Quality

[obsidian crescent]

Actually, how fast are the recycling recipes for various stuff? Is it a constant speed (so everything recycles at the same rate) or does it vary based on the recipe speed of what it's undoing?

[uncut helm]

I forgot but you need shittones of recyclers to process the scrap

[frank warren]

If you look at the context of SoulBurn’s message you might find it funny, he is reacting to me saying “JG uses speed mods sometimes”

So we have come full circle

[uncut helm]

[frank warren]

We do at least know the time for scrap recycling - it’s very fast.

[obsidian crescent]

It better be; you need to churn through a lot of scrap.

[frank warren]

Which means you have to be doing a lot of recycling.

It’s like 0.12 seconds per scrap per recycler already…

A base recycler can chew through 8.33 scrap/s

how much damn scrap are we gonna have to sort though

this game is literally garbage

[obsidian crescent]

If it takes 5 holmium ore to make a single science pack, then to achieve 500 SPM, you need to process 250,000 scrap per minute.

[frank warren]

roughly 4k/s

I think that 5 ore number has a lot of question marks around it though unfortunately

[obsidian crescent]

Oh definitely; that's a number I just picked to do a calculation. But holmium is almost certainly the limiting factor in science production.

[frank warren]

That’s 480 recyclers.

yeah maybe JG is not so crazy

[iron root]

Also don’t forget leveraging on stuff going up to science

With holmium likely being limited by mining/recycling speed, prod will be huge for boosting science output

…
+1 point for vulcanus first

[obsidian crescent]

Sure, prodding will definitely help, as will Bacchus/Foundry production of holmium plates. But at some point, you hit the cap.

[iron root]

There will be a cap, but it should make a significant dent in that ‘480 recyclers’ ballpark

[unborn flax]

i analysed one of the FFF-399 videos and decided that at the very least, recycling speed is not constant: #1215078107334057984 message

[daring siren]

Well it's all probabilistic, with about 50% to get anything

[unborn flax]

I mean the time taken to recycle an item appears to be constant per recipe but differs between recipes - it's not probabilistic AFAICT.

[daring siren]

yea it differs between recipes. What I mean is that something may look like it takes time X and but we see 2*X because the first one didn't produce

[unborn flax]

oh right. I didn't look at the outputs: I counted the number of inputs going in, then counted the number of frames the machine was on by looking at the little light

[daring siren]

Interesting point. The graphics for recycling scrap is a bit different than recycling other things

[unborn flax]

I'd be interested to know how recyclers do RNG. e.g. if you're recycling a LDS, does a recycler really call the RNG 20 times for each input ?

[daring siren]

No, it's called once

[unborn flax]

Do we know how that works? central limit theorem?

[daring siren]

When amount_min and amount_max are not provided, amount applies as min and max. The Expected Value simplifies to p * amount, providing 0 product, or amount product, on recipe completion.

See here https://lua-api.factorio.com/latest/types/ItemProductPrototype.html#probability

If you have min and max, it's a linear distribution

But most recipes are just item or nothing

[unborn flax]

If recycler is meant to output 25% of ingredients, that means max must be 50%

[daring siren]

Max is 100% of ingredients, 25% of the time.

Actually no

I'm lying

See 376

But they do come out all with the same quality

[unborn flax]

So in the context of that link, p=0.5 - which gives 0.5 probability of nothing and 0.5 probability of linear distribution between (0, max). Hence expected 25% of ingredients back.

[daring siren]

Where did you get p=0.5?

[unborn flax]

you need it to get to the "More specifically, for each crafted item, the recycler gives you 25% of the original ingredients back" from FFF-375 while allowing a linear distribution going up to max

[daring siren]

It's 4, 25% simplified to 1 * 100% I guess

I didn't see any time it gave out more than 1 speed module

[unborn flax]

Are you certain recycler outputs are implemented using the ItemProductPrototype / probability mechanic, or are you just going on the FFF-376 video?

[frank warren]

this makes sense computationally

[hoary current]

I guess for the recycling recipes they will have min/max if the amount is larger than 1

[hoary current]

did u see it ever giving no speed modules?

[daring siren]

Good question

If it does, then it must sometimes return more than 1

For the numbers to work out

[hoary current]

say an arbitrary recipe: (4A + B)x -> y
the the recycling result will be y -> (A+(B/4) * 100% * 1)x ?

[daring siren]

In jq I did it in a way that maximizes probabilities and variance

[hoary current]

I would prefer the result is as stable as possible

if the recipe take more than 4, the recycle result should give no less than 1

literally do the x/4 calculation, the interger part of the result is garanteed, the decimal part is probability based

[daring siren]

It is possible to do

[jaunty citrus]

My expectation would be that it behaves similar to quality rolling (if you have enough, it's guaranteed). Of the 25% of an items ingredients, it only rolls for the decimal.

So for some imaginary item that uses 12 iron plates and 7 gears, it would return 3 plates and 1 gear, with a 75% chance of returning an extra gear.

would be a good example. (10 , 1 )
It should return 2 every time, with a 50% chance of returning 1 and a 25% chance of returning 1

[hoary current]

yeah, that’s also what I expect, guaranteed interger part and probability based decimal part

[burnt reef]

huh, that's very different from what I would expect, which is just for each individual ingredient, a 25% chance of getting it back.

so for an iron gear, it's
0× - 56.25%
1× - 37.5%
2× - 6.25%

total expected value of half an iron plate, but distributed differently.

[jaunty citrus]

It is possible to do it that way, but in my opinion it would be better to do it how I described as it removes the extreme options of getting everything back or getting nothing back, which can cause unexpected results.

[burnt reef]

I agree that it would probably be better, just it's not what I'm expecting

[unborn flax]

I like this idea - it's numerically clean and has a nice low upper bound of ceil(n/4).

Also I think it may be what has been implemented: look at fff-399-scrap-recycling.mp4. There are three visible recyclings of processing units, which should output 20/4 green circuits on average. In each case I can see a full stack, followed by what looks like a stack of 1, i.e. it looks like exactly 5.

[daring siren]

It's not how it works with a single output, and they won't how multiple outputs

I agree it makes sense though

If I was a betting man, it would show the input recipe with 25% near every line

Though that is inconsistent with what we saw

So I bet they do some number compacting

Unless they change how it works internally

[unborn flax]

One advantage of this (now confirmed) method for quality cycling is that the low variances and low maximum help avoid jamming.

i.e. an upper bound of ceil(n/4) for recycling a single recipe means that unless your bonus productivity is close to 300%, a cursed RNG can never make you jam.

Things are almost as good for multiple-input recipes as the low variance helps keep the relative numbers of different items close together. And if they drift apart too much you can just void without worrying about jamming.

[Other recycling formulae, e.g. "0.5 chance of a constant distribution on the integers [0, n]" don't have these anti-jamming properties.]

[jaunty citrus]

Well I'm glad that it's confirmed now. Should drastically reduce the opportunities for jamming/edge cases.

I'd prefer if it was at least a little reliable that if I took 4 items and recycled them I could make 1 of the item again.

I do wonder what it looks like on the tooltip actually

Could you say that an recycles into 2-3 and 0.25 ?

I don't think you could currently but It seems it's getting updated between 1.1 and 2.0 to fix some issues with/allow recycling.

[frank warren]

how is it confirmed?

[half raptor]

Not sure why you'd want to recycle belts, even if you say "easier way to transport iron" it still feels like the investment isn't worth it

[daring siren]

Because you can put quality in them and recycle for quality gears.

[neat shard]

I'm not sure it even has a tooltip tbh

Recyclers work kind of like furnaces, rather than using recipes they just react to whatever you stuff in them

[jaunty citrus]

Yes, so when it's processing what does it say it'll produce

They still have recipes you just don't select them manually

[hoary current]

why do u need quality gears then

[obsidian crescent]

Gears are components of a lot of things that you want to make quality versions of.

[hoary current]

but u still need quality of all the other ingredients for those thing u want to make

[half raptor]

Quality gears into engines which eventually turn into bots

Quality bots have better battery

[rotund egret]

@obsidian crescent re: that one reddit thread about spidertron grids

assuming the rocket turrets will require fish, it's going to be slightly easier to get legendary spidertrons. both because the recycling recipe is cheaper and because you get another quality step

whether or not legendary spidertrons are a vanity project is another thing altogether

with how large bases will get, I don't think endgame spidertrons will be anything other than legendary, but it's going to be one of the last things you do

[obsidian crescent]

I don't see rocket turrets requiring fish, as this would require that fish would be something you can get on Bacchus (by process of elimination, since neither of the other two planets could possibly provide fish). Also, eating a legendary fish wouldn't be much of an achievement if you could quality cycle rocket turrets. The thing that makes it noteworthy is that the only thing you can quality cycle to get them is the second-most expensive craft in the game.

[rotund egret]

I think it was more or less hinted at that the new spidertron recipe is turret + 4x exoskeletons

and having fish on bacchus is very much possible, since it's a planet with lots of water

[daring siren]

• fish

#friday-facts message

[rotund egret]

well, that settles it then, muh reading comprehension

[daring siren]

TBF he said the turrets "have a fish brain", but later clarified with that

#friday-facts message

[kindred crater]

when crafting a recipe that accepts any quality, if you have a recipe which needs 2 gears, are you able to have 1 and 1 , or do they have to be the same quality

[charred gyro]

I would think any combination of qualities would work, but don't recall it being explicitly shown either way

[obsidian crescent]

No; there's only one input slot per ingredient, and different quality versions of the same item don't stack. So once a single Q2 plate is inserted, that slot won't take anything other than Q2 plates until it goes away.

Also, it's the machine that accepts any quality, not the recipe. All recipes can be made from any quality ingredients.

[kindred crater]

I figured that was the case, thanks

just thinking about fulgora and how to manage quality intermediates

ie, if are put in your main recyclers processing scrap

[obsidian crescent]

If the scrap has quality (from qual modules in miners), then you might get some kind of jam. However, since scrap recycling consumes exactly one scrap, any jams will be temporary; once the slot is empty, any quality scrap can be put into it.

No, the bigger problem is the outputs from recyclers. They only have so many output slots, and if it can't put them into an output, those can fill up with different quality stuff, causing the recycler to stop.

I suspect it will be easier to just shove them into an active provider and let the logistics bots sort it out.

[kindred crater]

yeah, that would certainly simplify things. as for quality scrap, I can just have that put into its own recyclers with quality modules, and the main production recyclers can be left alone. less volume of quality stuff, but at least its isolated from your main processing and won't jam everything worst case

fulgora is going to be a tricky planet to optimize production on, both for quality and non-quality

[obsidian crescent]

I expect to get some pre-Fulgora practice by building a circuit-based system for making certain quality stuff using quality ores on Nauvis. The idea is to make sure that you don't consume some quality intermediate if there isn't a certain minimum amount of that intermediate around. Something similar would be applied to Fulgora, but in reverse.

[kindred crater]

based on what I know for now, will be my main quality planet. I'm going to play a very long term game (months+ running nonstop), and I want to save as much resources as possible on nauvis for science. Not to say I'm not going to make quality stuff there (I am), but to upgrade tons of machines in a >5-10k spm base is going to consume a ton of resources. I'd much rather raze the other planets for that kind of stuff lol. And vulcanus looks prime for that. Perhaps I'm looking at this the wrong way, but there you're "only" consuming calcite for iron/copper/steel and it goes a long way, versus consuming iron and copper ore patches.

[iron root]

Vanilla factorio ore gen has the fun of just going farther from origin to handle increased demand.
If the actual mining rate becomes limiting then vulcanus might become strong if the calcite for lava processing goes a long way

[kindred crater]

yeah we'll have to see how the numbers work on lava vs raw ore

[obsidian crescent]

It's not really lava; it's calcite that it has to compete with. And the question there is how rich your calcite patches are (especially since the devs know that you will have the BMD on Vulcanus).

[gusty trench]

And almost certainly about 50% mining prod

[kindred crater]

I think I wasn't clear enough in what I meant- how the yields from calcite differ processing it from lava, versus from ore. making these numbers completely up, but with the 2 different steps for each, one could yield 100 plates and the other can yield 120 (or are both the same) from the foundry considering nothing else but base prod

[obsidian crescent]

Well, we know the ratio for ores: 1:75 (including base prod). Lava processing spits out two different molten metals. So lava processing could have a ratio of 1:37.5 for each output. But it also outputs stone. And since this can only really be done on Vulcanus (where calcite doesn't require interplanetary transport), I would expect the ratio to be a bit lower than that.

[kindred crater]

I was wondering the same thing as a light 'penalty' if you even want to call it that

[jaunty citrus]

I don't see there being a 'main' quality planet per se, but if I'm going to make something it'll probably be on Nauvis if it can be.

Nauvis will have the most space, easy water for nuclear power, it'll be where the majority of science is made so you can take advantage of that infrastructure as well

[kindred crater]

all strong positives for it

[half raptor]

considering Nauvis (Novice) is the beginner planet, it's reasonable to have it be the easy planet to develop

[burnt reef]

petition to retroactively make Nauvis the Roman god of beginners radioactivity

[obsidian crescent]

It's about miner counts. If on Vulcanus you can get the same amount of raw iron and copper with 100 calcite miners as you can on Nauvis with 500 iron/copper miners, then there's a pretty strong incentive to put more stuff on Vulcanus.

