Friday Facts #305 - The Oil Changes

[Katharsas]

Another nice solution avoided the elimination of the 3-output problem, but...didn't fully remove the complexity wall (since it requires cracking), despite also succeeding at enabling non-blocking chemical science until the player figures out advanced oil--and leaving the current chemical science recipe.

Only produce light oil in the first receipe?

That is actually a SUPER elegant solution. It allows you get started with a LITTLE bit of cracking without having to balance (but a new player could just mind-map this like a normal processing step like any other simple in->out receipe), while he later realizes what cracking is actually used for (balancing the outputs from advanced receipes). Since he already has done a little bit of cracking (only light to petro), it won't overwhelm him then (and he will understand heavy to light).

[Goose]

Asking because I honestly don't understand the reasoning. In V435000's oil changes mod (https://mods.factorio.com/mod/Oil-Changes), lubricant is locked behind flammables?

[mcdjfp]

Like many other "pre-release' games, Factorio was already better than most officially released games out in the market a couple of years ago. Yes, there were issues and bugs, but due to the massive number of beta testers they weren't all that bad. It seems to me that 1.0 is actually a kind of disease with early access games. The developers suddenly panic and a rush of feature cuts/hurried completions, modifications, fixes, and other panicked decisions ensue. Sometimes it even works out, after a buggy 1.0. Because even if the decisions were right, the balance that took years to form can't be rediscovered overnight. Factorio has been such a good game, and is probably the best optimized game I know, I just don't want it to suffer.

This change will require balance changes as 2 defensive options are being delayed, laser turrets and construction bots. Yes construction bots are defensive as you can give them a chest of walls, a chest of repair packs and let them take care of running everywhere maintaining the defenses. The biters will need a rebalance. I had one 0.16 game where for every 5 minutes of building I had 15 minutes of base defense repairs (desert map).

I said this earlier, in my opinion the core of the problem is the multi-output recipe. Off the top of my head I can think of 6 of them. These recipes reverse the lesson every other recipe teaches. For the majority of recipes, if your output is blocked, that is a good thing as it means you have enough and can move on to the next issue. For these 6 recipes it is bad to have an output blocked as it stops the other outputs which might be in higher demand.
Of these 6 recipes, only 1 is required.
2 Nuclear, optional anyways for the rocket, and Kovarex (which is even more optional)
1 miner, optional (any time a miner gets placed over two different deposits) as you don't have to mine everything.
3 oil, 2 optional, 1 required. While advanced oil processing is good it is not required and neither is coal liquefaction. Basic is required to get further in the game.

The choice seems to be to (one of the three)
1. keep multi-output recipes (and hopefully teach them as early as possible)
2. make all multi-output recipes optional
3. remove them all from the game.

Whatever you do, please minimize the collateral damage, and realize that many of us feel that Factorio was good enough to release several years ago.

Edit: I must add, I am still personally in the leave things alone camp.

[Reika]

Expressing agreement.

Count me in on that. Also I quite like this sentiment and phrasing:

1. propose a controversial half baked change
2. get furious feedback
3. propose a controversial half baked change with an cherry on top

[MiniHerc]

1. keep multi-output recipes (and hopefully teach them as early as possible)

This, and add it to the tutorial.

[mmmPI]

Expressing agreement.

Count me in on that. Also I quite like this sentiment and phrasing:

1. propose a controversial half baked change
2. get furious feedback
3. propose a controversial half baked change with an cherry on top

1. propose a controversial change
2. get furious feedback
3. propose a controversial half baked change
4?

I meant sure the 2nd change had less planning, even though there were many suggestions, maybe even BECAUSE, there were many suggestions, none of them is still implemented

[Reika]

I meant sure the 2nd change had less planning, even though there were many suggestions, maybe even BECAUSE, there were many suggestions, none of them is still implemented

You cannot seriously be claiming that because none of the suggestions was implemented - not even a true claim, as explained below - that must mean they were poorly thought out. There are many reasons that suggestions are rejected and/or ignored, and the primary allegation here is that the devs are fixated on and/or infatuated with their changes such that it blinds them as to their pitfalls.

Also, some pieces of suggestions were implemented, including my sulfur-in-blue-science idea.

