To Factory or not to Factory?
One of the critical parts of Minecraft modpack strategy is trying to get a stable supply of the hundreds of items in the game, and figuring out which items are worth setting up factories for. With the many variants of ores and woods, and all the alternate recipes for some items, this has turned into a real mess. Even with the nice NEI interface, we waste so much time looking up recipes ... only to find out that about half the time we're missing one of the ingredients. So, we leave the recipe incomplete on some autocrafting table and run off to find that last ingredient, only to get distracted by something else.
I used to make little "automats" to make all the key items. Each automat is just a chest, an autocrafting table, 2 pipes and a redstone engine: the items craft and go right back into the source chest (to save resources). One player picked on me, was too impatient to wait for stuff, and asked why I don't just craft the items by hand. If it's a simple recipe, sure, craft by hand. But it gets so tiring splitting some stack into 6 slots, grabbing some other stack and splitting it into 3 slots, then crafting, then taking all the unused junk back off the table. I'd rather set up the recipe once, throw stacks of items in the chest, and come back in a few minutes to get as many items as could be crafted from the piles provided.
But every engine, every pipe, adds some tiny bit of lag. Approaching one of those massive factories, the game just slows to a crawl while hundreds of blocks try to update 20 times per second.
After I while, I went more minimalist, and just started building what I call "blueprint rooms". Rows of autocrafting tables in groups of 4, with a sign for each group saying what recipes were on those 4 tables. Always organized in sensible groups, like the basic IC2 machines, generators, etc. It has an interesting feel to it. It's like dismantling the first copy of each item and keeping its blueprint on file. What's nice is that I can go and stock up all the slots of the most common materials. If a machine needs stone or cobble, I can fill those slots, then when I need the actual item I usually only have to add one or two items and take out the final product. A bit of a nuisance for things like the extractor which use unstackable treetaps, but still handy. And those autocrafters have no lag. More recently in FTB Ultimate, I started a few rooms whose walls and floors are almost entirely autocrafters ... feels neat walking in there, clicking some block, adding one or two items, and taking out a machine whose recipe I can never remember. Way faster than searcing in NEI and clicking all those slots every time. When there are alternate recipes (like for electronic circuits), I pick the one with the best output using the materials I have a good supply of, and usually have the cheapest recipe as well.
Some recipes are just one step, simple. But others are processes, like advanced solar panels. One trick with the blueprints is that for a multi-step process, you actually have to sacrifice multiple copies of the final product to setup all the recipes. Consider a 4-step process: you'd need recipes for step 1,2,3,4 side by side, but you can't setup the recipe for step 2 without first setting up the recipe for step 1 AND crafting a step 1 product for your step 2 recipe; in fact you have to craft step 1+2 to setup step 3, and step 1+2+3 to setup that step 4. These are the ones that call out for a factory, BUT odds are you will never need so many of them that a factory would be worthwhile! And you'd still have to sacrifice so many materials to set up the autocrafting recipes, whether you had engines attached or not.
Now, for single-ingredient recipes, I wrote a Lua program for a crafting turtle where I can put some stack of items in the first slot, and it arranges it in whatever common pattern I tell it to, and crafts as many as it can. That's nice, but typing some command -- in this case "makeme chest" -- is probably not optimal. Faster than an automat, and fun to watch, though. For a few special items, I keep a crafty turtle on top of my compressor, so for making coal balls or carbon plate or compressed plants (the compressor doesn't recognize many leaf types) I run a program that compresses the pile and drops it right into the compressor below. That's handy.
The real comedy comes in planning supply lines. I do have some autocrafting chains off my main tunnel, where I can throw surplus sand and have it fill barrels with all 3 kinds of sandstone ... only to find out that I never actually use those. Another autocrafter group turns stone (from my igneous extruder) into stone brick, stone slabs and chiseled stone brick, which I DO use regularly. Yes, I could have gone further and turned half the stone brick into slabs and stairs, but I'd rather make those on demand. I almost never use those. I can't know today which items I may need for a project next week. Even with a written TO DO list in a text file. If I turn all my sand to glass or logs to planks, oops, when I need sand or logs for a future project, I wasted it all on something I didn't need. Some things I try to keep a half a stack of, like advanced alloy and carbon plate. When that runs low, I throw a few more in the compressor. But you know how it goes, when you finally decide to make a treehouse and you need 200 planks, logs and fences, and you go to your vault only to find you turned all your tree farm output into chests and ladders, doh!
