If I make that one wire green, things quite literally go haywire, and I don't why -- but since this works, I at least know what to try if I see similar behavior in the future. Gradually I'm getting more comfortable and learning how these things work. I also like the little boxes on the mining drills where the wires are hooked up; the small green light there turns red when they switch off. Nice little detail.

Next major goal was setting up a similar operation for the ammo-supply belt. Except that we didn't have enough tin or solder to do that yet, as I'd used it all for the materials to get the coal-mining combinators going. So what else needs to be done? Welp, I had only 3 nanobot cylinders left, so I built another 30-some of those.




Then I put up this moderately spaghettified thing, which just makes the iron plates loop around to west so they are basically in the shape of an 'L', more or less. A couple poles, another inserter or two, and another assembler was all I needed, it was more figuring out the space issue - and this was the best way I saw to limit the amount of stuff that needed to be moved. Now our basic transport belts can get produced again. Let the iron-chewing resume!!




Seemed a good moment to take out that one closer nest the radar had spotted. The Chaingunner Squad actually got a real task, although not much of one as only a handful of biters emerged from it and it was quickly taken down.




Interesting thing here, as sending them past the wall has them trying to go through it at first. They think for a moment ...




And eventually figure out the correct way, heading east around the wall, just a few seconds later. Good job fellas. Long as I don't make them go through forests, they really do quite well. And there are no apparent forests round these parts. I think it's an interesting part of the AAI implementation -- you can just imagine the relatively basic logic in the navigational computing in each of the chaingunners whirring away, pondering ... and then figuring out it has to go east around the wall. .




The coal drills in the east have caught up. The yellow lights(love that Bottleneck mod) indicate they are not operational. Much of the belt is now just above the half-full point. Plenty of fuel for everything to grab something as soon as it needs it without waiting, no danger of overflows anywhere. Almost makes it worth the brain cells I burned getting it to work.




There's always more make-work tasks to do, it's just a question of what the best thing is to do next. Here I put down about 300 more wall sections, which gets that project up to approaching halfway done. Call it 40% I'd say. That used up all the nanobot tubes and the majority of the wall material, and took just enough time that I had the required tin I'd been waiting on.




With the new combinator logic up(another 10 of them required), this inserter is now given instructions on the supply of ammunition. I don't want that belt nearly as full; one per second is enough. Things are kind of screwy right now as parts of it have double that amount, parts almost nothing, and so far the turrets haven't had to shoot at all so it's really not going to correct itself much. But at least I can fill in the empty spots in case hostilities do commence.

After putting up the rest of the finished wall sections, I think we need 350-400 more. There are still a few more things to be done before I can start work on the minibus proper, but there's definitely a light at the end of the tunnel now.

IceMole
Aug 1, 2009
Each wire shares all of the signals it's carrying with each same color wire it's touching. It's a little hard to see the connections in a screenshot, but it looks like that wire would get the output of all three arithmetic combinators on the left rather if it was green than just the one the red wire is connected to.

Veloxyll
May 3, 2011

Fuck you say?!

Combinators :psyduck:

And this is why I will never run something as crazy as the belt overflow mod.

It would break my fragile brain.

Strategic Sage
Jan 22, 2017

And that's the way it is...

IceMole posted:

Each wire shares all of the signals it's carrying with each same color wire it's touching.

I do know this, but one of the arithmetic combinators doesn't have an output of the particular signal in question. I also didn't see any way the value should escalate to the millions within seconds. No more than several hundred should have been possible in the given timeframe unless I'm missing something(obviously was). It just sticks in my craw because one of the best pieces of advice I read in my circuit-network-searching was to make sure you fully understand what's going on, because any template is going to go wrong somewhere -- one wrong operator, wire, or signal is all it takes to completely screw it up. Can't troubleshoot it if you don't thoroughly grasp what's going on. But I'm getting better … slowly.

IceMole
Aug 1, 2009
I'm pretty curious now. Any chance I could take a look at the save?

