Quality

[grand igloo]

but we should be able to reach somewhere in the ballpark of ~150-175% prod on modules in the foundry alone

[kindred crater]

probably better to ship calcite to nauvis for this strat

[grand igloo]

Which reduces the loss from 75% to ~32.5%

Which is pretty damn good

that means we'll be looking at roughly 1 set of plastic for every ~3 cycles

[obsidian crescent]

Um... what "loss" are you talking about? Weren't we discussing making quality plastic first, then using it to make quality copper and steel?

[grand igloo]

yes

loss as in how much plastic you need to replace

[jolly swift]

Ok but on Vulcanus aren’t iron and copper already basically free from lava

[jaunty citrus]

Iron and copper is also basically free from the ground on Nauvis

[red kelp]

it costs calcite which is a limited resource

wait does it

[jaunty citrus]

Yes, it does

[red kelp]

ok good im not making things up

[kindred crater]

why would not making things up be good

[rotund egret]

calcite is more a catalyst than an ingredient (aka consumed in very very small quantities)

you can see this in the vulcanus FFF

[obsidian crescent]

If it gets consumed, it's not a catalyst.

[rotund egret]

I did say "more a", not "a"

I'm not sure if the recipe numbers have been leaked yet, but the consumption of calcite is very, very low

[jaunty citrus]

It was mentioned a while ago that it's something like 1 calcite to 50 ore, I assumed the ratio would be similar for lava processing, maybe slightly less

[kindred crater]

and if you even less

[loud prawn]

Do we know how much calcite fit in a rocket? I think the standard would be 2kg per unit so only 500, so its not really free or cheap

[jaunty citrus]

Probably 1000

We'll find out on monday

[full hedge]

calcite + acid => steam + stone cosumes much calcite

[kindred crater]

this was discussed a few weeks ago when the b-roll footage of the planets came out after the LAN event... I think it was said the recipe actually produces quite a lot of steam for how little calcite it seemed to consume

[obsidian crescent]

Note that this was with Fluids 2.0.1, which did a 1:10 water:steam heat capacity ratio.

[jolly swift]

is this recipe energy positive if you use the steam in turbines?

[full hedge]

yes I think

[kindred crater]

it better be, power will be a nightmare otherwise lol

[iron root]

vulcanus is well suited to solar at least

...
come to think of it. Steam condensation is meant to be how we get water on vulcanus right? But now water:steam is 1:10
and if it goes the other way, that means we either get way more steam from acid neutralization, or way less water

[jolly swift]

do we need water on vulcanus?

[iron root]

and reading more thoroughly it seems we got the "way more steam" option

[iron root]

coal liquefaction & oil cracking

[jolly swift]

coal liq uses steam

maybe water is the limiting factor for cracking

[iron root]

im considering steam and water pretty equivalent for now

funny enough all this implies that the devs were ok with the acid neutralization recipe just giving 10x the power it used to
which means its proooobably not a major source of power on vulcanus

or maybe it can be now and that was considered cool

[jolly swift]

or vulcanus uses a ton of power already and we need that boost

[iron root]

the only real contest i have to that possibility is that blending steam and solar without anything backing up (which would kill sulfur production) is hellish

[jolly swift]

if any planet is going to be hellish its going to be the one with lava and volcanoes

[iron root]

hmm.
Vulcanus gives us a solid void via throwing into lava, but we dont have a fluid void do we?

[jolly swift]

what exactly would back up with solar+steam

[iron root]

steam backing up would kill sulfur production (and vice versa)

and otherwise just the way solar panels and accumulators interact with steam means there's going to be wasted solar power unless you fine tune your power production

[loud prawn]

Just make water out of it, u need it anyway for making light oil

[iron root]

and what if water backs up?

[loud prawn]

Than produce more light oil

[iron root]

converting to water is just kicking the can down the road when the steam production is meant to power the base

[loud prawn]

(Also you can flush the tanks manually)

[iron root]

kick the can all you want but there's a stopping point eventually

[loud prawn]

The stopping point is too much light oil, make solid fuil and void it 😅

[iron root]

yeah, its probably going to end with something getting crafted into a solid and lava-voided

[loud prawn]

Or maybe instead quality recycled when it makes sense

[iron root]

if youve done fulgora by then yeah

[loud prawn]

I think i will go fulgora first, to have at least the revyclers

[iron root]

i suppose if we turn things around a bit and use that solid -> lava void thing
then everything on vulcanus is pretty trivially voidable except water/steam, which requires other materials to make it into something solid.
each oil fraction can be turned into solid fuel and voided, sulfuric acid is just directly mined and i dont think its a byproduct of anything, liquid metals can be casted -> voided

[loud prawn]

And at least they should be achieved really soon and easy

[iron root]

soooo
the production chain on vulcanus is likely going to be balanced around... water 😂

[loud prawn]

At least the advanced oil puzzle will be skipped 😅

[iron root]

i mean
coal liquefaction spits out heavy+light+gas as is, just in different ratios
its still there

i think anyway

i do remember seeing a coal liquefaction recipe that also output steam but i dont remember if that was just the mods i ran or not

[loud prawn]

Thats stuff from mods

But who know what they changed, they already said there will be a basic version of coal liquidification

[iron root]

ah it was heavy oil that was the cycled material

[loud prawn]

Yes u needed some to kickstart it

[iron root]

yeha, basic liquefaction removes the heavy oil requirement but i think it still needs steam

[loud prawn]

And maybe gives less different products like basic oil processing

[gleaming quartz]

I think the thermodynamics formula goes PV = nrT, or Pressure * volume = temperature.

Going out on a limb here, if the steam is clocked at a higher temperature, then its either because more volume and more pressure combined. If the pressures constant, then a higher temperature of steam correlates to a higher volume of it.

That said, i dont know if factorio dev team had that in mind

[distant eagle]

casually ignores nR

[gleaming quartz]

For discussion of ratios, its not particularly useful. So what if pv is t times a multiple of nr. The gas (steam) has a constant molar mass. The gas constant is just a constant. it doesnt change the fact that if p or v increases, then t is the only thing that should increase as well, unless the gas constant is changing or the chemical changes its molecular makeup through the act of condensing or evaporating

[obsidian crescent]

FWIW, basic liquefaction:

1. Requires calcite and sulfuric acid, not steam. Basically, it does the acid neutralization internally.
2. Only produces heavy oil. So if you want anything other than lubricant, you have to do cracking.

[jolly swift]

When water is boiled both volume and temperature go up

But also those are gas laws and not liquid laws

[gleaming quartz]

That seems like a dubious claim, so the equation is probably wrong. Its more likely how much volume a gas will occupy given a particular pressure and temperature

[jolly swift]

what do you mean

Steam is less dense than water and by definition it’s warmer

[obsidian crescent]

The 1:10 water:steam ratio is not about pressure or temperature per-se. It's about the relative ratio of the heat capacity of the two substances. That is, if you take the energy needed to raise X water 1 degree C, that energy would raise 10*X steam by 1 degree C.

[jolly swift]

The number of moles of H2O stays constant but the pressure, volume, and temperature change when boiling

[gleaming quartz]

if i boiled 1 cup of water, the amount of molecules in the gas that was boiled off is the equivalent to the amount of molecules in 1 cup of water. Otherwise, breaks conservation of matter and energy

[jolly swift]

number of moles is related to mass, not volume

steam takes up more space for the same mass of h2o

in any case I think its fairly obvious that the ratio of 10:1 was made for game balance purpose and not as an accurate reflection of real life chemistry

[gleaming quartz]

Im fairly certain that incorrect with what I just said

again, the amount of molecules in 1 cup of water is the same amount of molecules if that 1 cup of water was boiled into a gas

[jolly swift]

im so confused

what are we even talking about rn

[iron root]

The 1:10 is both a game a balance thing and physically is explained by volume expansion

No one ever said factorio liquid quantities were normalized with respect to anything

[gleaming quartz]

im replying to the seemingly irrelevant "number of moles is related to mass, not volume" comment

no where did i mention, or claim that.

[iron root]

This seems like a bunch of confusion over what’s what when dealing with liquids and gases

[gleaming quartz]

I originally speculated that maybe higher temperature steam provides more water assuming constant pressure

[jolly swift]

i said that water goes up in volume and temperature when boiled, and responded saying that was dubious- im so confused rn 😭

[gleaming quartz]

Which i said if going by pv = nrT, may be an incorrect equation.

You are directly increasing the temperature of water, but you arent adding or removing the water to the boiling operation. The equation I gave may apply more correctly to gasses not undergoing a transformation.

tl;dr; the equation i gave is probably incorrect for this situation

[iron root]

Physically I think higher temp steam in factorio is just at higher pressure to make it relatively constant density compared to lower temp steam, which at least squares with it making more power per unit when run through a turbine

[gleaming quartz]

good point.

[iron root]

And pv = nrt doesn’t really apply to boiling very well
But there is a very much unambiguous increase in volume

It’s just not governed by pv = nr * (delta) T

[jolly swift]

because energy generated is given in joules, which is a real life unit, we can actually determine what unit is used for fluids and convert to real life measurements

ill pull up the numbers rq

[iron root]

If you take factorios units at face value you get some very silly implications

Such as nauvis surface receiving at least 150/9 kW/m^2 (legendary solar power density) in solar irradiation

It’s uhh
Toasty

[gleaming quartz]

im curious about that calculation

off hand it looks right, but idk if people were accounting for the size of each individual panel

[jolly swift]

one unit of steam at 500C produces 100,000 joules
one unit at 165 produces 33,000 joules
67000/335 = 200J/u
200J/u = 4.184J/g
200g = 4.184u
u=47.8g
therefore factorio units of fluid represent just under 50 grams of fluid

*one water produces 1 steam at either temperature so we know the number of moles is constant

[iron root]

A puts out 150 kW during daytime, and is a 3x3 and therefore takes 9 m^2
Assuming it’s perfectly efficient sets the lower bound on irradiance, as any inefficiencies require the surface to receive more power for the panel to keep making 150kW

[jolly swift]

how does that compare to earth?

[iron root]

Earths solar irradiance is about 1.4 kW/m^2 from my searching

[jolly swift]

wait how do we know the solar panel is 3 meters by 3 meters

[iron root]

For a closer reference, mercury gets up to 14.4kW/m^2 apparently
That’s less than 150/9

[iron root]

Devs have stated that 1 grid tile in factorio is 1 m^2 I believe

[jolly swift]

gotcha

[grand igloo]

so nauvis' is 16.6, 11 times higher

[jolly swift]

absolutely unplayable

[iron root]

Nauvis gets more irradiance than mercury

[gleaming quartz]

I count 6 individual panels on this one. I assume 1 panel occupies about 1.5 by 1 tile location. The irradence is probably slightly lower than 150/9

[jolly swift]

the total footprint of the panel is still 1m^2

[iron root]

Considering you can indefinitely tile solar panels with 0 loss, each panel can’t take more than the 9 m^2 allotted by its 3x3 footprint

[jolly swift]

if the panel is making 150kw from less than 1m^2 that increases the irradiance

[gleaming quartz]

its 6 panels (which would mean each are producing 25/1.5 kW/m^2)

which is the same value

[iron root]

I guess
I’m just using the lump pieces the game gives me

[gleaming quartz]

kinda nifty the calculation worked out both ways tho

you dont see that a whole lot

[iron root]

Also it’s worth considering that physically, solar panels are limited to something around 45% efficiency

[jolly swift]

maybe the engineer is really good at solar panel design

[gleaming quartz]

i think that applies to the solar cell. Its possible to get higher than that with a different configuration

[iron root]

The efficiencies higher, but there is a physical limit based on re-radiation

[gleaming quartz]

pulling up the wiki again on this. I always re-read this page every so often

[iron root]

Okay it’s 86.8% for irradiation from all directions, and 68.7% for irradiation from the sun

[gleaming quartz]

current world record for solar cell efficiency is 47.6% using a concentrating photovoltatic cell (basically shining more light on the cell than what is naturally gained from the sun if im not mistaken).