But at the same time, you have to take into account oil products, which requires not only coal (which is less efficient than crude) but also steam and water, which draw on both calcite and sulfuric acid. I could imagine red circuit production in particular being bottlenecked on plastic.

[gusty trench]

Definitely prod, since the overall effect is exponential with the number of steps you can apply prod to

[obsidian crescent]

Qual modules would be more important for quality cycler setups. Prods are useful there, but getting your qual modules better probably buys you more than your prods if you have to choose.

Especially since the recycler can't have prods at all.

[burnt reef]

but 1. it's only polynomial in the value of the module, and 2. there's gonna be prod research reducing the relative value of prod mods, which will eventually reach 3. the 300% productivity limit.

[obsidian crescent]

While productivity research will certainly guide your hand as to what gets quality cycled, reducing the cost of everything that gets to that point is still useful.

[burnt reef]

absolutely, quality prod mods are still incredibly valuable, I just wanted to list some mitigating factors that make them slightly less incredibly valuable

[obsidian crescent]

From a quality standpoint, productivity research doesn't reduce the value of prod modules. It takes 13 levels of blue circuit research plus Q5 prod 3s to make an EMP hit 300% productivity. This makes any quality of blue circuit equal the recipe cost of a blue circuit.

That's the point where you're done with blue circuit productivity research. Yes, doing more research and swapping some of those prods with quals will make it take less time to generate high quality circuits. But given that it's an exponential research, there's probably more important stuff you could be researching (like LDS productivity).

[burnt reef]

productivity research reduces the relative value of prod mods. going from 0% -> +10% productivity is a bigger step than going from +100% -> +110%. and prod research eventually turns that 0% into +100%, making the additional +10% on top of that a proportionately smaller step. that's the point I'm trying to make.

(the actual numbers are a little different – +10% -> +25%, net +15% – but the point about relative value remains.)

anyway my first priority for quality quality modules will be the quality module assembler, followed by the prod mod assembler

(b/c that's what this question boils down to. s themselves can be manufactured en masse; it's the s you'll have to distribute carefully.)

(or. well. s at first. but still.)

[daring siren]

Which will be the EMplant once we get to

[neat shard]

Prod mods and qual mods are the most important things, because they generate value out of thin air with pretty much no resource input required to run and no impact on UPS.

Second most important are the tools that let you squeeze maximum value out of prod mods: Assembler-type buildings, speed mods, and beacons.

[daring siren]

Technically eff modules reduce cost of power.

[neat shard]

Power is cheap bordering on free in most circumstances. Outside of the context of space platforms, I don't expect this to be much different in Space Age.

[fallen urchin]

Even on platforms it's probably not worth it to use high level eff modules

[neat shard]

Considering the size of space platforms we've seen in teased footage, I'm not sure there's enough room to include enough power production to sustain buildings stuffed to the gills with prod and speed modules, and it's better to give up a speed mod or two for eff mods rather than forfeit the benefit of beacons altogether.

[obsidian crescent]

Really? Vulcanus has good solar but cramped land area until you get cliff explosives (which may not be on Vulcanus). And on Fulgora, you're going to have to subsist off of lightning, and even then, most islands will be electrically isolated from one another for most of the game's progression. Eventually both of these restrictions will be lifted to some degree, but I don't think either planet is going to be as power-rich as Nauvis.

Remember, a Foundry gulps down 2.5MW each; that's half the power output of a single steam turbine. And you're going to need a lot of Foundries to process stuff. Good quality efficiency modules, especially early on (even in beacons) will be pretty important there. And even as you scale up, those power costs start to get pretty extreme.

At the very least, on Vulcanus, I'd probably go for quality solar panel/accumulators or steam turbines (if that's useful) before anything else.

[fallen urchin]

I think there will be new power options in SA

But yeah, power on other planets might not be as abundant as on nauvis

[daring siren]

Like neutralization on and 🌩️ on ...

[final nimbus]

neutralization is mechanically similar to coal power.

becomes
becomes calcite

[obsidian crescent]

What's neat is that this fuel needs to be shared between power production and iron/copper production, much like coal-fired furnaces (and it is even shared with oil production via steam and water).

[gusty trench]

Fulgora has a semi-alternative method of power gen: because solid fuel from heavy oil is effectively free but ice is scarce, you can transport ice around for steam power.

Say that you do scrap processing on one island, and general manufacturing on another, transporting everything between them by train. Ice can have its own train for shuttling it anywhere, and solid fuel is made on site, so you can have power independent of lightning collection

[obsidian crescent]

Of course, that raises the question of stack density. If ice stacks to 50 and contains 50 ice each, then a single wagon can transport 100,000 water, 4 fluid wagons. But if ice only stacks to 10 (like barrels), then you're down to 20,000, less than a fluid wagon.

Also, assuming that melting ice requires more expensive infrastructure than unbarrelling water, you might want to use any excess power on your scrap processing island for melting water; that way, it only needs to be converted to steam.

[gusty trench]

Fair. You could also make steam centrally and then ship it, which is possible in vanilla but never particularly sensible

[kindred crater]

I can see the utility in power generation this route, but seems like its more earlier game for the planet, or an emergency backup

at least from what we've seen in the FFFs so far, Wube seems to be pushing us towards spamming as many lightning collectors as possible (placed intelligently, of course) to power your base

[gusty trench]

Yeah, hard to say when or if it would be practical. I'm thinking around the time when you've done the initial 3 planets but before you have the end game landfill. When you want to scale Fulgora production a lot and are limited by the amount of power and space available on a single island

I don't think it would be a ton of extra work even when you know that you'll get the oil sands landfill eventually. Processing scrap on a separate island from your main science production is a sensible division of manufacturing, and having a few trains that transport ice or steam is a fairly minor addition if you're sorting items already. Then you just need to add power production to all the small islands you want it on

[kindred crater]

yeah that definitely makes sense. plus you're getting as a recycling product as well. which lets face is, is probably meant to turn into rocket fuel for launches, but if you're not exporting anything substantial at that time, might as well use that and the ice for something, if it has no other present use.

going on a slight tangent, its been discussed before using beacons for , and FFF 409 even demonstrated it on platforms: https://cdn.factorio.com/assets/blog-sync/fff-409-platform-loop.mp4

if power restrictions really become an issue at some point, it would be very useful to have some kind of hybrid of / and , and all around, so you're getting the most out of your space, and scaling down joules/craft

[gusty trench]

buildings already decrease joules/craft by crafting faster at no energy penalty, and pairing that with modules and beacons further improves it. Even speed modules improve joules/craft by increasing speed more than energy usage

[kindred crater]

yeah it can get very impressive when the numbers start multiplying with each other. I'm really excited for this mechanic

[obsidian crescent]

Not just that: let us not forget the humble assembler 2. Power-per-craft is pretty good, and it's cheap enough to make in bulk. And quality versions of them only get better. For things that you really don't need in great quantities (and on Fulgora, that's a lot of things; most of the rest are made in EMPs or other 50% prod buildings), bringing a stack of Q2 or Q3 assembler 2s could really pay off.

Oh, and don't forget about our friend, the Q2 or Q3 beacon, which not only consumes less power, but also boosts machines more.

[gusty trench]

Space efficiency is another metric to optimize on for Fulgora, which can be important when trying to scale science production. Power efficiency can run against space efficiency at a certain point

[kindred crater]

all of these things really wouldn't be that hard to obtain with some 's either, its an investment that would quickly pay for itself on fulgora

yeah if you're hitting that ceiling and can't expand at that time for whatever reason, you'd just want to do as much as possible in a smart way

[gusty trench]

Practically, I think it would often be "don't want to expand to another island and set everything up again" rather than being genuinely unable to expand

If you're butting up against the size of your primary island

[kindred crater]

fair point

[gusty trench]

The islands are irregularly shaped, so it's far from a copy-paste of your existing factory

[kindred crater]

people like nilaus complain about that because they can't use their perfected bps every time now lol

personally I'm looking forward to the challenge

it will make every game feel fresh

[gusty trench]

I'm fairly skeptical on the takes of some big streamers. Not being able to use your existing blueprints for the expansion is a good thing, and the devs seem to have taken active aim at making all your old blueprints obsolete by the mid game

[kindred crater]

beacons getting buffed, , new known/unknown production stuff, productivity research, bulk inserters/higher belt tiers, probably more I'm forgetting... excluding some stuff like power bps, starter bps, train stations, tons of "good" 1.1 bps will be obsolete

[gusty trench]

Train stations are different now because of the changes rails, elevated rails, and new stack inserters. Additionally there's the train interrupts, circuit groups, station priority, and calcite logistics, all of which can change how stations are made. The very basics of loading and unloading items/fluids will remain the same, but so many details have been changed

[obsidian crescent]

I don't think it's about not using existing blueprints for them. It's more that they can't just slap down another 30 green circuits per second BP whenever they want without properly integrating it into terrain.

[kindred crater]

I don't disagree, I was more just talking about a basic/generic kind of loading/unloading station with fast inserters or bulk inserters.

[daring siren]

I speculate ice stacks to 10 and gives 50, like barrels.

[neat shard]

I strongly speculate otherwise. As usual, the go-to comparison is SE – in SE, ice is a million times better than barrels

In SE water ice stacks to 200 and gives 100, for a total of 20K water per stack of ice

I don't know if it's going to be quite that extreme in SA but I'd be shocked if it isn't way denser than barrels

[iron root]

i think it'll be a few times denser than barrels but im not sure its gonna be the 40:1 ratio in SE.
SE handles waterless planets mostly by just using condenser power architecture and then just shipping in ice to cover the costs
SA gives/makes waterless power architectures much better on said planets, and gives restricted sources of water on the planets themselves

so itll probably make shipping ice/water much less trivial to encourage use of the on-site resources

id spitball the total water per stack of ice being 2k-5k i think

[obsidian crescent]

The only way this kind of extreme would make even a little sense in SA is if you can't make ice; you can only pull it out of asteroids or dig it up out of the ground.

At the same time, they probably don't want to make early-game ice barges the go-to solution to getting nuclear power to work in power constrained environments. If one platform dropping ice on Vulcanus allows you to run 3 GW of nuclear power, then it's really difficult to justify using native sources initially rather than just importing nuclear fuel.

One simple way to mitigate that is to make the stack density of ice fairly low. A single stack may hold only 500-1000 units of water.

[iron root]

also its worth noting even if you got 20k water to a stack of ice
it still wouldn't last long at all in a steam setup in SA, because there aren't any condenser turbines

power uses a lot of water

[heavy ember]

Do we know what q5 turbines/ heat exchangers do?

If they get more power per water consumed the that might change things

[jaunty citrus]

We don't.

It's a bit of an issue as it determines everything about how useful quality is on reactor setups

[gusty trench]

It would be nice if quality heat exchangers made higher temp water, so you could get more energy per water used

[jaunty citrus]

It could be considered overkill if every step was multiplied by 2.5. Getting close to 8 GW from a 2x2 reactor would be pretty silly. Maybe that's what we should want from quality though.

I think it would be perfectly reasonable if reactors and exchangers had higher conversion rates, and turbines output more energy per steam

That way a reactor would produce 2.5x as much energy per water and per cell, but also consume the cells and water 2.5x as quickly, for 6.25x total output.

Then again, they could also just change the quality numbers for reactors, exchangers and turbines to not be 2.5 at

[uncut helm]

they do brrrrr

[jaunty citrus]

go brrrrrr

[daring siren]

They likely just input and output more

[jaunty citrus]

Probably, quality doesn't increase power consumption in most cases so there isn't a big reason to scale production multiplicatively.

[dense kraken]

that's gonna be annoying as pipes don't have higher throughput :/

[obsidian crescent]

It's not even a question of pipe throughput; it'd just be completely useless. OK, not completely; it would be a purely space compression thing. Which is not how quality works in most cases. Most of the time, having a high quality building is better than having multiple low quality buildings.

[dense kraken]

UPS

smaller nuclear reactor means higher performance

[obsidian crescent]

If you care that much about UPS, you're not going to care that you can reduce your nuclear UPS by 1/2.5; you're going to use solar.

[dense kraken]

else everyone would be their factory, which is actually far from being the case

that's subjective tho

why is there nuclear power in the game if solar is objectively better?

[obsidian crescent]

I guess I'm confused about the conversation at this point. You suggested that quality scaling of that sort would be better for UPS relative to base quality. I pointed out that if UPS is a problem, the best solution would be to move to solar. You then replied that nuclear doesn't need to exist if it's so bad for UPS? I'm not sure I understand what's going on.

If you have a UPS problem with you nuclear reactors, you can do two things: make a higher quality version of them, which will reduce your power UPS costs by at most 1/2.5, or... switch to solar, which will reduce your power UPS costs to 1/infinity. Neither of these is trivial (not only do you have to produce high-quality reactor parts, but you have to completely redesign the reactor; it wouldn't be an in-place upgrade), but one of them will solve the problem completely.

[dense kraken]

sorry my mad, I'll try to make it clear 😅
someone may prefer nuclear over solar for coolness reasons, but doing so may run them into UPS issue depending on scale and hardware
using quality is one solution to this issue, all I'm saying is that it thus makes the fluid routing challenging

wouldn't it be sad to "have to" use a solution you don't like just for performance reasons?

