Friday Facts #375 - Quality

[AirForce1]

I like the idea in principle, but the randomness of the quality seems contradictory to the nature of engineering. IRL, Nobody would tolerate building a car, which randomly attains the desired quality, and subsequently recycling multiple runs of it until it comes out perfect by chance.

Hilariously, that is/was the method of American car production.

I recall reading an article about American engineers going to a Japanese vehicle production plant for a tour and asking where their bone yard was. The Japanese were confused and everyone thought there was a translation issue.

It turned out that the American engineers would just toss numerous bad parts/cars off to the side. The Japanese didn't have such a situation. They said "when we produce something bad, we examine it to find out why it was bad and make corrections so that it doesn't happen again".

Yeah agree with that. That lead us again to the thing that factories should have quality control switch that set desired output quality. This feature would be activated only while you have quality module inside and higher the quality module the faster production it would be and better desired quality the more resources and time it would cost to produce. But as a new feature it could produce bad products by low chance (lets say about few percentage). So you would have to dealt with this. But i would do it differently. The bad pruducts would be marked as bad and you couldnt use this products right away for another production. You could try to fix the product but in lower success rate (but in desired quality) and chance to lose whole thing or you could recycle it but you would lose some input materials.

THAT is what I call a reasonable choice not just use quality and have advatage or dont and starve.

[adam_bise]

I don't get why the difference between assembling machine tiers and quality is confusing some people.

Assembling machine 1 and 2 are entirely different products by design. Like a 50W motor would have thinner windings and smaller magnets than a 100W motor. They are built from different parts.

A 50W motor may be rudimentary tested to see if it even works at all, with no quality control engineer involved. Whereas, a tier 1 quality control engineer might point out that the motors coming off the line are falling below 90% efficiency during QC checks and calibrate the manufacturing equipment. These calibrations may push the efficiency of the motors over the desired 100% baseline. These 50W motors would be slightly better than the ones without QC.

I do love this idea and am excited to see what it means for various blueprint designs!

I agree with the others that the naming convention doesn't fit, however.

People have been complaining about UPS and megabase scaling for a long time. Seems to me this change will help to address that issue.

[FuryoftheStars]

I like the idea in principle, but the randomness of the quality seems contradictory to the nature of engineering. IRL, Nobody would tolerate building a car, which randomly attains the desired quality, and subsequently recycling multiple runs of it until it comes out perfect by chance.

Hilariously, that is/was the method of American car production.

I recall reading an article about American engineers going to a Japanese vehicle production plant for a tour and asking where their bone yard was. The Japanese were confused and everyone thought there was a translation issue.

It turned out that the American engineers would just toss numerous bad parts/cars off to the side. The Japanese didn't have such a situation. They said "when we produce something bad, we examine it to find out why it was bad and make corrections so that it doesn't happen again".

Hence why American car companies are/were notorious for crap quality vehicles.

‐-------------------

I don't get why the difference between assembling machine tiers and quality is confusing some people.

I don't get why you think people are mixing the two up?

[mmmPI]

I don't get why the difference between assembling machine tiers and quality is confusing some people.

I don't get why you think people are mixing the two up?

Let say it's not straightforward to know which is the fastest and which is the most expensive between an assembly machine tier 2, epic, and a assembly machine tier 3 rare.

And it could feel worse for the inserters.

[lordxi]

Definitely not interested in the ridiculous randomness of the 'quality' mechanic. There is no shitty meat in my factory to pass a questionable part, only the purity of the machine.

I see why the devs lead with bot optimization.

[adam_bise]

Let's look at what we know.

1 assembly machine 2 with 4 legendary prod3

0.75 * ((10 * 1.5 + 10) * 4) = 75

1 assembly machine 3 with 4 epic prod3

1.2 * ((10 * 0.9 + 10) * 4) = 91.2

1 assembly machine 3 with 4 rare prod3

1.2 * ((10 * 0.3 + 10) * 4) = 62.4

So, yes. There is some overlap.

Edit: Nevermind. There is no having 4 modules in an assembly machine 2.

[lordxi]

To the people saying that it's not how real factories work... I wonder if you've ever worked on/in a real factory? I mean, neither have I,

That's not how factories work. Parts come off the line pass/fail, that's it. Pass? Goes in a box and gets sent to assembly. Fail? Goes in a box and gets recycled.