Probably gonna be logarithmic gains in the forseeable future.

[iron root]

Concentrated PV is just using lenses/mirrors to, well, concentrate sunlight on the cell

You know like using a magnifying glass to fry ants, except it’s to make power

[gleaming quartz]

its such a simple and hilarious concept. Want more power? just shove more sunlight on to it. EZ

just gotta make sure you dont fry the poor thing

[iron root]

One of the big catches of concentrated PV is it gets basically 0 power from diffuse light, so if it’s cloudy they don’t work, cause the optics are fucked

[grand igloo]

y'know what the clear best power option is?

building a giant straw and then sucking up the sun's plasma

[iron root]

And yes, cooling concentrated PV cells is another major issue

[gleaming quartz]

you know those cookers where they are like trays rotating on a (spit?)? Im imaginging something like that where the person is litterally just rotating the panels into a cold solution after sititng in the sun for some period of time

there are better solutions, but its funny image

[iron root]

Nah they’ve just got heatsinks, link the ones you attach to your cpu

[gleaming quartz]

i like the ones that use water. Decent way to boil water tbh

[iron root]

Ye. Solar thermal is interesting too

Solar thermal gets kneecapped by the Carnot limit which were finally starting to beat with PV

[gleaming quartz]

i was thinking more the "use water as your heat sink for panels" as opposed to "heat vat of water with reflectors to spin turbine"

[iron root]

I mean
You did specifically mention boiling

[gleaming quartz]

thats fair, but CPV's can get pretty damn hot.

[iron root]

They do yeah. Concentration ratios can get into the thousands iirc

[gleaming quartz]

and crist, at that temperature, just boil the water to make even more energy

[iron root]

Combined cycles

Though unfortunately solar cell efficiency falls off a cliff with increasing temperature
And Carnot efficiency falls off a cliff with decreasing temperature
So pick your poison

[gleaming quartz]

probably some smart cookie engineers out there have a good automated way to switch between the cycles

[iron root]

You can’t really

Cause keeping the solar cell cool means you can’t get the water that hot without violating thermodynamics

[gleaming quartz]

never mind the infrastructure would probably be a bit awkward

[iron root]

Iirc in cases like this you’d use the waste water as just a municipal/industrial hot water source

There’s a good few power plants that supply nearby areas with hot water from the waste heat they can’t turn into further power

[kindred crater]

good thing never has to take off their scifi suit

though actually

how do the oceans stay liquid

hmm

[gleaming quartz]

who said it was water

[kindred crater]

this emoji

and i always believe emojis

[gleaming quartz]

yea, but is it pure water?

[daring siren]

Water you on about?

[fallen urchin]

Can we use this recipe elsewhere to generate energy?

[solar osprey]

The sulphuric acid neutralization recipe is Vulcanus exclusive.

[dense kraken]

I think buildings are implied to be much bigger than they appear on the grid

I mean, 9m² is pretty small for a whole turbine or nuclear reactor

[jaunty citrus]

but a reactor takes up 25m^2 on the ground?

[faint blade]

The part that actually generates the heat in a real reactor isn't much larger I think

[dense kraken]

yeah but you get what I mean

9m² can be the size of an "inserter" irl

[jaunty citrus]

I bed there's 0 radiation shielding on the factorio reactor

[dense kraken]

then how do you not get radiation poisoning :p

[jaunty citrus]

Engineer built different

[half raptor]

so 8 module slots.... this means a 50% quality chance?

[solar osprey]

Though who knows what it can make.

[burnt reef]

49.6%

[daring siren]

You will, on Monday

[half raptor]

i know all about the rounding error, it's just easier in conversation

[burnt reef]

fair

[daring siren]

Also consider +200% productivity instead 😄

Or +1000% speed

[half raptor]

it all depends on what it makes

[wheat sluice]

I think the application will be quite limited, but we do know that quantum processors can be made in it, so we can get +200% prod on those

[half raptor]

or.... we can get quality quantum processors

[daring siren]

Or quantity qualium processors

[iron root]

#1215078107334057984 message
But there's still this issue
if solar panels were bigger than that 9 m^2, there would have to be losses when theyre placed next to each other

physically, anyway

though this all comes back to "taking units at face value makes things silly"
and they are definitely very, very silly right now

[wraith moss]

The answer of how solar panels can make so much power is "shut up it's a game"

[half raptor]

It's the same logic for why 3 + = you're ruining the immersion

[dense kraken]

no I mean, think about it like buildings in old pokemon games, they appear tiny on the outside, it's just a representation

[iron root]

so we have something even spicier than a nauvis getting more irradiance than mercury

[abstract condor]

guys it's obvious the solar panels have tiny RTG generators in each solar cell

[gleaming quartz]

Not gonna stop me. im gonna mod calcite production as a possible output from gleba manufacturing

And if needed, make the sulfuric acid neutralization available everywhere

[wheat sluice]

New machine new module analysis:

[distant eagle]

with recycling?

[iron root]

yeah that table has to include recycling

though personally id multiply everything by 4 because then it'd work for the inputs as well
well.
the math on that actually breaks down on anything that isnt 300% prod so nvm

[wheat sluice]

Yes, this is for a simple recycling loop

[distant eagle]

Seems strange that the shapes are so different

[merry spindle]

First one looks like right third of second one

[sage vapor]

do you also have modules in the recycler?

[wheat sluice]

ofcourse, of the same quality as in the machine

[wheat sluice]

Worth noting, that since Cryo plant mainly works with fluids, and probably uses coolant in most cases, the actual costs will be much higher, on account that fluids don't get recycled.

[stray marsh]

You forgot the new 10x dense water

[inner pilot]

what I'm reading is that legendary Quality modules are a trap
*I realize that that the second graph only looks as it does because of legendary Q in the recycler, iff recycling was using worse modules it would be different (though I don't know by how much)

[wheat sluice]

The recycler has the same quality modules as the machine

[inner pilot]

it says so at the top of the chart 😄
do you mind making one where the recycler has Q0 T3 quality modules and the modules in the cryogenic plant are legendary?

[inner pilot]

thank you!

[kindred crater]

thanks for putting this together, but what do the colors symbolize?

[inner pilot]

the darker the worse it is compared to the optimal module loadout for this amount of global productivity research

[wheat sluice]

yep, the color goes from black to white - worst to best cost multiplier in each column

[kindred crater]

ah that makes sense, thanks

its awesome how low some of the multipliers are, and with relatively low prod research. granted you'll need to obtain legendary modules beforehand.

[inner pilot]

the moment you can make something with 300% prod you really ought to be making it at quality is my takeaway 😄

[wheat sluice]

not really, if you can't utilize it then it would be a waste of energy, and huge waste fluids if there are any in the recipe

And I expect the +300% productivity on anything will be very difficult to achieve.

[kindred crater]

#friday-facts message this is before productivity modules and lab productivity (new biter lab/research), but yeah its a substantial amount of science

[wheat sluice]

Once you get your first set of 9 modules, you can keep making them at a cost of 23.48, which isn't even that bad. Further more, there's a bunch of room for chaining quality in the module production trees of modules, so the cost can definitely be brought way down.

[wheat sluice]

that's pretty costly, but maybe not as much as I expected,

[inner pilot]

so with a 1 million SPM base, like Kovarex said he was building, you could get the research done in just over a day?

(also, you don't need to get to 30 when you add a few good prod mods in the process. Depending on the prod research formula we might get to good enough prod for 300% in much much less time)

[kindred crater]

1000x faster than 1k spm would get you to 0-300% for a given research in less than 3 1/2 hrs, but its kinda circular reasoning because that level of SPM would likely require high levels of prod research to achieve. for maximum UPS performance I think you'd still want to achieve 300% (or close to it) because the module slots in those machines can then be speed instead of prod, cutting down the number of machines you need to reach that given level of output.

[inner pilot]

I would expect that you have a bunch of the prod% researches done by the time your base gets to 1M SPM, yes. Same as I expect most 10k+ SPM bases to have a bunch of levels in mining productivity today 😄

[kindred crater]

I do wonder how close kovarex is to his goal. The fact he actually makes it sounds reachable makes me think he's at least well into six figures.

[lyric yacht]

So in this the recyclers always have the modules? While only sometimes the cryo plant will have them instead of prod

Will these tables be added to the wiki for all machines?

[wheat sluice]

on the left you can see the 2 gray columns indicating how many quality vs prod mods there are, all slots always filled, just with different combinations.

and recyclers are always full of quality modules

[wheat sluice]

I'd be happy to upload these tables to the wiki if the wiki people deem it fitting for them to be there.

[kindred crater]

if factorio cheat sheet ever gets updated for SA, these would be great to include for quality

[lyric yacht]

It's always quality 3s right?

[lyric yacht]

Can you show them for assembler 3s, foundries and EM plants?

[distant eagle]

Lots of combinations possible, 1-3 tiers, 1-5 qualities, 2-8 module slots

[lyric yacht]

I feel like only the extremes matter

Quality items with low tier quality modules are probably not worth it

and if you can go for legendary quality then why ever not

[obsidian crescent]

Space Age is a much longer game than vanilla, and you'll have access to quality for a much longer period of time. There are also places with a lot of space and or power constraints where even mild quality can have a significant impact.

[lyric yacht]

What I'm wondering more is for things that aren't intermediates so you can't prod them, what's the best way to make them high quality, cuz I assume it'd be different with other ones

[obsidian crescent]

Also, the new beacons are so OP that even Q2 beacons are worth doing.

[lyric yacht]

Yeah but at the stage just plonk all the quality modules in, and any quality you get is better than nothing, but you're not like cycling over and over to get legendary or whatever your highest tier is

[obsidian crescent]

"Best way" is both to-be-determined and depends entirely on where you are in progression.

For example, doing some light quality cycling of beacons once you get the EMP and its 50% prod bonus sounds like a really good idea even fairly early on.