Fat Samurai
Feb 16, 2011

To go quickly is foolish. To go slowly is prudent. Not to go; that is wisdom.
Holy poo poo :psyduck:

Strategic Sage
Jan 22, 2017

And that's the way it is...
But Wait, There's More ... Combinators!

Video





The next step was another extension of the coal line to reach here. This is the bobmonium patch in the northwest of my base. For now, it will just send the ore south to add to the furnace operation in the southwest. I need more tin to do just about everything I need to do at the moment, and this should satisfy that need.




Not much different here.





Look at the top line of the graph(the blue one). That's coal, and you can see that for about the past 45 minutes it's been remarkably steady. That's about to change as it fills in the part of the belt that got extended for the bobmonium-mining operation, but it still shows the general equilibrium of the system.

Resource Overview

** Saphirite(356k, -5k) -- Had a bit of a stall here as I needed to move a couple of drills and didn't notice it right away. Still good for now, but we're probably down to less than 100 hundred hours worth with no additional source in sight.

** Stiratite(382k, -1k) - Didn't need to use much of this.

** Rubyte(196k, ??) - I reported this as 222k last time, which has to have been a mistake. In any case, we need to use very little.

** Bobmonium(332k, new) - As we just set this up, I'll need to get a handle on how much we are using. Happy with it for now though.

** Coal(14k, -10k; 1.8M, --) - We're in the final stages of clearing out the starting patch, and I'll move those drills out east when that happens.

Action Plan

We need to expand radar operations in order to find more saphirite. And also to have better coverage around our walls so we can see any nearby biters. But there's something I want to do first before making that jump from one to several radar stations, and the energy drain that will go with it.




We need to research the initial Steam Power for 50 red vials(sorry, 'sacrificial electromagnets'), but since we can do this one as well I'm going to get the first upgrade at the same time. Our current turbine gives 40.5% efficiency; entry-level steam power gives 50%; and this one gives 60% along with doubling the maximum power output per steam engine. Eventually we can get that all the way up to 90% and several times more capacity, but this is as far as we can go at the red-science level. The reason I haven't done this before is the cost - these things are not cheap and I'll get into that when I build them. We're talking about slashing the coal usage for energy production by half though, so I think it's well worth it. There's a lot of coal out there available ... but why not conserve it if we can?

That research then, was the next item on the agenda. As the tin started coming in more quickly, it became clear that the bottleneck was going to be iron again. I crammed one more furnace(five now) and drill(four) in, but I don't think there's a useful way to do more than that without a total rethinking of the setup. I think we'll suffer through for the time being.




Finishing up some basic circuit-network stuff was next. I'd been fiddling with a problem in the resin/rubber setup ever since I put it up really, and after devoting a bit of time to it I reversed it, putting the resin assembler after the rubber furnace in the production flow instead of before it. Which really just meant I got the chest in the middle clogged with rubber instead of resin, and didn't actually accomplish anything. Finally I just decided to heck with it, and put this additional belt out the 'back' or top of the rubber furnace to keep that mixing from happening. I'm sure there is a more elegant way of doing it but I reached the point of 'screw it, this works' for what is, again, a temporary and small element of the factory.

I also have the greenhouse hooked up; it won't stop producing until both the buffer chest here and the one at the main assembly area have roughly one full stack. After adding a simple one for the stiratite drill as well - which really is sort of superfluous since that pretty much runs non-stop and barely produces enough to supply our copper needs, it was time to pay some attention to the whole lead/tin/solder area.




The main problem here, and the basic reason that I didn't do anything with this before, is that I can't hook up the circuit network to a furnace. If I could, I'd just have a basic 'don't send any more of this item if there are plates waiting in the furnace' kind of deal. It does make sense of course since there are no electronics involved in the furnace, presumably they work on some sort of elaborate gearing or pressure-plate gizmology internally.




This 'fix' took me a bit to figure out and rig up. But I'm darn pleased with it, if I do say so myself. So, what's going on here?

