Friday Facts #266 - Cleanup of mechanics

[Ferlonas]

I'm sorry for double posting, but I'm deliberately doing it as this is an entirely different train of thought.

I can see many people saying "This is an example of how software development should be like." And in many cases, I do agree. You guys have done a splendid job so far.
In this specific case, it does not apply. Usually, you do software development for customers that want results from that software. A software for creating purchase orders should give the customer the ability to create purchase orders - the faster the better (and simplification helps with that). A software for managing firewalls should give the admin the ability to create firewalls, even if he isn't an expert in low-level networking. Simplification helps with that.
Even a tool for writing software is focused on the output: producing running code.

In Factorio, the result is what we strive for, but only because it gives us a sense of achievement. The easier the way is, the less of a sense of achievement you have. Going from that, you can say that the result is far less important than the way you get there.
And any change that simplifies that way takes from the players a part of that which made them buy the game in the first place.

I think I now have said all I can, and the only thing I can do is phrase it differently if I am misunderstood. So I will lean back and hope you will think carefully about what your final decision - and your reasoning for it - will be.

Regards
Ferlonas

[DRY411S]

If the motivation for changes is to make entry level simpler, I get it. Just look at the Steam achievement percentages to see the low proportion of people who stick with the game.

But do all the mining changes and the assembler ingredient count nerf meet that motivation? Mining, I don't think so. Energy, no. Both those are nuances you learn about later, and add depth rather than make game entry too hard. Assembler ingredients, yes but there must be a better way.

There is something wrong about Assembler 1. It works half the speed of the character, and cannot craft as much. If you need to make game entry level easier, fix that, please don't dumb the the whole assembler hierarchy down.

[Blacky007]

@ devs why do you give Bob and Angle this big punch in the face?
they made this game to what it is now - what would happen if they stop supporting this game?
all the high skilled streamer stop playing factorio and as a follow the sales will decrease.

[wartthog]

I love what you guys are doing to make the NPE more approachable! I love your courage to challenge every paradigm we've accepted as players in the name of perfecting gameplay! I love how you field ideas in the FFF so that we can all wrestle with them and discuss the mechanics we love and hate.

However I agree with the sentiment in these 7 pages. Feel free to simplify the NPE and vanilla experience, but many of us (my Steam says 1844 hours now) have turned to modders like bobingabout to give us our fix of complexity. Factorio is a class above most games for this reason. Taking away the modders' tools to this level feels like betrayal.

Upvotes:
* Ditch durability (but maybe keep it available to mods?)
*

Next simplification is to nix fluid temperature since it isn't modeled anyway.

* Changing AM1 neuters Lazy Bastard.

Things that haven't been said yet:
* Would it be better to restrict AM1 by limiting the specific recipes it can process? It gives you creators more power but you'd have to find a good way to help the user understand.
* Some people enjoy the early game in its own right. The Stone Age mod (https://github.com/StoneAgeFactorio/StoneAgeFactorio -- I can't find the page on the mod portal at the moment) extends this quite a bit. I'm totally fine with getting rid of axes in vanilla (as long as we keep AntiElite happy!) as long as you add mod support. Idea: Add MORE modding support for using an item. You can thrown grenades, fix things with repair packs, activate things with remotes... would you consider maybe making this more generic i.e. allow mods to make any item throwable or usable on entities they specify?

Final words:
Don't give up bob! We love you! <3

[wartthog]

Oh also, thanks for keeping 1.21 GJ. X'D

[Jan11]

My first impression is hmmmmmmm because the player needs something (goals) to move forward. With removing the "Assembling machine ingredient limit"
I think the game will lost a littlke bit. I am pleased every time if I reached an important point in the game. Evey new version of the assembling machine (because I can craft so much better stuff)^^ ; Laser turrets and much more.

But I think it`s good at the end of the day because I know that you created a really awesome game until now.^

[quyxkh]

I welcome most of these changes. Most people who dislike them do it because they have to change. would they have never known the "before", they would probably have loved the "after" even better when discovering it.

So, this is about gratifying "most people", with the idea that if they don't ever have to encounter grade-school math they'll flock to the game, then. Instead of a game that's easy and fun until you try to get 100% performance out of anything at scale, then it gets intricate and hard, depths and crevasses and stumbling blocks everywhere, tradeoff after tradeoff and fun in a whole new way, let's make it all nothing but easy fun and easy fun! Push a button, get a reward, repeat!

[QGamer]

TL;DR: The assembling machine ingredient limits are good because they help give a sense of game progression.