It's actually a very complex problem. I try to think of it in terms of "industries". An industry is a subset of related materials like ore processing or sand/glass or bees, which share some small group of machines. I keep each industry in a separate work area -- usually one industry per floor of my buildings -- partly for sanity and partly because it makes a town feel a lot more realistic. You'd never really need a bakery and a rubber extraction plant at the same time. But the way all the industries intertwine is tricky. If I send half of my glass to make glass panes (building industry), then switch industries to potion-making and find I'm out of glass bottles, oops; so I make a big batch of glass bottles then find a new oil well the next day and need glass tanks instead, oops ... but if I make 50 glass tanks and 100 glass bottles and never use them, that's a dead end.
I also try to keep some mods separate, so the whole flavor changes as I go into certain buildings or caves. For a few I try to locate the mod's workplace in a suitable environment. I always liked having my magical stuff in wizard towers well away from my main town, so there was a sort of pilgrimmage involved; then with Thaumcraft (which I still haven't started learning!) and its flux and will-o-wisps it turns out there's a good reason to stay away from home, to avoid corrupting your whole city. But even that's not so easy anymore. When I first heard of Thermal Expansion, I set up a little "Thermal Expansion Lab" and went to work figuring out those machines ... but now I like them better than the IC2 machines, so they've become the heart of many of my production lines. Railcraft used to be in a vault of its own, but now creosote can be burned as fuel in a (GregTech) generator, steel is used in some machines, coal coke is all-around handy fuel, and GregTech shifted some IC2 recipes to use the Railcraft Rolling Machines! Even funnier: it turns out I hardly ever use rail, now that I can make MystCraft linking books to get from place to place, now that the old Bukkit "warp" command is unavailable.
Wow. That takes some strategy. I have text files of recipes, one file for each major mod. I don't run the game full-screen, I can always see about 6-7 lines of my notes below it. I don't know any other game where I've had to take 100+ pages of notes. But for some reason, Minecraft keeps it interesting. Well, not Minecraft itself. The vanilla game is as dull as a rock. Did they really just add ugly hoppers to move stuff from place to place? Bleah. It's the ingenuity of the mod community, then the oversight of the modpack managers that make all the mods work together ... THAT is what makes this fun. I see when I log in that a bunch of mods have been updated, so when FTB Ultimate updates, there will be new stuff to unlock. And I have still not made a tesseract or a thaum golem ...
I used to make little "automats" to make all the key items. Each automat is just a chest, an autocrafting table, 2 pipes and a redstone engine: the items craft and go right back into the source chest (to save resources). One player picked on me, was too impatient to wait for stuff, and asked why I don't just craft the items by hand. If it's a simple recipe, sure, craft by hand. But it gets so tiring splitting some stack into 6 slots, grabbing some other stack and splitting it into 3 slots, then crafting, then taking all the unused junk back off the table. I'd rather set up the recipe once, throw stacks of items in the chest, and come back in a few minutes to get as many items as could be crafted from the piles provided.
But every engine, every pipe, adds some tiny bit of lag. Approaching one of those massive factories, the game just slows to a crawl while hundreds of blocks try to update 20 times per second.
After I while, I went more minimalist, and just started building what I call "blueprint rooms". Rows of autocrafting tables in groups of 4, with a sign for each group saying what recipes were on those 4 tables. Always organized in sensible groups, like the basic IC2 machines, generators, etc. It has an interesting feel to it. It's like dismantling the first copy of each item and keeping its blueprint on file. What's nice is that I can go and stock up all the slots of the most common materials. If a machine needs stone or cobble, I can fill those slots, then when I need the actual item I usually only have to add one or two items and take out the final product. A bit of a nuisance for things like the extractor which use unstackable treetaps, but still handy. And those autocrafters have no lag. More recently in FTB Ultimate, I started a few rooms whose walls and floors are almost entirely autocrafters ... feels neat walking in there, clicking some block, adding one or two items, and taking out a machine whose recipe I can never remember. Way faster than searcing in NEI and clicking all those slots every time. When there are alternate recipes (like for electronic circuits), I pick the one with the best output using the materials I have a good supply of, and usually have the cheapest recipe as well.