[lyric yacht]

I guess that might be why they make it so quality isn't all unlocked at once

So even if people want to quality cycle they can only go up to rare or whatever, so they don't spend so much time and resources on ot

but what specific makes the beacons so good to quality? I forgot

[obsidian crescent]

When they redid beacon scaling, they change the effectiveness of a single beacon to 1.5x instead of 0.5x. They also made it so that quality beacons have both lower power consumption and higher effectiveness by 20% per level (previously it was just lower power consumption). So a single Q2 beacon around a bunch of buildings gives a 1.7x multiplier to the modules it broadcasts.

So a pair of speed module 2s in a Q2 beacon increases the speed of all of those machines by 102%.

[wheat sluice]

I don't wan't to spam yhe chat, not unless others want to see it too.

[wheat sluice]

Yes, that is true, but I don't think that all the combinations have to be show, i can arbitrarily pick some significant ones, maybe once the game comes out, it would become clear which ones are significant.

[lyric yacht]

Yeah that seems good

[wheat sluice]

Ah, I was answering to your messge bellow with that. To answer that question, yes it's always T3 modules.

[wheat sluice]

Oh that's funny, the fluid cost multipliers are inversely proportional to the item cost.

obv now that I think about it, but I hadn't

[obsidian crescent]

What if you quality cycle grenades for coal, but only collect Q4 and Q5 coal, then feed the Q4 coal through a Cyroplant with 8 Q5 quality module 3s in it (recycling any Q4 plastic that comes out)? How much less costly would Q5 plastic be in terms of coal in that case?

[wheat sluice]

my calculator isn't that flexible right now

[half raptor]

the math is going to be all over the place during the phase of the game where people are gradually step upgrading their for and up

setting up something modular and expandable and upgradable is going to be the way to go

[grand igloo]

Yeah, I think looping intermediates is gonna be ideal for that stage

cuz it's easy and expandable

and consistent

[grand igloo]

can a cryoplant make plastic?

[obsidian crescent]

Outside of AOP or coal liquefaction, I cannot think of a more important recipe for a Cryoplant to be able to do.

[grand igloo]

why?

you'll have other buildings on aquilo for that

[obsidian crescent]

In a vanilla 1K SPM base, more chemical plants are devoted to the process of making plastic than any other chemical plant process. If the Cryoplant is going to have utility in replacing a chemical plant in at least some cases (which is why they gave it 8 module slots), then that's one recipe that it absolutely must be able to do.

[grand igloo]

is it supposed to be a chem plant?

I thought it was just that it didn't have base prod like the other machines so it got more slots to compensate

cuz a chunk of its recipes are loops like coolant

[obsidian crescent]

Right, but that only matters if it can use recipes that already exist.

Take the Biochamber. Because all of its recipes are new, the number of module slots and base productivity doesn't matter; you can just factor that into the recipe ratios directly. The 50% prod bonus of a Foundry only matters because it does comparable things to other buildings.

Also:

While the structure can use some chemical plant recipes

If it can't take over for the most important uses of a chemical plant (like the Foundry and EMP), then why bother making it useful off Aquilo at all?

[grand igloo]

I mean it's mandatory for a fusion setup

[obsidian crescent]

Yes, but so are Fusion Reactors and Fusion Generators. The developers made explicit comparisons to the Foundry and EMP, so it probably ought to live up to that to some degree.

[kindred crater]

i wonder why only some

not even most

[obsidian crescent]

It is a cryogenics facility; it doesn't really make sense for it to do everything that a dedicated chemical plant can. So picking specific high-value recipes makes sense. Cracking, plastic, holmium solution (I still can't believe that's literally what it's called), maybe carbon and coal making.

[red kelp]

has there been any word on a quality upgrade planner

like "upgrade all entities to rare" or something of the sort

[obsidian crescent]

I'm unsure if you'd want that. Unlike normal entity upgrades, almost every entity technically has a higher quality version, even if you don't mean to use it right now. Belts do have quality, but you'd basically never want to use them.

[red kelp]

perhaps something more specific than everything perhaps - like "upgrade all assemblers and modules"?

[kindred crater]

don't see why this couldn't be integrated into

[half raptor]

you can set --> in an upgrade planner, but what I'd love to see is a flag you could put on that prod module that tells robots to upgrade it until it is , even if it has to do it one rank at a time depending on inventory of modules

[obsidian crescent]

It's not a question of if you could. It's a question of whether it would actually be useful. Do you really want to upgrade the quality of every building, inserter, belt, pipe, power pole, etc in a boxed-selected area? Or do you mean to upgrade specific things, or not upgrade specific things?

[frank warren]

I can confirm this is a pain point

[half raptor]

though I suppose it does allow you to upgrade certain machines first.... for instance the machines making the quality modules should get the upgrades first and when your supply chest of normal grade starts to fill up, you can upgrade planner the whole base to uncommon, then to rare

the returning normal modules just goes into the shredder

[iron root]

i have a feeling i'm going to be making a blueprint book filled with just quality upgrade blueprints...

[kindred crater]

that's already optimal gameplay anyways

it's easy to say "why doesn't wube just add one that only upgrades the things you want to be quality!" but that varies so much

[iron root]

yeah, ive already got tier-up blueprints to separately do crafters, inserters, belts (even a version that doesnt boost underground belts because of weaving silliness)

so i guess i get to do it again for quality

[kindred crater]

sometimes you'll want so they last longer next to being spat on

the best option is just integrating with because it already lets you personally tailor it like this

[half raptor]

I can see making because it can be used for making engine units and other quality machines, but I doubt I'll want to make just for a little extra hp

[kindred crater]

then just replace with in that message

[distant eagle]

Trying to make Q1-Q3 T1 modules from ore.
Q3T1 modules

Runs fine, deadlocks after 6 hours without the overflow splitter

Anyone get something to work without recyclers?

[half raptor]

only in SA the mk 3 modules require an extra ingredient, so we'll only be able to craft mk 2 with

[kindred crater]

so holmium cable for , nauvis biter eggs for , has anyone figured out or yet?

[obsidian crescent]

Speaking of which, it's kind of irritating that none of the planet-specific intermediates get infinite productivity research. So there's no way to get quality version of them for "free".

[quick rampart]

confirmed?

[kindred crater]

all productivity research is visible on crowdin

[sacred totem]

trolled

[kindred crater]

there were a few new things we didn't know about, but nothing for planet unique resources

[quick rampart]

holy how could i missed that

i thought it would be some parameterised strings so i didn't notice

[kindred crater]

I think I just typed "productivity" and that returned the full list

[quick rampart]

now i know them

[distant eagle]

The modules match for each step for iron and copper. But it ends up with too much green iron plate going to green green circuits

[half raptor]

YouTubers are going to make some money tomorrow

[kindred crater]

I'm looking forward to maximum spoilage

[full hedge]

I'm looking forward to maximum spoilage

[daring siren]

Surprise: is 2%

I was surprised as well

[obsidian crescent]

But quality 3 is just 2.5?

Oh and while you're here, the big question: how does quality and power production work?

[daring siren]

Yes. It's weird

[half raptor]

Kind of makes sense

[half raptor]

One can assume quality fuel has more J of energy to give and quality boilers/reactors run faster

[daring siren]

Faster consumption

[obsidian crescent]

So just space compression, across the board. Thanks!

[half raptor]

Hey 4 is it really 24.8% or 25%?

[daring siren]

24.8%

[half raptor]

That's going to make it awkward in conversation

Wish module mk 3 just did 3% chance instead of 2.5%

[daring siren]

It breaks the 2:3:5 ratio standard. I asked the devs about it but didn't got an answer

[obsidian crescent]

Oh, did efficiency modules get some adjustments, or are they still 30/40/50?

[daring siren]

30:40:50

but many things are 2:3:5 e.g.

[kindred crater]

I'm not sure if this was said yet, but quality modules 1-3 have a flat -5% speed and no change to power consumption

https://www.youtube.com/watch?v=h-xxQnaIq3Q btw xterminator goes into quality more in his video starting at 38:26

[steel falcon]

Honestly perfect for early/midgame mall use, I was worried about increased power

[arctic gyro]

so with the infinite prod it seems late game quality science would be the meta because it multiplies

[solar osprey]

Well, no, as because if a quality science pack costs 5 times as much to make, but only gives 3x as much research, no matter what the research productivity is 3x is always less than 5x

[obsidian crescent]

I think it depends on what the packs take. Quality versions of most of the off-world packs will be hard to make since many of them can't take advantage of infinite prod for free quality.

[arctic gyro]

yeah that probably matters

depend how cheaply it can be made

[rotund egret]

uncommon packs are always worth it since they're 2x the value

[sacred totem]

No?

[kindred crater]

it might be 2x the value, but its more than 2x the cost to obtain

[sacred totem]

Since you’d need to put quality modules somewhere and legendary T3 mods can do +100%, so somewhere you’d be losing materials for that uncommon pack

Though if you have some op quality loop (cough lds cough) it could very well be worth it

Like notnotmelon is doing legendary yellow science…

[daring siren]

They can be a good sink for unbalanced quality items

[rotund egret]

it's worth it

[kindred crater]

!!!

we were all wrong

[obsidian crescent]

If Bioflux is quality-cyclable, then (especially with the Biochamber's brand new 50% prod) making quality Bioflux is easy. Making quality nutrients easy from that quality bioflux.

What that means is that if you can get just one quality ore bacteria, you can replicate it at that quality so long as you have equal quality bioflux.

So if you can consistently get Q5 bioflux... you only need one Q5 ore to get started.

So Gleba might be a good place to make Q5 stuff, including Q5 science packs. At least for red and green science.

[fallen urchin]

Sounds like a good way to get some high qual items without crazy qual cycling

[obsidian crescent]

Well, without quality cycling something that doesn't grow on trees.

[daring siren]

Bioflux is the old jelly-yum recipe, mash and jelly. Relatively cheap, but still non-trivial

That said, 1 bacteria + flux = 5 bacteria

which is a very nice ratio

[fallen urchin]

With some prod, it could be easier to qual cycle the flux than the final products

[obsidian crescent]

And you can prod it.

[sacred totem]

Seems like more potential for failure than the goofy lds setup

[rotund egret]

just cycle ore

it's really, really infinite

legendary drills only deplete patches 8% of the time

and with mining productivity you get more than you ever want

[obsidian crescent]

With enough Bioflux production, you're unlikely to run out of the particular quality you want. And bioflux has a long spoil time, longer with quality.

[daring siren]

1 to 4 actually

[obsidian crescent]

But what I'm talking about is something you can do after just Fulgora and Gleba; you don't need to have gotten Q5 before this is viable.

[rotund egret]

bioflux also cannot be recycled period at the moment

and I'm not sure it's gonna change before release

[sacred totem]

I think trees are more limiting on gleba than drills on other planets

You can’t boost tree production

[rotund egret]

yeah

agri towers don't have great throughput

[daring siren]

Also ag towers = sporellution

[sacred totem]

It’s certainly good but not good enough to justify burning craptoms of it on q5 stuff

[obsidian crescent]

Oh, BTW: how does quality affect Ag towers?

[sacred totem]

Range, I think?