Now that I think about it, the assembling machine ingredient limits are very useful for game progression. Please don't remove them. (In this argument I will abbreviate "assembling machine" with AM.)

Factorio is a game about automation. You could say a viable end-goal is to automate every single recipe in the game. But this is an end-game goal, and should not be possible until the end of the game (or close to it).

From a story perspective: When the character creates his first AM1, it has limited functionality. He doesn't know what he'll be making in the next few months/years on this new planet. So he creates the best AM he can, to his knowledge. It has limited functionality, and the character realizes this, so he decides that later, he'll make a better one that will suit his needs. But for now he'll do what he can with what he has.

Later comes, and his skills are better. He now knows what types of things he'll need to automate text, and so he creates a very good AM2 that can handle all but the most complex recipes. Because of this technological advancement, he can now automate the production of a lot of important machinery for his factory (including, crucially, AMs). Even later than that, he uses Speed Module technology to create an AM3 that is better than he is at crafting. He can now automate every recipe in the game (once he unlocks them), and his technology has surpassed his manual abilities.

My point is this: The ability to automate something should be a reward in and of itself. At first, the technology is limited, and so the AM1 has limited functionality. The player should be able to automate basic components (gears & circuits), essential ingredients (plates, science), and defensive materials that are needed in large quantities (ammo, walls). The ability to automate automation (making AM1/2/3) should come at a later time, and it should be a reward for beating the first era of the game (the burner/manual era, which is over by the time the AM2 is unlocked). It should mark a milestone in a player's Factorio experience, especially for new players.

The idea behind this mechanic was that better assembling machines can use more complex recipes. But the reality is, that there is not really a clear connection between the number of ingredients and the complexity of the recipe. Since it was yet another thing that had to be explained somehow, we decided to just remove it.

As for ingredient limits being "yet another thing that had to be explained somehow," explain that the simpler machines are more limited in functionality, and can only craft simpler recipes. The progression of the game from AM1 to AM2 is quite meaningful because it opens up a lot of new options for automation. The progression from AM2 to AM3 is more subtle: a few recipes, but 66.7% more speed and twice the number of module slots. I'd argue to leave the ingredient limits how they are. If you wish to change something, at least opt to keep the game progression that comes from unlocking a new tier of AM.

[meganothing]

The amazing (and to me, astonishing ) popularity of Bob's and Angel's

Only 5% of players use Bob's mods, and even less use Angel's.

5% of active players or 5% of players owning Factorio? Without that info your number is as useless as this one: 4563.

[DRY411S]

@ devs why do you give Bob and Angle this big punch in the face?
they made this game to what it is now - what would happen if they stop supporting this game?
all the high skilled streamer stop playing factorio and as a follow the sales will decrease.

Whilst I don't like many of the changes, that claim about Bobs and Angels just is your opinion. Back it up with facts. Here's some.

On the portal, bobs library has around 1/2 million downloads. Almost 18 months ago Devs reported 1 million sales https://www.factorio.com/blog/post/fff-192 . I'll leave it to you to estimate how many more sales since then, and what percentage of the mod downloads are actually used. I for example have them downloaded, but use them only for compatibility testing with my mods. So while active bobs players might be significant, they did not make this game, they modded a game in early access, with all the risks that entails about future changes to the core game.

Besides isn't one of the reasons that modders produce mods (I am one of those people) because they like modding?

Now they'll have a whole lot more to do.

[eradicator]

@Dev-Team: Can anyone please clarify on moddabilty of damage types?
If the ability to add new types is completely removed that is a major breaking change with far too many drawbacks. Currently the API doesn't even tell me what kind of weapon was used to cause an on_damaged event, so i *have* to hack around it by using named damage types (because without active scripting the damage system is too simplistic). And if there's not at least the 10 types of energy (kinetic, thermal, radiation, etc) as damage types even semi complex weapon mods would become impossible.
______
@ingredient limit: While i don't like the "count" mechanism as such i do agree that the current gating model should be kept by remodelling crafting categories. (For exaple by saying "assembler 1 can't make recipes unlocked by blue science" etc.)
______
@bob: Oh dear, hadn't thought of the fuel values at all. That seems like a lot of recalculation work if you can't automatically derive the new values based of current boiler efficiency ratios. Though the FFF mentions moddability so simply reverting it sounds like a quick fix.
______
Overall the discussion feels to me like we're arguing if it's nice to have some edges and corners (personally: oh yes!) or if we should polish away all the dangerous bits to make it idiot safe (hopefully the dev team doesn't aim for such an extreme path).

rough_vs_polished_amethyst_small.png (347.46 KiB) Viewed 6371 times

[TheVeteraNoob]

Getting my unread two cents in here. I feel like if the goal is to simplify a little than assembler 1 should have 3 ingredients. Now you can suddenly automate assembler 1s, inserters and splitters. Without being able to automate slightly less necessary in early game assembler 2s, steam engines, and fast inserters.

