Factorio Forums

It seems this issue has already been discussed, but mainly from the accumulator side rather than the solar panel side. In short, when you discover solar energy, energy becomes a no-brainer : no pollution, no resource consumption, no upkeep, just some land is required.

I've thought of two suggestions that, together, should maker solar panels a nice and useful addition but less easy to completely replace steam machines.

First, decrease solar panel output depending on local pollution (think of smog if you want a real-life inspiration ). This means that building solar panels near your base will be highly inefficient, though on the opposite this can also be viewed as an incitation to run a low-pollution factory.

However, the counter seems obvious : build your solar panel farm far away from your base. Here comes the second part : make pylons an active target for aliens, similar to turrets and the like. This means that managing this kind of remote power supply will be very annoying since aliens will regularly disconnect it from your base. So, if you want solar panels, you'll have to build them near your base, though they will be rather inefficient and likely require keeping some steam as backup and some surprises as your pollution grows if you try to rely only on solar panels.

But what about remote mining outposts? Those will be hell to manage without pylons! Well, not if you precisely use local solar panels to supply those outposts: they would be ideal for that since they are very easy to place, there isn't as much pollution there compared to your base, you won't need to many of them to provide power to a few extractors and inserters and the outpost is mostly fine with some occasional power interruption (unless you want to use laser turrets, of course, but that's why you have gun turrets).

So, solar panels would be less powerful to supply your main base, but gain an essential new purpose of decentralized power supply - incidentally also one of the most underrated (to the public; companies are well aware of it) usefulness of solar energy in real life.

I don't favor making power poles the target of alien attacks since it could have quite a lot in the way of unintended consequences. Lines to mining outposts need to be defended where before they could be made reliable for example.

Another solution is based on something from another discussion in How bright is this planet. Namely that solar panels produce a ludicrous amount of power. Indeed if factorio took place on an earth-like planet they' would need to be about 500% efficient. Conversely for them to be 50% efficient (better than anything we have) that planet would need to receive ten times as much sunlight as ours does.

Likewise, in the same vein it takes surprisingly little area to power your base with solar panels. Certainly it needs quite a bit, but a base can be powered by a structure roughly it's own size, or smaller. And, frankly, the amount of steam infrastructure needed for a similar power output is not much smaller.

So, what could bring solar panels down to earth is if they were made to require land in vast quantities. Let's say that solar panels now require 90 squares for a for 60 kW peak production would before have required 1 9m^2 leave solar panel.

We can easily keep material efficiency the same either by making solar panels come in runs of 10 (which I would recommend) or by cutting the costs by a fifth, to 1 steel, 1 copper and 3 circuits, and providing two solar panels per each run.

This way to power your base you require a significant safe zone in which to put all those panels, we must be first cleared and then kept clear.

The Phoenixian wrote:I don't favor making power poles the target of alien attacks since it could have quite a lot in the way of unintended consequences. Lines to mining outposts need to be defended where before they could be made reliable for example.

That's precisely an intended consequence: you will no longer build huge power lines up to the outposts, but instead you'll rely on locally generated power in the outpost. More interesting than just generating all your power in one place for everywhere, and several ways to do it: use steam power with coal from the outpost ? Bring coal or solid fuel by train ? If you have no access to water, build some solar panels and accumulators ? Or maybe just solar panels would be sufficient, outpost mining doesn't necessarily require 'round the clock power.

In addition to making solar power a bit less efficient and reliable (since it will decrease or increase based on your factory's pollution), that gives them a new purpose to supply outposts - well, it's already possible to do that, but it's building poles everywhere is so simple and no-brainer (it's always the same thing, no need to think about power demand or variation, just connect everything).

Just decreasing solar output might be relevant from a "realistic" point of view, but gameplay-wise it won't change that much (apart from making it more repetitive), you'll just clear a bit more land and put more solar panels, still the same almighty solution with reliable, infinite, pollution-free power that removes any challenge or thinking to power management.

I am absolutely against this suggestion, sorry.

You are suggesting to introduce an annoyance (as you say yourself), to discourage a playstyle you personally deem unbalanced. Yes, solar power is rather easy to handle. It is, however, challenging to get there and to automate its production so you actually use it on a grand scale. The game's mechanic here is not to provide a challenge to setup your power network, but to provide you with the tools to produce power and focus on other parts of your growing factory.