[daring siren]

Assembler quality is also space compression. But the huge difference is that assemblers get modules, and beacon effects

[obsidian crescent]

But assembler quality is not only space compression. High quality assemblers are more power-efficient per craft.

[daring siren]

True. Haven't thought of that.

Interestingly, inserters are faster, but not more power efficient per craft

What if boilers use the same amount of but convert more into ?

that also doesn't make much sense I guess

[fallen urchin]

Because Factorio is highly optimized, if someone doesn't want to go mega, nuclear won't cause UPS problems.

[dense kraken]

fair, you win

I had a though about it and I figured that I really enjoy going for more "powerful" stuff at the cost of increased complexity
nuclear vs boiler (or arguably solar) is one of them, but there are other examples

[median verge]

Productivity boilers? Nah, probably just speed and temperature

[obsidian crescent]

Hotter steam would reduce water consumption per kW generated. Of course, Q5 heat exchangers would need to generate 1227C steam.

[kindred crater]

if there are going to be new power generation techs added (outside of lightning rods) in SA, I think its going to be something thats also UPS efficient, but also space efficient. it could be expensive to offset the benefits and would still very much be worth it. it could be something like building a dyson sphere- the satellite can be very expensive as well as the launch needing tons of fuel, but it grants a permanent +50kW thats beamed to back to nauvis

its not a perfect example, but I could def see something along those lines

[obsidian crescent]

That being said, if quality reactors don't generate more heat from the same fuel, I'm not sure how useful it'd be to bother with them. Quality turbines and heat exchangers make sense, as they use less water per energy generated. But there isn't much advantage to using quality reactors without some kind of "productivity" boost. Even on space platforms, it'd be better to install two base quality reactors along with with high-quality exchangers and turbines than to use a single high-quality reactor.

[obsidian crescent]

The problem is that, once you solve the resource problem, you've solved the resource problem. Once you can afford the resources to add free power, then there's no reason to use anything else.

The thing about solar is that its land cost is always there. Even if you've trivialized (or disabled) biters, you still have to build on that terrain.

Even if this costs unique resources from across the solar system, once you've got the infrastructure to do it, there's no reason to use anything else.

Nuclear was not designed to be the UPS-inefficient option for power. That's just how it worked out.

[daring siren]

What if gets higher neighbor bonus?

[median verge]

I know, but implicit productivity with some way to return water from steam would create a positive feedback loop. Im looking at you, space platforms

[obsidian crescent]

Oh, I didn't think of that, but you're right. Steam condensation would allow you to make water from U-238.

So hotter steam is really the only way to allow high-quality steam generation to offer real benefits (ie: not just being space compression).

[obsidian crescent]

I don't know about that. Solar is not useful on Fulgora outside of bootstrapping. But it's likely going to be the end-game power generation method of choice on Vulcanus (especially since panels and accumulators are almost entirely composed of iron and copper). Unless some kind of enemy is there that specifically makes it hard to hold terrain. Nauvis's terrain is just free realestate by then too. And once you can make iron/copper cheaply, you can even invest in quality panels&accumulators.

Probably not. You're likely going to have to import U-238 for reactors there (and excavate ice).

Just a guess, given that it's supposed to be an iceball planet. Unless it has a unique power generation method, if solar is off the table, you're going to need some form of steam-based generation. Hence the ice harvesting.

[gusty trench]

My idea: quality reactors produce more heat per fuel cell, heat pipes have higher heat capacity, heat exchangers produce higher temp steam (provided you can get them hot enough), and turbines have higher electricity production. Turbines being a purely compressive upgrade means you might not bother with them on Nauvis, but they're way better in space constrained situations like space platforms or Fulgora. With full quality everything, you produce 2.5x power per fuel cell, and 2.5x power per unit of water, but you're not multiplying the two off each other, and might opt to not use quality in one part of a reactor because you're not constrained on the resource it saves.

Re: purely compressive stuff, solar panels are purely compressive. You get more power per panel, and nothing else, since that's really the only thing solar panels do.

Idk if walls would count as purely compressive

[daring siren]

I'd be surprised if heatpipes get anything other than HP

[kindred crater]

trains have at least which can help throughput substantially

[obsidian crescent]

One issue with exchangers having to be hotter to produce hotter steam is that reactors can only output 1000C, and 2.5x the power of 500C steam would require 1,227C steam. Which reactors can't produce.

[final nimbus]

Unless quality reactors go above 1,000C

[dense kraken]

can't wait to melt steel using nuclear reactor

[daring siren]

Going nuclear because rocket fuel is not enough for that

[uncut helm]

cant wait to melt nuclear reacotrs using quality nuclear reactors

[faint blade]

I think there is a mod that adds furnaces powered by heat pipes instead of fuel or electricity, which sounds kinda fun

[jaunty citrus]

I like the idea that quality reactors get a better neighbour bonus

Feels like the most sensible bonus you could give them

I've convinced myself that pipe throughput won't be a huge deal, and can be worked around well enough with mineable landfill

But heat pipes would probably need an adjustment as I don't see quality giving them anything other than bonus health

You can also just build the reactors close to water

the pipelines don't need to be that long

[obsidian crescent]

I don't want to build my oil setups on water either, but I do. It'll be even easier with remove-able landfill.

[jaunty citrus]

So if gets +250% neighbour bonus, you could (at a max) get ~850% power per instead of ~400% power per , or 600% instead of 300% in a 2x2 setup.

That would about double the throughput of water you need, which wouldn't be that bad to manage, and if quality turbines and exchangers have 250% throughput the reactor would also have a 20% smaller footprint for those entities.

[median verge]

Lower energy usage (i.e. fuel is more energy dense) is also an option

[jaunty citrus]

You know I always forget that efficiency is something you can change on boilers/turbines etc.

[heavy ember]

I'm not a fan of above 100% efficiency, or nerfing q1 turbines below 100% efficiency

[jaunty citrus]

There's so much that quality could affect on a reactor setup but the specific wording:

Nuclear reactors, boilers and steam engines have increased production.
Leads me to believe that throughput has to be considered in the bonus

Which would count efficiency out

[heavy ember]

Temperature on HEx/turbine could create some fun setups

[jaunty citrus]

idk changing the tempurature just doesn't sit right with me

You're capped by the 1000C temp of the reactor, unless that also gets affected. You'd need quality in multiple places for it to even be useful, and what about the steam from acid recipe?

getting higher temp steam from an exchanger doesn't do anything without a turbine that also accepts higher temp steam

[heavy ember]

Ah but you could mix lower and higher temp steam to get what you can use

Hence the "fun" setups

[obsidian crescent]

Well, the options really come down to these:

1. Productivity: generate more steam than you gave water (with steam condensation, you can now generate water from fuel with a higher-quality boiler).

2. Increased turbine efficiency (doesn't affect heat exchangers)

3. Increased speed (only acts as space compression)

4. Increased steam temperature (has trouble with max reactor temp, requires equivalent quality turbines to work)

Only the last one really seems to fix all of the problems: quality reactor setups don't have any fluid throughput problems that they wouldn't have had pre-quality (even the productivity one requires that you have extra steam).

Its issues with maximum reactor temperatures are fixable. And having to keep most quality elements in sync with one another is unfortunate, but I'd be willing to live with it.

[heavy ember]

Don't even need to keep them in sync.

What if you drop quality on items further from the cores.

It would increase the temperature band you can work with

Running at full tilt would mean the furthest exchanger couldn't get as hot as the closer ones meaning you wouldn't need to have them at a higher quality

[obsidian crescent]

You need a higher quality heat exchanger to produce the hotter steam. If all of the heat is being absorbed by X exchangers, then having X+1 isn't helpful. A lower temperature exchanger for that "+1" will only serve to reduce the power produced by X, as it would draw heat that could have been used in higher temperature steam.

Using a lower-quality heat exchanger would be like putting a hole in a high-pressure hose; the pressure would drop.

[heavy ember]

If the last exchanger in a line is below the maximum temperature of a lower quality one, you could use lower quality one and save resources without it affecting anything

[obsidian crescent]

How would heat reach it?

[heavy ember]

The heat was reaching the higher quality one as well.

You're not adding an exchanger, you're replacing one

[obsidian crescent]

If the heat is not reaching the last exchanger in a line, then it's not reaching it because it's all being absorbed.

[heavy ember]

Take vanilla reactors now.

With a proper setup the first exchanger in a column could be 900 degrees.

But the last one is 501 degrees.

All of them work.

But lets say you also had a cheaper exchanger that would work between 500 and 800 degrees

You could replace that last 501 degree temp exchanger without affecting anything.

But you couldn't replace the first one. Since it would waste heat.

And adding it after the last one wouldn't work since it wouldn't get hot enough.

[obsidian crescent]

Heat exchangers don't have a maximum temperature. They have a minimum temperature, but not a maximum.

[heavy ember]

They do and it's 1000 degrees in 1.1

ALL vanilla heat entities have the same 1000 degree maximum

[obsidian crescent]

But with the kind of quality changes we're talking about, this would have to change. It would not change by lowering this maximum for some entities based on quality. You'd just raise the general maximum for most heat entities, then raise the heat cap on quality reactors as needed.

That is, if you mixed Q5 reactors with Q1 exchangers, the exchangers wouldn't break just because they got over 1000C.

[heavy ember]

What I'm trying to say is that if you have q5 reactors you don't need to have all of the exchangers be q5.
You can mix exchangers from q5 to q1 if you're smart about it.

You'll probably need some q5 exchangers, but not all of them to be that high

They don't break, they just waste a lot of heat if the adjacent heat entity goes above their maximum temp

[gusty trench]

What if there was a new item added, a heat pump: cools down heat pipes behind it to increase heat ahead of it, at say 80% efficiency. Increases the length you can send heat to, at a cost, while also allowing higher temp heat exchangers further from the reactor

Also requiring power to run

[obsidian crescent]

The only entity that would need to have a quality-based heat cap is the reactor. Everything else could have whatever heat is available. Because there are no ways to direct heat, it always flows from the hottest point in the system. If all reactors in a heat system are capped, then all heat entities share that cap, even if they don't technically cap themselves.

So a Q1 heat exchanger will not by itself waste heat even when paired with Q5 reactors. What will eventually happen is that heat flow to the exchangers will be too slow to actually allow all the heat to reach them. So the reactors will hit the cap and waste heat.

[gusty trench]

You can't have space unconstrained reactor setups because trying to pump heat through multiple pumps quickly eats into your production, but one or two steps through an 80% retention rate isn't unthinkable. Gives more design space for reactors, and allows for easier integration with quality nuclear stuff (if relevant)

[heavy ember]

Hmm what if it takes, say, 10% of the MW moved to work, pulled from the electricity grid.

But 80% of that is heat that goes back to the heat system and 20% is lost

[gusty trench]

If you're throwing quality into the mix, a given unit of heat doesn't necessarily translate into a fixed amount of power. It's probably easier to give it a high but ultimately manageable electricity cost, since you potentially want a lot of them

Like, if the running cost is too high, it really constrains where they can be used. Maybe something like 1-2 MW is reasonable

[heavy ember]

I don't think unit of heat should change the energy amount.

A heat exchanger that would output hotter steam would take the same water intake but more MW to work

But yeah static power cost reasonable regardless

[gusty trench]

It would be something to decide after the existing nuclear stuff has defined quality progression. If 1°C translates to 1 Joule always (or whatever conversion), then you can raise the power cost of the heat pump without huge issues

[heavy ember]

Now that I think about it, I think reactors and heat pipes store different amounts per degree at the moment

[gusty trench]

(realistically Factorio power units start at kilojoules, even for adding two numbers together)

[gusty trench]

They do, true. I guess total heat energy independent of specific heat capacity is also the Joule

[gusty trench]

(An arithmetic combinator uses 17 joules at minimum to add two numbers together)

[loud prawn]

The more i think of it, merging neighbor reactor inventories would be really nice and would change the 2xN architecture to N×N architecture

[heavy ember]

You can already do nxn
Just need to have a speaker yell at you after 2.5 hours

[burnt reef]

that only works for reactors within reach

[loud prawn]

Not impossible just impractical 😅

[daring siren]

Every time I see 2.5 I think

[heavy ember]

you can run between them

[fallen urchin]

That sounds horribly inefficient

Maybe it's mainly IO cost

[frank warren]

Is this analysis factoring for each?

[fallen urchin]

I think it costs same energy whether you use each or not

So it's probably IO cost

That analysis is from the power consumption of 1kw divided by 1 operation per tick

[neat shard]

Let's say you have a small supply of and an equal supply of . Both give +10% productivity, but the is slightly superior because its drawbacks are smaller. So... what determines where the best place is to insert the as opposed to the ?

[opal haven]

in conjunction with , as no speed modules could be added to properly offset the drawback

[half raptor]

If the MK1 uses less power but has the same effect then the MK1 is better

[burnt reef]

it also has a lower speed penalty (5% vs. 15%)

[daring siren]

4 is -60% speed then? That's humongous

[neat shard]

Pretty sure they're referring to 4

4 have -60% speed, which is indeed humongous, but you can easily make up for the loss with speed modules, because prod mods don't have anti-synergy with speed mods the way qual mods do

[burnt reef]

4 only have -20% speed, which is significant but not huge.

