Friday Facts #305 - The Oil Changes

[posila]

The problem, though, is not one of immediacy. It doesn't matter if the player spends 3 hours or 3 days shuffling through the chemical sciences, at some point he/she will encounter the advanced recipe and say "Really? You're making me redo all that?" and possibly "I wish I could have just fixed this right from the start." (or more likely, something more emotional and less reasonable.) It's not something they could (reasonably) be expected to have seen coming, nor is it something they could avoid. Even if such a player builds an advanced refinery somewhere else, not only is this forcing players to redo a system they may have put considerable design time and effort into making, there is pretty strong pressure to get rid of the old setup anyways to not waste crude on the old, far less efficient (by about a factor of 2) recipe. For an analog, consider electric furnaces, which some people don't even bother with as a matter of course. In that case, there is, again, a higher tier of building that breaks previous setups unexpectedly due to spacing if not known about beforehand, and it similarly draws the ire of players as a result. The big difference, though, is that players have the option of simply not switching to electric furnaces if it bothers them enough, with no great loss. Not so with oil.

I claim "you're making me redo all that" is going to happen at most as much as with anything else in the game. Players doesn't know what is coming at any stage, so they don't know at what scale they need to build anything (and build in small scale). It's pretty hard to come up with some organized system like bus in the first playthrough, because it is solution to a problem players don't know exists.
Using inefficient BOP instead of AOP, is not problem as much as not using electric furnaces is not a problem. If player has enough oil, they won't mind using BOP, if they are short on oil, AOP can help them solve that problem.

[Adamo]

For the record, the comparison with steel/electric furnaces is not apt. It is more efficient energy-wise to use steel furnaces than it is to use electric furnaces, but electric furnaces give access to modules and beacons, so with furnaces, it is a choice, rather than a forced linear progression.

[posila]

For the record, the comparison with steel/electric furnaces is not apt. It is more efficient energy-wise to use steel furnaces than it is to use electric furnaces, but electric furnaces give access to modules and beacons, so with furnaces, it is a choice, rather than a forced linear progression.

I don't see it as unreasonable comparison - electric furnaces allow you to use solar energy, which is essentially free in long term, as opposed to burning resources you might be low on. And with productivity modules the furnace will have higher output/input ratio than stone furnace.

[Theikkru]

I claim "you're making me redo all that" is going to happen at most as much as with anything else in the game. Players doesn't know what is coming at any stage, so they don't know at what scale they need to build anything (and build in small scale). It's pretty hard to come up with some organized system like bus in the first playthrough, because it is solution to a problem players don't know exists.

My counter to this is two-fold. First, is that it is not necessarily true that the player has no idea what is coming. Factorio is known for being a factory building game, and it is possible to gather from simple screenshots and the trailer (as I did before my purchase and subsequent first playthrough) that plenty of expansion will be required. This is enough information to motivate (certain) players (including myself) to try and plan ahead for said expansion beforehand even during a first playthrough. (Some people much prefer doing things right the first time around, even if it ends up being slower.) Second, the cause of the rebuild or redo is more important than how often it happens, because it determines the lesson being taught. In cases where the player doesn't plan for expansion (in a game all about building and expanding a factory), he/she has only him/herself to blame for building small and boxing things in. In a case such as basic to advanced oil, however, there is not enough information available to the player to predict that the oil setup will need additional (and distinct) connections, so the rebuild is caused by game factors outside the player's control, which feels bad. Further, the more effort the player spent on basic oil, the worse it feels to have to ditch it.

Using inefficient BOP instead of AOP, is not problem as much as not using electric furnaces is not a problem. If player has enough oil, they won't mind using BOP, if they are short on oil, AOP can help them solve that problem.

There are two issues with this too. First, a problem of degree, in that electric furnaces are only "better" than steel ones by their module slots, which require further investment to utilize, and arguably their fuel source, though this could go either way depending on the player. All stacked up, the difference isn't that big. The difference in efficiency between basic and advanced oil, though, is over 50%. Therefore, there is a lot more pressure to switch to AOP than there is to switch to electric furnaces. On top of that, the second factor is that AOP must be built for progression, while electric furnaces don't have to be. This means players are forced to go through the trouble of setting up AOP, whether they want to or not for efficiency reasons, and at that point the BOP setup just starts to look inefficient by comparison. The decision is effectively made for the player by the game.

