Quality levels completely break factory

[NineNine]

In the original game, major systems were changed over time. The way science was made changed - no more collection of alien stuff when killing bases and different recipes. Things were tweaked. Factories broke and needed to be redesigned. And this made the game better.

But the game wasn't live. The era you're referring to was just beta testing. Major design reworking is fine when you're still putting the thing together. But once you go live, once you have a proper release, you can't just go break everyone's base on a whim.

Nobody's base was broken unless they updated Factorio to a new version while they were playing an existing game (which is a bad idea for the reason you stated). If you choose to break your game, that's on you. Factorio doesn't force anybody to change versions.

[Alfonse215]

In the original game, major systems were changed over time. The way science was made changed - no more collection of alien stuff when killing bases and different recipes. Things were tweaked. Factories broke and needed to be redesigned. And this made the game better.

But the game wasn't live. The era you're referring to was just beta testing. Major design reworking is fine when you're still putting the thing together. But once you go live, once you have a proper release, you can't just go break everyone's base on a whim.

Nobody's base was broken unless they updated Factorio to a new version while they were playing an existing game (which is a bad idea for the reason you stated). If you choose to break your game, that's on you. Factorio doesn't force anybody to change versions.

You're basically saying that you should never expect a Factorio update to preserve your save. That your current run is entirely ephemeral and it's going to constantly be destroyed and remade. And the only way to avoid that is to stick with a particular version, eschewing any bug fixes or features added to the new ones. And you must do so in perpetuity so long as you want to keep that base around.

This is the fundamental difference between "early access" and "release". Early access builds are expected to be constantly changing. The developers and playerbases both understand this and players expect it. With a release build, there is an expectation of stability, that periodic updates are not going to just break your base.

I can't think of other released games where minor updates are expected to overhaul large parts of the game to the point where you have to restart or retool vast parts of your base. MMO expansions may encourage you to make new characters, but you can just use your old one. Even Terraria expansions, which often add content that requires new world generation, still keeps your old worlds functional. You may not be able to access new content, but the old world is still there.

Even Space Age did not overhaul the early vanilla game or its intermediates and production structures. It made very targeted changes, but they didn't change anything that they didn't really need to.

[NineNine]

In the original game, major systems were changed over time. The way science was made changed - no more collection of alien stuff when killing bases and different recipes. Things were tweaked. Factories broke and needed to be redesigned. And this made the game better.

But the game wasn't live. The era you're referring to was just beta testing. Major design reworking is fine when you're still putting the thing together. But once you go live, once you have a proper release, you can't just go break everyone's base on a whim.

Nobody's base was broken unless they updated Factorio to a new version while they were playing an existing game (which is a bad idea for the reason you stated). If you choose to break your game, that's on you. Factorio doesn't force anybody to change versions.

You're basically saying that you should never expect a Factorio update to preserve your save. That your current run is entirely ephemeral and it's going to constantly be destroyed and remade. And the only way to avoid that is to stick with a particular version, eschewing any bug fixes or features added to the new ones. And you must do so in perpetuity so long as you want to keep that base around.

Yes, so what?

What are you suggesting as an alternative? To never update the game, ever? Or, do force everybody to update with each update, and the developers can write code to update everybody's factories? I'm trying to understand what, exactly the problem is. Factorio is not an MMO.

[quineotio]

2.1 breaks some factories.

[coffee-factorio]

You might be surprised if you go to look at the original blog post. Kovarex explicitly laid you how he thought the strategy would play.

I am.

In practice, a small number of quality items is almost completely useless - the exception being personal equipment (and maybe space platforms). For example, you have 20 assemblers for green circuits and 2 of them are uncommon? It doesn't matter - it's easier to just build more normal assemblers. You design around what you have, so either it's normal quality or some other. So in practice, if you're going to engage with quality in a meaningful way, you need a good level of reliable production - which you CAN'T have until after you leave Nauvis and get the recycler.