** Buffer-chest method at the furnaces, so I can hook up the circuit network to them and read their contents. If we get enough stored, the rubyte or bobmonium drills, as appropriate, will shut down.

** The solder(middle) furnace is controlled by the output inserter; if we have enough in storage, it won't output more which stops the system there.

** Added another chest by the rubyte crusher because we need to empty all that stuff out. The filter inserter moves the crushed stone on to the second chest while leaving the ore where it is. Eventually I'll get everything out of there, but that's going to take a LONG time with almost 2k rubyte stored up. A lot longer than I plan on having this whole thing in place.

** Two pairs of decider combinators. The ones in the east, north of the furnaces, check for tin and solder storage. If we have enough of both, no more bobmonium.

** The second pair, to the west of the furnaces, checks lead and the belt-readers. This tells the inserter emptying the rubyte chest when it's clear to empty more. Otherwhise it would overwhelm the capacity of the lead-smelting furnace. This is a good example of using the belt-readers in 'Hold' instead of 'Pulse' mode, so that they continue to count anything that's on the belt, not just when it enters that section.

** Same two belt readers also send their data to the inserter emptying the crushed-stone chest. Like the other one, it will only empty more if there's no stone on either consecutive belt segment. That makes the flow steady, but not so much that it will overwhelm the operation at central storage.

There is yet one thing left to do ...




This has just become a tangled mess of wires. Ok, an even worse tangled mess of wires before. Wire spaghetti > Belt Spaghetti fo sho. But basically, each of three decider combinators that you can probably barely see through the haze is responsible for a specific item(iron, steel, carbon) of the three that I want the filter inserter taking in. When storage needs those items, that filter will activate ... until I don't need them anymore. Then it will shut off.




Here's the iron one just for show; if we have more than the desired amount, it outputs nothing. Otherwhise it outputs iron at 1, which tells the filter inserter to activate the iron filter. I didn't realize until I did the video recording on this that only the iron one here is even necessary. There's no reason to not have steel and carbon filters on all the time, since nothing else is using those products(unlike iron). But whatever, it was worth figuring out and rigging up for the exercise of doing it.

And now the entire factory is finally running on a smart, or self-regulating basis. Other than correcting errors as they may come up, the only thing I need to do is keep an eye on the mining drills and relocate them when they run dry. Otherwhise I can focus pretty much everything on building whatever I need for the next project.

So, about that whole steam power thing ... Most of the infrastructure could be quickly made or was already built. I'd need a lot of transport belts because I want to have a separate coal supply for the power as I did in the previous vanilla run, to immunize it against shortfalls as best I can. I would need a couple of things that take longer though:

** MK2 Boiler -- 16 Iron Pipe and 4 Stone Furnaces for the standard one, then an additional 16 Steel Pipe and 20 Steel Plates. That's not too bad really.

** 2 MK2 Steam Engines -- 24 Electric Motors, 20 Iron Gear Wheels, 80 Iron Plates for the standard. Then we add 40 Steel Pipes, 20 Steel Gear Wheels, 40 Steel Bearings, and 40 Steel Plates for the upgraded versions.

Well then. That'll take a bit of doing. It seemed prudent to boost our supply of steel, particuarly since the ammunition line was caught up after spending almost two hours to get there, leaving us with limited need for iron. A brief interlude for putting up another 250 wall sections got that mostly finished but not quite completely. Naturally I'm now questioning whether we have enough room in the base, which means we almost certainly don't. I got the boiler built, but had barely enough steel for that -- not nearly what we'll need for the steam engines. Which is really fine, because there are nanobot tubes to make, and iron gear wheels, and electric motors, and the copper cable for the electric motors, and ... Still, I switched over a second furnace to steel, probably later than I should have, in order to speed things along while I did all of that.

Proper steam power is coming soon, but it's not ready just yet.

