Friday Facts #304 - Small bugs; Big changes

[zero318]

The way I see it, the problem with oil isn't that it's hard to balance, but that the need for that balance and the methods of handling it are presented poorly.

Currently, the player receives three new fluids from the refinery and likely wants to find something to do with them. However, unless a new player has read ahead in the tech tree and found the later recipes, the only use they'll see for any of this is solid fuel. Since all three solid fuel recipes are presented simultaneously, it isn't obvious that one of them is better than the others and players are likely to just read the first one and assume it applies to all of them. This is the only tech in the game that introduces multiple recipes for the exact same item, so there's no precedent for the player to not make that assumption. Despite that, the player will probably start producing solid fuel from some of the fluids unless they've switched to solar.

By the time players research plastic, sulfur and lube, belts with solid fuel are likely to be backing up and fluids just collecting in tanks. There's really no obvious use for large quantities of heavy oil except lube, which will require further research in order to utilize. Thus the refineries back up because the output tanks are filled with a seemingly redundant resource that can only be made into solid fuel or another less useful fluid. Without having taken the time to read over the entire tech tree, there's no way of knowing the future value of the other oil products and players are likely to start removing tanks to void it.

These problems could potentially be solved by changing when the solid fuel recipes are obtained, adding more uses for oil products, and rearranging the techs that unlock uses for oil products. If only the heavy oil recipe for solid fuel is unlocked at basic oil processing, players would be less likely to have jamming refineries and understand that the other solid fuel recipes are upgrades since they're further down the tech tree. Even if players have gone solar, they'll still need solid fuel for chemical science and understand the need for lube by the time the more efficient recipe for light oil is unlocked.

The various techs could also be merged to make it more obvious which products are being used, such as "petroleum gas processing" unlocking both plastic and sulfur simultaneously, "light oil processing" unlocking its solid fuel recipe and as a prereq for flamethrowers, and "heavy oil processing" unlocking lube and electric engines. This would allow players to research solutions to any backlogs and discover the uses for each product simultaneously.

Of course, there are definitely flaws with this idea too, but it at least seems like it would help smooth the transition to oil and naturally guide the player towards both solutions to their problems and higher efficiency.

[mm.lion]

I love new oil tech. But when I first sow it I totally hated it. Now I completely change my opinion. Here is the reason: newbie will face techs and puzzles in the following sequence:

1. BOP ---> newbie start working with fluids. It is simple, but it is in the beginning of the game
2. AOP ---> it is NOW the ONLY way to get lubricant. Newbie faces multiple output puzzle (middle game, newbie is already familiar with a lot of things).
3. Coal Injection ---> Obviously, first two techs are NOW produce NOT ENOUGH heavy oil! So player will use coal injection (with its interesting puzzle)! Before the change of BOP, the Coal injection tech was very often unused, since BOP produced a lot of heavy oil.

So I like new tech very much.

[Sander_Bouwhuis]

I'm thinking of overlapping ore patches here: It happens quite often that a miner will load two different ores onto a belt. You have to separate these or else they will break things up further down the line. Of course, there are people who avoid this situation by placing their miners in such a way that they only mine one kind of ore -- but why would they do that? Obviously because they know that mixed ores on one belt could break things, so they are either experienced players or noobs with a clue. These players shouldn't find it too hard to figure out that overload on one oil processing product will block production of the others. Those new players who place their miners on a multi ore patch unawares will learn that lesson the hard way early on and should be able to apply it to oil processing as well.

I don't fully agree with that, because a miner outputs all the mining results at the same spot. New players immediately notice the output belt, which is shared by all the different products, is full and that's why the miner has stopped mining at all. The refinery on the other hand uses different output "slots", so the players don't immediately see that another product is blocking the production. Since the output for PG is not blocked, new players might not think of another output at another spot being blocked as the reason for the refinery stopping to work. The problem is to point the players to these output slots as the cause of the problem.

Completely agree. It's a user interface problem. Just TELL the player something is wrong. A poorly designed UI is NOT the same as a puzzle.

