Friday Facts #375 - Quality

[aka13]

doesn't look like anyone's brought this up but with quality assemblers, does this mean there's going to be overrides on productivity for some items?

Generally speaking, the quality assemblers giving a small boost to the productivity of all recipes is a unique and powerful bonus but....

Fluid Barreling is going to be broken with productivity, even 1% productivity, barrel then unbarrel in a loop and a little fluid and 1 barrel can become as much fluid and as many barrels as you like with only a time cost, and it's relatively scalable. I know you've accounted for this with uncrafting but I think it needs to be thought about for all recipes

I'm not sure if this is the only infinite loop but I've seen mods that have had to forgo productivity on entire recipe chains because 1 element in that chain would have this sort of feedback loop. I think bobs or angels has cryocoolant as a recyclable recipe ingredient. It'd be nice if a given output could be marked as productivity restricted/immune both in Vanilla and Modding while allowing the rest of the recipe to benefit from this additional boost.

It might be 'okay' for the barrel loop to produce a small amount of extra fluids but making more barrels is nonsense (or may cause issues especially if people design closed barrel loops)

Where do you have the info from, that quality assemblers boost productivity?

[Koub]

Generally speaking, the quality assemblers giving a small boost to the productivity of all recipes is a unique and powerful bonus but....

Unless I'm mistaken, high quality assembling machines get speed boost, not productivity.

[XT-248]

Where do you have the info from, that quality assemblers boost productivity?

There is research that does boost recipe productivity, and I don't think there would be one for barreling/unbarreling fluid.

Example:

From this blog: https://www.factorio.com/blog/post/fff-376

[aka13]

That has aged like milk, seeing as how there is no RCU anymore

[Loewchen]

That has aged like milk, seeing as how there is no RCU anymore

Maybe it now applies to processing units, that would be even more valuable.

[XT-248]

That has aged like milk, seeing as how there is no RCU anymore

As Loewchen said, RCU are likely not the only one to get the productivity from research treatment.

[aka13]

I know what it applies to, but barrels are not in that list
What I initially meant, is that is a very far-fetched assumption that such research would be available for all/most recipies, instead of a few targeted ones.

[TomyTheBest]

Hey All,

Do you know guys how the recycler works? You will get back the 25%, ok. Is it rounded up or down?
If you recycle a higher tier quality item, what would be the outcome?
For example what you get from a recycled Epic Solar panel? (what would be the quality of the outcome? based on the quality of the item you recycle?) And how it changes when you put quality module into the recycler?

I might miss some interesting detail about recycling, was it a topic already?

Thank you

[EustaceCS]

> what would be the quality of the outcome?

Epic by default. Haste makes waste...

> And how it changes when you put quality module into the recycler?

Increases with some %, based on Quality modules inserted. It and above is basically showcased in Quality FFF's Recycler video.

> You will get back the 25%, ok. Is it rounded up or down?

It's chance to get specific batch of resources rolled per batch of resources, so there are no need in rounding (akin to Uranium refinement % roll), no?
Per-item roll for updated quality is needed for output anyway.
So it's probably 25% per individual item to be salvaged, likely.

[bppleman]

I tested the production difficulty of Assembler T3-Legendary in Sandbox mode.

Pasted Graphic 1.png (4.25 MiB) Viewed 1055 times

As you can see, I have produced 2991 assemblers of T3-basic quality, but I have only got 1 legendary one so far. So if I want to plan a 10,000-bottle base, then I may first need to plan a larger base to produce legendary assemblers.

I personally think that the design of "quality" should be a failure. When we design to achieve certain goals, we should not ignore the time spent on producing assemblers, belts, and robotic arms.

If the purpose of legendary quality is to expand vertically to achieve higher production capacity (produce more products per unit time), and I only need to spend less time using basic quality assemblers, and produce the same number of products at a lower rate, the only difference is that I use more basic assemblers. However, it is foreseeable that the time and resources required to mass-produce legendary assemblers are very huge, and the benefits brought by legends do not make up for the consumption of producing legends themselves.

The impact of "quality" seems to be just to make me design the "RNG grind" shown in the picture, and then consume my ore crazily, forcing me to mine farther away, render more map grids, and drag down my UPS at a time?