[Light]

I have to agree with posila on that one. Electric furnaces have many benefits which entice you to switch which does in fact require you to redo your entire smelting line.

But... It's not essential unless you're going for a high science pack count or megafactory. Steel furnaces actually go a long way in a vanilla game, which with abundant coal sources nearby don't make the switch a priority during standard play. That doesn't make the point incorrect though, just more specific to the players preference.

[Adamo]

Yes, in the context of free magic solar panel energy, you are correct. But what you are saying is that the electric furnace's "lack of superiority over steel furnaces" is only solved by magical solar power. Nobody in their right mind and with knowledge about energy thinks that solar panels are reasonably balanced. Now, the other argument is nuclear power. I haven't computed if, by using nuclear fuel, the steel furnace still keeps up with the electric furnace in terms of energy efficiency. I'll check it out.

Edit:

If we simplify it down to the key factors: steel furnaces use half as much energy (unbeaconed), and that ratio only improves toward steel furnaces with modules on the electric furnace, so we use the worst case of the steel furnace being twice as efficient. The electric network is considered to be 100% efficient. We consider U-238 to be a waste product, so don't worry about its number in any recipes, and the solid fuel is accountable for 10% of the energy in the nuclear fuel. This results in U-238 having an 8 GJ energy value as an electric producer and about 900MJ as a chemical energy producer, so we'll call that 1/8. So, clearly, the steel furnace is not viable here, having just under 1/4 the energy efficiency of the electric furnace, unless you really need to burn off some of your fuel, since while you would have worse power efficiency using modules once you're up to +400% power usage, at this point you have clearly made the choice to use more power for more productivity. So I think nuclear power supports your argument.

One key difference, here, though, is that the efficiency is "in the background", not part of the recipe. New players wouldn't typically be aware that their energy is being used more and less efficiently here or there, and they're certainly not aware of modules. For new players, the electric network either works or doesn't work, and most don't really know why. They are studying the recipes, though, which are right there in their inventory. This makes concepts in this realm very different from abstracted things like energy efficiency, which seem solved to a new player once "things go in, thing comes outs". So, the way the player experiences "efficiency" in terms of recipe ingredients and results is different. And come to think of it, we're really not giving them any tools to know which ones are better. We have the numbers, but, most people don't know how to just use those right off. My friend, in jest, but there's some seriousness to it, suggested I put a big red "2X" over the light fuel oil recipe icon.

[Theikkru]

If we simplify it down to the key factors: steel furnaces use half as much energy (unbeaconed), and that ratio only improves toward steel furnaces with modules on the electric furnace, so we use the worst case of the steel furnace being twice as efficient. The electric network is considered to be 100% efficient. We consider U-238 to be a waste product, so don't worry about its number in any recipes, and the solid fuel is accountable for 10% of the energy in the nuclear fuel. This results in U-238 having an 8 GJ energy value as an electric producer and about 900MW as a chemical energy producer, so we'll call that 1/8. So, clearly, the steel furnace is not viable here, having just under 1/4 the energy efficiency of the electric furnace, unless you really need to burn off some of your fuel, since while you would have worse power efficiency using modules once you're up to +400% power usage, at this point you have clearly made the choice to use more power for more productivity.

I'm not getting what you're saying here...I see you say you're disregarding U-238, so are you comparing U-235? I understand U-235 counting for 8GJ in a reactor, but what is the 900MW number? It's not the right units.

[Adamo]

I'm not getting what you're saying here...I see you say you're disregarding U-238, so are you comparing U-235? I understand U-235 counting for 8GJ in a reactor, but what is the 900MW number? It's not the right units.

1.21GJ - 120MJ from the solid fuel that went into it equals ~900MJ. Just wrote W instead of J by mistake. Will edit. I could have made it even lower if I had included the light fuel's extra 10% of value.

[Reika]

Maybe turn up the emissions on that one. (Yes, that's an option parallel to fuel_value, you probably already know that). Sulfur dioxide (the product of burning sulfur) is nasty stuff.

Oh, I did. Eightfold. Should I triple that further?

Here is the mod, by the way. https://mods.factorio.com/mod/TurmOil

It is still very beta - I just released it an hour or two ago, and coded it a couple hours prior to that - but I have tested it and it seems to work.