[daring siren]

ah yes that's better

[gusty trench]

And improving the speed of the machine you have quality mods is a super obvious thing to do first, so you quickly compensate for the speed penalty

[daring siren]

It's multiplicative

so it doesn't compensate directly

[neat shard]

Speed and qual have anti-synergy

Speed mods reduce quality

[daring siren]

There's still the idea to use a single for your d machines

[gusty trench]

Yeah, but getting green quality basically cancels out the penalty, and everything after is bonus

[cunning sleet]

quality sounds like a cool mechanic

[tall sandal]

true!

[half raptor]

quality makes you have to design a system that can handle RNG. Train Interrupts seem perfect for this.

Example, you have a factory line filled up with in slots, you will get a 25% chance (24.8 if rounding doesn't work) for an upgrade and 75% will just be normal circuits. 10% of upgrades go up 2 steps, 1% of upgrades goes up 3 and 0.1% goes up 4. With a train filled up with any combination of qualities, you can have stations specifically asking for each quality with filters on inserters set and your train will sort everything out for you.

Easy mode is to just build a quality loop for the final craft step of the item you want to be legendary and feed results into a recycler to go for higher qualty ingredients

https://cdn.factorio.com/assets/blog-sync/fff-375-quality-build.mp4

[obsidian crescent]

How do you do that, exactly? If you have multiple qualities of stuff waiting at the train stop, how do you decide which one to put onto a train? Equally importantly, high quality stuff will take a long time to fill up the stop, compared to low quality. So it might be many hours before you have even a 1-wagon train full of Q5 greens. Do you wait for a full 8000, or do you send partial shipments? And how do you decide how many to send?

[tall sandal]

i think the idea is stuff whatever fits and then have something like 5 seconds inactivity as the depart condition

[obsidian crescent]

But then you have multiple items in the cargo wagons. The generic destination trick only works with one of them.

[tall sandal]

how come?

[obsidian crescent]

Isn't that how the system was specified to work? You name all your receiving stations something like "<icon> drop". And then in the condition and destination name, you use the generic name "<generic> drop". And "<generic>" gets filled in with the icon of your cargo. But if you have multiple pieces of cargo, then it picks one of them in some way.

But that only gets you to one station.

[tall sandal]

I think it departs again and then gets interrupted again?

[winter saffron]

Yesn't

The drop off logic has to filter only the drop item

if the wagon has Y items and in the last slot X item

X interrupt will trigger after all Y are removed

[tall sandal]

honestly the whole thing is a little confusing to me and i dont think i'll fully understand it until i get my hands on it

[winter saffron]

essentially

  evaluate if wildcardDrop at slot[i] exists
    GoTo drop
  else i++

[hoary current]

if u have A and B in the same wagon, the interrupt system catch one of them and go to the corresponding dropoff, say A. After dropping off A, theres still B left in the wagon, and the interrupt system will go to the B dropoff because the interrupt is updated whenever the train leaves a station.

same applies for more than 2 item mixes

the train will go for one mix pickup and multiple dropoff

[half raptor]

Best way to view it is that and are treated as different items. Yeah you'd set up "cargo full or 5 sec inactivity" and then it would detect those two different items and would deliver to a station and then a order depends on the order you put your interrupts on the list

[cunning sleet]

Since quality is opt in, you don't have to design a system that handles RNG, I think? Well, you described a system that does upgrading?

I think what I'm gonna do, I'm gonna ignore quality initially, then I'm gonna build loops that produce everything of higher quality (but not the highest quality) and I'm gonna upgrade my individual sub factories with higher tier machines, in iterations

This way I can calculate how many machines I need in sub factories or what design they can use

Although that sounds boring, nothing really new in this approach from standard Factorio

[half raptor]

That's what a lot of people will do. Good strategy is to not do quality at all and just make quality upgrade loop for personal armor equipment until you have a base that can start handling quality

A good place to set up a quality loop is to try to produce

[daring siren]

We originally theorized about putting quality in every step, but it turned out too complex, and not worth it compared to productivity.
It seems like the consensus is putting quality where you can't put productivity, mostly in end products. Some argue for quality in miners, and working from there.

[opal haven]

completely ignoring quality at the start sounds like a mediocre idea - just don't use loops before mid-late game

Some argue for quality in miners, and working from there.
this one I 100% will try decently early in the game

[obsidian crescent]

In SA, you don't really have a choice about that, since recyclers are off-planet.

For me, it's easy enough to get quality ores, separate them, and then have a small setup for making a few key items with quality plates/plastics/etc.

Also, putting quality modules in the green and purple science precursors (except for rails and belts). That way, you'll get a healthy amount of mid-quality prod 1s for either direct usage or eventual consumption in making mid-quality prod 2s.

And your early space platforms will probably enjoy higher quality electric furnaces.

[opal haven]

yup, getting will totally pay off, while engi is spacefaring

[fallen urchin]

When do we get beacons in SA?

Without beacons, putting quality in miners means no eff1

[opal haven]

sounds like a reasonable tradeoff

[daring siren]

It's not too bad, you can also combine them

[opal haven]

the earlies reasonable amount of cheap quality stuff acquired, the earlier i can upgrade all my miners on starting patches to extract 30% more from them. sounds like huge stonks

[obsidian crescent]

I don't think Q2 or even Q3 miners will have a 30% depletion bonus. It probably goes up by 10% each level (if that), with Q5 miners being equal to Q1 BMDs.

It'll still be useful to take those drills to Fulgora or wherever else has thin mineral patches. But it's probably not going to be that easy to stretch your patches that much.

[opal haven]

looks like it's around 30%+
from vulcanus FFF:

Additionally it has a special unique property, the reduction in resource drain by 50%. This new stat means the resource underneath only gets its richness depleted half of the time. In other words, this directly multiplies with the productivity of the drill and you can imagine how crazy mining productivity can get. Better yet, the resource drain % is affected by quality of the drill, at Legendary quality a Big mining drill only drains the resource 17% of the time. Such a drill is mighty expensive but that resource patch you're about to place them on sure is going to last for a while. 

it's open to interpretation, bit if legendary big drill can cut 66% of ore consumption (200% more ore from patch), then even uncommon regular electric drill must be capable of reducing the number quite a bit

[opal haven]

I hope to stretch at least stone and coal, iron and copper gonna need some help from outposts anyway

[loud prawn]

Do we have access to all quality tiers from the beginning? I think there was something mentioned that they are behind some planet science locks?

[sage vapor]

You can get up to on iirc

[obsidian crescent]

You get up to Q3 at first. Q4 is on one of the first three planets, and Q5 is on Aquilo.

[opal haven]

, the rest are locked

[obsidian crescent]

It's going to be interesting designing a setup that won't break when you upgrade quality tiers due to unexpected materials coming out.

[opal haven]

which is quite cool - less stuff to deal with at the early stage

more reasons to visit Q4 planet when you're ready

[loud prawn]

I am thinking about easy sorting Q1 and rest, and just recycle all above Q1 till max or something...

[open halo]

That's a LOT of wasted resources

Tiers 2 or 3 are perfectly reasonable for a good chunk of the game

[loud prawn]

Or just prod and only recycling on endproducts

[opal haven]

I think goal for first like 10h+ is to get more stuff, and mostly ignoring anything above it

[loud prawn]

With max i mean max available stuff, so just make q2 to q3

[open halo]

You only want Q5 for the specialized stuff - armor and the like - since it takes so much to produce

[opal haven]

still wasteful imo, chances are not great

[obsidian crescent]

Are we talking about a point where the recycler is available, or is this before that?

[final nimbus]

It's been noted that focusing on Quality too much instead of just progressing the game is something of a trap.

[loud prawn]

Good question, not really sure if going deep into quality makes sense before fulgora

[obsidian crescent]

I figure that, by putting qual modules in miners, gathering quality resources is something that's pretty easy to do during periods of downtime. Waiting for your 4 nuclear reactors and attendant reactor parts to be built. Something to do while claiming biter territory and expanding. While designing your space platforms and waiting for their parts to be built. Etc. Quality resources can just build up on their own, and then you can use them to make whatever you want at the time.

[loud prawn]

And if Q modules on intermediates are worth the hassle compared to just prod and Q only at endproducts

The best storage for raw materials is let it in the ground imho

[obsidian crescent]

I don't really see much value in trying to blitz through SA. There are going to be a lot of parts of a run where you've researched what's available and are chewing on some infinite techs while dealing with a problem or starting on a new planet or whatever. Mining 5% extra resources that go into quality stuff isn't exactly a burden.

[loud prawn]

Inhave the feeling that its easier for me to just butt 3x eff1 into the miners like i am used to it and just try to make some quality stuff at the end of the chain, just to shift complexity a bit, but maybe i change my mind when trying 😅

[half raptor]

Gonna have two SA runs where in one I go full prod with a quality loop at top tier machines and another run where I attempt to go all in quality with recycle stations to eat the inevitable imbalances. As we research prod bonuses for intermediates, all of the math changes

Also early game to midgame don't really plan to do either qual or prod until I unlock tier 3 and the higher quality levels

I think the most sane route is to put your money on prod and perhaps transition to quality as your researched prod bonus exceeds the cap

Simple quality loops at the end product will give a decent number of legendary machines and modules if fed with a high quantity of intermediatea

[loud prawn]

...and would keep logistic really low

[half raptor]

Going for quality on every step is more of a challenge run... Like Doshington's beltless or all burner run

If you want early game quality though, make a bus that only puts normal on the bus, put MK1 quality modules into machines and have a filter splitter move any quality off to a storage chest. That way you wind up with a couple stacks of rares over time and you can make rare early game personal armor and defenses

As much as I want to put my faith in I'm still more confident in the power of

[obsidian crescent]

I don't see how this quality ore siphoning thing became a competition between prods and qual modules. It just doesn't compete. You weren't going to prod your miners anyway. And you're not shoving quality modules anywhere to turn base quality stuff into higher quality stuff. You're just taking the quality ores as inputs to the process.

[opal haven]

As long as you have only first 3 tiers of quality - logistic challenge should be pretty tame, especially if you ignore for the start, and/or immediately separate common ore. so, I'm leaning more and more to do some form of quality all the way as my first attempt, and starting before visiting first planet

[half raptor]

Same if I have the patience for it. My method is going to set up a city train grid with inputs and outputs on the inside. I'll design it once we get the new rail turns and elevated rails, but the idea is that a module requests on ingredients with a priority value assigned based on how full the station is so that trains will stay more or less balanced. I will then have 2 outputs, one for expected quality and then another for upgrades. With train interrupts being a thing, a train will just wait for inactivity before going off to deliver the upgraded intermediates to quality specific city blocks. In the event that a station is full because imbalance will be very common for a quality base then it would go to a recycle train station to junk the excesses and hopefully upgrade the ingredients. I have a few more strategies that will turn into a rant but ultimately the train interrupts allow for easy quality sorting and eventually as you expand horizontally you'll need more blocks for your uncommon production and maybe another rare

That one is for my quality save

Probably doing a prod save first go around

[loud prawn]

Sounds not so bad, but i have had negative experience with mixed trains and want to avoid a train for each quality

[obsidian crescent]

I wonder if there's a way to have a generic train schedule work for any quality other than base quality. That is, you have a stop for base quality ores, and you have another stop for all non-base quality ores of that type.

[loud prawn]

Pretty sure that works

Or do u mean mixed ore but same quality?

[obsidian crescent]

Same ore, mixed quality.

[half raptor]

You can totally have dedicated trains for upgrades but the way that the generic train stops can work, you really don't need to

[loud prawn]

Will work but "loading" will be harder u can not use "till full"

[half raptor]

You can tell it to go to dropoff if it has that first then as it goes down the list the dropoff is next and it'll drop that off. Just gotta make sure your inserters have filters set

[half raptor]

Full cargo or idle 5 sec

[obsidian crescent]

What I'm thinking about is having all non-base quality ore go to a single stop, where it will be sorted and processed accordingly.

[loud prawn]

Yes and than it clocks when one unload station is full

[half raptor]

Can do that too

Love a game where there's multiple correct answers

[loud prawn]

On the other hand, just sorting it after processing at least to plates might be easier

[obsidian crescent]

The way quality stacks makes it difficult to run a non-uniform quality belt through a furnace stack. At the very least, the stack would be less efficient than if the belt had uniform quality.

[loud prawn]

True

What endproducts do you guys think gives big advantage in early mid game when it has more Q?

[obsidian crescent]

The best advantage early-on might be space platforms. Quality thrusters, quality asteroid collectors, quality crushers, and quality chemical plants/furnaces could combine to make for much smaller or much faster platforms. And quality gun turrets of course. And maybe quality solar panels.

How many arms does a base quality asteroid collector have?

[loud prawn]

Oh one arm more per quality seems extremly good yeah

Solar should be enough to just have quality in the endproduct assembler, and use the normal ones on nauvis

I have a feeling that big power poles will make the start on fulgora much easier

[obsidian crescent]

But that would require building a lot of solar. Then again, you do have to make satellites for a while...