Alkydere
Jun 7, 2010
Capitol: A building or complex of buildings in which any legislature meets.
Capital: A city designated as a legislative seat by the government or some other authority, often the city in which the government is located; otherwise the most important city within a country or a subdivision of it.



Screams internally

...and externally


This is gonna be one of those LPs that takes 2-3 years at this rate if you're just now starting to setup the infrastructure to mass produce proper level 1 electronics. :psyduck: This belt overflow mod is pure madness.

Strategic Sage
Jan 22, 2017

And that's the way it is...
Famous last words, but I'm expecting and hoping to be able to have more regular updates most of the time at this point. But aside from that; yeah, pretty much.

Sage Grimm
Feb 18, 2013

Let's go explorin' little dude!
Your efforts in Bob's Mods had inspired me to start my own trek into the blackhole of modded Factorio with a similar mod set, barring belt overflow for obvious reasons. It's oddly compelling to be fiddling around pre-green science and still have things to experiment with.

Veloxyll
May 3, 2011

Fuck you say?!

Alkydere posted:

Screams internally

...and externally


This is gonna be one of those LPs that takes 2-3 years at this rate if you're just now starting to setup the infrastructure to mass produce proper level 1 electronics. :psyduck: This belt overflow mod is pure madness.

He'll be able to use blueprints to avoid some wiring solutions. Just has to change the detected resource.

Of course, he also then has to consider circuitry in his "Why is this running short" thinking.

Faylone
Feb 18, 2012
The thing that hurts most about the belt overflow mod is that the devs had decided belts were ALREADY NOT USEFUL ENOUGH and upgraded splitters in .16 so that it would be worth using them more. The idea of "let's nerf them even harder than they were before" is almost obscene.

Strategic Sage
Jan 22, 2017

And that's the way it is...
The thing about that is, I respectfully disagree with the devs on that issue. Strongly. I don't think the splitters particularly needed to be buffed -- they just didn't want everyone using bots and only bots in the late-game. Which a lot of them are still going to do. So … Personally I don't see anything wrong with belts gradually getting phased out of a factory in favor of bots as you progress later into things.

Also, of course I'm obscene. That's literally half the point of many things I do, as far as gaming is concerned :P

Sage Grimm posted:

It's oddly compelling to be fiddling around pre-green science and still have things to experiment with.

Here's hoping you don't finish your run before I even know what half the resources in the early game are for and crawl my way up to green science.

Veloxyll
May 3, 2011

Fuck you say?!

Thotimx posted:

The thing about that is, I respectfully disagree with the devs on that issue. Strongly. I don't think the splitters particularly needed to be buffed -- they just didn't want everyone using bots and only bots in the late-game. Which a lot of them are still going to do. So … Personally I don't see anything wrong with belts gradually getting phased out of a factory in favor of bots as you progress later into things.

Also, of course I'm obscene. That's literally half the point of many things I do, as far as gaming is concerned :P


Here's hoping you don't finish your run before I even know what half the resources in the early game are for and crawl my way up to green science.

One of the problems is that bots are like 5x times better than belts. And it's just not as interesting to have a sea of bots over your base instead of belt spaghetti.
Since the game is basically belt spaghetti simulator.

If all I have to do is build PPCs and Requesters for my machines it loses a bit of its charm.

I usually have a mix in my base come late game. Hmm. I need to go scare up the revised modlist you're using and start a game. And make a horror of trainspaghetti.

Jabor
Jul 16, 2010

#1 Loser at SpaceChem

Thotimx posted:

So … Personally I don't see anything wrong with belts gradually getting phased out of a factory in favor of bots as you progress later into things.

The reason it's terribad from a game design perspective is that it removes literally all the logistical challenge from the game, and all that's left is an endless sea of beaconed assemblers hooked up to logistics chests.

In order to really scratch the engineering itch, putting time and thought into coming up with a clever solution should result in something that's better (in at least one dimension) than the no-effort obvious solution. If the no-effort obvious solution is better in literally every way than any other solution, it's all kind of pointless.