[blueblue]

I like the direction of these changes. Obviously factorio is a complex game but these changes mostly dont take away from the main complexity and instead allow players to put more focus on those and just remove distractions. I do not watch people like AntiElitz play because they are so good at calculating mining drill production rates.

That said, making numbers like mining drill rate more accessible makes it easier to do the calculation on every assembly line and build everything according to ratios. I do not think that this is the developer intent but players might feel forced into it. As I said, I like the changes but I hope the new player experience makes it clear to the player somehow that this is not the intended gameplay.

With the changes to the assembling machines, the first two tiers are now almost indistinguishable. Getting a better assembling machine could be a satisfying moment of the new player experience but with how similar they are, that moment is diminished. I dont really have a good idea for how to distinguish them better. One way would be to add arbitrary restrictions on which items the assembler mk1 can craft. For example allow the mk1 to only craft items that are available via red science. Similarly an mk3 assembler could be required for everything that is unlocked via blue science, for example processing units and rocket control units. This appears to be what the ingredient restrictions originally intended but failed to accomplish. Another idea is to add an item that is hard to obtain in the early game to the recipe of the mk2 assembler in order to increase the crafting cost, similar to how the assembler mk3 requires speed modules. But there is no good candidate for this, mostly because engine units are later in the tech tree.

With the pick axe gone there are now only two items that require iron sticks, so it seems like an unnecessary item, especially with how similar to copper wire its recipe is. It could be removed or added into more recipes (for example engine units, red belt).

As another leftover mechanic, there is the somewhat arbitrary restriction that engine units cannot be handcrafted. Again, this should be either removed or fleshed out, for example I keep being surprised that red circuits can be hand crafted and if I recall correctly it is also possible to hand craft rocket control units and rocket fuel.

[meganothing]

Generally I think the existence of a mod using some functionality should not be a reason for the developers to constrict themselves in EA. Modders will find other ways if the modding interface is powerful enough.

1) So damage type removal is fine if it doesn't add anything to vanilla gameplay, but there should be a way to add damage types in the mod interface. More power to the mod interface I say.

2) For mining hardness Kovarex already outlined the alternative modders have.

3) Pickaxe removal is really more of an UI change. Even if some mods used the functionality for circumventing other limitations it would be much better removing the limitations than retaining this useless UI element.

4) Assembler limit though is tricky, it simplifies gameplay itself. There are good arguments on both sides here. As long as AM1 has no pipe contact I think I'm ok with it. But generally a game about complex factories should not not try to streamline complexity, it is the wrong direction.

[bman212121]

But the reality is, that there is not really a clear connection between the number of ingredients and the complexity of the recipe

As far as Factorio goes the increased number of ingredients DOES make it a more complex recipe. We have to feed all of that complexity into the machine somehow, and it's a whole lot harder to put 6 items into an assembler than it is to put 2 items into an assembler. Just because the intermediate items might have been more difficult to assemble, that doesn't mean the next step is going to be hard. You can argue that building an atomic bomb is really complex in life, but in Factorio is quite simple. But that just means that if you were trying to portray that as something difficult, you need to increase the number of items it takes.

That said, the entire process might be complex, but certain steps are not. So if you were building a computer, making a CPU to put inside is crazy complex to make that component, but the actual step of putting a cpu into a motherboard is not. But if one of the steps of the process was take all 7 parts and put them together at once, all of the sudden that one step goes from being simple to being very complex. That said the "difficulty" of taking 7 cpu parts and putting them together is far less difficult than takes 3 parts and making a CPU wafer, but difficulty isn't really represented in Factorio. Complexity doesn't equal difficulty of the step, but complexity does in fact go up directly proportional to the number of items involved.

[DRY411S]

If you must delimit ingredient counts, how about

Assembler 1 makes anything that does not need a an oil based product (plastic, acid, lube)
Assembler 2 makes anything and is faster and allows more modules
Assembler 3 is even faster and supports even more modules

[Oktokolo]

I can see many people saying "This is an example of how software development should be like."