Don’t think I saw anything else

[rotund egret]

no effects except health

[sacred totem]

Not at computer so can’t check…

[rotund egret]

range would have been nice but IIRC there were graphics issues and no one wants to touch the tower code

since the arm itself is only so long

[obsidian crescent]

So not even speed. Meh.

[rotund egret]

it doesn't need speed

trees grow on their own terms

[sacred totem]

Also I tried using stack inserters on agri towers but the item stacking just caused the fruits to stay in the belt longer and be more spoilt

[rotund egret]

the tower itself only plants and harvests, there's not much it could do other than range

[sacred totem]

Not the case for other production because I’m using flow through for everything else

[rotund egret]

it has uses if you're using many towers all outputting on one belt

that leads to a train

[sacred totem]

Yes, just saying you can’t simply assume stack inserter is always better

[rotund egret]

gleba stuff is infinite but you're gonna want a lot of land

of course

[daring siren]

You also need artificial and overgrowth landfill to make more plantations

[rotund egret]

and you can't put those everywhere

[obsidian crescent]

Overgrowth requires biter eggs, right?

[rotund egret]

yep

[half raptor]

I'm assuming that if the bacteria only has a minute left before biofluxing that the 4 products also have 1 minute before spoiling?

[daring siren]

Yes

That said, spoilage averages across everything

So fresh bioflux would increase the lifetime of the output

This is unlike what we thought in the past (bad apple ruins the bunch)

[half raptor]

Neat, so as long as you have fresh bioflux you can run an iron empire off of one bacteria?

[daring siren]

Fresh quality bacteria, if you want quality

[half raptor]

Does the bioflux have to be quality too?

[daring siren]

There's also a recipe to turn plant into 10% bacteria

[rotund egret]

of course

[half raptor]

The possibilities are spinning in my head

[arctic gyro]

we werent wrong, they changed it

[half raptor]

So if you start amassing a lot of bacteria but don't have the bioflux to go with it, eventually it'll just turn into I'm assuming

[arctic gyro]

i think so

[daring siren]

And bioflux is not trivially cheap

[half raptor]

I assume not

But anything can be scaled up

Neat though. I'm more excited for Gleba after seeing the recipes. All the FFF would give us is "there's a biolab.... It does biolab things"

[sage vapor]

asteroid reprocessing with quality modules allows you to get legendary asteroids

only loose 20% of an asteroid per cycle

and you get tons of ore per asteroid

• asteroids are very abundant in space

-> you can get , , (from carbon + sulfur), (from using calcite on vulcanus to turn lava into molten iron/copper)

[sacred totem]

oh thats broken

i hadndt considered that

[sage vapor]

and with that legendary coal you can get legendary plastic which turns molten iron/copper into legendary LDS

recycle that LDS and you effectively turned a little bit of + molten iron/copper into and

[half raptor]

Considering the cryo lab can make plastic at 49.6% quality, quality plastic won't be hard to get

[daring siren]

Or +200% prod

Also does kinda grow on trees on

[sacred totem]

that makes the cryo lab the cheapest 300% prod in the game, doesnt it?

since you onyl need 100% prod research instead of 150%

[solar osprey]

170 minutes at 1K SPM

[obsidian crescent]

Yes, but you can't quality cycle most of the item productivity stuff you make there.

[half raptor]

Well when it comes to cracking oils, that cryo lab is gonna be king, but what I'm really interested in is all the biolab recipes. I can't find any reddit threads or anything with all the recipes

[obsidian crescent]

Wait: can it do cracking?

The Biolab is a lab; it doesn't have recipes.

[half raptor]

I meant the bio thing on gleba

Chamber

[sage vapor]

cryo lab has no cracking

[obsidian crescent]

The Biochamber does cracking, as well as rocket fuel from light oil and solid fuel. I believe it also does solid fuel from light oil, but I'm not sure about that one.

[sage vapor]

cryo lab

[sacred totem]

acid neutralization in a chem plant must be insane

[sage vapor]

oh interesting, those are quite recent changes

[obsidian crescent]

That will definitely mitigate water consumption on Vulcanus, but without also replacing cracking.

[sage vapor]

the issue with the biolab is that you have to feed it

[daring siren]

Yummy!

[obsidian crescent]

True. So you'll need shipments of bioflux (which can make lots of nutrients) to keep the setup running.

Or you can locally source fish or biter eggs to make nutrients.

[sacred totem]

*cryo plant

it already does 2k/s in a (unmoduled) chem plant

so a cryo plant could do 18k/s

[obsidian crescent]

Oh, it actually hits the pipe cap.

[sacred totem]

easilly

deserves a nerf imo but the devs have seemingly decided otherwise

[sage vapor]

oh wait fish farming recipe changed

no need for nutrients anymore

yeah you can get nutrients from fish with tree seeds now

[sacred totem]

(outside of the nutrients for the biochamber, itself of course)

[sage vapor]

interesting

[obsidian crescent]

I think it's much more about reducing the amount of calcite you need. 200% prod means 1/3rd the calcite for water.

[sage vapor]

well you need to kickstart it once

then it can just run

[sacred totem]

oh, i see what you mean

no need for nutrient import

[obsidian crescent]

Wait: what does the recipe need?

[sage vapor]

1 fish to 20 nutrients

[sacred totem]

still gotta do bioflux for biter eggs anyway tho

[obsidian crescent]

The fish breeding recipe.

[sacred totem]

it used to be 2 fish to 3 fish

[obsidian crescent]

Oh, you need tree seeds. But you can make it in an chemical plant.

[sage vapor]

heck you don't even need nutrients to feed a biolab for fish breeding

just do chem plant breeding

[obsidian crescent]

But making seeds does require the biochamber, right?

[sage vapor]

nope

[daring siren]

That recipe changed!

[obsidian crescent]

Wood processing is a chemical plant recipe too?

[daring siren]

It used to be nutrients for fish

[obsidian crescent]

Oh... I think I know how to make a group of quality fish last indefinitely 😉

And that means an indefinite supply of quality nutrients.

Oho...

[sage vapor]

hmm interesting

[obsidian crescent]

You'll need a lot of wood production to make the quality tree seeds, but who cares.

It's wood.

And you want that to cut down on pollution anyway. And it's free.

[sacred totem]

agri towers arent exactly pinacles of speed

[sage vapor]

yeah

[sacred totem]

but theyre highly spammable on nauvis...

[sage vapor]

tree farms new meta

[sacred totem]

also also eat pollution.....

[obsidian crescent]

So basically Ag towers on Nauvis are Forestries from IR3?

[daring siren]

While on they produce pollution

[sage vapor]

the towers don't

breaking the trees does

ah I guess they do 4 spores / minute

but the gleba trees are 15 / harvest

[daring siren]

Fish breeding doesn't accept quality

So you'll have to make 20 quality fish from spiders, and then you can multiply them

[sacred totem]

yes

[frank warren]

It was never said to work this way ever

[daring siren]

That's why we assumed

But then a dev came in and said it's average

Single stack was always average

I'm talking about between stacks of different ingredients

[frank warren]

Oh, you mean in a crafting recipe?

[daring siren]

good bioflux can give lifetime to nearly spoiled bacteria

[half raptor]

So this is how we craft

[frank warren]

wait*

bacteria do not inherit, do they?

I think bacteria are always fresh now

[half raptor]

Bacteria spoils into ore

[daring siren]

ah yea they said that some time

[iron root]

yeah, i remember it being said that recipes can choose between refreshing or averaging spoilage

[daring siren]

recipes can choose a set output or ingredient based output

[iron root]

well, whoevers making the recipes can do that
we cant as players

[sage vapor]

no that recipe does not accept quality

to increase fish quality you need to recycle loop spidertrons

[half raptor]

Oof

What if you recycle fish

[daring siren]

Can't recycle fish

[half raptor]

Not with that attitude

[rotund egret]

you do not have to feed the biolab, it's a regular building

[daring siren]

But fish breeding can definitely help multiplying

[sage vapor]

you do have to feed it, it does not take electricity

[daring siren]

You need to feed captured nest though

[sage vapor]

it will not break or anything but to run the biolab needs nutrients

[rotund egret]

?!?

[sage vapor]

ah I meant biochamber

sorry

[half raptor]

Can't breed with which is a blatant act of classism

[daring siren]

Fish can only breed in their class.

No mobility.

[iron root]

thats how you get inbreeding, which should reduce quality

time to introduce a wholly separate quality system for breeding and genetics

[daring siren]

We rimworld now

[kindred crater]

it dawned on me just now... speed modules with quality modules have a really good niche. the routes to aquilo (and beyond) have a lot of asteroids with a lot of material, and you're seemingly tossing it much of it overboard, especially with more damage research and higher quality prod modules. You can put quality modules in the asteroid processing buildings that accept them, and from there you can either ship all of the uncommon+ stuff wherever, or you can quality cycle it on site to whatever level you want.

but getting back to my initial statement, if you're limited on space and need a small setup, a beacon or two around your machines can greatly increase output. and the losses wouldn't really matter, since this is an infinite stream of material. since I've never played SA and don't have a good feel for asteroids on for the routes to the first 3 planets, this could be just as viable for there as well with a larger craft. for cargo groups that are lower priority, non-spoilable, and lower quantity, you can put them on big barges that continually go back and forth to collect/process quality plates/plastic, so these platforms can have some useful role beyond that.

its a guilt-free way to process quality however you want since you're tapping into a truly endless source that can never run out

[iron root]

the speed/quality combo does require qualitied up modules I think, so that you can actually overcome the penalties pretty well
but i see where you're going with this and it could be neat

[sage vapor]

or just make more platform space for more machines

[neat shard]

So... one thing I've never been quite clear on... does this imply with four Q1 qual 3s and Q1 inputs, an assembler has a 11.11% chance to produce an output better than normal quality?

[jaunty citrus]

You could interpret it that way, but you'd be incorrect

[iron root]

the tables wrong, the mulit-rank pulls from the lower level chance

[jaunty citrus]

That's the chances it gets an item of that quality or better

[iron root]

so it should be 9, 0.9, 0.09, 0.01% for the first row

[neat shard]

Alright, gotcha. That definitely felt to me like that's how it should work.

[inner pilot]

I've seen a quality selection for resources on assemblers, can assemblers only take one specific level of quality then?

[wraith moss]

Yes

[inner pilot]

That feels a bit awkwardly restrictive... But good to know! 👍

[wraith moss]

it got changed to remove an exploit with productivity. By crafting a bunch of normal items, then crafting a single legendary item when the prod bar is nearly full you could get 2 legendaries for the price of one

[inner pilot]

that... yeah, I can vibe with that

[stable lagoon]

I couldn't find if it was said if machines could be made to take random qualitys of items or not? like you branch off purple and legendary and throw all the rest in the same thing to craft does that work? Or is it one quality per preset recipe kind of thing?

[inner pilot]

you need to feed the machine completely with the same quality of resources.

can't put a Q1 circuit into the Q0 recipe and vice versa

[stable lagoon]

thx.

[half raptor]

Q0? Everything starts at first quality doesn't it?