Also, I find your idea flawed:
first you say that players will be discouraged to build solar panels away from their base, due to the pylon attacking thing and thus you conclude that players indeed will build close. Now with pollution reducing the effectiveness, this just means I need to build more. This is just boring and doesn't add any depth to the gameplay. You might as well save the pollution-dependence and make panels less effective then.
Furthermore, you suggest powering outposts with local solar panels. I am not sure how this makes them any less efficient. But I do want to say that in my games, the mining drills are one of the biggest pollution drivers, so outposts would cloud their own panels.
Also, I need outposts to run all the time or I would not be able to satisfy the resource demand of my base.

So I still ask myself:
1) How would improve on the balancing other than requiring players to just build more and do more micro-managing on the outposts (which actually discourages the outpost model entirely)
2) How does it improve the fun the game creates? I fail to see the benefit

I dont use solar almost at all, in part because i dont like the cheatyness of them, in part cause i like the task of feeding boilers.
and i DONT like this suggestion. it would force me to use solar panels in my outposts, because i could no longer use pylons along my rails to the outposts, and theres no water for steam near some of the outposts at all.

I do kind of like the phoenixian's idea, but i think in this case we'd still need something about the solar panels to agravate biters. otherwise, it'll just make the fields bigger, and not change anything else.
But in the end, i'll probably only really start using solar, when there is some sort of maintanance/cost associated with them, one that can be automated, but one that is continuous. be it repair packs flown in by robots (solar panels deteriorating over time) or some sort of material that needs to be fed into accumulators to store energy for night time. I think no matter what you do to make solar panels less attractive, as long as they provide free energy for a one time investment, they will be cheaty and worth it in the end (for a base that lives long enough).

The heart of the solar energy problem, I think, is that in a game like Factorio, any new technology that makes your old problems simple to solve should do so in a way that introduces entirely new problems to tackle. And solar panels don't introduce many interesting challenges IMO. Thus, while needing maintenance is an option, it's not the one I favor as I think you can make solar energy work as a "free" energy source that relies entirely on infrastructure, you just have to make setting up said infrastructure an interesting problem. But in the end --- be it by maintenance, clearing all that land, or setting up infrastructure --- you do need an interesting problem, and that's what I think solar lacks.

Still, off the top of my head, there is a way to have a continuous material cost that might be okay. It requires a pretty big change though. On the plus side, it does feed boilers.

Pulling once again from reality, (it's not that reality is better, just that it provides good inspiration) there are several types of solar. Factorio obviously uses a form of Photovoltaic, but we can also consider Solar-thermal. It's been suggested before (might've been by me actually, but I think I just helped a bit after it was brought up) but the gist of it is simple:

Solar panels don't produce electricity directly. Instead, each panel contains a pipe and works by heating up the fluid in that pipe. That fluid can then be used to power steam engines.

Now that alone just makes you go and find a new way to boil water but the other thing you can do is have a type of heating oil that gets much much hotter than water can on its own. (either as an upgrade, or exclusively) You just need a heat exchanger to turn that into usable power.

By itself that already imposes a few infrastructure requirements on Solar power: You need a chemical plant to set it up, and you also need a nearby steam turbine setup which means nearby water too. There's a few interesting questions too: Do you set up a storage system to hold hot water or oil for nighttime? Can you get a countercurrent exchange setup to use more of that heat? What's the most efficient chain the panels together? And so on. Lastly, if you want, you can still make it less efficient than it is now so you need both a nearby lake, and a very large flat region. Which itself brings up the need for an efficient pumping setup.

Where the maintenance angle comes in is another kind of inefficiency if it's desired after all that setup. Every operation in the heat exchanger can lose a little bit of that oil. So if for every 100 units of oil run through at 500 °C to turn water into steam, if only 99.9 units of oil come out of the process, the system will gradually lose it's oil and will occasionally need to be replenished.

Of course, all in all, Solar-thermal is a pretty massive change and there may be better ways, but I think it's worth bringing up.

This type of discussion is comming every 2-3 month. My answer is simple: Change the solar input in a way, that it is more random.
In the moment you do that Solar Panels are no no-brainer anymore. You need to calculate: Will the next solar eclipse be so big, that I come through with my solar power?