Yes. That is exactly how software development should be like.
Wube only removes artifacts of early experiments wich don't add to the vanilla game while not crippling mod potential.
They will probably adjust the Lazy Bastard achievement.

5% of active players or 5% of players owning Factorio? Without that info your number is as useless as this one: 4563.

Don't knowm how many active vanilla players there are, but the last Update of Bob's core library was 4 months ago and since then there where around 100k downloads of it. This is also an all-time high for that mod. So i would assume that to be the upper bound of his active player base.
Angel's Refining never had an update with more than 38k downloads, so that is the upper limit for his active player base.

[bman212121]

- Should pistols be removed from the game? Right now there's no reason to ever craft one. On a free game, you never really have to use them. In multiplayer games it's just something every player will pile up in random chests.

I'd say yes, but I don't decide that. I've said for years pistols are worthless and I use them to troll people by dropping them all over/filling their inventory with them.

- Is armor durability necessary? They can't be repaired, and breaking a Power Armor doesn't sound like a very appealing mechanic. It doesn't seem to have any kind of interesting interaction from the player, other than worrying that your armor might break one day.

No, armor durability was removed from the base game in 0.17. You can still use it with mods but base game armor doesn't use durability in 0.17.

The issue with the Pistols is not that they are worthless, but they were designed to be used for a very specific task. The problem is that once you get past that task, they are STILL part of the game and haven't evolved, which is what makes them useless.

The pistol makes a lot of sense for the very beginning phase of the game (the burner phase). It is quite useful when a small biter is trying to attack you or you need to do a little bit of exploring. You will probably find use for it and the 10 magazines it comes with for about the first hour of the game. It loses it's usefulness at the same time you move out of burner phase and into building a real base. The pollution helps create enemies that are too hard for the pistol to handle. So at that point it needs to disappear into the background much like the burner miners do.

The actual problem is that pistols do NOT phase themselves out, regardless of how much time is on the map. Every time you die, you get 1 pistol and 10 yellow magazines. This is the real problem with the game. Even if you got rid of the pistol, the 10 yellow ammo is still quite useless as well. Obviously the intent of giving you those items for "free" was to make it so you don't get completely stuck in the beginning. But that safety is what's actually causing the problem. It doesn't scale with the game. And you probably don't want it to scale to the point where it just gives you a free SMG + 100 uranium ammo, because that could be abused. So rather than removing the pistol you just need to remove that safety, and you won't keep getting a pistol and 10 useless ammo in your inventory each time you die.

[Philip017]

The problem here is that the devs are mostly correct about the problems they've identified, but have chosen bad fixes to those problems.

Problem: Assembly Machine 1 isn't useful enough because anemic ingredient limit, which confuses new players because their fancy new toy can't even make most things.
Bad Solution: Remove ingredient limits on assemblers (removes tech gating and infrastructure improvement needs, which are core mechanics)
Good Solution: Change Assembling Machine 1 ingredient limit so that it can craft initial recipes and/or change initial crafts to fewer ingredients (2 -> 3 ingredient limit for AM 1 and removing iron plates from Assembling Machine 2 recipe would basically fix this on its own)

Problem 1: Making a pickaxe first thing is problematic
Problem 2: Rebuilding pickaxes is an annoying, mostly pointless time sink
Problem 3: Axeless mining speed is punishing and no longer fits game design vision
Bad Solution: Remove pickaxes completely (removes early and notable feeling of tech progression that directly effects player character, which is a core mechanic, loses character customization slot for modding and misc)
Good Solution: Remove durability from axes, improve initial mining speed, start character with makeshift axe/digging rod to retain progression element, make iron axe a researchable item (attach it to armor research 1?) as an introduction to character equipment customization.

i have to say i like these suggested changes.

if keeping the limits, for AM1 it probably should be able to make AM2, after all AM2 can make AM3. it would be really nice to be able to make green science from AM1 as well. yellow inserters come to mind as a really desired item to make with AM1, instead i have to rush to AM2 before Logistics or craft ALOT by hand.

i am not opposed to removing the pick from vanilla, but should still be a thing mods can add back.
otherwise the good solution listed here is a welcome possibility. although making the pick even more complex seems like an area for mods.

[AvitusNerva]

I agree everything because I think it will really clean things up. I do agree that the number of ingredients isn't the best mechanic for assembly machines. However, my initial thought is to replace this with the "technology level" of the items it can make (example below). It looks like a couple people had the same idea.

AM3: can make everything
AM2: can make everything unlocked by only science pack 1,2,3, military, and production
AM1: can make everything unlocked by only science packs 1 & 2