[sacred totem]

It’s so sad that we didn’t zero index quality

[half raptor]

Using the notation Q1 = is confusing since there are 2 dots

[limber basin]

but then legendary quality icon wouldn't look as cool

[daring siren]

Q1 is

It's 1-based

[iron root]

has 1 dot
therefore its q1

[half raptor]

Can you recycle u-235 and u-238 to get better quality?

I think someone said you cant quality on the Kovarex process

[obsidian crescent]

You can. But once you have 40 U-235 of a particular quality, you just need U-238 of that quality to make more.

[half raptor]

Right I figured as much. I'll have to fiddle with my idea but the thought was that I Kovarex enrich any and all u238 into u235, matching natural qualities and once I've primed the line with u235 of all 5 lines I just send any excess uranium u235 product through a recycler to hope for an upgrade. I could try crafting it up into fuel cells first as an extra chance for quality before recycling which would give it two rolls of the dice before losing 75% on the ingredient instead of getting only one roll of the dice by straight up recycling.

Though.... Do nuclear fuel cells have a researchable prod bonus?

[obsidian crescent]

Nope. Which is good, because with Q5 prod 3s, burning fuel cells is very close to providing a net gain of U-238.

[charred gyro]

Does anyone want to use more than 5 quality levels via mods? I've seen people joke about it but am not sure if that's something people will seriously go for or not.

[keen igloo]

i have an idea for a 'unique' quality i'm going to try, where you get a chance at a unique building, but you can have only 1 on each surface

[obsidian crescent]

It depends: how much can mods alter the way quality bonuses on machines work. The global 10% change of a higher level of quality than the bonus makes having lots of quality levels a pain to deal with. But if you can alter that, then it could be interesting to have more levels.

I just don't see the point myself.

[half raptor]

If we are going for higher than q5, might as well go for q255 for the meme

Instead of +30% per quality level, you could do 1-3% per level and to account for a wide variety of levels you could go from a 1 in 10 chance to upgrade again to a 1 in 2 chance, and be allowed to craft a recipe with ingredients + or - a few ranks of the goal quality

It would be very janky and very stupid, but could be a fun mod

Even a 1% gain would be good because it could have an end bonus of +254%, we don't get 255 because the first quality is just base

[sacred totem]

remember that time when i said "power poles are not beacons"?

that was cuz beacons are like the only case of a quality effect being configurable in the prototype

99% of quality effects are hardcoded

[daring siren]

Yes. I was very sad when I saw that 😦

I tried to find control for quality, but nope. It's hard coded for performance

Luckily Melon is a mega-modder and made this https://mods.factorio.com/mod/quality-rebalance

[half raptor]

After reading the description (also it discussed spoilers, so read at own risk) it just seems like it removes some of the downsides of quality modules but those limitations and downsides were put there to give people a choice with consequences and sacrifice. I think a base built with vanilla quality is going to be much more interesting than a base built with modded quality

I like the idea that some of the machines do NOT benefit from quality because it means you have somewhere for your normal intermediates to go

[daring siren]

It basically replaces buildings when you put them down, with versions that have different bonuses. It's quite similar to Janky Quality

[obsidian crescent]

I was actually surprised that offshore pumps don't get faster with quality. That really ought to be in SA.

I really disagree with all of the capacity upgrades though. Those sound like UPS traps.

I also am very unsure of anything scaling the radius of something, such that upgrading them in-situ can cause problems.

I shouldn't have to rearrange my farms if I want to use higher quality Ag towers.

[half raptor]

I'll be honest I'm a little shocked we didn't get bigger train wagons with quality, but some things are balanced the way they are and would throw the game balance off if buffed

I'm sure someone is going to make a "everything is legendary" mod and quality will just not be fun anymore. My impression of quality is that it's a feature that will get better as we unlock mk 3 and higher qualities. Once we start to accumulate legendary quality modules, the base is going to really start cooking

[obsidian crescent]

Also, while I would love the idea of quality seeds bearing quality fruit... that would turn Gleba into a quality fest. Get one Q5 seed, and that will multiply itself naturally.

[half raptor]

You'd have to say that quality seeds might have a higher chance to bear quality fruit, but genetic mutation can happen and you'd add a chance for fruits to downgrade

Honestly better to just nip it in the bud and say "all fruit are created equally"

but some fruit are created more equal than others

[neat shard]

I spent the afternoon playing with recipe numbers on a spreadsheet, and I realize purple science in particular may actually be worth the trouble of making in above-average quality.

How much extra value does quality science give over normal science, again?

[iron root]

1, 2, 3, 4, 6x for normal->legendary

[neat shard]

Focusing our attention just on quality levels 1 and 2 for simplicity, one quality level gives you +100% value.

So a 1% chance of +1 quality level gives you, on average, +1% value. So 1% quality can, in a way, be used as 1% productivity.

If you look at the ingredients for purple science, all three of them are intermediate products, and so none of them take prod mods.

And of those three, rails and electric furnaces absolutely guzzle resources – very high costs relative to the crafting time. They guzzle resources so hard that 1% productivity for them gives RoI comparable to 4% productivity on something simple and decent like gears or circuits.

The prod mods that go into purple science aren't quite as good a candidate for quality, but I think the overall package benefits quite nicely from quality, such that it justifies the little extra complexity of routing items to a science assembler.

[wraith moss]

It's multiplicative with productivity too. 10% prod and 10% quality is 110% number of packs worth 110% value, or 121% total value

or with more extreme values, 100% prod and 100% quality are 200% number and 200% value for a total value per input of 400%

[neat shard]

This is true, but if you can insert prod mods in a building, it's probably just best to use all prod mods. The little extra gain from the multiplicative effects doesn't make up for quality percentage points being 4x as scarce as productivity percentage points.

[obsidian crescent]

The thing is, I can think of a lot of useful things I could do with higher quality furnaces (even Q2) and prod module 1s. And by the point where those things don't matter anymore because I have enough resources to make module 3s in bulk as well as Foundries... I won't care about the minor boost that the added complexity of making purple science this way will cause.

Consider the sheer infrastructure needed to make electric furnaces with quality. That means no speed modules or beacons. At 1K SPM, you need 16 assembler 3s making these things. For 10K, it's 160.

Prods get a pass because the EMP is there to both add productivity and be faster.

But rails? You need 23 machines for 1K SPM, so 230 for 10K. That's a lot.

All for a marginal chance at a 2x better pack, that you have to stuff into special labs so that you don't get packs stuck.

I'd rather have the smaller production center.

[neat shard]

The incompatibility with speed mods is pretty unattractive for megabasing, granted, but a smaller base working on getting to space for the first time might benefit from throwing qual 1s into their purple science production line.

[neat shard]

I'm not sure I can visualize how this would happen in practice. You should just be able to mix common and uncommon science bottles without any problem?

[main ruin]

quality for prod science isn't probably aimed at 1k SPM bases

for early game it makes sense

I think you will overflow with quite soon

[obsidian crescent]

Quality electric furnaces and prod modules are useful in the early game. They're better than high quality purple science.

[main ruin]

but I playtested quality waaaaaay too far back to remember it exactly

I just remember uncommon stuff being devalued pretty soon

[obsidian crescent]

Even at 100 SPM (which is a lot for "early game" science), you only make 24 prods per minute. If you're using 4 quality module 1s, then 4% of them are higher than base quality. That's 1 per minute.

It will take 40 hours for a single steel chest to overflow with that.

[main ruin]

you don't use anywhere close to steel chest worth of prod1 modules in the entire game lol

what an argument

or any lvl 1 modules lol

[obsidian crescent]

You said "you will overflow with" them. I pointed out that you will not.

[main ruin]

you will overflow after a few stacks

[obsidian crescent]

Furthermore, they can be used to make Q2 prod 2s assuming you can make quality blue and red circuits.

[obsidian crescent]

What does "overflow" mean besides exceeding the storage capacity of a chest?

[main ruin]

maybe sooner if you qual up circuits so you actually get multiple sources of qual'd modules

[main ruin]

technically correct is the best kind of correct, I see

I just meant you will run out of space to put them

you will get some rare stuff when you also qual up some circuit assemblers

[obsidian crescent]

In any case, later on you can harvest any extra prod 1s and electric furnaces with a recycler to make Q2/3 circuits and steel.

[main ruin]

so might not be that valuable

[obsidian crescent]

I'm not putting quality modules in circuit assemblers. Not unless it's already hit 300% prod.

[main ruin]

yea putting it into chest and jumpstarting recycling loops might be also an option, and easier to set up

[main ruin]

and some people will put quality modules in miners! The horror!

People are trying to think about different strategies, no need to tell them that your is obviously superior

[obsidian crescent]

And I'll be one of those people. For a while.

[inner pilot]

Y'all don't know that it's Q0, Q1, Q2, Q3, and Q5, based on how many quality bonuses the tier gives 😤 🙃

[kindred crater]

actually so real

amazing level of basedness

[inner pilot]

I am joking, but that is also 100% how I'm using QX for now, until outside forces change me XD

[jaunty citrus]

That is both disappointing and makes no sense

[earnest oar]

Unpopular opinion: I don’t like that productivity research trivializes the cost multiplier for high quality items …

I would like it to be expensive till late game so high quality items mean something special and you need to solve the problem with large scale production.

[obsidian crescent]

You will not be getting 13+ levels of item productivity research before "late game".

[earnest oar]

I meant i want it to be expensive through late game not just before late game.

[neat shard]

The thirteenth level of an item productivity research is about 130x as expensive as the first, so take that into account when evaluating what's possible when.

[earnest oar]

I just don’t like item prod research in general, also competes with prod modules and makes them feel not as powerful.

[obsidian crescent]

But It doesn't really matter. It doesn't matter how cheap Q5 blue circuits are if prod module 3s require biter eggs, which is almost impossible to produce in quality without recycling. This is true of a lot of the SA-specific items: Biochambers, Foundries, EMPs, all of these use intermediates that don't come from blue circuits or LDS.

[jaunty citrus]

The selection of items that have prod research is very limited, and mainly exists to make rockets cheaper the more you play

[iron root]

yeah, they are very much targeted towards rocket production. The only ones that significantly break from this steel and plastic of, which dont recycle back to their ingredients

theres asteroid prod too but thats more like mining prod

[arctic gyro]

same with scrap recycle prod

[obsidian crescent]

Sadly, no fruit productivity.

[sacred totem]

yes

the biochamber now has four modules slots + 50% prod and you still want more? greedy

[obsidian crescent]

Hey, the recycler has 4 modules slots too. Sure, none of them can have prod modules, but they're still there 😉

[half raptor]

You should only use prod modules on anything you can't quality, like things that are already or fluids

[obsidian crescent]

Oh, BTW: the question came up a while back about whether forbidding quality in catalytic recipes was an engine thing or a per-recipe setting. Is it something the engine does when processing a recipe or is it something that's done by the maker of the recipe (thus allowing mods to have quality with catalytic recipes)?

[daring siren]

Maker of the recipe

Kovarex specifically has "allows_quality" false

[obsidian crescent]

And all of SA's catalytic recipes forbid quality?

[daring siren]

Also, they changed the limitation system to go from the recipe side.