There's another thing companies do to get rid of their excess sulfur that isn't so polluting (because in real life all of these processes leave us with *far* too much sulfur than we need): make concrete out of it.

What ingredient does the sulfur replace? Or it is just added, making..."sulfonated concrete"?

[Theikkru]

TurmOil

XD

[Reika]

TurmOil

XD

I like punny names, as can be seen with some of my others: Oreverhaul, Factor-I/O, NauvisDay, TRe(e)Plant....

[Adamo]

Oh, I did. Eightfold. Should I triple that further?

I agree with that number at least until you know for sure it isn't already too high for gameplay. Incidentally, I balanced my gas-fired boiler the same way... if you use the natural gas or methane, it pollutes half as much as burning petroleum products.

What ingredient does the sulfur replace? Or it is just added, making..."sulfonated concrete"?

Sulfur concrete doesn't need cement! Or, rather, it turns out, when mixed with stone, sulfur has cement-like properties. It's mixed with stone and sand aggregate directly. It cures quickly, and it has great acid resistance, so, in modern industry, it is used when fast curing or acid resistance is needed. I think it has a sort of orange-yellow color, and sometimes dendrites of darker colors.

[Reika]

I agree with that number at least until you know for sure it isn't already too high.

You agree with eight or 24?

Sulfur concrete doesn't need cement! Or, rather, it turns out, when mixed with stone, sulfur has cement-like properties. It's mixed with stone and sand aggregate directly. It cures quickly, and it has great acid resistance, so, in modern industry, it is used when fast curing or acid resistance is needed. I think it has a sort of orange-yellow color, and sometimes dendrites of darker colors.

So - in Factorio recipe terms - it replaces some or all of the stone or iron ore then.

[Adamo]

I'd stick to 8 and go higher only after testing, but going higher wouldn't be UNREALISTIC, just worried about game mechanics when I start approaching multiplying anything by 10 or more.

So - in Factorio recipe terms - it replaces the stone then.

Yes, and sulfur concrete doesn't even take water. Oh, but it needs to be heated to 140C. I think I had trouble making furnaces take multiple inputs, which is why I had to make a "kiln" building.

[Reika]

I'd stick to 8 and go higher only after testing, but going higher wouldn't be UNREALISTIC, just worried about game mechanics when I start approaching multiplying anything by 10 or more.

Well, considering NauvisDay adds a 96x multiplier to refinery pollution (and more if you go to Bob Mk2/3/4)...

I think I had trouble making furnaces take multiple inputs, which is why I had to make a "kiln" building.

Furnaces are hardcoded to one input; I had the same issue implementing my steam-powered furnaces. That was back in July 2017, but I doubt things have changed.

[Adamo]

Well, considering NauvisDay adds a 96x multiplier to refinery pollution (and more if you go to Bob Mk2/3/4)...

I wrote up a quick math for this, since you have tested numbers that go plenty high -- but remind me again what your exact plan is? Just burning sulfur directly? Also note I added more info in a previous message.

[Reika]

Well, considering NauvisDay adds a 96x multiplier to refinery pollution (and more if you go to Bob Mk2/3/4)...

I wrote up a quick math for this, since you have tested numbers that go plenty high -- but remind me again what your exact plan is? Just burning sulfur directly?

Yes, primarily as a way to non-cheatily void it.

[Adamo]

Yes, primarily as a way to non-cheatily void it.

OK.

If you can operate under existing test solutions then let's think about it for a moment. Sulfur content in fuels is anywhere around 1-5%, but is essentially the worst part of the pollution, so I'd say we call it that sulfur accounts for 30% of the pollution in the existing solid fuel. Let's say sulfur accounts for 2.5% of the fuel in the game.

This means we would need to multiply 30% of the pollution by 40 to get the new pollution from just the sulfur, or, 12 times.

What I'm not sure about is what the fuel_value should be.

[Reika]

This means we would need to multiply 30% of the pollution by 40 to get the new pollution from just the sulfur, or, 12 times.

Done.

What I'm not sure about is what the fuel_value should be.

I went with 600kJ (relative to coal's 4MJ).

[Adamo]

I went with 600kJ (relative to coal's 4MJ).

I think that's a good guess, if it was a guess. If it's from a source, please share, because I couldn't find one. I'm glad you went very low with it. I think that's right. That's less than wood, right?