[daydev]

I did try to point out why each of the solutions I have read and described in my post are not better than the proposed solution. Which solution would you prefer and how can you answer my problem with it

What about my proposal to leave both oil processing recipes alone, but introduce an earlier "primitive refining" oil -> gas for the chemical plant, similar how burner drills are the precursor to the "real" mining? As far as I saw you didn't address that. This way newbies get get into oil gently as you want, but you don't have to wreck other things to make it happen (robots, flamethrower, solid fuel).

[XkyDiver]

I did try to point out why each of the solutions I have read and described in my post are not better than the proposed solution. Which solution would you prefer and how can you answer my problem with it

What about my proposal to leave both oil processing recipes alone, but introduce an earlier "primitive refining" oil -> gas for the chemical plant, similar how burner drills are the precursor to the "real" mining? As far as I saw you didn't address that. This way newbies get get into oil gently as you want, but you don't have to wreck other things to make it happen (robots, flamethrower, solid fuel).

This, in combination with a shiny new "Oil Refining" Tutorial would be stellar.

If the change literally only affected Basic Oil Processing, it'd be one thing--but if you make the proposed change, you've got to change recipes, pre-requisites, and upset the order of a lot of things. Making Flamethrower Fuel out of Petroleum Gas just feels *weird*!

I like the idea of somehow using the Chemical Plant as a "Primitive Refining" thing to do Crude-->Petroleum Gas.

[conn11]

I did try to point out why each of the solutions I have read and described in my post are not better than the proposed solution. Which solution would you prefer and how can you answer my problem with it

What about my proposal to leave both oil processing recipes alone, but introduce an earlier "primitive refining" oil -> gas for the chemical plant, similar how burner drills are the precursor to the "real" mining? As far as I saw you didn't address that. This way newbies get get into oil gently as you want, but you don't have to wreck other things to make it happen (robots, flamethrower, solid fuel).

This, in combination with a shiny new "Oil Refining" Tutorial would be stellar.

If the change literally only affected Basic Oil Processing, it'd be one thing--but if you make the proposed change, you've got to change recipes, pre-requisites, and upset the order of a lot of things. Making Flamethrower Fuel out of Petroleum Gas just feels *weird*!

I like the idea of somehow using the Chemical Plant as a "Primitive Refining" thing to do Crude-->Petroleum Gas.

any solution without the potential of replacing the main bulk of PG production with a heavily simplified version of oil processing seems to be preferable

[mcdjfp]

What about my proposal to leave both oil processing recipes alone, but introduce an earlier "primitive refining" oil -> gas for the chemical plant, similar how burner drills are the precursor to the "real" mining? As far as I saw you didn't address that. This way newbies get get into oil gently as you want, but you don't have to wreck other things to make it happen (robots, flamethrower, solid fuel).

While it would bug me that the oil refinery has a flare stack while the chemical plant doesn't, this might work. If combined with delaying solid fuel from petroleum to push the use of the oil refinery (maybe) and a poor enough recipe that it is clearly a temporary stopgap like the burner mining drills. As long as it avoids the temptation to keep using the primitive recipe the whole game. I really want to like this idea, but a part of me still feels like it disrupts one of the games key challenges and risks luring new players down the wrong path. The longer the proper solution is delayed, the more difficult (or at least time consuming) it will be to implement (for the player in his factory).

I still think a tutorial where some experimenting is possible without outside pressure would be very useful.

[VFaalcatnodriiro]

One of the less complicated but still contributing factors is that suddenly there is 5 different fluids that you need to do something with, and they go into

5?

I only count 4 (for new players): Crude, Heavy, Light Oil and Petroleum, that's it. But that's just nit-picking.

I don't really understand the argument of "it's making the game simpler" and I feel like many people take it in the "git gud" way - "we understand it, make newcomers suffer too"

It has nothing to do with suffering.
But here's a suggestion: Change the Oil-Refinery so that if Basic Oil Processing is active, the "flare stack" that's already to be seen at the current refinery-model is actually been used. I can think of two ways of doing this:
* Just "burn" everything that is not being transported out of the Refinery. That takes the pain for new players away of HAVING to deal with all three outputs, but they (and we) can, if we want to.
* A little harder: Check if an output-pipe is connected and "burn" everything that has no output. If everything has an output-pipe, leave the behaviour like it is now.

I've read many times the difference between getting solid fuel from petroleum gas and light oil. I think you massively overestimate how much a new player cares about efficiency, and how much should he.