Which makes more sense

[molten bobcat]

Do we have a video how quality really works and how the best way to get legendary stuff?

[daring siren]

We had like 300000 discussions on it

[molten bobcat]

xD ok

[daring siren]

And there are many different ways to go at it

[obsidian crescent]

"Best way to get legendary stuff" is something that will take a long time to figure out.

[molten bobcat]

😄

[obsidian crescent]

The fact that there are so many ways to engage with quality is one of the best things about it.

[daring siren]

It's not "solved" in any way, shape, or form

[daring siren]

Speedrunners will grok that for 100%

[wheat sluice]

do T3Q5 qual mods have 6.2% quality or 6.25%?

[distant eagle]

6.2

[obsidian crescent]

6.2%. So yes, the rounding thing is still there.

[wheat sluice]

😦

[grizzled lagoon]

Yes it has so many layers. Subproblems are quite easy: which modules are optimal to quality cycle, or for multi step quality. But which item is optimal on which planet through which process has so many edge cases that it seems near impossible to me to say, e.g., 'for quality iron, go to Gleba'. I'm even considering a fully fledged quality base everywhere and then just distribute whatever each planet can make to the others. That way, I'm guaranteed to mostly use whatever process is the fastest

[iron root]

quality ALL the ores
and then be sad when things like holmium have to go through a fluid step

[sacred totem]

I wonder how viable it is to use quality moduled drills on fulgora

[obsidian crescent]

Well, it really depends on what you want. If you want quality holmium derivatives, that's not helping. If you want quality other stuff, it can help.

Just remember that you're slowing the miners down, both by not using speed modules and by the speed penalty of quality modules. So it'd probably be better to mine the stuff and shove (at least some of it) through quality recyclers.

Though that just shifts the speed problem elsewhere, since you need to recycle a lot of scrap.

[iron root]

yeah, i think nauvis is going to be the only place i do significant amounts of quality ore. Could maybe do it on vulcanus too (lds cheese go brr).
Fulgora will need recycling most likely due to holmium and gleba is...
uhh... no.

[obsidian crescent]

Gleba does have potential for quality production. Once you get a single QX ore bacteria, so long as you can continuously produce QX Bioflux, you can make as much ore of that quality as you like. The question is how much effective you can be at making quality Bioflux.

Fortunately, Bioflux is really shelf-stable, and higher qualities are moreso.

[iron root]

I'm just not going to bother with gleba quality until postgame I think.
Maybe quality cycle (and ) though

[kindred crater]

at least from my current understanding, gleba looks like the best place for quality plastic, which will subsequently be transported off planet to vulcanus which the bulk of my iron/copper heavy recipes will be quality cycled

[neat shard]

"The best way to get legendary stuff" is not a uniform thing. It varies from item to item.

Some items are theoretically somewhat amenable to naturally gathering high-quality ingredients; others you really don't have any choice but do a recycling loop.

[gusty trench]

Do quality personal weapons do anything different?

[iron root]

range

[gusty trench]

Nice

[iron root]

more range on personal railgun to destroy more of your factory when you misclick "c"

[neat shard]

Just unbind the key

[gusty trench]

Rebind it to the space bar

[civic pivot]

I tried to make formula how many times it costs more to make an item to at least better quality :
X - input cost
Y - item with better quality
Q - machine quality (0 to 25%)
P - productivity ( 0 to 300%)
x = y * Q + x * (1 - Q) 0.25 (1+ P /100) ,so :
y * Q= x - x * (1 - Q) 0.25 (1+ P /100) , finally :
y = ( x - x * (1 - Q) 0.25 (1+ P /100) )/Q
in conclusion:
an assembler with 4 legendary quality modules and 0 productivity would need 3.25 input to make an uncommon item ,so 3.25^4 is 111.5 more resources is needed to make legendary item ( assuming only + 1 quality stuff is made)
also with 300% productivity y = x so it cost the same to make an legendary quality stuff

[obsidian crescent]

Quality isn't nearly that simple. There are many ways to go about making quality, and the one you're using (where you only consider one step and pretend that all of the non-quality items vanish into the aether) is the least practical one for making very high quality goods in most cases. It's not even considering quality cycling, where you get 25% of the inputs back for lower quality items (along with another chance at a quality boost).

[half raptor]

I'm going to base my quality balancing using train logic. Each of my stations reads its buffer to determine the priority score for that station so that will go to where its needed most, and this priority score (red wire) along with station count (green wire, value of 1 per station) will be used for measuring averages across the base elsewhere, but I can use these priority scores for other purposes too.

Each production leaf off of my rail stem will have a station for each ingredient, sometimes 2 depending on throughput demand and have 1 station for product and 1 station for product quality upgrade (referred to as upgrade station from now on). At the upgrade station I will use filter splitters to fill 1 chest with legendary, 1 chest epic, 1 rare, and 3 for uncommon. At production lines where a higher quality is the base quality, this station will be changed to fit the logic. Piggybacking off of my wire that is measuring my base's priority scores, I will only load if there is a demand for legendary being read for that item. This is so that when I unlock legendary and I haven't built any stations that can accept it, the inserter will not read a demand value for that quality and thus will not be enabled when the train comes for the upgrade items. For my epic items, it will check to see if there is a demand for either epic OR legendary. The reason it's checking for higher qualities is because there's the chance discussed below for a quality upgrade and if I do NOT have a demand for either of those two qualities, I don't want to load it. Each inserter responsible for each quality will do a similar check, also checking for demands on qualities higher before loading, all the way down to normal items.

Once loaded onto the train, which will check to see if it's at least half full and inactive for 3 seconds. This is so that partial stacks of quality items will not make the train wait for full.

Then I'd let the train interrupts go out and deliver as needed with higher qualities delivering first across the base. However if a train encounters a Destination Full error, I will then make it wait 10 (maybe 5) seconds. This is just in case a station frees up in that time trying to deliver. If the destination is still full for 10 seconds, it'll go to the Recycle station. I hope my idea will work but I'd want it to read the signal off the train and I'm hoping it'll read the signal for the interrupt it's on to set the filter on the inserters so it'll ONLY take off items for that train signal. If it doesn't work this way, I'll just make sure to deliver higher qualities first since it'll probably be normals and uncommons going through the recyclers most often. The output for the recycle yard gets picked up by the same schedule train as what deliviered it and will distribute items again for further crafting.

The logic is that i wouldn't be stuck in a loop of constantly recycling my uncommons unless there's room for rare, epic, or legendary variants. if those are full and there's no demand signal for the uncommon item, then the inserter will be disabled and will not load onto the train. It will only do so if the logic says there's room for upgraded. I'm also going to set my input buffers very high. Six chests can hole 7.2 train loads and I'm requesting 6 train loads be delivered because when it fall below 5.99, it'll have enough room for that full train to deliver without going over the capacity. With such a high request buffer and hopefully fast belts delivering, the only time an item would be cycled for recycling is if I'm producing too much anyway or there's a shortage of a secondary ingredient. This will hopefully be covered in my alert idea which will notify me if a) I have enough ingredients to craft more and b) there's a strong demand for more product

When I first start my run I'm not enabling any "go to recycle station" conditions and will add them in manually as I go just to see if such an idea is even necessary. Might be fine just letting trains sit in stations until they're needed, but I feel like it's a good opportunity to recycle excesses because crafting more will yield occasional upgrades anyway and I don't want to stop crafting. Might be wasteful use of resources, but I'll save that thinking for a run when I'm more focused on prod than qual

gah 4.5 days!

[distant eagle]

so, with a ~25% chance to not be normal quality, you think you need 3.25 attempts to not be normal quality?
also, if the chance to be uncommon is 25%, the chance to be rare is not 25% of 25%, it's 10% of 25% because the 25% is actually 90% of 25%. but if you have legendary, then rare is 9% of 25%. (input for 1 step)^(steps) doesnt make sense

pretending that 4 quality modules will give you 25% quality...
the number of normal produced is (input)(1-.25)
uncommon is (input)(0.25)(0.9)
rare: (input)(0.25)(0.09)
epic: (input)(0.25)(0.009)
legendary: (input)(0.25)(0.001)

(4000)(0.25)(0.001)=1
but it's slightly higher due to 24.8%

[gleaming quartz]

Do quality wagons have more inventory slots?

[kindred crater]

just more hp

[jaunty citrus]

My guess is not very outside of niche circumstances.
The only viable way I've seen quality be used in drill is for ore, where you can filter for only uncommon items by trashing the rest, which you can do because of ridiculous drill yields and speed from productivity.
You can't exactly throw away scrap like you can ore, it turns into a bunch of other stuff when you recycle it.

If you're going to instead use it just to make the highest quality items possible and not use it for science/at scale, then you're going to be competing with all the other strategies those requirements make viable. I don't know which one is 'the best', and there probably won't be one

[steel falcon]

I can't wait to see the setup for cycling quality nuclear and fusion reactors

[obsidian crescent]

That reminds me. I'm surprised I'm not saltier about quality power generators being nearly useless across the board (with the exception of solar and accumulators).

[grizzled lagoon]

If you just want the results, quite a few people made exact calculators that take everything into account. Most of them just focus on a simple craft, recycle, repeat set-up. If you only have 4 module slots, use legendary modules and no productivity bonus, it takes about 40x as much material to make something legendary. Here is the one I made, though it needs some more work as I didn’t cover cases where you first craft x times before recycling.

https://docs.google.com/file/d/1GpszMksTmaaIV4opnKTx_e7JrrBPRaSp/edit?usp=docslist_api&filetype=msexcel

[civic pivot]

Thanks for answering, yeah i very much oversimplified it. By ^ steps I thought of recycling higher quality products to use better crafting recipe. I'm gonna test it in creative when SA releases

[kindred crater]

the 8 module slots in the cryo for some some/most? of the fuel's and coolant's crafting recipes makes me feel better about that

[steel falcon]

It's just kind of awkward. Quality Solar effectively makes more power for free, whereas a Quality boiler or steam engine consume resources faster. From an objective standpoint comparing to assemblers it makes sense, you're increasing craft speed, but from an intuitive standpoint one would maybe expect them to make the same power with less resource input.

[half raptor]

Quality solar just consumes sunlight faster

[kindred crater]

it just happens to be consuming from an infinite source you don't have to manage

[supple storm]

is it too late to make sure if people say Q1, Q2, etc. those correspond to the number of pips so Q0 or Q1 are both normal quality, and Q2 is where it is actually upgraded?

[inner pilot]

Yes (no)
One way will win, and it's probably the way of thr pip number

[daring siren]

99% of people say Q1-Q5. The only ones I see saying Q0 are those mistaking level with per tier bonus

[supple storm]

ah kk

[steel falcon]

Obviously the notation should be Q0 Q2 Q3 Q4 Q5 ^^

[gleaming quartz]

The programmer in me will refer to pip 1 as Q0

[frank warren]

There is an argument for Q0, Q1, Q2, Q3, Q5 though...