[KuuLightwing]

I don't get why the difference between assembling machine tiers and quality is confusing some people.

It's not confusing to have both Assembler 2 and Assembler 3 and quality on top, it just adds unnecessary granulartiy because quality is a tier system sitting on top of existing tier systems that serves largely the same purpose. We already have tiers, now we have other tiers on top of existing tiers, and also modules, which also have two different tier system too. IMO this is just a rather clunky design, especially since quality is just a small icon on top of an existing model.

[FuryoftheStars]

I don't get why the difference between assembling machine tiers and quality is confusing some people.

I don't get why you think people are mixing the two up?

Let say it's not straightforward to know which is the fastest and which is the most expensive between an assembly machine tier 2, epic, and a assembly machine tier 3 rare.

And it could feel worse for the inserters.

Let's look at what we know.

1 assembly machine 2 with 4 legendary prod3

0.75 * ((10 * 1.5 + 10) * 4) = 75

1 assembly machine 3 with 4 epic prod3

1.2 * ((10 * 0.9 + 10) * 4) = 91.2

1 assembly machine 3 with 4 rare prod3

1.2 * ((10 * 0.3 + 10) * 4) = 62.4

So, yes. There is some overlap.

Edit: Nevermind. There is no having 4 modules in an assembly machine 2.

That's not really what I asked, though. I asked why the thought that some people are confused by the differences of the two systems.

[adam_bise]

I have been looking forward to more complexity, especially this kind. It means the best possible blueprints will be tricky to discover, while at the same time being optional. I don't want to spend the whole game worrying about which combination is faster. But when I get to the point where I am making my final final final smelter blueprint. I DO want much more of a challenge than before. So I say to Wube: THANKS!

[adam_bise]

That's not really what I asked, though. I asked why the thought that some people are confused by the differences of the two systems.

Just the sense that I got from skimming through the posts. I don't want to go dig up the specific posts. Perhaps I misread them anyway. I'm just excited about the added complexity.

[KuuLightwing]

I have been looking forward to more complexity, especially this kind. It means the best possible blueprints will be tricky to discover, while at the same time being optional. I don't want to spend the whole game worrying about which combination is faster. But when I get to the point where I am making my final final final smelter blueprint. I DO want much more of a challenge than before. So I say to Wube: THANKS!

Well for final final blueprint it's probably maximum quality with maximum quality productivity modules and beacons with maximum quality speed modules and max quality inserters. I don't know if there's much challenge there.

[mmmPI]

Let's look at what we know.

1 assembly machine 2 with 4 legendary prod3

0.75 * ((10 * 1.5 + 10) * 4) = 75

1 assembly machine 3 with 4 epic prod3

1.2 * ((10 * 0.9 + 10) * 4) = 91.2

1 assembly machine 3 with 4 rare prod3

1.2 * ((10 * 0.3 + 10) * 4) = 62.4

So, yes. There is some overlap.

Edit: Nevermind. There is no having 4 modules in an assembly machine 2.

That's not really what I asked, though. I asked why the thought that some people are confused by the differences of the two systems.

I don't know for other people, and i'm not sure what you asked, i said something that confused me, and i'm still confused.

Do you think the legendary fast inserter will be faster at moving item than the uncommon stack inserter ? and the energy cost ? x)

And i have another one if you can solve this one, let compare with the classic 24 stone furnace making iron plate and using yellow belts, and then i transition to steel and red belts, that's easy and straightforward, then i redo the furnaces when i have electric, and i put prod3 module in them surrounded by beacons with speed3 in them. That's the pre-expansion me.

And then i unlock the "QUALITY",

What would you change first ? beacons quality ? the furnaces quality ? the prod modules quality ? the speed module quality ?
And would you go directly to quality 5 if you can avoid using intermediate quality level ?

It's paralizing, maybe because it's too early to know the cost and bonus of all things, but it makes me negatively anticipate a situation where instead of organically improving the furnace lane, with a quality upgrade on a different compornent here and there that are progressive, i may resort to just using a quality 1 blueprint, and then a quality 5 blueprint, as it feel simpler, but also ruining part of the system as it may be intended.