Maybe. But why would you FORCE and TEACH them (and us) to use inefficient designs?

Adding cracking with basic oil processing and having basic oil processing only output heavy or light I don't find to be a good solution. It would mean it just extends the refining process by one (in case of light oil) or two (in case of heavy oil) rather basic steps - when we already have a lot of basic steps.

Correct. But it would make more sense in a "Basic" vs. "Advanced" sort of recipe. Why would a "basic" recipe do all the cracking for me when the "advanced" recipe forces me to deal with fluids I maybe don't care about?

Allowing the player to select individual recipes (heavy or light or petroleum gas only) would allow to completely avoid using Advanced oil processing. That's surely dumbing it down significantly. Adding flare stacks has the same effect of making Advanced oil processing completely optional additional complexity for more resource efficiency.

You'll achieve exactly this. Advanced oil processig will be skipped for coal liquefaction. The "common" factory will have a coal liquefaction-build turning coal into lubricant and lots of Basic Oil Processing for Petroleum. Why would/should anyone bother with AOP anymore? It's no longer needed.

[daydev]

If combined with delaying solid fuel from petroleum to push the use of the oil refinery (maybe) and a poor enough recipe that it is clearly a temporary stopgap like the burner mining drills.

If it were up to me, I would make the primitive processing give 100 oil -> 20 (maybe 25) gas. It doesn't seem beyond reason. Burner drills, stone furnaces, yellow belts, yellow inserters are two times slower than their upgrades, so let it be about the same for the first oil, plus you're losing out on other products. And then even make solid fuel take 30 gas (so 10 from light, 20 from heavy, 30 from gas, because c'mon, it's gas) to make it absolutely clear that this way of doing things is no-good.

[conn11]

You'll achieve exactly this. Advanced oil processig will be skipped for coal liquefaction. The "common" factory will have a coal liquefaction-build turning coal into lubricant and lots of Basic Oil Processing for Petroleum. Why would/should anyone bother with AOP anymore? It's no longer needed.

Futhermore AOP only gives 10 HO, coal liquifidation (netto) 65 clearly the go to way. Of course there is the 45 LO in AOP but beside cracking it to PG, whats the use, elimated from flamethrower, a little bit more efficent in flamethrower turrets (with infinite flame damage research) an SF. 10 LO or 20 HO/PG not that much of an difference, considering you will gain 50 PG (40 by default +25% bonus) from the new BOP. the only really practical use of AOP is for your rocket fuel demanding RPM Megabase, rather a niche application. In the long run with this solution LO seems like deletable all along. Not to mention the new blue science repice is somewhat led ad absurdum.

[SpiffyTriffid]

Here's an idea, have a flashing UI indicator over the blocked outputs in the GUI. Flash a localized text box as well, somewhere under or above the progress bar, that says "output pipe (x) [full/backed up/etc.], cannot continue processing" or similar. This is fundamentally a problem of telling players about a mechanic in the game (systems stop all outputs when blocked) and regardless of whether or not a tutorial level for oil will ever be produced (time, effort, .17 is going to be 1.0, etc.) a GUI will help either way in bringing the cause of the problem to the player instead of removing the problem to be solved.
There are still posts asking about why the refineries aren't producing anything, so the UI indicators we have now are evidently not enough. Instead of abandoning the UI indicator path and simply tearing out complexity, could you consider more UI indicators before going down that path?

[morsk]

I just realized. If oil is so frustrating that they're willing to break 1/2 the early game over it, maybe they should just let players walk through pipes. What's really causing the frustration, hm?

[FuryoftheStars]

I just realized. If oil is so frustrating that they're willing to break 1/2 the early game over it, maybe they should just let players walk through pipes. What's really causing the frustration, hm?

Yeah, really, where’s that noclip cheat code.

[BlueTemplar]

Flame turrets are available much earlier, and you can plop them down even before you have a pumpjack placed and powered !
(And looks like flamethrower ammo will be much easier to make too now!)

Well, technically, I think they only made the change to the ammo used by handheld? Still a bad change, imo, and I hope they didn’t do it to turrets, too.

They are available much earlier (well, considering the whole green science phase) than laser turrets, already in 0.17.