(this is not my opinion, but there is an actual argument here 😉 )

[steel falcon]

Call them by their right full names: vendor trash; quest greens; dungeon blues; epicssss; thunderfury

[frank warren]

alfonse can't tell me no

[jaunty yacht]

Thunderfury, Blessed Blade of the Windseeker?

[steel falcon]

Blessed Gear of the Circuitseeker*

[jaunty yacht]

I'm trying to wrap my head around quality but I think I just need to play around with it myself and see the numbers one SA releases

But if you goal is to get something legendary, is it generally better to upgrade ingredients, or just mass-craft them and hope for the 0.01%?

[half raptor]

Coming from an RPG background, I see no issues with calling them what they are

[steel falcon]

There are arguments for both, it's not really a solved system yet which makes it interesting and fun

[half raptor]

If you start with quality ingredients, it will 100% come out at least that quality

It's up to you on how you want to process items

[steel falcon]

Wait did you just repeat a Papa John's commercial

[half raptor]

Didn't mean to

🍕

[kindred crater]

this is cool! construction robot energy capacity is 100% per level, 200% to legendary

[supple storm]

aye

most people seem to be sleeping on the effect of quality on the lightning collectors

[kindred crater]

what does it do?

[steel falcon]

converts more or bigger areas? most streamers I've been watching seem allergic to reading tooltips

[supple storm]

both.

it puts them right up with the grabby arms in being worth upgrading

[steel falcon]

yeah that's pretty big

[supple storm]

yet I've seen seas of legendary accumulators...with normal quality lightning collectors amongst them

[kindred crater]

isn't there 2 tiers of them too?

[supple storm]

yup

[kindred crater]

look at that range lol

[steel falcon]

It will be interesting to see if people just brute force Aquilo with regular bots or invest in epic conbots for it

[supple storm]

range, you say

[daring siren]

😄

[kindred crater]

do quality enemies come from quality eggs?

[supple storm]

quality spawners lol

quality capture rocket -> quality spawner -> let it go rogue -> it can create expansion waves with its quality

you'll need a mod to separate the different qualities onto their own forces to out-compete the other biters though

[kindred crater]

that is awesome lol

hopefully griefers never find out about this stuff because theres so many more creative ways to mess with bases with these bio mechanics

[supple storm]

some subtle time-bombs for sure..

[kindred crater]

another 10/10 place to put quality modules

[iron root]

huh, range doesnt scale in the 1.3, 1.6... etc sceme

it scales less

[frank warren]

this is for a reason you know? you are almost never collector limited on fulgora.

[obsidian crescent]

Collectors require some holmium derivative, but do rods?

[frank warren]

no

[obsidian crescent]

So it might be worthwhile to make high-quality rods off-world and ship them in (assuming Fulgora can't make them itself).

[supple storm]

isn't it possible to reduce one's storage demands some if one increases the periodic production capacity some?

like if solar is productive enough you don't need as many batteries since the net positive window widens?

or is the cutoff for the lightning collection so sharp it doesn't have that effect and instead it's purely the staying power challenge?

[sacred totem]

you cant make the rods off world

both lightning fixtures are fulgora only

[iron root]

can still ship in quality material to make a few initial quality rods

[supple storm]

either way you could have fewer of the collectors if they have more reach

I don't know if that also means they grab everything from their areas, too, meaning you get purely an increase in power even with fewer of them

[obsidian crescent]

Ugh. Well, maybe quality materials can be shipped in instead.

But you do make them in the EMP, right?

[supple storm]

yes

well, the second tier, first is more flexible

but a high quality first tier is comparable to a second tier

[half raptor]

?

[supple storm]

yup.. from rogue spawners

[half raptor]

So they don't do quality naturally?

Good

[daring siren]

You can, but also in assemblers/by hand

More reach means they reach farther into the oil oceans

[reef saffron]

do quality fish retain their quality if put back in water?

I want a legendary lake

[daring siren]

No. They lose it. Same as tiles.

[steel falcon]

wait I thought putting fish back was just a running joke, but it actually works

wtf

[charred gyro]

catch and release

fish are friends

[daring siren]

Resets their freshness

[iron root]

I still wanna know what happened that one time someone put a fish in the Aquilo ocean

[daring siren]

Not sure if possible. Will check when I'm near the game

[quick rampart]

recycle the final product and craft until legendary, raise quality gradually in the crafting process, which strategy is better

[wraith moss]

Third option: quality the miners and sort the resulting ores.

[quick rampart]

well, my goal is to get legendary items...

[iron root]

its more or less always gonna be recycling final product for forcing legendary i think, because trying to quality everything in the process just limits the item by its shortest chain

[wraith moss]

And you can send the 90% normal quality stuff to your science assembly line, and use the 10% quality stuff as a baseline. essentially removing one quality tier from the equation entirely.

any excess lesser quality stuff can go into quality science packs, while you save the best of the best for personal stuff

[faint blade]

What does operational time mean here?

[quick rampart]

draining excess resources by science packs

[wraith moss]

That's either a construction robot or a logistic robot. The operational time is (i think) referring to the battery capacity (the time spent flying)

[mortal tapir]

Is there a list/wiki of what all the quality upgrades do?

[wraith moss]

It's in the ingame factoriopedia. Most of the effects are listed on the original FFF for quality: https://factorio.com/blog/post/fff-375
I don't know about any compiled list of everything, but there is a set of videos by a Japanese player where you can see all the effects as of the time of recording: https://www.youtube.com/playlist?list=PLmEpv00ZHGv-yw_niV_Y8b3qud9a1ZWOd (although it is in Japanese, but you can see the actual numbers being affected)

[mortal tapir]

Sadly don't have access to that yet, guess I'll have to wait a few more days 😢

[stable lagoon]

I was watching krydax stream earlier and he found out you couldn't add quality items to his rocket silo. Is that because the silo itself wasn't quality? or that it couldn't be used it at all. The second option would suck.

[main ruin]

were all 3 ingredients of the same quality?

maybe the exploit fix messed it up that way

[stable lagoon]

he removed all ingredients and made sure there were no in progress parts but it wouldn't let him put it in even manually

[main ruin]

well that's unfortunate, I hope he reported it

[stable lagoon]

he did

[supple storm]

wait, why? you wouldn't want to make a rocket out of quality ingredients if it doesn't have any benefit

[obsidian crescent]

Buildings have a filter that says what quality of inputs they take. It defaults to Q1, but it's just a drop-down you can select. But if the rocket silo doesn't have such a dropdown, then you can only use Q1 ingredients.

[half raptor]

Seems like you route your to rockets and your quality chips to module production

[main ruin]

E.g. if you use qual modules in scrap recyclers on fulgora you can have excess qual solid fuel

Generally having excess of quality items will happen as you move up the levels (unless you ignore quality before cycling for legendaries at the end), so using them for rocket launches is actually useful. Just voiding in recyclers just feels bad.

[supple storm]

quality solid fuel you can make into better fuel for trains

[main ruin]

Trains have miniscule consumption compared to base output though

But yeah it is kind of a niche

[heavy ember]

Out of curiosity: with quality enabled in 2.0, where does the recycler unlock land in the tech tree?

[kindred crater]

I think its actually very early. some stream I was watching yesterday briefly showed it, it could've been a trigger tech from mining scrap

[heavy ember]

NOT with SA. I'm talking about 2.0

since you can separately enable quality from SA if you have the dlc

[kindred crater]

oh now I understand your question. I don't know

[obsidian crescent]

Purple science. Makes sense. So it doesn't require blue circuits in vanilla + quality.

[heavy ember]

Thanks

[obsidian crescent]

So... shouldn't the research require blue circuit research?

[supple storm]

could be an oversight

anyone doing that particular combination of mods is likely to know what they're doing though

[charred gyro]

Doesn't blue science require that anyway? No need to have a redundant connection if it's needed to get to purple science in the first place

[sacred totem]

No

[merry spindle]

Why?
It's not like you actually create blue circuits at Fulgora

[obsidian crescent]

This is the research for vanilla+quality, not SA.

[half raptor]

You do if you're quality cycling 😉

Seems like you should set up delivery stations for from scrap and then a recycle station when the destination is full

With quality inserters, I don't suspect train stations will be occupied for very long

[merry spindle]

then recycle station when the destination is full
Hmmm...

[half raptor]

You should absolutely abuse your train interrupts

[iron root]

Oh, what are the power cost modifiers on quality 1/2 modules?

[wraith moss]

Quality only increases the main benefit of the module. i.e. a quality speed module is even faster, but has the same energy consumption as a normal speed module. It doesn't change the energy consumption, except for efficiency modules, where the consumption reduction is the main benefit.

[obsidian crescent]

As I understand it, none. Directly. They make the machine slower, but that increases the cost per craft. But they all have the same indirect power cost.

[iron root]

So quality module is just straight +x% quality/-5% speed and nothing else

Unlike speeds/prods which add energy costs

[obsidian crescent]

They really don't need much of a downside. The opportunity cost of using quality modules is, in many cases, non-trivial. So if you're using them, you're not using something else.

[iron root]

Not having a power modifier will be really nice for quality miners

I don’t get efficiency mods but base miner pollution is still… manageable

[clear scarab]

you can still use efficiency beacons with them

[iron root]

I could, it’d be awkward until bmds though

[molten bobcat]

I thought that belts have no quality. The speed of the belts increases drastically 😄

[clear scarab]

They only get health.

[molten bobcat]

oh okay 😦

[stable lagoon]

Yeah I don't understand that one. You would thing higher quality belts would approach the next tier of belt and then the final tier would have a good legendary with the most speed. The health is just disappointing. I hope a mod brings that after release.

[obsidian crescent]

It's better to not have 20 belts with slightly different performance characteristics. Maybe if belts didn't pre-date quality, there might only be two kinds of belts. But as it currently stands, using quality to do what belt tiers also do isn't the best thing.

[stable lagoon]

But then you could say that about every machine in the game that gets a speed increase. a 15, 16 , 18, 20 ,22. 25 for belts wouldn't be that out there especially since we still use normal tiers as a baseline such as t1 assembler or t2 , t3 etc. Now there will be several quality's in there that modify their speed (not to mention other ways to modify speed) but we will still see the next tier as an upgrade when looking at it from a speed point of view. I guess you could argue power consumption for those but that's not that huge a thing to work around.

[obsidian crescent]

There are exactly two machines in the game that get tiers: assemblers and furnaces. But even then, there are real differences. Assembler 2s can do fluid recipes; 3s have 4 module slots. Electric furnaces can use modules.

Outside of underground belt distances, the only difference between belts is speed.

[half raptor]

the issue with belt speed is how the game engine works. Every tick, a moves an item 1 pixel, and each upgrade adds 1 more pixel to the travel per tick. It's difficult to move items 1.3 pixels since each item is tracked in movement every game tick

so you'd have to upgrade it in terms of whole pixel values, which right now 2 pixels is just a straight up red belt

Only way I could see belts working is perhaps the item spacing could be more compact

[stable lagoon]

thats fair but couldn't that mostly be resolved by false moving like how smoke and bots use?