See
https://forums.factorio.com/forum/vie ... f=80&t=183
https://forums.factorio.com/forum/vie ... f=6&t=6181

If solar panels power output is decreased in any way then a new way of creating power should be made to accommodate megabases. Setting up solar panels already takes tens of hours when building megabases and i really don't think there is any reason to make it an even bigger annoyance to setup a power supply. Right now to setup a solar farm you have to do the following things:

• Find a large land area which is usually far away from your base
• Kill all biters -- Expensive and time consuming
• Remove all trees and other obstacles -- Expensive and time consuming
• Setup a train line and station to supply solar panels to the now cleared land -- Expensive and time consuming
• Place solar panels -- Time consuming
• Place accumulators -- Time consuming

Even with blueprints/roboports/grenades/posion capsules/destroyer capsules, it still takes a ton of time to it all and when you need more power then you have to do it all over again.

Any kind of maintenace cost will severely discourace you from using more power than absolutely needed which will discourage you from using productivity modules, speed modules, roboports and beacons because it will just drive up the maintenace cost way too much. If the maintenance cost doesn't encourage that then i really don't see the point behind it and would just se it as an annoyance that has no purpose.

tldr: In my opinion solar panels are not overpowered as they use up way more space than steam engines does and you have to use the above steps to clear it which more than pays for the maintenace free power you get afterwards.

@keyboardhack: I believe that your statements are very true for megabases. Only for megabases though and I would vote not to balance for megabases. If any balancing makes creating megabases annoying, this can easily be fixed by a mod.

@ssilk: I still do not think that this is a good solution. Making things less reliable (e.g., random power generation) will just force players to build more. You basically calculate with the minimum. It's the same as increasing the cost.

@The Phoenixian: splendid analysis and suggestion. I think the problem has never been described in such a good, concise way with a clear angle from game design and not emotions. Can be raise this to the attention of the devs?
Of course, with solar-thermal, steam engines would have to stop producing pollution (I never understood why they produce pollution anyway...).

SirRichie wrote:@ssilk: I still do not think that this is a good solution. Making things less reliable (e.g., random power generation) will just force players to build more. You basically calculate with the minimum. It's the same as increasing the cost.

You don't know the minimum, that's the point! (*)

Consequence: All those threads about "What is the right amount of solar panels vs. accumulator-ratio?" are then useless.

The sense of this is also to make other forms of power more attractive (wind, nuclear...), cause/but they have other dependencies.

So as a player you have a choice: You have a very reliable power generation, with extreme costs, cause - if you want to make it safe - you need to produces 4-8 times more energy than yet (and/or store 4-8 times more) and that still doesn't make it sure that you will come through with it, or - also a good choice - not so reliable power generation, but either a good defence, that is able to hold down enemies for a while, until you get power again. Or - much better! - intelligent power management: You switch off the parts of the factory, that are not needed - with 0.13 we will have that new element.

It's then a play with chances: Will my current power generation break down and how bad will it come then or can I come through with this intelligent - but minimal - solution?

[I don't know how it is with others, but I like such games, cause it's not just luck, it's playing with probabilities and possibilities. You come into some kind of operative hectic, which is quite inspiring, cause you find then the best solutions.]

What I'm really going with this whole discussion is, that the current handling of energy is in the long term boring, after you get solar. Steam engines are much more appealing.


(*) It can be that there is an eclipse like taking some seconds. Some minutes. Or a whole day. Or two. An game-hour? That can happen, when the planet is going around a gas-gigant. Also for the current type of wind: I think it should change much, much slower. It can be, that there is a day long no wind. And for nuclear energy: There will be no guarantee, that you find uranium nearby. Such things will influence the handling of power strongly.

ssilk wrote: What I'm really going with this whole discussion is, that the current handling of energy is in the long term boring, after you get solar. Steam engines are much more appealing.

And here is where the crucial point. It is boring in a low-maintenance sense. For a game about automation, I do not see a problem here, but as said, I like The Phoenixian's suggestions.

Also, I still find randomized power generation not that of a good idea. Either it is just a way of making solar/accu more expensive as I outlined. Or, if you introduce heavy randomization, as you just suggested, then it becomes a game of chance. This could very well mean that my 30+ hours factory is laid to waste because I was unlucky with the power generation. I think that would be flawed game design. Of course you can say that this is the price for solar, but then why have it at all?

Also, I still find randomized power generation not that of a good idea. Either it is just a way of making solar/accu more expensive as I outlined

That's not exactly true. In the grand scheme of things a single cloudy day per week is not a huge loss of energy. However it makes a HUGE difference in the viability of night time energy storage.