[mmmPI]

What i meant is that so far none of them made it to my current game, i have tried 2 differents mods from the previous FFF thread, in test game to see the tech tree and in my game without saving, just to see how it would change things.

There are so many different suggestion and directions that were proposed in FFF304, that even the morning before FFF305 some propositions were still discussed, some of them indeed made them through the FFF305 so that means some are heard, not others i can understand that you might refers to that as going in the same direction again, (stubborness ?) i am for my own not satisfied with the current proposition before i even had time to test it, so i'm not expressing it with argument i would need to test it first, but i think the next step is the feedback on the proposed mod, that is a good indicator that things aren't pushed on us lightly that it is a mod optionnal, good to give feedback, maybe/probably on a hot topic that is susceptible to still receive some attention the next weeks .

I was convinced there was no oil changed needed at first, then i changed my mind, i think i understand why the change is attempted. Hopefully things work out so that it's better for more person after some time.

[Barhandar]

1. propose a controversial half baked change
2. get furious feedback
3. propose a controversial half baked change with an cherry on top

And the cherry is unripe too. I agree with the whole sentiment.


Wube, this change is a bad idea inherently. Doubling down on it (making players need to belt sulfur and then, in the very end, also making them add piping for light oil to produce rocket fuel) makes it a worse idea. Instead of piling up things on a bad idea in hopes that it'll stop being a bad idea (it won't), please go back to square one instead and rethink it from scratch.


Also, I agree with some of the comments on reddit, and likely in this very thread too, namely, that oil processing isn't even the issue with chemical science. The problem is the recipe being made out of a bunch of complex things that, at that point, aren't continuously necessary, while all three of the lower science (red, green, military) are made out of stuff the player is going to be using en-masse, and hence automatically making, before even touching the science in question. If you want to fix that, introduce something continuously necessary that uses engines/red circuits and is available without chemical science.

Hint hint that's construction robots.

[MiniHerc]

1. propose a controversial half baked change
2. get furious feedback
3. propose a controversial half baked change with an cherry on top

And the cherry is unripe too. I agree with the whole sentiment.


Wube, this change is a bad idea inherently. Doubling down on it (making players need to belt sulfur and then, in the very end, also making them add piping for light oil to produce rocket fuel) makes it a worse idea. Instead of piling up things on a bad idea in hopes that it'll stop being a bad idea (it won't), please go back to square one instead and rethink it from scratch.


Also, I agree with some of the comments on reddit, and likely in this very thread too, namely, that oil processing isn't even the issue with chemical science. The problem is the recipe being made out of a bunch of complex things that, at that point, aren't continuously necessary, while all three of the lower science (red, green, military) are made out of stuff the player is going to be using en-masse, and hence automatically making, before even touching the science in question. If you want to fix that, introduce something continuously necessary that uses engines/red circuits and is available without chemical science.

Hint hint that's construction robots.

Perhaps Chemical Science needs to be renamed.

[bman212121]

I feel like at this point you are trying to over engineer the game. And you will keep doing so until it is no longer the game I fell in love with.
All under the guise of "making things better".
I'm not saying all updates are terrible.
I'm not saying do not try to improve.
But sometimes, notably this time (as, sometimes in the past), it just feels unnecessary.

Basically this. Not every game needs to be linear and have everything spelled out for you. That used to be the point, you actually had options and would require you to figure it out on your own. It's been far too long since the first time I remember figuring out oil, but clearly we figured it out. This game is basically already complete, so trying to fix things aren't broken is just going to make things worse. Especially when you start trying to wall things off to hand hold players.

Anyone who knows what they are doing is going to get annoyed because the game is basically baking a tutorial into itself. Oh you didn't want to use belts on your map? Too bad, the game is now 100% linear and you're forced to setup production for them now. That's the path this game is starting to head down, and it's very concerning. Pretty soon you'll be setting up a ton of tech you don't care about, and every time you play you'll basically be forced to go through the "New Player Experience" tutorial on every new map because the game was designed to force you down one path.

[fusionfan]

Long time lurker, but I made an account to post this. Background: I can speedrun 0.16 under 2:30 for my own fun, and I have built Nk SPM megabases as well. 0.16 is my cocaine.