Strategic Sage
Jan 22, 2017

And that's the way it is...
I think that's true if that's happening in the base game, but I don't think from what I've seen that it really is. I.e., by the time you've got nothing but bots everywhere, you're basically into the post-game spot and working your way towards a megabase, beyond the scope of what really applies as good design for the 'crashlanding-to-rocket launch' phase. Maybe there's more than I think but I haven't seen a bunch of vanilla no-belt, bot-only bases that arise before the rocket launch. Not really trying to recreate the bots vs. belts thunderflaming here that happened at the official forums of course … that was a sight to behold. And not a good one. .

Veloxyll
May 3, 2011

Fuck you say?!

Everything is terrible: The Internet.txt

Also fooey, couldn't find SpaceX and Space Satellites. And now I'm suffering mod fatigue so a new game will have to wait till tomorrow.
(Also didn't go with Rampart or Beltmod)

Strategic Sage
Jan 22, 2017

And that's the way it is...
Steam Power

Video


So it turns out that I overlooked one thing: MK2 Steam Engines have five ingredients. I can only make four in the yellow assembler. And as you might guess, the next tier in automation requires green science, among many other things. So that's not going to happen ... but I can still use the MK2 boilers. It's not a total loss, and that's where the efficiency gain comes from which is what I was most interested in.




Held pretty steady at just over 400kw, or a fifth of the turbine's capacity, with one modest dip. No concerns here.




Dropped to almost nothing for a bit while I was working on the solder-furnace area and all that circuit network fun joy. Aside from that, our top items - coal and saphirite - saw a modest increase. Stone probably won't stay that high after we work through the rubyte-crusher's backlog.

Resource Overview

** Saphirite(347k, -9k) - 77 hours to go at that pace. Which will accelerate.

** Stiratite(382k, -1k) - Still just a relative trickle.

** Rubyte(195k, -1k) - Plenty.

** Bobmonium(331k, -1k) - Nothing to worry about here either.

** Coal(5.3k, -8.7k; 1.8M, --) - Main thing here is just finishing off the nearby patch.

Resource Action Plan

Thotimx posted:

We need to expand radar operations in order to find more saphirite. And also to have better coverage around our walls so we can see any nearby biters.

Yep, still need to do that. Said this at the beginning of the previous update, and proceeded to diddly squat about it. I had good reasons, but I'd be well advised to not take any longer before doing this than necessary.




So here's basically how I want this to work. The steam plant will be in the northwest of the factory. Leaving plenty of room for greenhouse expansion if I need it to the south, and eventually as we grow this can grow as well to the north. Bobmonium patch is out that way, but we can fit in around that if need be. Anyway, offshore pump brings in the water which runs along the south end here. Boilers in the east, two steam engines for each one of those extending back to the west, and then I can duplicate that moving north as much as I need to. The coal supply isn't hooked up at all, but I'll be doing that next. I've got a buffer chest there because I want to maintain a reserve, and try out the alerts system to let me know if it gets too low. Not a fan of the idea of running out of power, particularly with this modpack. VeryBadThings(tm) will happen due to a cascade/domino effect, so I think it's worth some extra front-end effort to make sure that doesn't happen.




So now I need to get the coal to get this up and running. This is the north end of the east coal patch, or actually just past it; there's a small smidge here with a few thousand units. I've got the main factory drills on the south end, so they'll both work towards the middle. Long-term I want totally separate patches for each and all that but this will work fine for quite a while. I've built a temporary line from the existing coal loop to prime these machines, and a burner inserter to get the initial bits on. That won't be needed for long.

Several hundred basic transport belt sections and several dozen nanobot tubes saved up proved to be insufficient for the task of getting the new coal belt loop finished, so I had to wait for the iron to get more. Meanwhile we'd built up enough rubber and resin for the time being so the greenhouse was almost done with it's work, and I did some just-because stuff like getting stone tablets and wood squares put together. Even with that, there was still some while while there was nothing to do but hurry up and wait. Anything else useful I could do consumed more iron, which just meant the belts taking longer.