Solar panels and accumulators work because the factors are always known. Once you get the ratios down everything works without a hitch. As soon as you throw a curveball into the mix, you don't know what you need anymore. What if there's one bad solar day? You suddenly need FIVE TIMES the accumulators to cover the gap. Two days in a row? You need too many to count. On a good day those excess accumulators aren't doing anything useful, making them a ridiculously expensive way to brute force the solution. Eventually loading up so many "emergency" accumulators gets so wasteful that the solar/accu strategy falls apart, it simply can not be sustained. You NEED a backup plan, which means you need to automate a real solution. That's another puzzle to solve, and puzzles are a good thing(tm) for factorio.

What if a player just wants to shut off their factory on bad days? That's certainly a possible solution. What if they want to switch to coal backup? That's another solution. What if they want to run seperate grids for low and high priority energy networks? When solar isn't a guarantee, you suddenly have plenty of things to think about.

Solar energy dependent on pollution

This has kind of nasty repercussions. When you make more pollution you get less solar energy, which means you need more coal power, which makes more pollution, which makes it harder to break the cycle.

bobucles wrote: What if a player just wants to shut off their factory on bad days? That's certainly a possible solution. What if they want to switch to coal backup? That's another solution. What if they want to run seperate grids for low and high priority energy networks? When solar isn't a guarantee, you suddenly have plenty of things to think about.

I very much agree that this makes for interesting puzzles. I am, however, concerned that this may very well take a different direction, for two reasons. One: if solar causes so many things to deal with, forces me to come up with complex construction setups, I might very well just bruteforce the solution. Yes, 5 times the accumulators is expensive, but in the grand scheme of things, still not that much to deal with, if I get free, easy power for it. Two: If I want enough backup power to sustain my factory even when my accumulators have run dry, I can simply ditch solar completely. Sure, in the times to sun, I will not pollute the planet that much, but let's face it: if you have a large factory, pollution is rarely a concern.

SirRichie wrote:

bobucles wrote: What if a player just wants to shut off their factory on bad days? That's certainly a possible solution. What if they want to switch to coal backup? That's another solution. What if they want to run seperate grids for low and high priority energy networks? When solar isn't a guarantee, you suddenly have plenty of things to think about.

I very much agree that this makes for interesting puzzles. I am, however, concerned that this may very well take a different direction, for two reasons. One: if solar causes so many things to deal with, forces me to come up with complex construction setups, I might very well just bruteforce the solution. Yes, 5 times the accumulators is expensive, but in the grand scheme of things, still not that much to deal with, if I get free, easy power for it. Two: If I want enough backup power to sustain my factory even when my accumulators have run dry, I can simply ditch solar completely. Sure, in the times to sun, I will not pollute the planet that much, but let's face it: if you have a large factory, pollution is rarely a concern.

I've been struggling to put my finger on what my problem was with eclipses for a while now but I think this here encapsulates it:

With an Eclipse blocking light for a period of time, solar is no longer a core power supply. Rather, solar and accumulators instead become a power supplement. In theory, you can run a base off of of only solar power. In practice, this requires either a massive expense, or a very complex solution. Moreover, while requiring a complex solution is a worthy goal, in this case it's not a complex solution in order for advancement but instead that complex solution is still consigning yourself to having a worse base because of it. (via turning parts of it off.)

Now I'm fine with solar being massively expensive, I'm fine with it being massively complex. I'm fine with it being both. The problem is that the effort you put in should match what you get out and, while solar as it stands is too little effort for too much gained, this goes too far in the opposite direction.

If you do a lot of work, you should receive a significant reward. This is especially true if all that work is done within the framework of the game's values: That of automation and complexity.

Which isn't to say that any sort of light variance is bad, but I don't think eclipses are the way to do it.

Good points.