On the same trend i don't think i will ever be making blueprints with anything else than quality 1 blue/green/filters inserter or quality 5 green/filters inserter because i don't think i will ever bother sorting and even less remembering which one is the fastest/most energy efficient/cheapest, between the so many that will exist.

I feel a little like an idiot saying this, but having a quality 5 alternative feels enough to me, no need for 5 level of qualities i'm not that obsessed lol. Maybe i could do with quality 1 quality 3 and quality 5 if there is a need for an intermediate step to make the transition smoother for other things than inserter, like beacons, or assembly machines, mining drills but in that case it will most likely be always the highest tier that i will try to levelup in quality. (

When not enough choice => brain wants more, when too many choice => brains want less, brain never happy when thinking too much, sorry, i think i need to play the expansion already, so that it becomes less of a speculation. There must be more things related to the different planet, that would complete the puzzle, such as "when" or "where" you unlock the different qualities maybe would give incentive to use intermediate quality levels.

[FuryoftheStars]

Let's look at what we know.

1 assembly machine 2 with 4 legendary prod3

0.75 * ((10 * 1.5 + 10) * 4) = 75

1 assembly machine 3 with 4 epic prod3

1.2 * ((10 * 0.9 + 10) * 4) = 91.2

1 assembly machine 3 with 4 rare prod3

1.2 * ((10 * 0.3 + 10) * 4) = 62.4

So, yes. There is some overlap.

Edit: Nevermind. There is no having 4 modules in an assembly machine 2.

That's not really what I asked, though. I asked why the thought that some people are confused by the differences of the two systems.

I don't know for other people, and i'm not sure what you asked, i said something that confused me, and i'm still confused.

Do you think the legendary fast inserter will be faster at moving item than the uncommon stack inserter ? and the energy cost ? x)

No, my question is not about the systems or the differences themselves, but rather why adam_bise felt as though some people didn't understand them.

The original piece from adam_bise's post that I first quoted ("I don't get why the difference between assembling machine tiers and quality is confusing some people") gave me the impression that adam_bise read/saw something somewhere to make them think that other people were not understanding the differences between tiered machines and the Quality system proposed. I was basically wanting to know what/where adam_bise read/saw to make them feel that way.

----------------------------------------------

That's not really what I asked, though. I asked why the thought that some people are confused by the differences of the two systems.

Just the sense that I got from skimming through the posts. I don't want to go dig up the specific posts. Perhaps I misread them anyway. I'm just excited about the added complexity.

Hmm, ok. I know there are a lot of pages, so a lot of information to process (and retain!), and I know I skimmed through a lot of the earlier posts because they were coming in too hot and furious to realistically keep up, so it's possible I've missed/forgotten something, but I know from these last 10+ pages the basic feeling I've been reading (of those not liking the system) have seemed centered around the randomness of the new system or, as this person put it:

It's not confusing to have both Assembler 2 and Assembler 3 and quality on top, it just adds unnecessary granulartiy because quality is a tier system sitting on top of existing tier systems that serves largely the same purpose. We already have tiers, now we have other tiers on top of existing tiers, and also modules, which also have two different tier system too. IMO this is just a rather clunky design, especially since quality is just a small icon on top of an existing model.

[mmmPI]

The original piece from adam_bise's post that I first quoted ("I don't get why the difference between assembling machine tiers and quality is confusing some people") gave me the impression that adam_bise read/saw something somewhere to make them think that other people were not understanding the differences between tiered machines and the Quality system proposed. I was basically wanting to know what/where adam_bise read/saw to make them feel that way.

I felt like i could have given such impression to adam_bise with some of my previous posts, because , although i understand the difference between the 2 different system, i have mentionned the fact that's it's potentially confusing, such as for the inserter knowing which one is "best" becomes not so straightforwards, when either the tier and the quality system can improve one item and as such, the low tier high quality and hight tier low quality may overlap, or make it difficult to understand the cost, and impact of an upgrade on a lane of furnaces without a few math.

[FuryoftheStars]

The original piece from adam_bise's post that I first quoted ("I don't get why the difference between assembling machine tiers and quality is confusing some people") gave me the impression that adam_bise read/saw something somewhere to make them think that other people were not understanding the differences between tiered machines and the Quality system proposed. I was basically wanting to know what/where adam_bise read/saw to make them feel that way.