I activate the system having spent, going back to the last update, about an hour and a half working up to it. It's not finished, but the basics are in place. Here, the loop enters from the west, adds whatever these drills produce, and exits to the east.




This steam-supplying coal line heads along the north end of the base; I diverted it that way so it wouldn't be in the way when I want to build other things. Overall, the project required over 100 nanobot tubes, approximately a thousand belt sections, and pushing 6k iron -- which came from four burner mining drills. That rather emphatically emphasizes the point that we need to expand those operations.




Here's our latest pollution cloud, which overall should go down a bit once this is running. The spread of it hasn't increased a whole lot, but you can see we definitely have some hazardous conditions near the factory.

So now I want to complete the steam power-plant system while I wait for coal to percolate through. That means two things:

** Another set of combinators to read how much coal is on the belt and turn the drills on/off.
** I also want to try out the custom alerts thingy, which I've never messed with, so I'll be warned when the buffer chest over by the steam engine themselves is getting low. I really want to do everything I can to insulate against the possibility of blackouts.




Re-introducing the Programmable Speaker! This lovely toy I mentioned briefly in the vanilla game and then never seriously considered using it.

** Global Playback controls whether I can hear what it's doing anywhere on the map or just when nearby the speaker. I definitely want to know no matter where I am.

** Show Alert has it not just play a sound but also display something visually in the GUI for me to see. Under the Alert Settings at the bottom I tell it what do do there -- what icon and tooltip message to display.

** Polyphony is useful for playing music, it allows multiple sounds to play at once.

** Circuit Connection Settings in the middle tells it when to activate and what sound to use. Here, I've got this hooked up to the chest, so when the coal is less than the Black signal(set to 50 currently), I'll hear a Siren and get a visual as you can see here. You can also do a more generic alarm, or a ringing tone, or various musical instruments(strings, piano, drums, etc.) There's a miscellaneous category for using some of the game's sounds, like what it does when research is completed or an achievement is unlocked, etc.

Now I have an added motivation to get enough coal coming to make it shut up, since until I do that the siren will not stop. It'll just wail, and wail, and wail. Most people who know me are stupefied by my ability to block out sounds that are generally considering aggravating, but this is fairly annoying to even me. That's pretty much a necessity, since it needs to be something that I'll notice or else it's not much of an alert.




Right now it's only coming sporadically, enough to barely keep the steam engines trickling out a bit of power, and have the boiler flash the 'out of fuel' icon regularly. But it's working doggone it! It's progress!! To really get this rolling, I need more drills, more inserters, more combinators, and more nanobot tubes to put it all in place. Back to the assembler I go.




I'm not sure if the grid is trying to draw equally from both the turbine and the steam engines proper, or if it's just using the turbine for what the steam engines can't provide. I suspect the former by the way the bars look here. Either way we are already getting an advantage from this; the less the turbine has to do, the more efficient our coal usage is going to be. Clearly there is much work to do before I can take it down entirely.




By the time I had the necessary materials together, the existing operation was already making progress. Each steam engine was almost up to 20% of it's capacity in stored steam, and that amount was slowly increasing. Still not enough coal to keep the boiler satisfied, and over half our power is still coming from the turbine. Once again I make the run across the desert to the coal-mining area ...




Where I discovered that three of the initial five drills, placed as they were on rather spare supplies, were already exhausted. That means our progress isn't lasting long if I didn't do something.




This re-routes the line and adds more, with proper spacing as it doubles back so that I can keep moving drills along and expand to quite a bit beyond the current amount. Soon I had eight drills running, though most won't last long in their present locations. Then I just needed to painstakingly recreate the combinator setup I'd used twice(the other coal line and the ammo belt), and everything was set up.