There are two things, that should be considered:
a) You have more possibilities then. Which is always good. More possibilities without introducing more complexity (there is just this little randomnes) into the game is in my eyes always a good way to go. The rest can be done as the player wants to solve it. He can make a very complex scenario with mixed power and swiitches etc. or just a simple steam engine setup.
Or you can see it also differently: Now, after research of solar/accu, one of the best way to solve the game is to go into that direction. Then - by introducing this simple randomness - this isn't so clear anymore, because it depends on so many factors, that nobody can say: "This is the right way".
Which is good, cause there should be no "This is the right way to play Factorio".

b) Other parts of Factorio will be also developed. For example the blueprints. For 0.13 there is a blueprint-book implemented. I admit that it is completely unknown to me, how bpb's in 0.13 will work, but I'm sure, that we then will have soon blueprint-books. And blueprints around "energy". Complex circuits for example that are able to switch your factory partly off. Simple setups for solar-fields. Solar-panel production. Etc. All that makes the "work" easier. A lot easier. Of course you can do that on your own, but if you don't want to, just download some bpb's (blue print books).
Or new mods: A "weather forcast combinator", that is able to look into the predictions for weather/eclipses etc. and turn on your steam power long before that, so that you come better through.

So you can decide to use such stuff, or not. You have nearly no disadvantage, if you don't use that but you have a lot more fun, if.

What's such a downside to having an energy star[:)] compliant base? You're just going to run every device on permanent hot standby? How will you expect to expand into CPU breaching territory if you won't cut CPU cycles where they count? It's practically a requirement for everything nowadays.

The problem is that the effort you put in should match what you get out and, while solar as it stands is too little effort for too much gained, this goes too far in the opposite direction.

Actually it only needs two components to make the setup REALLY EASY:

1) A circuit breaker to manage electrical networks. As long as it can respond to automated input it should be fine. The devs have actually shown art for a circuit breaker so it's definitely on the way.

2) An electrical sensor to detect when energy is misbehaving or high or low or whatever. A sensor to detect solar levels may supplement this, but is not required. Players have run clever setups to do this in an ad hoc way but an official sensor would still help.

That's all you NEED. When energy gets low, disconnect a factory circuit. Or connect a steam power circuit. Or just brute force it with more everything. Do whatever you want.

Weather events are definitely game changing enough to be a meaningful map option. With normal settings having a bad solar day 10-20% of the time isn't going to break solar. No. It won't. It's still a lot of solar. On a high setting having 75-90% shady days will absolutely devastate solar strategies.

Don't forget you can also have a "full moon" or "nearby nova" type of event or a "mirror satellite" that allows solar to generate during the night. Modders are clever, it doesn't have to be one way.

Other parts of Factorio will be also developed. For example the blueprints. For 0.13 there is a blueprint-book implemented. I admit that it is completely unknown to me, how bpb's in 0.13 will work, but I'm sure, that we then will have soon blueprint-books.

The simple option is to have a blueprint book act as a carryable "chest" for blueprint pages. That way a large number of blueprints only occupy one inventory slot, even if it has to be a special equipment slot.

The best option is to let players import and export blueprint pages from their hard drive. All that effort that went into building your favorite base can now be carried between games.

I still think any variant we come up with either is just a little nerf for solar, in which case the easy fix (from the player's perspective) is to build more.
Anything that cannot be solved by building some more (not excessively more) solar panels entirely breaks solar and renders it near useless.

Maybe The Phoenixians's suggestion from a few posts up is the only one which I've seen that makes solar more interesting, without introducing just more annoyance around it to discourage its use.

Personally, I think solar is fine as it is at the moment. It is a wonderful reward for advancing in the tech tree. Yes it is easy to use once you have it with nearly no drawbacks. But really, so what? If that makes the game to easy for you, just don't use solar.

The best solar solution is going to affect the net value of panels as little as possible, while achieving the maximum output in how players have to rethink their gameplay. I don't think that's an unfair statement to make. That being said, there have been multiple solutions brought up in this thread and earlier ones:

1) Change the raw data values of solar panels/accumulators.
This is a direct nerf to one of the components of the solar system. Either the components cost more, or they do less on a regular basis. This nerf is overcome in a direct 1:1 way by simply putting more resources into the solar system. In the end nothing changes except maybe how annoying it is.

1b) Cause pollution to reduce solar power
This is along a similar vein to a direct nerf. Players will try to sneak solar power way out to the fringes of their base. Productivity modules are super polluters so solar will become vastly more difficult in that regard. In the end the player pays a higher tax on their solar power and just liek builds moar in a 1:1 way.

2) Solar panels suffer decay/upkeep of some variety.
The energy you get from solar panels must be maintained in some way. In the long term, this means that iron and copper have to be constantly reinvested to generate energy. Basically, iron and copper become coal Mk. II.