Everyone says that "quality" is optional, but if "quality" becomes a mainstream (I don't think so, but it's possible), would you really be willing to turn it off? After all, "quality mod" is part of space age, which is not a free project.

[meganothing]

I tested the production difficulty of Assembler T3-Legendary in Sandbox mode.

Pasted Graphic 1.png

As you can see, I have produced 2991 assemblers of T3-basic quality, but I have only got 1 legendary one so far. So if I want to plan a 10,000-bottle base, then I may first need to plan a larger base to produce legendary assemblers.

I personally think that the design of "quality" should be a failure. When we design to achieve certain goals, we should not ignore the time spent on producing assemblers, belts, and robotic arms.

If the purpose of legendary quality is to expand vertically to achieve higher production capacity (produce more products per unit time), and I only need to spend less time using basic quality assemblers, and produce the same number of products at a lower rate, the only difference is that I use more basic assemblers. However, it is foreseeable that the time and resources required to mass-produce legendary assemblers are very huge, and the benefits brought by legends do not make up for the consumption of producing legends themselves.

The impact of "quality" seems to be just to make me design the "RNG grind" shown in the picture, and then consume my ore crazily, forcing me to mine farther away, render more map grids, and drag down my UPS at a time?

Everyone says that "quality" is optional, but if "quality" becomes a mainstream (I don't think so, but it's possible), would you really be willing to turn it off? After all, "quality mod" is part of space age, which is not a free project.

What does "mainstream" mean? That you need to use it because others use it?

[all12jus]

Is there a way to make it so we can do one of those upgrade chains but have a constant combinator set the base item recipe, and then all the assemblers get their retrospective recipes based on the quality some how.

I'm working on a mod that allows you to retry the quality to improve it. https://github.com/all12jus/SimpleUpgrade

The other question I have with this, what research is that "recycler" locked behind?

[EustaceCS]

The other question I have with this, what research is that "recycler" locked behind?

Fulgora > mine scrap from some ruins

. No actual research required.

[orzelek]

You missed one bit in your test - you are using basic T3 quality modules.
One of first things you need to do is to make legendary (or highest tier you can) Q3 module. That will reduce the amount of needed cycles significantly (it will still be a lot without modding and I think thats by design).
There are also other ways to make high quality assembler that might be cheaper in practice.

[Makka77]

I think the FIRST thing to produce with quality is Quality Modules, because all else will flow from this...

Edit:

... closely followed by labs, if they have quality. I'm off to check now

[Acacel]

Well I think if you are planing to get verything to legendary Q just don´t start with the final produkt start with miners to improve Quality in every singel step. so if you have green ore you more likely get blue plates and so on, should reduce the needed input, but will eat a lot of space.
Also after a few hours I´m sure only a few buildings realy need the high Quality

[Ranakastrasz]

Drills are probably the most useful overall, but really, it follows the same logic as productivity modules, in that you want to module whatever uses the most raw materials per time unit. Also, that if you increase the quality of some type of materials, you have to stockpile the high grade components, until such a time as you can spend them perfectly matched with all other components for something you need. Well, unless you are voiding anyway. Which vanilla now has several automatic methods.

[Wockes]

Well I think if you are planing to get verything to legendary Q just don´t start with the final produkt start with miners to improve Quality in every singel step. so if you have green ore you more likely get blue plates and so on, should reduce the needed input, but will eat a lot of space.
Also after a few hours I´m sure only a few buildings realy need the high Quality

I went a step further and just cloned my whole base for every quality. The green base makes stuff only from greens, blue only from blue and epic only from epics(I haven't unlocked legendary yet because I'm slow). Then I use filters, transport belts and bots to transfer the higher grades to the other bases with recyclers only used on end products at the moment.

[Junorus]

I made 2 tier base.
Quality starts from mining and intermediate, but I filter all products for quality (except like copper ore, as it is only for plates). Quality items goes to storage (in future uncommon items will have their bus), normal quality rolls on to regular science production base. There is many items that do not benefit from quality (rails, belts etc gets just health), they are made without quality.
Then the quality mall is a programmable assembler. There is no point to have so many assemblers idling 99% of time.

( Science assemblers works on regular quality ingredients, but have quality modules. I will calculate in future is that is better than productivity)