If new players are finding oil to be so difficult, just do what's been done with everything else - put the dumbed-down hand-holding version of it into the New Player Experience campaign and leave the base freeplay game alone, for people who weren't born yesterday.

It another game, this might work, but Factorio has a freeplay reputation (like its predecessor : Minecraft). A lot of players don't even bother with the campaign (maybe if they'll make it awesome enough... the new not-tutorial seems promising !), but I wouldn't bet on it.

Please, don't make silly comparisons. Trains only need stone, steel, green circuits and basic engines.
While conbots also need lubricant and batteries, and 45 red circuits per roboport to actually use them. (And personal roboports need 50 blue science.)

Ingredients aren't the point; if techs were available later, they'd have different ingredients. I was being generous with trains. There's no need to build trains at all. I've seen I've seen reddit posts by people who've played 200 hours and want to start learning trains. And yet the train is available in green science, for anyone who wants to start using it.

And guess what, blue science also needs red circuits... and in 0.17.0 solid fuel, which is probably the easiest of all petrochem products.
One could even say that blue science is easier to set up than conbots !
(As suggested multiple times already.)

Blue science is easier to build, but if you build it first you have to build it without bots, and then you still have to build the bots. Good choices like this should be preserved, not discarded.

Ingredients are the point (even more so for new players). By the point that you have set up conbots, you only need to add solid fuel for blue science. In a pinch, you can even hand-craft blue science from red circuits, engines and plastic.
(It's harder for bots since the flying frames take so long to craft.)
While if you had set up blue science first, you still need to set up sulfur, acid, batteries, lubricant, and electric engines for conbots !!
No wonder that most players forget that bots are in green science !
(However, I don't really care whether bots are in green or blue...)

Re the oil change, imo it is pretty simple: Anyone who has played the game up until you need oil processing is the kind of person that likes this type of game and will not be put off from a slight difficulty bump. The people that don't like this type of game will quit the game before they even need oil processing. The only thing that needs to be in place is an up to date wiki that gives guidance when you get stuck.
[...]

Latest stats (for non-modded games) :
~58% of players have researched Oil Processing (but might have quit before setting it up).
~42% built (and placed?) a locomotive (would new players prefer to use one for first oil transport, or car, or just a loooooong pipe?)
~39% achieved 16.66.. advanced circuits per minute. (which need oil)
~28% got supplied by a logistic robot
~26% built 100 machines with conbots
https://steamcommunity.com/stats/427520/achievements/

Allowing the player to select individual recipes (heavy or light or petroleum gas only) would allow to completely avoid using Advanced oil processing. That's surely dumbing it down significantly. Adding flare stacks has the same effect of making Advanced oil processing completely optional additional complexity for more resource efficiency.

You'll achieve exactly this. Advanced oil processig will be skipped for coal liquefaction. The "common" factory will have a coal liquefaction-build turning coal into lubricant and lots of Basic Oil Processing for Petroleum. Why would/should anyone bother with AOP anymore? It's no longer needed.

Depends on the relative efficiencies...

I just realized. If oil is so frustrating that they're willing to break 1/2 the early game over it, maybe they should just let players walk through pipes. What's really causing the frustration, hm?

Nooo, do not strand me in the pipe maze ! XD
No wonder that squeak through is one of the most downloaded mods. I remember it being the 2nd mod I ever installed - and the first was RSO, so it kind of had to come first anyway...
(It gets better with experience...)

"Breaking 1/2 of the early game" seems to be a curious over-reaction, considering how little is going to change in practice... (most of the experienced players beeline to Advanced Oil Processing anyway, don't they?)

[VFaalcatnodriiro]

I don't really understand the argument of "it's making the game simpler" and I feel like many people take it in the "git gud" way - "we understand it, make newcomers suffer too" (I don't assume all of people who posted this argument think along those lines, but I'm getting that feeling from many posts). Advanced oil processing is still completely mandatory to go through, it's just a bit later. The thing that changes is that you can access anything unlocked by Chemical science pack, but you can only get robots after you get Advanced oil processing going. You could say the puzzle and the reward is exactly the same as it was then - except cracking provides a solution to the oil puzzle other than adding storage tanks to fix the problem temporarily.