If you want to upgrade your factory to quality using the recycle loop method only, that's really, really boring and wasteful. And when I say wasteful I don't mean of resources, because who cares (resources are unlimited). It's wasteful of your time and effort. And even if you do it this way, you're eventually going to get all you need, and then you've built a bunch of infrastructure that doesn't do anything any more, which goes back to my point about quality in the lategame.

And then there's the other trap for quality, which is that your builds will break if you didn't account for additional tiers after Gleba/Aquilo. I can't help but feel like it would be better if all tiers were available at the same time, and there were ways to limit which outputs you got. Having to account for 5 different outputs every single time you put a quality module anywhere is also annoying, especially after recyclers where you're getting 5x the number of ingredients outputs.

Quality items themselves are fun, but so many things make the system unwieldy. It's just so strange that other areas of the game have had so much polish and quality of life and bug fixing, but so little has been done to address probably the most complained about thing in the game.

Part of me, tries to imagine what the game designer was thinking. For some reason, when people do that, it's a weird daemon. Because it is a daemon, it knows all.

Is not a man, with finite knowledge and energy.

One of things I figured out in the map editor is, you have your own story of what the gambling mechanics are. But my story is "optimization" and not "gambling mechanics". A lot of the things that I think are appropriate criticism of Wube's approach is that they are not being clear that the quality system is supposed to have the traits that I associate with an optimization process.

Ambiguity, complexity and context sensitive problems are all parts of optimization. When you look at it from that perspective, they meet the criteria for the system. There's a requirement I feel that is enforced where the quality levels are to make a safe sandbox rather than one where things truly break, which makes things feel too similar till you start upcycling on multiple planets.

I can have my opinion that it is a dishonest engine meant to make you deal with context sensitive problems that yield context sensitive results. I'm not sure I can defend the developer's approach or should.

And I have to add, the problems with quality (and some other things) arise in part from the lack of iteration upon game design choices in Space Age. It's one thing to design a system that works in theory, but when it comes into contact with the broader playerbase, issues arise.

And this is one of my longstanding complaints about Space Age. The game was released and then there was silence for 1.5 years. There was a dire need for iteration based on player feedback, but this didn't happen. And now, after almost 2 years, the design has become baked in, and now it's difficult to change anything.

Yeah. I interpret this as a standing assumption that players will design something that corresponds to a prompt that makes sense. Make a Red / Blue choice (and blue is evil? I dunno...) . The reality is that you optimize the blue pill, which now has a decorative five dot label. And I still don't know if the blue pill is evil because ambiguity is, part of the design criteria.

[coffee-factorio]

Here are some changes that I think will help smooth the transitions between before and after introducing of the respective mechanics tremendously.

Some ingredients could have been given the "spoilage" treatment on Nauvis to ease new players into Gleba without Quality. Repeat for quality mechanics; after all, OP indicated they had spent considerable time building out factories without quality modules before deciding to do a small experimental run and go all-in on quality modules. That kind of experience ultimately turned the OP away from using quality modules, which is a problem.

Then, in the end, players can take what they learn and combine it into a functional design that incorporates one of the mechanics, or even both. Nothing currently exists that could resemble an introductory experience for each world's unique mechanics, with one small, noticeable exception: Vulcanus (liquid/petroleum handling).

This is an interesting idea, but it lacks specifics (as it relates to quality). And those specifics are kind of important.

For example, I feel that the Tips&Tricks entries on quality are not just poor introductions to the concept, but they are actively harmful to properly building setups to produce quality goods. The biggest offender is the setup showing a bunch of gear makers with quality modules in them that put their outputs onto the same belt, with no filtered splitters in sight.

The first setup players see for making quality intermediates should not brick their base within 5 minutes of trying them using the provided example. So there is a lot to be said for improving player guidance with regard to the quality mechanic.

In that tutorial there isn't a demonstration of how the outputs fill up a logistics line. There isn't a demonstration of how the outputs fill up a warehouse. There's a demonstration of some ratio of parts and an assumption players will build on that ratio. There isn't a demonstration of new build concepts like automalls which make it so that you can make on demand production work on 10 buildings versus a traditional mall where it was 1 building per entity.