This major change to oil doesn't make sense. What, we just miraculously extract petroleum from oil? What happens to the other 55 fluids? The whole point of the procedure IRL is separation into components, i.e. the old recipe. Currently we can just make solid fuel out of light and heavy and burn it for power or in trains. Easy. Dealing with pipes wasn't ever an issue, you have to figure it out exactly 1x in your life. A million people figured it out already. Can't? Internet exists. Just look up a build or blueprint. It is 2019, you should know how to search the internet if you get stuck in a game. Besides now the piping comes at advanced oil, so the problem (which isn't a problem really) is not solved, just delayed. Except now you have to rebuild everything.

Also, I am that person who likes to rush construction bots ASAP. Then I can have my bots just scale up the whole base as I go to more advanced tech. I like to design blueprints, make a mall, pop things down. I hate building furnace lines by hand, so boring. Now for every new map it will take even longer to get to automating construction. I thought automation was the point of the game!

[wobbycarly]

*snip*
As for oil products, my sugesstion is to keep current science, flamer fuel and rocket fuel recipes as is, but change following:
1) Change Petroleum Gas input in Sulfur recipe for Light Oil. This, along with Solid Fuel, gives Light Oil a major resource sink, so backing up should be less of a problem. This way, Plastic uses Petroleum, Acid uses Light, and Lubricant uses Heavy Oil.
*snip*

Having read every word on the posts about the oil changes (FFF#304 and FFF#305) I came to a similar conclusion to this as well. At the very least, have the refinery spit out at least 2 items - definitely not 1 - and the player is more likely to have some room to add a third output. I actually think HO and PG (leftmost and rightmost) outputs are optimal, as the player is likely to be hinted that something will come out the middle, especially if they've looked any way down the tree to see that LO has a use.

Having nearly 2000 hours in game (since version 0.14.x), I have to admit I don't remember my first oil build well, but I know I launched a rocket in my first game (eventually!!) and went to youtube to get some ideas on how to optimise it.

[BorisTheBastard]

I don't understand the notion regarding oil processing being complicated or confusing... The refinery has 3 outputs, just store the light and heavy oils in tanks until advanced oil processing is researched and the oils can be utilized.

When I start a new map I bee-line for construction robots so that I can start laying down blueprints easily. Moving them further down the tech-tree is going to make the initial grind even longer.

[rantingrodent]

I have several hundred hours under my belt with Factorio but also on my very first encounter with oil processing I was immediately like "this is too much to figure out at once". It was the first time that I chose to look at other people's designs before trying it myself because it was obvious that the feedback loop on doing the right thing was going to be useless.

The changes outlined here smooth out the learning curve for sure. You're much better equipped to tackle complex refinery setup if you've done at least some non-trivial fluid handling and piping liquids into assemblers. This separates learning how to handle fluids from the bigger challenge of how to handle a bunch of fluids in the same space.

This won't actually make advanced oil processing any easier to grok, though, I think. The oil refinery building in and of itself is the reason advanced oil processing is such a pain in the ass for new players. Having a single "oil refinery" building that encapsulates this 2-in, 3-out process is what forces you to learn too much at once. There's pretty much only one "right way" to set up such a complicated building. You can't muddle around with suboptimal solutions, you can only fail until you hit the one right solution or look at the wiki.

If oil processing was broken down into a couple of more discrete steps with simpler individual recipes, you could muddle your way through with a bad refinery design that at least didn't get fully blocked up, and if one product was bottlenecking the system, that would be happening at a machine that handled that product, rather than one product out of three produced by the same building being the bottleneck, which is difficult to surface without being intrusive.

At the moment the oil refinery feels like the least Factorio part of Factorio to me; complexity that should be manifesting in my factory design is all tied up in this one building and I have to deal with all of that complexity at one time to get any value out of it.

[Domanoc]

I don't like the changes in the basic oil processing as it feels like dumbing down the game. And i don't feel like it makes it easier for new players to learn oil processing.

Als the current changes now propose we are going to base the production of science across all the packs more on petroleum gas. Changing the solid fluid requirement to sulfur looks as a no deal change at first to me but then I realise that until the rocket you take a way the need to use light oil (based on an optimal use to craft the solid fuel). Heavy oil use stays the same in the need of lubricant in electric engines.