I am in the camp that it won't necessarily make the game simpler, but it won't distribute the challenges any better because it just shifts the hump to advanced processing. I think it's important to introduce the concept of output blocking during basic processing, because if you don't, then (new) players will assume oil refining to be a simple re-skinned assembly line, and plan their setup accordingly (i.e. not leave space for piping water, heavy, or light oils). This means that when advanced processing rears its head, players are faced with the prospect of tearing up their whole refining setup in order to restructure everything to accommodate all the extra fluids, and redesigning the chemplant piping to run off the 3 different outputs, all without bots. I don't see this as any less strenuous than having to set up oil in the first place. Delaying the introduction of output blocking to this point would therefore be no less frustrating, with the only saving grace being the immediate availability of cracking to fix it once discovered, and, as has been raised repeatedly, the biggest stumbling block is discovering the problem, not solving it.

This'll be an huge issue, I agree. If they don't tear down everything they have ALL of the problems they now have with BOP with AOP. Blocked outputs, multiple fluids AND no bots to help them doing stuff. All they learned with BOP is how to handle two more fluids (and pumpjacks...). But even worse, the fist Oil setup will most likely be built into their bus (or where-ever), with no space for expansion. So they have to deal with one basic-setup (for petrol) and another setup for ... for what? Lubricant. And nothing to do with the light oil and a whole spaghetti pumping of the overflow-petroleum to the first build.

You are creating new problems while NOT fixing the initial problem.

[conn11]

"Breaking 1/2 of the early game" seems to be a curious over-reaction, considering how little is going to change in practice... (most of the experienced players beeline to Advanced Oil Processing anyway, don't they?)

a simple 1 in 1 out PG repice, no balancing needs, therfore the possibility of very compact setups, directly at your main bus or depending on playstyle red circuit/ bot/ PU production, no dependency on water, just the need to ship or pump crude oil in. It isn't more efficent but oil is an infinite ressource after all. A expirienced player would rather stick to AOP out of nostalgia than actual practicability (if not playing a modded "classical" version). Making the great riddle of oil a temporary thing to get a miniscule amount of HO from AOP.

[TatsuZZMage]

PLEASE LEAVE to Basic oil to being a mod, But DO keep the 1 input for oil PLEASE.

[BlueTemplar]

It isn't more efficent but oil is an infinite ressource after all.

So are other resources, for most practical purposes, if one goes far out enough / does enough productivity research.
Maybe the 20% (of initial or min) yield of oil should be lower ?
Why was it buffed so much in 0.13 and then in 0.15 ?

[VFaalcatnodriiro]

Allowing the player to select individual recipes (heavy or light or petroleum gas only) would allow to completely avoid using Advanced oil processing. That's surely dumbing it down significantly. Adding flare stacks has the same effect of making Advanced oil processing completely optional additional complexity for more resource efficiency.

You'll achieve exactly this. Advanced oil processig will be skipped for coal liquefaction. The "common" factory will have a coal liquefaction-build turning coal into lubricant and lots of Basic Oil Processing for Petroleum. Why would/should anyone bother with AOP anymore? It's no longer needed.

Depends on the relative efficiencies...

Since oil is infinite, not really. It's easier (faster?) to search yet another oil-patch than to fiddle with the ratios.

Step 1) Build BOP -> PG/SF till you need lube.
Step 2) Set Up AOP for one barrel of HO.
Step 3) Build Coal-Liquefaction for HO/lube (at this point, the refinery may even directly output lube, since HO is (almost) obsolete). LO goes 100% into SF, replacing the output from 1) via priority splitter - once it's backed up, it's cracked into PG and the highest priority-source for PG-products.
Step 4) Expand BOP if you're low on PG.

HO/LO as well as AOP are basically obsolete at this point if this goes live on monday.

PS: I'm curious, can you at wube see how many of us have pinned their version to 0.17.58?

[conn11]

It isn't more efficent but oil is an infinite ressource after all.

So are other resources, for most practical purposes, if one goes far out enough / does enough productivity research.
Maybe the 20% (of initial or min) yield of oil should be lower ?
Why was it buffed so much in 0.13 and then in 0.15 ?

With the difference of beeing quite late game or even post rocket launch then. I doubt that any first resource field (except oil of course) will still not be mined completly by then.