[half raptor]

instead of moving items faster, you're just able to compress them tighter so instead of a new item appearing every 8 ticks (8 pixel spacing) it only takes 7 ticks at uncommon, 6 rare, 5 epic and 3 legendary

potentially

though your items will probably look VERY squished

[sacred totem]

thats not how it works though

[half raptor]

also you'd have to change the base code for how belts work to accomplish this.... and it kind of goes against the design principle of belt throughput mechanics

[sacred totem]

each tile is 32 pixels, 7.5 items per second is a rate of 240 pixels per second, or 4 pixels per tick

...that math actually works out, huh

[half raptor]

that's the whole reason they gave us slightly different shaded items when they gave us mk 4 belts, because moving 4 pixels with an 8 pixel gap made it look like it was flickering back and forth, so the slightly different shaded items gave the illusion of motion instead of the flickering

[sacred totem]

a full belt with 60 items per second would look stationary

not flickery

[half raptor]

in the stacker inserter and belt reveal FFF they showed how they had to give us the illusion of motion because of the higher tick movement

[half raptor]

items move 4 pixels per tick, every 2 ticks, an item has moved across a tile per side of the belt, meaning every 2 ticks we move 2 items (both sides) for an average of 1 item per tick. 60 ticks per second, 60 items/second

belt speed is pixel based

that's why back in 1.0 development one of the things they fixed was belt throughput issues by working on item spacing, it used to be 9 pixels between items until they changed it to 8 so that belts will always have 8 items on them instead of going back and forth between 6 and 8

[sacred totem]

belts are 32 pixels long...

[half raptor]

and a spacing of 8 pixels, sorry I meant to say travels from one starting point to the next on a belt

which is why you can have items be 4 items long per side

the reason it is hard to see if the items are moving is that items move 4 pixels, and then the next tick after moving 4 more pixels, the item is moved to the same position the previous item was in

hence the flickering animation, and why they shaded the items to give the appearance of movement

if you had a belt that could go 5 pixels per tick, a compressed belt would appear to flow backwards at 3 pixels per tick (blue belt) without that

[sacred totem]

you are making sense, but you have the wrong pixels per tick

[half raptor]

yellow is 1, red is 2, blue is 3, green is 4 from what I understand

[half raptor]

belts are 32x32 pixels, item spacing of 8 pixels

[sacred totem]

wait, maybe my math is wrong. let me think about it a little more

[half raptor]

why does it sound like we are agreeing

[sacred totem]

oh, we are

just not on the exact numbers

[half raptor]

okay lol

anyway the original point I was making is that if you were to have a quality belt, it's not practical to increase speed of the belt because of how the animations work

[sacred totem]

yeah, the speed increase would be awkward to work with

[half raptor]

if you wanted to increase belt throughput without changing the animation speed, you'd have to look at the item compression on the belt instead of the speed

[sacred totem]

and theres no way thats changing

[half raptor]

no, belts are hard coded for an 8 pixel gap, you'd have to change that purposefully in the base code, not as a mod

if devs had that intention in mind, they could make the belts able to have a tighter item gap for higher quality belts, but truth is that I doubt they're interested in that

the ability to stack on a belt is probably the best innovation we're going to get in terms of upgrading belt throughput

now..... quality undergrounds..... that is a different story. Underground length is an easy to mod quality stat

[sacred totem]

it would look weird

[half raptor]

as long as you code the rule that can't pair with a

it would look weird, it'd break game balance, and belt weaving would be broken as heck, which is why they probably just said "nah, quality is only an hp boost"

personally I kind of like the fact that we have machines that is useless to go for quality on, because it gives an excuse to put our normal materials down one path and any quality intermediates we get down another

[iron root]

okay yeah watching a stream i can confirm that as of right now quality modules dont directly increase energy costs
nice

[distant eagle]

Since speed can't change, now have +1 lane

[small egret]

Has anyone tried to mod in negative quality modules? What happens?

[wraith moss]

Negative quality modules? You mean speed modules? They reduce the effectiveness of quality modules but don't cause items to downgrade in quality.

[small egret]

Ah you’re right I forgot about that. I was kinda hoping negative quality could downgrade things.

[stable lagoon]

I wouldn't mind a true random mod where the quality had an even chance to upgrade or downgrade

[warped obsidian]

I still think it would be cool if quality underground's reached further

[half raptor]

I'm hoping to set up a system that will roll for quality on every step of the way, using some circuit logic to only load different quality of intermediates onto the train depending on demand across the base and a separate recycling train stop. I will probably control one item at a time for having that stop in their interrupts, but I think eventually the system would stay pretty balanced as long as the engineer is monitoring and correcting as needed. The one thing I'm nervous about is focusing on another planet for 3 hours, coming home to Nauvis to find a traffic jammed system

[distant eagle]

yes, i remember

[half raptor]

[half raptor]

so a poll for the quality enthusiasts. What stage of the game are you investing heavily in quality production?

[earnest oar]

Nothing before fulgora, I will try to make quality^2 modules in fulgora for a few select things, then unlock legenedary on aquilo, go for a second push for some legendary, then automate consistent science production, finish all non-infinite research, and go for a final quality push for anything that makes sense before trying to scale up science production / megabasing / post-endgame stuff.

[half raptor]

probably a good notion.

[main ruin]

what do you mean by heavily

having quality modules in 1 out of 6-8 GC assemblers counts?

and similarly across prod

[obsidian crescent]

I plan to invest quality modules initially in module makers (this will likely be my first or second try at doing circuit-controlled recipes, switching which module is made based on how many are available), as well as assembler makers, probably miner crafting, and maybe one or two other frequently produced things.

And then put some in miners, filter out the quality ore, and hold on to it until I have nuclear power and electric furnaces to start smelting them for other quality goods.

Then invest in purple science, where quality modules can go into the furance and prod makers so that I can have some quality furnaces for my eventual trip.

[earnest oar]

Do you guys think sushie belt with recipe-quality-filters, some recyclers and some circuits will be a good way to produce quality in mid game? That's the type of setup I'm imagining for the first quality^2 modules I will be making with EM plants.

[main ruin]

sushi belts and quality mixing seems above my paygrade

I am going to just sprinkle quality mods across Nauvis base, aiming mainly for better stuff for personal equipment / space platform / Fulgora jumpstart. Then on fulgora maybe some first light cycling of stuff that will massively overflow. First real cycling after epic quality on G, no idea what but I bet I will want something upgraded by that point. Legendary is kinda too far for my plans, I am not rushing to the end

[earnest oar]

you can force the assembler to select one quality tier as its input, so it will ignore anything else from the sushie belt, this seems promising to me. Shouldn't be too hard, just need to make sure belt doesn't fill up or clog up, that's the main challenge with circuits.

[earnest oar]

This approach is most resource efficient to make quality, but the amount of spaghetti it will make dissuades me from considering it. I don't want inventory clog, and increase my base size too much just to get some quality.

[main ruin]

ye "shouldn't be too hard" is something I can't say about that lol. I will first try recipe switching on platform for asteroid processing, maybe

[main ruin]

why inventory clog. Belts for main production, bots for quality / mall

no need for the stuff to spend almost any time in your inventory

[earnest oar]

while building, you may have to pick up some stuff manually just to fix bugs etc. I want to keep the quality stuff away in a small modular area.

[obsidian crescent]

Working sushi requires strict controls over the ratios of products on the belt. You have no control over the ratio of what comes out of a quality-producing machine.

[main ruin]

there won't be many bugs (before Fulgora, there it might be fun), filter splitters are simple. Also logistic trash for quality > common might be the play for a while

[main ruin]

yea I am not talking about sushi for my quality plans, those will be dumb filter splitters and red chests

[earnest oar]

every filter splitter and belt dedicated to certain quality tiers is increasing the area needed for the build, potentially causing spaghetti issues.

by bug I meant errors in the layout

[main ruin]

if that is the spaghetti causing issues for you I have no idea how do you handle advanced circuitry for sushi belts lol

that's even simpler than main bus splitting / re-balancing

also potentially there won't even be splitters, just maybe some advanced filters on inserters for assemblers with qual modules

but I have no idea how easy will that be to setup in spage

and quality bot mall is dead simple

and if it won't be, I can still pivot to the quality-miners route 🙂

[earnest oar]

Can't explain my design idea with words properly. Will share if I'm successful in a few days 🙂
My goal will be to create a quality startup build. I'll optimize for small area, less belts, less machines, rather than raw resource efficiency and speed.

[main ruin]

oh you are talking about your build

[half raptor]

I have my plan for end-game, but I think for early game i'm just going to put my best s into my machine factories and just rely on my normal machines, using my upgraded ones on the space platform and such until I've unlocked the pieces I need to do some quality cycling

[main ruin]

I think another quality poll question might be "how soon are you going for rare+ stuff in your personal armor"

and I have no idea about the answer for that one, because I will prioritize platform and tank before armor

[half raptor]

I saw a few different takes on interplanetary platforms, going for some quality power and quality asteroid collectors seems like a good investment in quality

3 arms at rare

[hoary current]

funny that a T3 quality modle is not as good as a T2 uncommon quality module

[obsidian crescent]

It's similar to how a Q2 eff1 is almost as good as a Q1 eff2. I still think the +2% quality is a typo.

[hoary current]

prod module is 4->6->10% from t1 to t3

speed is 20->30->50%

eff and quality modules breaks the pattern

eff is 30->40->50%

quality is 1->2->2.5%

I wonder if it really should be 2->3->5% for quality modules. I never know how they balanced the numbers

for eff maybe 30->50->80%

[half raptor]

yeah, definitely worth keeping a supply of epic mk2 modules around

"we have quality modules at home"

[iron root]

as soon as i can

will be throwing quality 1s in my miners basically the moment i unlock them, though i dont think ill get the chance to go too quality crazy until i finish the EM science pack and begin infinites with prod science after, because I wanna get that rush to space achievement

[burnt reef]

imo it should be 1->1.5->2.5% for quality

but I'm sure they have their reasons

[iron root]

Level 2 being 2% is all the better for quality spamming

[autumn magnet]

How’s it going folk, I have and idea that must be stupid, so need someone to convince me that it is. Early game why shouldn’t I quality up my miners and siphon from them the resources for deterministic uncommon/rare early game space station?

[clear scarab]

1. Significantly increases complexity with logistics
2. using quality modules means you're not using efficiency modules

[autumn magnet]

Ah, the natives… bummer

[jaunty citrus]

Quality modules also come with a speed reduction, so all your production suffers

[autumn magnet]

Yeah, but it’s mostly needed so I can push out 2 stations (science & travel to next planet) quickly, wouldn’t be a long-term solution

[jaunty citrus]

If you just want to push out some platforms quickly, then quality doesn't matter

Best use would probably be quality solar panels, which you can get just by making a lot of them with

use the normal ones to power your nauvis base

[dire cosmos]

can you make quality quality modules

[jaunty citrus]

Yes

[dire cosmos]

time to quality everything

factory will be 50% recylcers

[cold wadi]

||are there quality fluids?||

[jaunty citrus]

no

[dire cosmos]

How would that even work 😂

It’s all gonna get mixed up in the same pipe