[loud prawn]

I use normally a lot of solar, so getting about 10-20 rare should be quite easy

Thrusters and crushers will be much harder without piling up a lot of them

Lot of different ways and no real silver bullet, i think they do a real good job 😎

[obsidian crescent]

One of the biggest advantages of any intermediate-based mechanism for quality generation is that you get to pick and choose which quality things you want to make, rather than being limited to only the end products that you're making a lot of. This is especially important in the pre-recycler phase.

[half raptor]

Putting any quality products into a storage chest does indeed let you focus your rares into making rare space platform and personal armor pieces

[loud prawn]

Definitely true, but working with backpressure and quality seems hard and you will have to manage the stockpile

[half raptor]

If your stockpile fills up, remove the modules

[loud prawn]

Still manageable you can just makw a lot of storage too

[half raptor]

Granted it requires so it's a post oil problem

Have they said anything about cargo wagons having more storage with quality? I don't know if they're getting a bonus beyond hp

[loud prawn]

Nothing yet, i think there was also nothing about trains itself

[obsidian crescent]

I'm pretty sure they said that containers won't get storage boosts with quality.

[obsidian crescent]

One steel chest can hold 2400 plate. At 20 Q2 plate per minute, it will take 4 hours to fill up. Use two of them, and the 8 hours it takes should be long enough to research requester chests and make the problem go away entirely.

Once you can just request stuff, it'll be easy enough to say "if I have too many Q2 iron plates, make some Q2+ green circuits" or something. And building that circuit logic will be good practice for doing the same thing on Fulgora to manage stockpiles.

[loud prawn]

And then the inbalance begins with one step more from copper cables and so on... in theory I like it but i think it's just drastically more complicated compared to focusing on specific endproducts with recycling loops after fulgora visit

[obsidian crescent]

As I pointed out, you're going to have to build that setup on Fulgora to manage scrap products. Maybe as well build v1 on Nauvis where the consequences for failure are much lower.

[half raptor]

When the imbalance occurs in the early game (pre recycler) just remove the quality modules for a while from that production line

[loud prawn]

I would manage fulgora the other way around, recycle blue when low on red, instead of recycle blue when too much blue because of compression

[obsidian crescent]

You have to do both. If you need reds, you need to recycle blues. But if you have too many blues (somehow?), you have to get rid of them. But also, if you have too few blues, then you can't just recycle them to make reds because you need some to do normal blue manufacturing.

It's the same idea as before, just with a different input/output relationship. If you need X, you must consume Y to make it. If you have too much X, you have to get rid of it by producing Y.

[loud prawn]

But maybe it gets much more parameter

[half raptor]

Yeah, Fulgora is the oddball where even people going prod focus is going to have a good recycle system

[loud prawn]

Yeah i love the idea of fulgora and in my imagination it is the planet for me to really start making modules, on nauvis just eff1/eff2 and prod for rocket

[half raptor]

Vulcanus is more efficient ore processing (there it isn't ore, just lava) Fulgora is where we get more efficient electronics and I'm guessing Gleba is going to be efficient fluids

[loud prawn]

Fluids and chem stuff like plastic i think

[half raptor]

We might go peat mining

[loud prawn]

I we think we will have at least a (quality) mall on each planet

[daring siren]

For planet exclusive buildings for sure

[half raptor]

I'm hoping we get a space elevator at some point as an alternate way of "launching" materials

Most of the reveals have been midgame reveals. No clue what we will get on last planet

[obsidian crescent]

With rocket part costs going from 340/600 iron/copper ore (with no prods anywhere) to 18/40 iron/copper ores with EMP/Foundry processing and mid-quality prod 2s, I don't see why we need an even cheaper way to launch stuff. Put Q5 prod 3s in there, and they're down to 4/10 iron/copper ores per rocket part, around 2-3% of the original costs.

[half raptor]

I just think it'd be cool

[loud prawn]

Still no idea which planet first after nauvis 😅

Excited what gleba offers

[daring siren]

Missing just one

[final nimbus]

might be the closest planet, and if the basic ores are present it could very well be the least difficult to figure out.
One of Gleba's major draws might just be that it's easier.

[daring siren]

has enemies. probably too. is devoid of life.

[obsidian crescent]

Gleba will never be as easy as smash-and-grab on Fulgora. That's where you bring along (the components of) a rocket silo, go down, mine just enough holmium to make 100 EMPs, maybe build 100 recyclers with local materials, and peace-out to return later to make its actual science packs.

[final nimbus]

You could, but why not stay a little longer and set up science if you need to do that anyway?
At least set up bots and a basic mall so you don't have to go back, then you're all good.

[obsidian crescent]

Bots can't expand across the oil seas. Doing any significant science is going to require pulling scrap from a lot of places. So you'd need to come back if you're doing smash-and-grab from only a couple of islands.

[final nimbus]

Save yourself the trouble of running between planets all the time and tap a few rich deposits.

[daring siren]

Stay a while, and research!

[obsidian crescent]

Yes, but then it wouldn't be a "smash-and-grab". The whole point is to get a good payoff from a minimal time investment.

I have no idea how effective something that this would be for getting through the game quickly. But sprinkling a few EMPs on Gleba could speed it along if the place isn't exactly overflowing with ore.

[loud prawn]

100 emp s will take some time i think

[half raptor]

Don't try speed running your first time through. Stop and smell the mycelium

[frozen fiber]

Quality belts help with quality labs and later, once you have the recycler, it's a good source for quality gears and iron plate. You craft tons of them. Seems like a logical place for quality modules to me. No?

And the quality tiers you don't want you can use for green science or red belts no problem

Those only take one yellow belt to craft

[half raptor]

Remember too that we can craft quality sciences. Each rank adds 1 pack worth and legendary adds 2

[frozen fiber]

Sure, but green science is so cheap, I'm not sure I'll bother

Purple science on the other hand might be worth the trouble. It would be a lot of trouble though

[neat shard]

You can, but... why

[final nimbus]

The "optimal" way to go about things is to use where- (my connection briefly dies in the middle of typing this) - it's allowed and everywhere else.

[gusty trench]

It makes sense for green, gray, and purple science

[frozen fiber]

Purple will (or would, i guess) be pretty complicated though, what with the 30 rail of varying qualities and also the furnaces and prod modules you wanna syphon off.

[obsidian crescent]

You wouldn't bother qual-ing rails. And siphoning off higher quality stuff is as simple as just having an appropriately filtered inserter dumping them into a chest nearby.

[frozen fiber]

Then you can't make quality purple science

[final nimbus]

Chest fills up. what do?

[neat shard]

The main problem with trying to do high quality science is you're just asking to get deadlocked when you get quality items in the wrong ratio.

[frozen fiber]

Which I was talking about

[obsidian crescent]

... even ignoring that it would take 2400 prod 1s to fill up the chest, just use an active provider.

[chrome mauve]

wrong ratio is problem for any player that didnt left newbie phase of Factorio

[frozen fiber]

If you use a bit of circuitry and recipes, you can combine different quality tiers and probably make it work while still getting some benefit

The higher tiers are so rare as to not make much of a difference to overall science yield anyway

But it's still a hassle

[obsidian crescent]

The problem is labs. Different qualities don't stack, even in labs. So if there are only Q2 packs waiting for insertion, the lab will have to burn all of the Q1 packs before picking those up.

[frozen fiber]

Depending on how you feed your labs, that's no problem

It wouldn't work with sushi belts though

If you use chests it should be fine

[frozen fiber]

I calculated it. Assume the assembler makes 25% quality products. If all of them were q2 (I know that's not actually possible) you'd get 25% additional science yield. If they're q2 through q5 in the known ratios, you get 27.8% additional science yield. Not much of a difference

[iron root]

and nowhere near as effective as the +100% you could get by swapping those quality mods for productivity mods

[frozen fiber]

That's not possible for green, grey and purple science

I was talking about the assemblers making the ingredients, not the science pack itself

[iron root]

i suppose that yeah, sciences that are assembled entirely from items that can't be directly prodded can have assemblers meant for quality materials to make quality packs

that does make the siphoning game a bit more interesting though

cause if youre pulling quality inserters for the factory, but not quality belts, then theres a net imbalance ther that needs to be compensated for

[frozen fiber]

Unless you use the recipe for science. Then you're all good when it comes to green science

If you don't want quality science that is

[iron root]

i imagine it'd be a sequence of like, science -> science -> science machines

really just haivng a at the end to mop up imbalances (maybe it should go -> -> ?)

[tall sandal]

I think you'd need to loop the inputs

[frozen fiber]

If only assembler needs to deal with the quality ingredients you can put them in a chest

[iron root]

there's a non looping solution in here i think but it might get a tad messy

[frozen fiber]

The problem is that the quality tiers appear randomly. That can clog up the belts. In a chest there's a much larger buffer, which will most likely prevent deadlocking the whole thing

[dense kraken]

[neat shard]

Of those, may be a junkyard, but is the one that really got trashed

[burnt reef]

it deserved worse

[sage vapor]

I see no difference between this

and this:

is the final planet!

[tall sandal]

i want those science emojis!

[sage vapor]

those are all from the seablock discord 😅

or do you mean the new ones for SA?

[tall sandal]

i mean those ones

[cunning sleet]

Purple but pinker, green but darker, yellow but more orange

These are the official names of sciences

[kindred crater]

devs have been really quiet since fgardt dropped this

[dense kraken]

coincidence? I think not!

[half raptor]

long live Bwuhuo

[pure spire]

wait what happened wdym trashed

[jaunty citrus]

long live Bwuhuo

[pure spire]

just that the name isn;t what we thought it was?

same planet right just not the same name that we thought it would be/renamed from what it used to nbe?

[kindred crater]

the devs are just confused, thats all

[burnt reef]

"Bwuhuo" has finally been revealed as the lie it's always been, yet its cult persists

[frank warren]

haha yeah

[obsidian crescent]

It's not that simple. From the FFF, they made it clear that the planet used to be more foresty, but is now much more fungal. That is, the name was changed because the nature of the planet changed.

This is also part of why it's so WIP; when you change the nature of a planet, you have to change a lot of graphical stuff to properly present that nature.

[pure spire]

yeah

i wish it was more of a jungle rather than poisionous waste

[jaunty citrus]

What makes you think it's poisonous waste?

It's more fungal/coral inspired growth

[pure spire]

well

its green and purple in some spots

[jaunty citrus]

Pretty much the opposite of 'waste'

[pure spire]

well rather less wasteland more thriving with icky

[jaunty citrus]

Well apparantly it started as more of a jungle theme but the gameplay/vibe was too similar to Nauvis, so they changed it

Seems pretty good from what they showed us, and now I understand why they waited to show us this one last. (I don't expect to be shown anything about Aquilo but that's just me)

The amount of stuff you need to make to pull of to make this style planet work is pretty immense.

[charred gyro]

I'm looking forward to them revealing the major features of Gleba, and how acquiring/processing resources will be unique compared to the other planets. It looked like copper, iron, and coal stone were all present as natural deposits so I'd think that would mean the same processing chains as Nauvis would be available. So maybe it's just the oil-related stuff which is different?

[burnt reef]

I don't think so. In the video, I only see stone and copper patches, and they're not that big. I think there's another source of basic resources.

[charred gyro]

Er, I meant stone instead of coal

I also heard that several ore patches got deleted to make the video more cinematic

[burnt reef]

That makes sense.

Even so, "the same processing chains as Nauvis" seems to be something the devs are trying to avoid, given what they said about Fulgora.

And, again, the resource patches we see are small. Maybe they're there to start us off, before we build the planet-specific infrastructure?

[charred gyro]

That's why I'm looking forward to the reveal so much. I didn't really come up with any ideas for unique systems that include iron/copper/stone as raw ore patches. I'm not convinced that the smaller sizes we saw are reflective of the actual sizes, or that it'll be a limitation. So I'm curious how they'll end up making things different enough with those resources available.

[jaunty citrus]

I expect some sort of weird biological catalyst type processing tree to get more minerals than what are immediately available

the patches do look really small

it's very possible though they turned down the patch size to get a nice path for the cinematic

No coal and from what we can see no oil is quite interesting though

Lots of interesting stuff to be done with those

[burnt reef]

I trust the devs to make it good. I can't wait to see what they've come up with.

[jaunty citrus]

"I trust the devs" is a pretty safe take in this case I imagine

[burnt reef]

it's worked out so far!

[heavy ember]

The list of companies that I dare to do this on is dwindling fast.
But wube is still on it

[neat shard]

I'm sure there will come a day when Wube decides to start liquidating the community's trust in them. (Tis the lifecycle of all good brands.)

However, today is not that day.

[kindred crater]

at least so far, theres been no visible indication of that lol

I don't agree with all of their design choices, but the way their dev team works is pretty unprecedented for games, and that definitely earns my trust + respect

[unborn flax]

Picking this up, I was thinking about builds. The motivation is that EMP stages that take prod modules are much better at quality cycling than those that don't. So the idea is to quality-cycle the intermediates that go into a to tier X, then do the rest of the cycling to in the final stage.

This is the number of outputs produced by starting with 1 input and:

• quality-cycling to the given tier in an EMP with a recipe allowing prod
• then quality-cycling to in a chain of 3 EMPs with a recipe not allowing prod
  0.591599903180316
  0.642393256049745
  0.610934717543026
  0.463586474789975
  0.255026671443717

This says that the optimal level for quality cycling of intermediates is Uncommon, which I find pretty plausible.

