Arch666Angel wrote:I'll have to read through all the chat at the weekend, there are some lengthy posts here which demand a proper reply
For the bio processing stuff remember I put it back to alpha for two reasons:
-New Stuff getting in and stuff already there changing
-No balance yet, or waiting for feedback or a proper testplay to balance stuff out
So neither the wood nor the other production chain have seen a real balance sweep, they are set to whatever I felt right at the moment so you have to be patient with me
With that said, I have a new idea for the arboretum/wood production stuff, so expect that to change to something more drawn out than it is now.
No worries!
I did an up to date 0.15 playtest and now a 10 hour 0.16 playtest and I must say this last one is one of my worst bases ever. Normally, I would restart, but it will be interesting to turn in around instead and make it great.
I wanted to test out a few things and here are my findings and also a summary of the discussions.
Is 1 to N sorting any good?
Short answer: no! Not for high demand ores. For low demand ones it is probably good, or at least easy to set up as a temporary setup that will last you to end game and then you tear it up and be done with 1 to N forever.
First let's define the ore need. I'm usually in need of really large amounts of iron early game, but I've seen that not everybody plays like this, so I'll settle on a "very low" iron production: 4 yellow belts, to be updated later to 4 red belts.
Using 1 to N sorting, 4 ore sorting facilities will eat a yellow line of saturated crushed saphirite. This will output so little ore that not even 2 modules of Smelting, that is 6x2 casting machines can be 100% fed. Even at full load, 12 casting machines do not saturate a yellow belt, so when they are starved for iron ore, they produce less. A yellow belt is 800 iron/s, but 12 casting machines give you 720 iron/s but with one input line of crushed saphirite you will have around 550 iron/s.
So you need 4 fully saturated lines of miners. When you upgrade to red lines, you will cover up with this your entire initial saphirite patch. This means extra early game train. Then you need 4 full lines of crushers to give you 4 saturated lines of crushed saphirite. This will give you eventually 4 lines of iron that are hardly saturated, but with a 4 to 3 ballancer you will get 3 lines of iron.
So you get 3 lines of iron, tap 50/100% of your initial saphirite, while constantly producing 1/1.5 lines of copper. Which you need to warehouse because you have nothing to do with it.
So I stand by my conclusion: 1 to N sorting is not any good and just a stepping stone tech. It is enough for ores or setups where you put down 1-4 ore sorting facilities than output more than you need. For the rest, that is not any good.
Especially if you would like 16 express lines of iron.
Now there might be some late game smelting recipes that I don't know, iron, nickel and cobalt from 1 to N sorting and produce insane amounts of iron, outpacing anything else, including N to 1. I don't know. All I know is that until you reach that point, you will just suffer. In these 10 hours, never once have I been not iron starved and never has the iron backed up.
On the other hand, you can use N to 1 sorting, need only a few miners on saphirite, only a few miners on jiv and done, only two lines or crushing, one for each ore, you can easily get 8 red lines of iron and then you can easily and naturally upgrade them to express once you get lubricant.
All while not getting any side-product ores than you need to store.
Now some might argue that the initial low output is OK, because slowly, little by little, you will upgrade your tech. You won't increase your raw initial stage output, but build more other complicated systems on top and eventually those 3 lines will turn into 8 lines. The problem is that this is far too slow of a process. By the time I've turned my minuscule 1 to N sorting into something that would satisfy my original needs, I'm really expecting the equivalent of 24 express lines of iron.
1 to N, in all my testing ever, is one order of magnitude lower than needed for any "real" base.
Now some might argue that this whole point is not important because there are so many ways to produce iron. But my question was not "Can you produce tons of iron in Angel's?". Because the answer for that is: you can produce crazy amounts with ease. Far more than in vanilla and even more than in Bob's. The question was "Is 1 to N any good?". And the reason I care because if you don't use 1 to N, you are ignoring a large portion of Angel's. I wanted to know once and for all if I'm missing anything. All I'm missing is a large low throughput base .
N to 1 setup time and complexity
So 1 to N is so bad and N to 1 is really good right? Well, N to 1 is not easy to set up. The early game obstacles were BEBs, steel and trains in 0.15.
Now in 0.16, like I detailed in a post, there are far too many steps and setups you must do before getting N to 1. I feel this is too slow.
I like the addition of bricks, so that can stay, so the other obstacle is sulfuric acid and mineral catalysts. Way to much stuff to set up before you can get iron.
So there are two solutions: remove catalysts from T1 ores (like in 0.15) or add a small amount of jiv and crot to your starter area. Just a bit. Enough for about 4 hours of production. This way you can build up your initial base with plenty of iron and not have to wory about trains until your jiv runs out.
Paper processing
There are (at least) two early game ways of getting boards if not using greenhouses: one is good enough and the other is horrible. Paper processing is new, so I would get rid of the other older path, since early game is already insanely saturated, but if Angel feels that it is good to offer choice, no problem: make both have the same output and input use. BTW, make paper as fast as using sodium hydroxide, not the other way around
)! You can't have a incredibly large iron setup that barely produces anything AND and huge setup of wooden boards which barely produces anything, which of course you can't feed, because you have not enough crushed stone, not even with 2 fully saturated yellow lines of crush stone.
While some people might like it, I doubt the majority want a base that takes 30 hours to feel like an OK green science tech base. Angel's != Marathon!
And if the two recipes are not made equivalent, speed up wooden boards by 4 while reducing input needs by 4 to keep things in balance. So two assemblers would not produce 0.75 wooden boards/circuits second, but 3, which lines up pretty well with 3 bob's electronic assemblers producing circuits at a speed of 1 and 4 copper wire assemblers.
Me personally I would increase the factor by at least 8. The goal is not to make basic circuits a pain in the ass because Bob already takes care of that by spiting up T1 circuits intro 2: the basic ones and BEBs. Early game circuits are complicated enough as they are, I don't want to put down 60 algae farms to produce boards.
Building overload
In all my bases, I use a 9x9 grid of assemblers with bots to create all the building. Angel adds a ton of building and they mostly have 4 tiers, so it is easy to see how 81 assemblers is not enough to cover all tech buildings. And a giant pain in the ass to build. Especially without bots. Looks pretty epic though.
I've been bothered by this forever, hence I can up with the 9x9 grid. And its size is carefully chosen too.
But we need a better solution. I'm pretty sure this can't be done with Factorio as it stands today, but I'd like to see a new special module slot in all buildings that only accepts one single special speed modules that apply multiplicatively on top of other speed modules. This means only one level of building and 3 special upgrade modules, with 3 different speeds. probably the 33%/100%/266%, but you can only have one. So would would go though speeds of 0.75/1/1.5/2.
But if this can't be done, I really like your idea of mining buildings and getting back building blocks. I'm very curios to see your solution, but I would go with 3 basic building block module types for a T1 building (and not using all 3 for T0 buildings, like stone crushers, probably just one) and 2 upgraded module types per tier, one using gears or pipes from some higher tier ore, one using higher tier circuits, and maybe bricks? So a progression of stone bricks/clay bricks/concrete bricks and iron/steel/brass gears and so on. With all 4 tier buildings, you would need 9 modules. Enough to put on 5 lines. Higher tier buildings would not use lower tier buildings. So you set up your assemblers in a row and produce all T1 buildings. When time to upgrade, you add a single new line with 2 modules and change the recipes in all the assemblers and get all new buildings. And you repeat this 2 more times and have all the buildings. Picking up buildings would give you the building block modules back and modules stack to 999, so picking up every single building even in a large base will eat up only 9 inventory slots.