The siren shut off just as I was starting to wire up the mines and set conditions on them. Heading back to the production area, I took down the steam turbine for good. It has served us well, but that time of service is now over. Steam power will take over the energy duties for the forseeable future. Even with the engines taking the full load, the siren still remained silent. In contrast to the vanilla experience of having it available from the start of the game, there is a real sense of accomplishment here in getting steam power fully up and running.




The sound of silence was a beautiful thing. I could not be satisfied though, because a single burner inserter was just fast enough to keep up with the boiler, despite only using a fraction of the potential that the two engines could provide. It looks like, for now, we'll need to have a boiler for every engine. I added two more buffer chests, one on top and one on bottom, of the existing one. That should provide plenty of throughput for a full steam engine, and we'll need to add more with another boiler if additional power is required. I could always go with fast inserters, but that would defeat the goal of keeping the feeding operation entirely independent of the power grid. It does still rely on it to power the combinators, but all that produces over here is the alert.

For a significant period of the steam-power setup, the factory has been churning out steel to stock up on that. We're now ready to get more radars going, after which there'll be time to move on to bigger things - the minibus beckons. Eventually.

Strategic Sage
Jan 22, 2017

And that's the way it is...

Veloxyll posted:

couldn't find SpaceX and Space Satellites. And now I'm suffering mod fatigue so a new game will have to wait till tomorrow.

Assuming you are using the in-game mod downloader thingy, they are listed as Space Extension Mod and Satellite Radar.

Olesh
Aug 4, 2008

Why did the circus close?

A long, chilling list of animal rights violations.
Bots aren't required, sure. Your basic logistics robots in vanilla only require red and green science to research, but you can't properly convert your factory from belts to bots until you've researched Logistic System, which gives you the rest of the logistic chests that make bots work industrially. It also requires purple and yellow science, meaning if you can research it you could also be researching (and building) the Rocket Silo to win the game.

On the other hand, bots make handling logistics so much simpler and easier that there's very little practical reason to expand the belt system ever again once Logistic System is unlocked, which is a shame because optimizing and spaghettifying belts is a huge part of what gives the game its appeal.

Strategic Sage
Jan 22, 2017

And that's the way it is...
I've come to the conclusion that I need to put this LP on indefinite hiatus. Short version is that my graphics card(4-year old GTX 960) just isn't up to the task. It works fine for most things, but I'm already starting to see significant framerate issues. I've tried everything I can think of, and can't get rid of them without lowering the graphics quality to what I consider to be an unacceptably low level for a Let's Play. I think it's a combination of using a fairly mod-heavy setup and the fact that Factorio's assets have become more detailed(and thus more graphics-intensive) over the last couple of versions. Much later vanilla saves have virtually no issues.

I apologize sincerely to all those who took the time to read, watch, and participate. Thing is, if I'm having these problems NOW, by the time I get even halfway through the game the thing would be groaning horribly. I hope to someday return to this or do a similar 'go-nuts' playthrough, but that would require a considerable hardware upgrade that just isn't going to happen anytime soon. My other LPs will continue, and I have many other ideas for stuff I want to do so you'll probably see a new one from me to start fairly soon. I'm going to chalk this up as a learning experience in making sure I do enough research before beginning a new project, as I hate abandoning something in the middle. At this point though, I'd just be postponing the inevitable if I didn't.

Thanks again to all who participated in this.

Veloxyll
May 3, 2011

Fuck you say?!

Doh. I was going to throw together a train madness update when I got to that tech level in my playthrough.

Because I am pretty sure the madness of sorting all the train inputs is going to be a nightmare all on its own. I've seen the smelting flowcharts and down that path lie only horrors.

hoonigan_neil
Feb 25, 2014

Thanks for taking us as far down the rabbit hole as you did. Would love an update of any capacity, should technical hurtles ever be overcome.

Adbot
ADBOT LOVES YOU

SettingSun
Aug 10, 2013

This LP inspired me to pick up and learn the game proper so it will be missed.

  • Locked thread