There's little difference from instead early cycling to or , the former meaning no cycling at all. But from upwards you start to regret wasting an EMP slot on a qual module when it could have been a prod module, given you're about to go through three EMPs full of qual modules.

The number is equal to 1/13.2339099314357 * 1.5^3, where the 13.23 is recognisable as the number of needed to produce one by cycling in an EMP that take prod modules.

If you only want to do quality in the very last step then the number is 0.2055651166784894 (which is 1/30.0999512951288 * 2.75 * 1.5^2, where 30.10 is recognisable as the number of needed to produce one by cycling in an EMP that does not take prod modules). This is only a third as efficient due to the fact the EMPS for prod1 and prod2 don't contain qual modules. The main benefit of this strategy is simplicity - you won't need to provide mixed-quality intermediates for the other module ingredients, RNG-permitting.

[jaunty citrus]

This will probably get more complicated when considering the "extra" ingredients that T3 modules will probably require in Space Age. Current example being superconducter wire in .

[sage vapor]

modules are a 1 time cost tho

[jaunty citrus]

Yes, but you need them in every build and playing the game means expanding/building.

You'll always need more so long as you're actually playing the game

[unborn flax]

The superconducter wire is one of the intermediates you'll have to quality-match to 4 levels - yes. Though there's a sense in which it's simpler than say , which you'll need to produce at all quality levels in multiple different recipes along the way.

[obsidian crescent]

Red circuits are easy though. Blue circuits have productivity research, so quality cycling them becomes increasingly preferable. And if you manage to hit productivity level 13, then you hit the productivity cap and blue circuits of any quality are now just the recipe cost of a blue circuit (ignoring sulfuric acid). Which means quality red circuits are pretty cheap: make blue circuits, then recycle them into reds.

Now, if one of the planet-specific intermediates have a productivity research, then that could be a target for cycling.

[jaunty citrus]

Lucky for us, sulfuric acid is easy to ignore on vulcanus

[unborn flax]

You may have ruined my day (in a good way). I was about to say something like "yes but that's very late endgame". However I suspect your strategy may become useful much earlier than you suggest as it doesn't have to be almost free, only cheaper than the normal way of quality-cycling reds.

Starting with normal ingredients for reds, assuming no infinite research for reds and X% infinite research for blues, I believe the point at which it becomes cheaper to produce reds by recyling blues is 15.4% infinite research for blues:

"Normal quality cycling of reds with P4Q1 in an EMP": 0.0755634582055459 legendarys per normal input (which is 1/13.23, i.e. recognisable)
"P5Q0 in red EMP, then quality cycling of blues to legendary with 15.4% bonus prod, then recycle back to red": 0.0755674043798983 legendaries per normal input

I need to go now but I'll double-check the numbers later. They're a bit off in a couple of ways: 1) I have crafting as the last stage of the blue quality-cycling instead of recycling, so I'm missing the opportunity to have qual modules in the final recycler. So the real threshold will be less than 15.4. And also 2) you have to 'borrow' the other ingredients of blues (i.e. the green circuits); these are up-cycled to legendary as a by-product but there's (maybe?) always a use for legendary greens.

[fallen urchin]

Another problem is, crafting things take time, and currently setups use speed in beacons

Mixing qual modules with prod means you can't have speed modules anymore, which means you need a lot more machines

[half raptor]

[neat shard]

Only the finest quality garbage

[iron root]

huh
that does make me wonder how the hell quality is going to be integrated on gleba
I guess because factory buildings sort of have to be non-spoilable, it's fine, and it just means you can't buffer quality components for later, you'll need to immediately get them to the quality level needed and then make the building

the inability to buffer does mean that the rate at which you can make quality materials is extremely important for gleba, as you can't just accumulate them through a low-chance-high-volume case like on other planets

[daring siren]

Kovarex touched on this on reddit

There are non-perishable items requiring perishable items as ingredients

When you recycle they may gain the spoil bar again

[chrome mauve]

my first idea would be if I wanted anything that have spoilage just make it one demand - separated production from rest of factory with intent that o lot of it would be lost

[heavy ember]

At what % would be my question.

[winter saffron]

I assume recycling into a higher tier won't reset decay. Therefore one would need to put the recyclers as early as possible to roll for high quality

[daring siren]

100%. K answered that

[heavy ember]

Oh.
Well that's a way to refresh sub 25% items back to 100% and gain something

[unborn flax]

Slower spoilage for quality items would be cute. This could give a reason to make quality Bwuhuoic science.

[chrome mauve]

I think it was already confirmed that higher quality have larger time till it spoils

[jaunty citrus]

Yup #friday-facts message

[unborn flax]

Excellent - so the two systems are joyously intermingled. Did I fail to RTFFFF again?

[winter saffron]

I doubt there's a way to filter for spoilage rate.

[loud prawn]

At least inserter can pick the lowest/highest spoilage

[winter saffron]

wha?

lowest is 0 = spoiled item and highest is freshly created.

Filtering for spoilage would require additions to the inserter UI

I doubt they'd add that only for this purpose

[jaunty citrus]

Well, they did

[winter saffron]

^ source?

[jaunty citrus]

#space-age message

https://media.discordapp.net/attachments/1215073493142474833/1248621594729910302/image.png?ex=666454fa&is=6663037a&hm=7f535a60e7f5a4d9e0ee7556d4d501d910831c78770cf86406b0e6f7a8a3a82f&=&format=webp&quality=lossless&width=726&height=905

[winter saffron]

So, they choose between left/right belt. I remain sceptical about its usefulness.

[jaunty citrus]

left/right belt?

it chooses the most or least spoiled item

[faint blade]

I think it's mostly for chests or other inventories

Belts would probably be a performance nightmare given that inserters are one of the bigger UPS hogs already

[unborn flax]

Quality gleba science looks mostly not worth it. From a back-of-an-envelope calculation, making reasonable but unconfirmed assumptions that science ingredients are made in a 4-module machine with 0% bonus prod, and that the durability of items increases by a factor of the normal quality multiplier (i.e. 1.0 to 2.5):

becomes worth it when science is 81% spoiled.
becomes worth it when science is 118% spoiled.
becomes worth it when science is 148% spoiled.
becomes worth it when science is 170% spoiled.

I can show working if people are interested.

[kindred crater]

so it seems like all of the starter planets are good for in different ways

especially with petro products for

even if it takes up more real estate to do it (farms, processing, etc), you can't argue with truly free materials

[sage vapor]

free power aswell, just spoil stuff and burn it

[obsidian crescent]

The problem is that you can't recycle plastic (except into itself). So your options are to get quality ingredients and/or use quality modules. But since the ingredients only have one step (that's probably non-recyclable?), you're not going to be getting much out of them.

[frank warren]

I can. I think all materials are free.

[kindred crater]

with zero need to ever expand

[obsidian crescent]

Even if the inputs are free, you're going to be stuck with a lot of low-quality plastic.

[kindred crater]

all of that can be recycle looped for better quality

which of course will require visiting and grinding out enough good but thats another story

[obsidian crescent]

But if you've already been to Fulgora, there's a much easier way to get quality plastic. Quality cycle grenades to get quality coal, then feed those into chem plants.

... good point, no need to get iron involved.

[kindred crater]

maybe we're weighing the good and bad things differently. the fact is actually all free, unless its grossly more inefficient, I'd want to maximize. plus, plastic from bio plants could very well take advantage of productivity research

[jaunty citrus]

Because if you quality cycle coal itself, you only get one quality step per recycle step, if you craft them into grenades, you get to attempts to raise quality per recycle

yes you finished typing before I did, well done

[unborn flax]

i think coal made through grenade cycling has 8.8x efficiency of pure coal recycling, though this ignores the 'wasted' iron that also gets bumped to

[daring siren]

Only reason to make is for though

Or for explosives I guess

[half raptor]

is important for a lot of things

[kindred crater]

explosives too to a lesser extent. if you can get a bunch of rare/epic explosives from some setup here, you can have a nice amount of powerful rockets/tank shells for personal use

[frozen fiber]

So, since the foundry recipe for LDS was changed, we won't be quality cycling them to get legendary ones, I guess?

But instead we only need legendary plastic

[frank warren]

:( seems so yes

yeah that is also true

[frozen fiber]

That changes things

[frank warren]

problem: our solution to the plastic bottlekneck was, in part, LDS.

[frozen fiber]

Makes it easier in a way too

You might want to make legendary plastic on gleba

It's basically free there

[frozen fiber]

Considering how little plastic lds use, and how long they take to craft (and presumably, recycle) this change, and gleba's recipe, might make it ultimately easier

[daring siren]

or we just don't make rockets on .

Bring everything and research there

[frozen fiber]

Don't you need a rocket to get off the planet, at the very least?

[frozen fiber]

But that probably also means that you can't make quality plastic this way on any other planet

Hm

[daring siren]

You can make from , on other planets

[frozen fiber]

Yes, but coal is not infinite

If you set it up on gleba, it will keep working forever, no need to expand

But true, we can of course make quality plastic elsewhere

[fallen urchin]

I wonder whether we can bring those soil tiles elsewhere

[daring siren]

It's been suggested that the green area on is plants

[frozen fiber]

By whom?

[chrome mauve]

Im on Phone and cant look back at That FFF

But it can be possible that other planets have their own thing

[daring siren]

By V 🙂
#friday-facts message

[frozen fiber]

Ok

I'm not sure if I'm gonna believe that

[daring siren]

I mean, he said "ecologic center" as joke, but considering the shape and the color?

[frozen fiber]

Could be the truth camouflaged as a joke

Who knows

We'll hopefully know by the end of the year

[daring siren]

By "October at the latest"

[frozen fiber]

I'm aware that he said that

He might be wrong though. It wouldn't be the first time

I'm sure you recall how many times the original release was postponed

[daring siren]

I know haha

but eventually the game was preponed by a couple of weeks

[frozen fiber]

Still

[daring siren]

They honestly don't seem that close to release

[frozen fiber]

We'll see. I expect it to come out this year

But if it doesn't that's ok too

There's not really any rush

Only that I'm a little impatient to finally pay it haha

[frozen fiber]

It's really hard for me to tell. There's a lot that can be done in four months

[obsidian crescent]

Also, plastic doesn't spoil, so as long as you use the ingredients before they completely spoil, plastic production won't care much about your spoil times.

[frank warren]

yeah.. plastic is very notorious for not spoiling

[hoary current]

iirc theres prod reseatch for plastics

[fallen urchin]

Maybe spoiled ingredient gives fewer outputs in that case, so it's still important to get the items there in time?

[obsidian crescent]

They said nothing to suggest that this would be the case.

[daring siren]

Spoiled ingredients are just and can't be used for the recipe

We do know that bioscience loses durability with freshness

[chrome mauve]

when I was at work I wondered if durability loss is proportional to its freshness or lets say that we can lose like up to 50% of durability when it comes to almost 100% of spoil

[daring siren]

It's linear

[iron root]

...this is the second time ive heard of a new foundry lds recipe
where can i find this info

[obsidian crescent]

I think JG mentioned it. Basically, the Foundry's LDS recipe now takes molten metals instead of plates. We don't know the exact amounts of molten metals though; it could be less than the original recipe ratio (like casting steel).

[iron root]

huh
yeah that has some interesting quality implications

between not being able to recycle the lds for plates (possibly, depends what wube makes the recycle recipe), but more importantly needing only quality plastic to make quality lds

[obsidian crescent]

Maybe Gleba can export quality plastic to to Vulcanus and Nauvis for use in making quality LDS.

[obsidian crescent]

You can recycle iron gears despite using molten metal to make them.

[fallen urchin]

Will quality be important in rockets?

[iron root]

for making rockets to launch stuff probably not

but who knows

[daring siren]

Exactly. And you can recycle it to the main recipe i.e. to plates.

We figured that you can upcycle the free on and then mix it with molten iron and copper to get , and then recycle it for and

[frozen fiber]

Depending on how much of a problem energy will be on vulcanus, I might want to make all kinds of quality items there

Except plastic, of course

[gusty trench]

Coal is fairly plentiful, it's honestly not a bad place for quality plastics. You need coal for all your oil needs, so quality miners can send regular coal to oil processing and quality coal to make into plastic

[frozen fiber]

I just feels kinda wasteful to me to quality cycle using finite resources, even plentiful ones

Also, you'd lose all the petroleum gas in the recycler

[gusty trench]

Plastic recycles to plastic, not coal

[frozen fiber]

Ah, true

The first point stands though

[frozen fiber]

If it can be avoided, like in this case, apparently

[gusty trench]

The main point is that you can start with at least green quality plastic

You can alternatively simply up the quality level of the coal you use for making quality plastic. Throw green coal into oil processing when you only want blue or better plastic

No quality cycling at all

[frozen fiber]

That strategy would result in a very slow trickle of legendary plastic

[loud prawn]

Oh is this an additional recipe or was it changed to only molten metals?

[obsidian crescent]

It was changed, so it's a full-fledged alternate recipe.

[gusty trench]

So would quality cycling for legendary plastic

[obsidian crescent]

It would be better to quality cycle grenades to get high-quality coal, then put that in your chem plants for plastic.

[frozen fiber]

But you could scale that up as much as you want

[loud prawn]

Thx

[frozen fiber]

Why would that be better?