If I go down the rest of your strategic assumptions in this post. I can give you counter examples but conceptually you're right so why argue? Some of the solutions sound so trivial they come off as cynical. For instance, the steel furnaces problem: just run a gig of nuclear and don't use steel furnaces. Out of context it sounds like I'm arguing against solar panels, in context it's even more hilarious because I'm buying solar panels as part of a command economy.

I run another command economy of landfill and throw the result into a lake so I have a box of parts for a spaceship and oil refineries. It looks like Yolo's rather than an Edge Lord (oh, political extremists got that label too... what do you call someone who just enjoys hard content?) running a series of small command economies that he worked out on his lunch break.

Edit: And just to take it a step further. Those are all potential failure points where suddenly adding byproducts generates operator errors. And it is pathological, to the quality system. That doesn't stop, just because you have a big economy. You just change where you experience it.

[Alfonse215]

2.1 breaks some factories.

It breaks small parts of some factories. Specifically, hazard concrete recycling, asteroid cycler platforms, and certain cases of assembler recipe controls. These are non-trivial changes, which is why they waited until a point release.

But the level of changes we're talking about would break significant parts of basically any factory that engages with quality at all.

[Alfonse215]

In the original game, major systems were changed over time. The way science was made changed - no more collection of alien stuff when killing bases and different recipes. Things were tweaked. Factories broke and needed to be redesigned. And this made the game better.

But the game wasn't live. The era you're referring to was just beta testing. Major design reworking is fine when you're still putting the thing together. But once you go live, once you have a proper release, you can't just go break everyone's base on a whim.

Nobody's base was broken unless they updated Factorio to a new version while they were playing an existing game (which is a bad idea for the reason you stated). If you choose to break your game, that's on you. Factorio doesn't force anybody to change versions.

You're basically saying that you should never expect a Factorio update to preserve your save. That your current run is entirely ephemeral and it's going to constantly be destroyed and remade. And the only way to avoid that is to stick with a particular version, eschewing any bug fixes or features added to the new ones. And you must do so in perpetuity so long as you want to keep that base around.

Yes, so what?

What are you suggesting as an alternative? To never update the game, ever? Or, do force everybody to update with each update, and the developers can write code to update everybody's factories? I'm trying to understand what, exactly the problem is. Factorio is not an MMO.

To do what every other game does: when you release the game, you don't keep developing on it like it was still in a pre-release state. You respect that your audience bought a released game and you don't massively change it out from under them without a version change. And even then, you respect your audience by doing your level best not to change major system unless it is absolutely essential to do so. You make the minimum changes necessary to solve some problem, and only if that problem is substantial enough to warrant it.

Released games are not in early access, and game developers ought to respect the difference between those two things.

If WUBE wanted to do an early access version of Space Age, that would have been an acceptable environment for them to redesign a major system like quality. But they didn't; they released it. That effectively promises that there are levels of changes they aren't willing to make anymore. At least, not without a major version release.

And again, 2.0 didn't make major changes to the core recipes, intermediates, and the like, except where it was needed by Space Age.

[XT-248]

This is an interesting idea, but it lacks specifics (as it relates to quality). And those specifics are kind of important.

For example, I feel that the Tips&Tricks entries on quality are not just poor introductions to the concept, but they are actively harmful to properly building setups to produce quality goods. The biggest offender is the setup showing a bunch of gear makers with quality modules in them that put their outputs onto the same belt, with no filtered splitters in sight.

The first setup players see for making quality intermediates should not brick their base within 5 minutes of trying them using the provided example. So there is a lot to be said for improving player guidance with regard to the quality mechanic.

The problem is how you do it without actually changing the nature of the mechanic.

See, quality is a very open mechanic; almost every recipe participates in it. So you have a lot of choices for how to engage with it. But at the same time, quality engages with one of the most difficult elements of Factorio's logistics: handling byproducts. By putting a quality module into any intermediate's production, you are opting into a process that produces unwanted byproducts.

A player who is new to Factorio, who hasn't even set up advanced oil processing before? That person is never going to get into quality intermediate production early on. And that's the big catch-22 of quality.