Some recipes are just one step, simple. But others are processes, like advanced solar panels. One trick with the blueprints is that for a multi-step process, you actually have to sacrifice multiple copies of the final product to setup all the recipes. Consider a 4-step process: you'd need recipes for step 1,2,3,4 side by side, but you can't setup the recipe for step 2 without first setting up the recipe for step 1 AND crafting a step 1 product for your step 2 recipe; in fact you have to craft step 1+2 to setup step 3, and step 1+2+3 to setup that step 4. These are the ones that call out for a factory, BUT odds are you will never need so many of them that a factory would be worthwhile! And you'd still have to sacrifice so many materials to set up the autocrafting recipes, whether you had engines attached or not.
Now, for single-ingredient recipes, I wrote a Lua program for a crafting turtle where I can put some stack of items in the first slot, and it arranges it in whatever common pattern I tell it to, and crafts as many as it can. That's nice, but typing some command -- in this case "makeme chest" -- is probably not optimal. Faster than an automat, and fun to watch, though. For a few special items, I keep a crafty turtle on top of my compressor, so for making coal balls or carbon plate or compressed plants (the compressor doesn't recognize many leaf types) I run a program that compresses the pile and drops it right into the compressor below. That's handy.
The real comedy comes in planning supply lines. I do have some autocrafting chains off my main tunnel, where I can throw surplus sand and have it fill barrels with all 3 kinds of sandstone ... only to find out that I never actually use those. Another autocrafter group turns stone (from my igneous extruder) into stone brick, stone slabs and chiseled stone brick, which I DO use regularly. Yes, I could have gone further and turned half the stone brick into slabs and stairs, but I'd rather make those on demand. I almost never use those. I can't know today which items I may need for a project next week. Even with a written TO DO list in a text file. If I turn all my sand to glass or logs to planks, oops, when I need sand or logs for a future project, I wasted it all on something I didn't need. Some things I try to keep a half a stack of, like advanced alloy and carbon plate. When that runs low, I throw a few more in the compressor. But you know how it goes, when you finally decide to make a treehouse and you need 200 planks, logs and fences, and you go to your vault only to find you turned all your tree farm output into chests and ladders, doh!
It's actually a very complex problem. I try to think of it in terms of "industries". An industry is a subset of related materials like ore processing or sand/glass or bees, which share some small group of machines. I keep each industry in a separate work area -- usually one industry per floor of my buildings -- partly for sanity and partly because it makes a town feel a lot more realistic. You'd never really need a bakery and a rubber extraction plant at the same time. But the way all the industries intertwine is tricky. If I send half of my glass to make glass panes (building industry), then switch industries to potion-making and find I'm out of glass bottles, oops; so I make a big batch of glass bottles then find a new oil well the next day and need glass tanks instead, oops ... but if I make 50 glass tanks and 100 glass bottles and never use them, that's a dead end.
I also try to keep some mods separate, so the whole flavor changes as I go into certain buildings or caves. For a few I try to locate the mod's workplace in a suitable environment. I always liked having my magical stuff in wizard towers well away from my main town, so there was a sort of pilgrimmage involved; then with Thaumcraft (which I still haven't started learning!) and its flux and will-o-wisps it turns out there's a good reason to stay away from home, to avoid corrupting your whole city. But even that's not so easy anymore. When I first heard of Thermal Expansion, I set up a little "Thermal Expansion Lab" and went to work figuring out those machines ... but now I like them better than the IC2 machines, so they've become the heart of many of my production lines. Railcraft used to be in a vault of its own, but now creosote can be burned as fuel in a (GregTech) generator, steel is used in some machines, coal coke is all-around handy fuel, and GregTech shifted some IC2 recipes to use the Railcraft Rolling Machines! Even funnier: it turns out I hardly ever use rail, now that I can make MystCraft linking books to get from place to place, now that the old Bukkit "warp" command is unavailable.
Wow. That takes some strategy. I have text files of recipes, one file for each major mod. I don't run the game full-screen, I can always see about 6-7 lines of my notes below it. I don't know any other game where I've had to take 100+ pages of notes. But for some reason, Minecraft keeps it interesting. Well, not Minecraft itself. The vanilla game is as dull as a rock. Did they really just add ugly hoppers to move stuff from place to place? Bleah. It's the ingenuity of the mod community, then the oversight of the modpack managers that make all the mods work together ... THAT is what makes this fun. I see when I log in that a bunch of mods have been updated, so when FTB Ultimate updates, there will be new stuff to unlock. And I have still not made a tesseract or a thaum golem ...
Comments
Post a Comment