[obsidian crescent]

Well, "better" in that it's faster. You consume more resources, but you get higher quality more quickly than with skimming.

[gusty trench]

Making coal into grenades and then recycling those adds another step that can add quality without losing material

Recycling plastic voids 75% of it every time you try to add a step of quality. Coal -> grenade -> 25% as much coal has 2 steps of possible quality bumps

[frozen fiber]

I'm proposing to build a dedicated legendary plastic setup on gleba, not skimming

I guess it depends on whether you prioritise speed or efficient use of resources

[iron root]

Re quality cycling for plastic
It might be better to leverage 300% prod via emp/recipe prod to quality cycle blue chips, then recycle to red and then plastic for quality plastic

Sulfuric acid is free on vulcanus, after all

Would be overall a 75% loss to get to legendary plastic, cause red chips cycle at no loss due to being a component of blue chips

[frozen fiber]

What du we need legendary plastic for anyway?

[kindred crater]

advanced circuits, explosives, low density structures

[frozen fiber]

We can use it to make legendary LDS, sure. But what is that used for? If we need it to obtain legendary copper and steel, then fine. Because that process uses up (practically) no plastic (assuming a productivity close or equal to 300%)

[iron root]

Legendary lds could be used for damn near anything right now

There’s so many high tier buildings that are post lds unlock in SA

And anyone whose played SE knows how much E like making lds an ingredient in things

[obsidian crescent]

It would make sense for a lot of space infrastructure to use LDS. Asteroid collectors, crushers, thrusters, the hub, etc.

[neat shard]

Is it actually going to be practical to hit 300% prod in practice?

If recipe prod research cost scales exponentially – and I fully expect it does – that could take staggering amounts of science

[kindred crater]

138 IRL days @ 1k spm to reach lvl 30 for a single productivity tech #friday-facts message

[tall sandal]

Level 30 prod research costs 200M science

[kindred crater]

however... in labs and lab productivity research can dramatically slash that

[neat shard]

Is there going to be lab productivity research in SA?

[kindred crater]

yep, which is some of the most powerful research that we know of so far

I'm sure it gets stupid expensive, but apparently it doesn't have a limit either

[iron root]

we're thinking for EMP/foundry with it will be

especially EMP

[burnt reef]

for the EMP, it'd require 13 levels, which is ~1000x cheaper

so like 3h20m @ 1k spm

hey, is prod research per recipe or per item? do they apply to alternative recipes?

[kindred crater]

I don't think thats been clarified by devs yet. but at least the way it appears based on the wording and how its been described, its for the item in general

[burnt reef]

hm. then, if Foundry productivity is uncapped, it might be more efficient to use it for cable/gear casting in the very very very long run.

you know. once productivity gets above +1600%.

wait no I did my math wrong, it still isn't, sorry

[kindred crater]

productivity for crafting actual items is all capped at +300% so you can't have net-positive recycling loops

I just had a moment of self-reflection... I don't think I've ever memorized so much stuff about a game before that hasn't even come out yet

[burnt reef]

right, but foundry recipes are non-recyclable.

[kindred crater]

hmmm

[obsidian crescent]

Items are recyclable or non-recyclable, not recipes.

LDS is, as far as we know, recyclable. And that's really important on Fulgora.

[burnt reef]

what I mean is, recycling the output of a foundry recipe does not give you its ingredients back, so a recycling loop isn't possible.

(at least for a lot of foundry recipes, if not all of them.)

[kindred crater]

if your foundry is making copper cable, I think that just means recycling it would yield copper plates, even though that might seem strange

[obsidian crescent]

It doesn't give you its molten metals back, so yes, you'd have to send what it returns to an assembler.

[burnt reef]

and since you can't use a foundry recipe as part of a recycler loop, you don't need to cap its productivity to prevent a positive loop.

[obsidian crescent]

All buildings have the same cap. It's a global thing.

[burnt reef]

okay. I didn't know that for sure.

[fallen urchin]

If mining productivity don't have much higher prices in SA, then you can reach far higher levels than in 1.1, thus making coal far more plentiful than it seems

1kspm for MP probably won't even need to mega in SA

[obsidian crescent]

Yes and no. It's important to remember progression in SA. You get prod 3s off Nauvis. And quality 4 is off Nauvis as well. All of the cool super-productivity stuff you can bring back is off-world.

While you can eventually start really burning through many levels of mining productivity, it will take a while for that to happen.

[fallen urchin]

Patches draining probably isn't that big problem before that

[obsidian crescent]

It is when you're not around to expand 😉 Not unless you decided to overexpand before leaving.

[daring siren]

BMDs though. 50% consumption at normal quality

[chrome mauve]

securing resources seems like lesson that people need to learn either by thinking whats happening in their factory or by experiencing shortage of ore

[half raptor]

Sirens and alerts is a great way to measure your buffers. If my ores on average get to about 40% capacity, I have an alert go off. Thing about buffers is you're either tending toward 100% full or 0% and catching it on its downward trend is a great way to stay ahead of your problem

One reason why I'm loving because you can sort your demand signals in order of priority, especially if you set a conditional combinator that says to only output said demand signal if the ingredient demand is low.

[obsidian crescent]

Here's an interesting thought. If you're making 500 SPM of science (pretty reasonable after completing a planet or two), then you're making 500 purple science per minute. That involves making about 120 (factoring in productivity) prod 1s and electric furnaces. If you can get 5% quality in those processes, then you can skim off about 6 Q2+ prods and furnaces every minute.

But with recyclers, you can recycle them to generate quality versions of their components, particularly quality red circuits. Even with the 25% penalty, the furnaces each cough up 1.25 quality red circuits as well as a bunch of quality steel and stone brick (for concrete or oil refineries).

So that's 7.5 quality reds per minute. Not too shabby.

[final nimbus]

An endgame build could have up to a 31.25% quality chance using EM plants with
Having to filter out nearly a third of them does force you to scale up production somewhat. Aside from recycling, it wouldn't be a bad idea to use some for directly.

[fallen urchin]

The problem with is it's slow, and qual will exclude speed modules

[final nimbus]

Another thing I thought of related to the fact that are not cheap;
Unless you are so far into post-endgame that it doesn't matter anymore, you want the machines with your best Quality Modules (or any of them really) to be running as much as possible, yes?
That means trying to prevent them from backing up and stopping. If you aren't researching something that uses and it backs up, keep the machines running by diverting the output to recycling.

[fallen urchin]

For endgame setup, it might be better to loop

are not cheap, and I think this means they should be put on those high throughput recipes first

[final nimbus]

This is more for a midgame build by a hypothetical player utterly obsessed with quality.

[obsidian crescent]

But they wouldn't have Q5 qual module 3s.

[final nimbus]

And?

[obsidian crescent]

The funny thing about high-quality qual modules is that having more of them makes them cheaper.

[fallen urchin]

That's true

[final nimbus]

Come to think of it
If you really want to maximize how much quality stuff you make, it would help to divert anything that starts to back up into recyclers fitted with

[fallen urchin]

Having better qual modules makes it cheaper to get high qual things, and this affects themselves

[daring siren]

Remember that EMP is also faster than normal assemblers, and quality EMPs are even faster

500 SPM is definitely not a midgame base, at least not in 1.1.

That said, with the new infinites, it means going for early high SPM of some sciences may be worth it

[obsidian crescent]

What's interesting is that some infinites require space science and some don't. So you can design your base so that it has higher SPM output if you're not launching rockets for space science without increasing your resource consumption.

[daring siren]

Technically is a tech more than anything.

Of all sciences, that's kinda the exclusive tech because of uranium.

The "main" 6 sciences are generic, then the 4 planetary sciences (+space as a planetary science) and final post endgame science

[final nimbus]

What if i just launch a ton of 🛰️ and ignore uranium?
Then I don't need Nauvis

[daring siren]

That is correct, but expensive

[final nimbus]

No-uranium Space Age speedrun category

[obsidian crescent]

Doubly so in that you can't use any productivity outside of the rocket itself.

[fallen urchin]

The rocket itself become much cheaper, so the cost of 🛰️ is relatively significant

[daring siren]

🛰️ isn't cheap. 100 100 100 5 radars 50 100 .
But I see that as a good thing. From all those and , some are bound to be high quality, good for your first platform.

[fallen urchin]

Do boilers work in space?

[daring siren]

No

Burner entities for a bit of realism as there's no atmosphere in space.

[half raptor]

Only reason why nuclear works in space is that you don't need oxygen or atmosphere to boil water via radiation

[tall sandal]

i feel like burner entities should work if you use rocket fuel

[gusty trench]

You probably wouldn't want boilers in space anyways given the low power density

[opal haven]

do we know, that satellites will have same cost?

[daring siren]

Was suggested that yes

[opal haven]

quite cool then
a bit sad, that we not gonna launch a lot of them - steady trickle of would be nice

[dense kraken]

factorio's vanilla rocket fuel doesn't contain an oxidizer tho

[tall sandal]

its slightly abstracted. the oxidizer comes from the air during manufacturing. clearly rocket fuel works in space because the satellite uses it as fuel

[dense kraken]

damn there's rocket fuel in the satellite

u right

I'm glad I put dinitrogen tetroxide in my recipe...

[half raptor]

You want realism? Quit putting iron into your electronic circuits

[median verge]

Did i miss something? Source?

[daring siren]

Space science in space requires U235, which is likely exclusive to

[obsidian crescent]

That is, you can make space science in any (space) location, but for all useful purposes, that platform either is parked in Nauvis orbit or will have to make regular trips there.

[dense kraken]

Place an offshore pump by the shores.
Now get water, concentrate salts using reverse osmosis then distilate the result to get sodium chloride, using high temperature electrolysis you separate sodium from chlorine, and with some pure water back from the reverse osmosis you got yourself hypochlorous and muriatic acid, but you're still missing the most important part: oil. By setting up a pumpjack you get some dirty oil that you must wash before performing first a fractioned distillation then a vaccuum distillation, giving you a lot of petroleum fractions, but sadly they are so full of sulfur, oh that nasty sulfur, all you have to do is electrolyse some pure water, and shove the resulting hydrogen within those fractions to get hydrogen sulfide that you must get rid off elsewhere somehow and more importantly clean fractions that you can turn into precious olefins and aromatic hydrocarbons, the former combined with chlorine makes allyl chloride, add some hypochlorus acid to make epichlorohydrin, mix some other olefins with oxygen to make bisphenol A (yeah that nasty compound) and with epichlorohydrin you get your holy grail: epoxy resin, you know, what makes the PCB's FR4 material.
Now you can have a more realistic substrate for your electronic circuits

will I ever mod this hell into factorio? fuck I might, check my stuff out in maybe 5 to 10 years

[half raptor]

Most factory games are just logistics puzzles that use commonly recognized materials to inform logical production chains

Crafting in Factorio happens in a box with a progress bar. The real game is the logistics

[iron root]

soooo
angel's petrochem.
Got it.

i haven't actually played angel's mods but i swear thats what it looks like

[dense kraken]

yeah same, but from what I've heard it looks like that

[iron root]

oh yeah
and have fun with the leftover sodium

maybe just chuck it back into the ocean idk

[dense kraken]

I was wondering about using it for the accumulator recipe

following IRL's sodium-ion tech

[iron root]

i like it

[dense kraken]

and iirc I already have some uses in like caustic soda and sodium chloride stuff

[wary beacon]

do miners mining scrap produce quality scrap?

[kindred crater]

if you put quality modules in the miners, yes

[iron root]

its trash
But its quality trash

[fallen urchin]

Is that Pyanodons?

[iron root]

py definitely comes close. I'd need to review the py recipes again but it might not get quite that bad (then again, i've only dealt with turning raw coal to coke so far)
aside from everything burning into fucking ASH

[dense kraken]

no, but since I don't know py's recipes it might

that's just roughly how it's made IRL

[neat shard]

Ship in some to go with your trash

[hoary current]

stinks 1.5x more!

[loud prawn]

2,5x 😅

[hoary current]

same as +1.5x

[loud prawn]

Legendary smell 😅

[median verge]

Do you have a source for the uranium requirement? Like a FFF or smth?

[obsidian crescent]

The space platform FFF. They showed us the recipe and talked about how it uses uranium.

[daring siren]

https://cdn.factorio.com/assets/blog-sync/fff-381-science-platform.mp4

[jaunty citrus]

This platform can combine crushed asteroids with imported uranium from Nauvis to produce science packs in much higher quantities.
With 1 rocket load of enriched uranium you can produce 1000 space science packs in space which makes it 100+ times more efficient than sending satellites into space.
This is effectively splitting the space science technologies in two distinct tiers.

[teal wren]

Will resource amount per area will be distributed accordingly to their consumption by all technologies?

Because at the moment you do not need so much coal and uranium

And stone is also a little to often

[daring siren]

You need a decent amount of uranium ore per 1 . Not trivial at all.

We don't know exactly how heavy uranium is to launch, but 1 rocket gives back 1000 (ignoring prod)

[teal wren]

Can that give us a hint on how much does a science weight? Or any intermediate is just 1 kilo?

[daring siren]

Science specifically is 1000 per rocket

Weight system

We decided the axiom to be that the rocket can carry 1000kgs of goods, and balanced the rest around it. We started by setting ore weight to be 2kg per item, so 500 items/10 stacks can fit in a rocket. Then we created an automated formula to derive the weights of items based on their recipes. Many items also have specific hand-picked weights defined in their prototypes.