The simplest, earliest means of producing quality stuff is just putting modules into end products and hoping for the best. But in order to understand later means of producing quality stuff, you need to have a grasp on a concept that that method never uses. Namely, that quality intermediates can be used to produce equal quality outputs. This is the core of any quality cyclers setup, as well as any method of making quality that relies on intermediates.

But making quality intermediates early-on is a trap. Not unless you know about quality mining, and even then, you'll need to know how to find out if your buffers are filling up and to add more.

First, you can't put quality modules in steel furnaces, so at the moment QM1s become available, you can't use them to make quality plates (without knowing about quality mining). Getting quality gears is a bit easier, but many thing that take gears (or even circuits) also take plates. Which you don't have.

And even if you do get quality intermediates, the ratios almost never line up. You'll always have too much of one thing and not enough of the other. You need some kind of escape valve for having too many intermediates of an undesired quality, some way to prevent items from backing up and blocking something.

Quality mining is a good idea, but the issue is that players need to know about it and try it. And even then, backing up can be a problem unless players are able to buffer with more and more chests. The ratios still don't line up, and there's not a simple answer to solve it. Driving a train through a car loaded with the excess is a good way for a skilled player to automate disposal, but an unskilled player almost certainly won't think of that.

The only thing I can think of is out-right forbidding the player from using quality modules in any intermediate recipe until they get recyclers. Note that mining drills don't execute recipes, so they can still use them.

This would have the effect of limiting the player to exactly 2 modes of interaction with quality: end products or quality mining and crafting up the chain. As these are the two most effective ways to engage with quality early on, it makes sense.

Of course, this has a number of negative side-effects. It nerfs Vulcanus's ability to make quality stuff; not entirely (resources are still cheap and you can still dump things in lava). But 2/3rds of Vulcanus's resources are impervious to quality mining (you can still make quality tungsten carbide and plate from quality coal and quality tungsten ore). Gleba becomes even more hostile to quality without Fulgora.

How to improve, or even add introductory experience to quality while retaining the original "openness" of the quality modules, is a hard question to answer or even come up with an idea for.


I could answer the question of how to introduce spoilage much more easily because wood is one of the most disposable materials we have in Factorio.

[NineNine]

But the game wasn't live. The era you're referring to was just beta testing. Major design reworking is fine when you're still putting the thing together. But once you go live, once you have a proper release, you can't just go break everyone's base on a whim.

Nobody's base was broken unless they updated Factorio to a new version while they were playing an existing game (which is a bad idea for the reason you stated). If you choose to break your game, that's on you. Factorio doesn't force anybody to change versions.

You're basically saying that you should never expect a Factorio update to preserve your save. That your current run is entirely ephemeral and it's going to constantly be destroyed and remade. And the only way to avoid that is to stick with a particular version, eschewing any bug fixes or features added to the new ones. And you must do so in perpetuity so long as you want to keep that base around.

Yes, so what?

What are you suggesting as an alternative? To never update the game, ever? Or, do force everybody to update with each update, and the developers can write code to update everybody's factories? I'm trying to understand what, exactly the problem is. Factorio is not an MMO.

To do what every other game does: when you release the game, you don't keep developing on it like it was still in a pre-release state. You respect that your audience bought a released game and you don't massively change it out from under them without a version change. And even then, you respect your audience by doing your level best not to change major system unless it is absolutely essential to do so. You make the minimum changes necessary to solve some problem, and only if that problem is substantial enough to warrant it.

Released games are not in early access, and game developers ought to respect the difference between those two things.

If WUBE wanted to do an early access version of Space Age, that would have been an acceptable environment for them to redesign a major system like quality. But they didn't; they released it. That effectively promises that there are levels of changes they aren't willing to make anymore. At least, not without a major version release.

And again, 2.0 didn't make major changes to the core recipes, intermediates, and the like, except where it was needed by Space Age.

"Every other game" doesn't let you play any version of the game indefinitely. WUBE lets you do that, so your complaints don't make any sense. Nobody is forcing you to change versions mid-game. That's 100% your decision.