I have a feeling the current solutions are suffering a bit form tunnel vision. All the solutions to make the process more understandable that are presented try to do this by changing the basic oil processing. And while only producing 1 output make it easier at first glance. The player hasn’t learned anything about maintaining the different oil levels. This simply moves the players that are stuck to the point where they are going to need heavy oil for electric engines. Where by simply moving from basic to advanced processing they are going to halt the production with the same problem of backing up oil products.

I propose the change should be made in the consumption of oil products instead. Currently proposed the chemical science pack only needs petroleum gas or incase of the previous recipe you can provide the solid fuel with light or heavy oil. But you are not forced to use it. If instead we change the recipe to use all oil types near the rate basic oil processing provides. For example: advanced circuits use petroleum, use electric engines to use heavy oil/lubricant (or let engines use lubricant directly), change the sulfur/solid fuel to something that can only be made with light oil (Has to be new since this doesn’t exist now and it’s still too early to introduce rocket fuel).

While this won't stop the oil products from potentially backing up it wil move it to a point after the player has created some chemical science and thus unlocking the advanced oil processing to fix it with cracking. It also makes the player more aware of the 3 different oil products and the need to maintain throughput for all of them.

To further prevent the player from being stuck the cracking tech could be separated from the advanced processing research into its own tech unlock based on automation and logistic science. This allows the player to fix the problems without needing to fully complete the chemical science pack production.

While iam aware that this solution doesn’t solve everything (for example the need to use pumps and circuit conditions to effectively manage oil cracking) I want to promote thinking in other ways to solve to oil introduction.

Thank you for reading.

TLDR: I want to promote thinking in other ways to solve to oil introduction.

[BattleFluffy]

I'm having doubts about these changes again.

In a megabase scenario, Mining Productivity research can reach well over lvl 100.
At level 100 this means pumpjacks which have reached their minimum output and contain 2 speed modules will produce crude oil at a rate of 40 per second. That's a lot of oil from 1 pumpjack.

With this much crude available, Basic Oil Processing is definitely going to be the go-to solution for all things petrol.
Instead of a deliciously complex puzzle problem in lategame with trying to balance the fluids of a complicated refinery, we can just make everything into petrol and we're done. This way we have far fewer active entities and far fewer fluid boxes, so it's better for UPS. We only need a tiny AOP refinery for the miniscule amount of lubricant that is required.

This decrease in complexity guts the oil system of any design interpretation, particularly in ultra-lategame. There is simply an optimal way to do it, and it's a depressingly simple way. Basic Oil Processing should not be the "best" way to acquire oil products, if anything it should be the worst.

Between this week's FFF suggestions, last week's FFF suggestions, and the current system, I still like the current system the best.

[Optera]

Simplifying basic oil processing breaks bases relying on a mix of basic and advanced oil processing to have perfect ratios for heavy, light, and petroleum. It basically screams use circuit controlled cracking instead of pre-calculating usage and build in perfect ratios.

Rocket fuel requiring 10 solid fuel + 10 light oil is a 10% increase in cost (excluding productivity), requiring not only rebuilding the rocket fuel assemblers but very likely the whole light oil production.

I'll probably use (or make) a mod reverting these changes to what we where used to.


On topic of dealing with outputs. We used to be able to just pipe excess heavy and light oil into steam turbines. It made sense to "burn" excess waste products, that is exactly what real refineries do.

[Reika]

Reading through these threads again, something stands out to me:

The response to the oil changes from the general playerbase is what Steam would call "Overwhelmingly Negative". Five to ten, or even more negative replies for every positive one.

However, the response from mod developers is almost unanimously negative. While I am not going to claim that mod developers are inherently better or have more worthwhile opinions, they nonetheless are generally highly knowledgable about the game design and balance, far more so than average.

So if the general response to a proposal is so strikingly negative, and seems to get worse the more well-acquainted with Factorio the people are, I would strongly suggest that this is a red flag the size of Asia.

[IronCartographer]

It is certainly true that ensuring players engage with a tutorial / play a campaign could go a long way toward making this all irrelevant alleviating the problem. However, the game mechanics that the tutorial would explain directly influence how approachable the tutorial is. A tutorial doesn't change the problem itself.

I do like the implications of having a belt of sulfur available--but that says nothing about the refinery process itself skipping straight to gas.