This created some kind of rough base for testing, and we then modified many things a lot. Most non-intermediate utility stuff was cut to be at most 1 stack per rocket, many things were rounded up, etc. In some cases, we bent things a lot, science packs are expensive, but they can't be recycled, so we allowed 1,000 to fit in a rocket. Modules are expensive, but a whole stack can fit a rocket, because recycling modules is just silly.

[fallen urchin]

I thought gravity would vary per planet, causing rocket capacity to vary too

[daring siren]

We all did, but it doesn't affect neither capacity, nor rocket price.

[hoary current]

I guess u can’t store a lot of circuits inside high tier modules anyway for transport, because to get all the circuits from T3 modules u need to recycle them multiple time for the T1-2 module ingredients

[neat shard]

I have to wonder how overhauls, both the ones we know and new ones that pop up going forward, will design around the quality mechanic...

[keen igloo]

So i installed the janky quality mod, and made a few mods to simulate the earlier infinite techs you get. First thing I noticed is its really easy to make uncommon and rare science labs, and they are monsters. So far the best place to put quality mods is in your mall, and just use the better stuff randomly around your base as needed. Fast long inserters are funny. Steam turbines and boilers are very nice to have with quality mods, though its speculative if they will use the same amount of fuel for more power(i say yes because why not.)

[daring siren]

reach beyond assemblers, so 1 for 6 buildings 🙂

[keen igloo]

yea those are awesome

the rare substation i got covered my whole oilfield

They are right though in the fff, it can be a mega trap to go hunting for it, I think the only place you should do it is quality ore(+coal for plastic) and a quality part maker for labs, then just leave it until you get fairly far.

[daring siren]

Glad you're liking the mod 🙂

[keen igloo]

Yea it's neat, it does change the early game, and you notice it, and I can see how it becomes a mega trap. Quality seems for those people who don't like beacons too, because they aren't really useful with them.

[daring siren]

ah I haven't added the beacon transfer bonuses

[obsidian crescent]

Without also adopting the scaling mechanic, that could get really out-of-hand.

[keen igloo]

yea i was playing with settings and just figured to leave the beacons alone in case I broke something, but the earlier newer beacons will be way faster. I think there is a janky beacon mod out there now

[iron root]

I do think this is the way
Quality mod just your mall for true casual quality, and if you want to invest, also throw quality in the miners and divert quality ore to mall/overflow back to science

I personally plan on going the quality ore route. I like low frequency / high richness and size so slower miners due to quality will have some consequences

[keen igloo]

you get an awful lot of power out of uncommon/rare boiler/steam engines, it makes for a real nice boost as you progress, and they are dead easy to make as well

[daring siren]

How come? They just take up less space

[iron root]

I mean, if you’re rigging your systems up so quality just happens, then yeah you need less of em

[daring siren]

Gives a reason to automate everything, including things I usually hand/ad-hoc craft

[iron root]

Even labs

True lazy bastard run, enforced by… quality

[half raptor]

You could do lazy bastard with quality. Build a bus that only handles materials but with on your machines. Yeah that means you'll have 3-4% chance for quality, but with a priority splitter redirecting quality intermediates off to storage chests. You can then use your random intermediates to make mid-game level armors and vehicles

if you go for lazy bastard run, you have my greatest respect

[obsidian crescent]

FYI: you can't hand-craft quality goods with quality ingredients. So lazy bastard doesn't really relate to quality.

[fallen urchin]

This means the most effcient way to make use of would coincide with lazy bastard, so it could be fun to do them together

[rotund egret]

lazy bastard should be mandatory

[daring siren]

tbh, they could have added a selector to hand crafting, but they haven't

[jaunty citrus]

I'd be fine if they don't. Useful maybe 0.1% of the time, just takes up inventory UI space the rest of the time.

[daring siren]

Same. Just saying it's a reasonable UI

[teal wren]

Probably quality falls under the same category as engine units, to complex to hard with hands

But honestly one selection slider won't be that overwhelming for ui

But maybe something like setting to use high quality items and first or low quality ones may be good, as you don't really have control what resources will go for hand crafting

[jaunty citrus]

It's not that it would be overwhelming, it's just another thing that isn't useful 99.9% of the time

[daring siren]

No fuzzy. Only exact.

[keen igloo]

happiness.

4% chance

[frank warren]

Seeing people use soul burn’s mod hurts my head every time

[daring siren]

Actually unlucky, as it doesn't neighbor with non quality reactors

[wary beacon]

it's ok, i'll get more

[frank warren]

It will in actual SA though, you think?

I am guessing different reactor kinds can't neighbor in current game, so that is why you didn't impl it like that

[fallen urchin]

I wonder what will be the effects of quality on reactor

[wary beacon]

in the quality mod the labs don't seem to like the quality packs, is that a bug

[jaunty citrus]

Sounds like part of the jank in "Janky Quality"

[wary beacon]

yea

[daring siren]

In real SA of course it will work 🙂

[daring siren]

Can't do it in a mod. That's what you use the deleveler for

It returns the correct multiple of science packs for the quality level

[wary beacon]

ahhh i see

thanks!

[daring siren]

I'm pretty sure it's written on the known issues segment

[normal heart]

Sorry if this has been asked already, but with the introduction of recycling, how will the recyclers handle items that have multiple possible crafting recipes? Like for instance if a mod adds alternative recipes to circuits or something.

[frank warren]

They don't

[normal heart]

Like they can't recycle those items?

I don't know if it's even something that matters in practice, but I've just been wondering about it lol

[burnt reef]

they avoid giving items multiple recipes to prevent this confusion

well, that's not quite true, b/c recyclers only reverse assembler recipes, so those are the ones they don't have multiple of

we've seen alt. recipes in the foundry and the biochamber

[normal heart]

That makes sense to me. Is this the confirmed behaviour?

[burnt reef]

I'm pretty sure it's been confirmed, but I don't have a source on that

[daring siren]

Will use the "main" recipe

So e.g. will recycle to 25% of 2 regardless if it came from plates or molten.

[normal heart]

Works for me, I'll take your word for it

Thanks, the answers are much appreciated

[final nimbus]

builds are a lot more attractive now, even if it's all just throughput increases.

[daring siren]

Yea it really helps to not think about pipe throughput. Especially when confirmed they are efficient in rows

[grand igloo]

gotta rejoin this…

[unborn flax]

Quality boilers conversation from #friday-facts message : even if they're just faster, this is an effective reduction to pollution for one of the most polluting buildings early game. They're dirt cheap to make so before leaving Nauvis for the first time you could mint some boiler upgrades (and just store the rejects as you don't have a recycler yet).

[final nimbus]

I can't think of a use case for this unless Gleba relies on steam power, which it might.

[obsidian crescent]

If they're faster, but consume the same fuel per energy generated, then they must produce the same pollution per energy generated, right?

Also, by the time you can get quality in any significant quantity, you've moved on to nuclear power and/or solar.

So quality boilers and steam engines that are just space compression just won't be viable on Nauvis.

[unborn flax]

Tooltip for boiler pollution says "30/m". So if it's faster, it's polluting less per energy generated.

[keen igloo]

producing quality products that only take one core input(stone,copper,iron) is dead easy, quality mod your miners and quality mod the assembly machine for a shot at a rare one. Quality boilers and even laughably quality stone furnaces are stupidly easy to make, it's just a question of how necessary are they really.

[pure spire]

Oh yeah, what do quality miners do?

^ i checked and it said depletes slower

Is that the only benefit? Still good, just wondering

Also, do we know where quality is unlocked?

Nevermind

[daring siren]

Depletes slowers. That's it.

Unlocked on with all other modules.

[pure spire]

But the recycler is not unlocked till Fulgora

So isn’t quality functionally worthless on Nauvis without recycler?

Making 1000 assemblers to get 1 blue rarity assembler wouldn’t be very resourceful

Ig you could recycle the 999 later down the line but it seems worthless imo

[daring siren]

Not at all

[opal haven]

you can easily get 30+ from 1000 with worst modules, quite decent to use on space platform

[daring siren]

Recycling is a late game solution for overabundance of bad quality items

You're not supposed to make 1000 assemblers for 1 blue assembler

[gusty trench]

No. You can either put quality modules in miners to get quality ore and turn that into better stuff, put quality modules in stuff for science so that non-quality stuff is consumed by science and the quality stuff is for personal use, or just put quality mods in the mall and accept that you will only get a fraction of your stuff as quality

[daring siren]

You put in your and skim the quality intermediates

[opal haven]

in is totally the way to go, no need to use wasteful recyclers for quite some time

[daring siren]

Also in your mall assemblers

[gusty trench]

There's reasonable alternate ways to do it. Mods in miners is expensive and requires extra logistics and infrastructure, but gives you the most flexibility in making end products

Skimming quality science ingredients is easy and cheap, and gets some of the important things to have quality on (electric furnaces, prod mods, inserters) without sacrificing much of anything

I somewhat wonder if any Nauvis science recipes have been changed to better facilitate quality skimming. I know a number of recipes have been confirmed to remain the same, but requiring gun turrets in military science or electric miners in chemical science (things that were required in the past) would actually be somewhat beneficial

Or speed mods in yellow science, since we no longer produce speed mods for science with the RCU removal

[final nimbus]

used to take as well

It would be boring now because there's still a T1 module used for science

[iron root]

there are ways to sneak around not having a recycler, notably using the science pipeline as your normal item sink

so you can quality module things like the belts, inserters, and furnaces that go into science, and just keep the high quality stuff for building
If you take this a bit farther, you might even start looking at quality modding your miners (looks like people have said this already) and letting the normal ore go to science, the quality stuff? well.......

quality mall and just keeping the quality stuff for space platforms is super low investment, quality ore lets you make everything at least but might require more miners, because now your miners have some speed penalties and cant take speed modules

[obsidian crescent]

The main issue I have with "quality modules in the mall" is that most mall items aren't produced in sufficiently high quantities to make this viable. Especially before you leave the planet. You might be able to get an assembler to +2% quality early on, so if you want to make 10 Q2 chem plants, you need to make 500 chem plants. But you don't need 500 chem plants. And that's true of most buildings.

Miners and maybe assemblers are buildings you need in sufficiently high numbers that getting some quality items is worth doing. But even then, is 10 Q2 assembler 2s all that meaningful?

Now, quality modules when making modules is a much better use case for mall items.

[daring siren]

Naturally modules is where you put your first Q modules, but putting them in and will make a nice number of quality items

of course is for and later recycling, but isn't bad

[final nimbus]

You need other ingredients of matching quality before turning into or into

[obsidian crescent]

Given that quality mining requires quite a few modules, it's pretty easy to slip a couple into the green science producing line to skim quality stuff there. So really, you're doing both and thus you can lower the cost of quality labs and fast/bulk inserters.

[full coyote]

i'm excited for the new dimension of things™ quality will open up.

[half raptor]

In the very early game when you only have you can slip them into your and have a filter splitter take anything or and store it in a chest.

This allows you to keep your bus filled with normal materials and you can have another assembler make your rare power armor MK1 and your rare personal solar panels

Before recycling, you'd just throw unsuccessful attempts into a chest for recycling later

[daring siren]

I'd recommend setting the chest limits using inserters with conditions, set with param BPs. Specifically to work only when < SET_LIMIT. You can then get any , but not explode on too many .

Here's also hoping for stack size operator on param BPs 🙂

[sacred totem]

(also putting this here, for reasons that are hopefully obvious)
quality trim forum post time
https://forums.factorio.com/viewtopic.php?f=6&t=113994

[daring siren]

Kinda cute!

[charred gyro]

It's probably more appealing than the alternative of having quality icons on everything, but I imagine I'd just disable such icons for normal gameplay anyway. If the maintenance/modding issues that go along with it were addressed (by having the trim replace the icons if and only if the trim exists for an entity) then I could see myself using something like that.

[tall sandal]

i love it!

[winter saffron]

I prefer that

But there are no devs in this thread. So they won'T see it here at least.

[obsidian crescent]

Why not? It readily communicates, both through shape and color, the quality.

[winter saffron]

Less noise is more IMO

[obsidian crescent]

3 dots means Q3. 4 dots means Q4.

[burnt reef]

they need to be distinguishable without color for accessibility reasons

[winter saffron]

fair

[burnt reef]

my own thought is distinct shapes – dot, vertical line, plus, triangle(?), gear

[obsidian crescent]

They're distinct shapes already.

[burnt reef]

(the gear would have a hole in the center, to make it distinguishable from just a circle at a glance)

[burnt reef]

yeah but they're boring

that's not a bad thing, really, but it could still be better imo

this is not something that needs to be improved, but it'd be nice if it was

[obsidian crescent]

For something you might need to see everywhere, boring and unobtrusive are good qualities to have.

[burnt reef]

"boring" and "unobtrusive" are two different things.

my proposal leaves the colors and size untouched, and I think the complexity of the icons is comparable

so it'd be about as obtrusive, I think

vs. is the complexity comparison I'm least sure of

[obsidian crescent]

So why would the gear mean Q5?

How would I know that?

[burnt reef]

yeah, that's a good point, it's less clear