I felt like i could have given such impression to adam_bise with some of my previous posts, because , although i understand the difference between the 2 different system, i have mentionned the fact that's it's potentially confusing, such as for the inserter knowing which one is "best" becomes not so straightforwards, when either the tier and the quality system can improve one item and as such, the low tier high quality and hight tier low quality may overlap, or make it difficult to understand the cost, and impact of an upgrade on a lane of furnaces without a few math.

Ah, now that you mention it like that, yes, I do vaguely recall seeing some posts like that.

Ok, I guess my interpretation of those posts was less that they were confused on how the two systems actually worked and interacted, but more about a lack of clarity on which could be better without having to break out the calculator/spreadsheets compared to before.

[blazespinnaker]

Super awesome!

So much more you could have done with this but probably good that you didn't. This will be an endless source of complexity as is.

However, for SE a very cool addition might be - machines which temporarily and significantly degrade in quality and only a combinator setup will summon a robot to repair! But how to detect degradation...

To save on combinators/wires, perhaps you could have the robots visit all the assemblers in an area.

Or some kind of timer setup, where you get the robots to visit every 1000 produced products.

I like the idea in principle, but the randomness of the quality seems contradictory to the nature of engineering. IRL, Nobody would tolerate building a car, which randomly attains the desired quality, and subsequently recycling multiple runs of it until it comes out perfect by chance.

That's exactly how it works IRL, though the probabilities are a bit different. IRL, you build things to a specification, and some (ideally small) percentage will be crap, though that probability will vary over time due to various factors. As to how much recycling goes on, I suppose it depends on the product.

[blazespinnaker]

Reviewing some stuff from all those pagess:

Overall I'm really looking forward to it, but there's a couple things that need improving. The first is the quality names. What is this, an RPG? They should be given industrial names, like "basic", "refined", etc. The second one is that quality for inserters improves swing speed. It should be improving stack size so that the player doesn't pay a larger UPS cost for higher quality inserters for the same items/sec throughput.

+1! The names are totally silly. 'Epic'? Really?

Here's some ideas: non conforming, defective, conforming, meets requirements, exceeds requirements, best in class, state of the art, ...

This is interesting !
- I wonder if the quality info (the colored dots) visibility on entities will be togglable within the alt view, or if it's linked to alt view. The big green circuits setup is a bit hard to read now with so many information on screen.

I assumed that obviously the quality dots wouldn't show on the ALT layer, but on a new layer. I mean, wow, the visual clutter would be totally irritating for sure and that alone might make me think twice about using it.

I am really warming up to the idea... I have long hoped to see more "recycling loops" in factorio, ala science cards in SE!

Exactly, byproducts are a big deal in all the most intriguing factory designs. This bakes byproducts into everything. I admit, I'm a bit stunned that they added this. I thought it was just going to be simple stuff only. In some ways, this is more advanced addition than some complicated mods like SE. Raising the bar..

[adam_bise]

At any rate, here is how I envision this would likely play out for the average player.

At some point during normal gameplay the player unlocks quality and then is faced with the question of what to try to improve.

I think a mod pointed out that it would be cheaper to do intermediate items rather than costly late items. Since the quality properties will have a trickle effect.

Say you start out by putting quality modules in lead plates. Improved items will start showing up almost all across the board, at a snails pace. Then what?

I imagine people will wind up starting at raw and intermediate items, ignoring the rare improved costly items in their logistic design. They would get used all the same because those belts, splitters, filters and so on are still set to any quality. They might start actually doing something different with improved items at the lower end of the material chain. I imagine there will be many upgrade planners placed with >= uncommon on lower end factories, followed by lots of waiting for these to be fulfilled, followed by more and more improvements on how to sort and handle above normal items specifically.

I imagine the decision to use a filter other than >= uncommon would not come until much later when more and more improved items are being made. Perhaps at some point you would decide to alter your handling of improved items to >= rare, when you start seeing enough to actually work with.

There will be points in the game where trying to get x quality items will simply take an unfeasible amount of time and raw resources. These things will be learned by playing through the game. I imagine many of the people who are not liking the idea now will change their mind once they have a little experience actually playing with them.

[blazespinnaker]

If I had to guess, the factorio folks will strongly encourage folks to complete the base game first before doing the dlc. So things like quality probably won't be too much of a shock.