If you thought finding coal for a gigawatt base was a problem, now try finding spare iron and copper to do the same! Productivity modules get screwed over big time because they turn energy into extra resources, buuuut you need to spend resources to get the extra energy. It's a vicious cycle. Players will have to discover a hundred new ways to shut down unused factory bits and conserve energy like never before, lest they burn all their material goods on solar power. Racing to get resources just to keep a megabase's lights on does not seem to me like the spirit of Factorio.

Expect players to dramatically cut down on solar use and go for the next best infinite energy source: fuel oil. There's no material upkeep on fuel oil, so should we aim to nerf that as well? Why should solar panels decay when nearly everything else has moving parts but never needs upkeep? Is that not an inconsistent message to deliver? Maybe the devs just hate renewable energy? This solution does bring up a bunch of weird things that will have to be addressed in some way.

3) Eclipse/weather obstructions causing intermittent system loss
Solar systems lose their 100% reliability. When the system goes down it becomes useless or nearly so until it is operational again. The actual numbers here are fungible, but the sweet spot is probably somewhere around a major solar loss 5-15% of the time.

Players here argue that you can simply build MOAR to overcome a complete system loss, and that is indeed probably true. But no number of solar panels can overcome a complete loss of the system. Some quick math determined that you need about 5 times as many accumulators to survive just a single day on battery power, and if there is an extra day it can be 10x or more. Solving a solar shortage is no longer a direct 1:1 solution to just liek build moar. Players who brute force the solution could instead be paying 5x, 10x, or more resources just to survive a 5-15% loss in overall solar power. That's an obscene solution and totally inappropriate for a serious base builder.

A more appropriate gameplay solution is to figure out how to survive the rare days that solar doesn't work. Maybe you switch over to steam. Maybe you cut off parts of the base. Maybe you turn everything off. It's all up to the player to figure out what to do. That's what we like to call a "puzzle", and one with a rather modest difficulty. As soon as the player solves it they move on to the next puzzle in Factorio.

Besides, you know it's going to be great when a new player gets rolled by biters during an eclipse. It will be a dumb mistake, albeit a very fair one as long as the game tools allow players to solve their energy issues.

Okay, a few points,


One: this is a bit of a nitpick but

bobucles wrote:The best solar solution is going to affect the net value of panels as little as possible, while achieving the maximum output in how players have to rethink their gameplay. I don't think that's an unfair statement to make.

The first half of this isn't really important, except as a thing that emerges from a dozen other factors. The value of solar panels can be changed a lot so long as,

1: The effort the developers put in is worth the engagement of the players out.
2: The value of solar panels as a solution is weighted against the engagement player has setting up solar panels, vs setting up any other power system.

So in short, I would say that the value of solar panels should be balanced against the difficulty of the puzzle they provide. ("Interestingness of the problem" in my previous words)


Two: This bit?

bobucles wrote: That's what we like to call a "puzzle", and one with a rather modest difficulty.

Is very condescending. For these forums at least. Seriously, if we didn't not only know what a puzzle is, but also decide for ourselves that Solar as it stands is a boring puzzle, we wouldn't be complaining about it the way that we are.



And lastly, but certainly not least, because you skipped my bit, I'll note that while I opened up solar-thermal as way to make a resource cost work without repair packs (by having a very small oil cost at the highest efficiencies instead), in actuality it falls under a very different category:

4) Increase the system complexity of solar.

Fairly simple: a more complex system give the player more principles to work with, more pieces to the puzzle as it were, and they think harder to set things up. Once things are set up however, everything is fairly simple to copy into a blueprint and paste.



So, allow me to go into this in a bit more depth:

Solar-roadwaysthermal. Step one: The basics.

The reason I proposed Solar Thermal, and the beautiful thing about it, is that its complexity scales with its efficiency: You can very easily have an offshore pump at one end, run pipe through a few dozen panels, run a steam engine off the heat, and call it a day. You'll run into problems with losing working time of course as the fluid needs to evaporate each morning, but the imaginative player has many ways to counter that, from simply adding backup boilers to, accumulators as we know them, to storing hot water produced by day, for use by night, though that last one will likely require turning off the flow at the right time. Something much easier with coal power.


Solar-thermal: Step two: Heat exchange.

But of course, this is only the beginning. With heat exchangers and heating oil the system is more complicated to set up, but once it works, it works perfectly. (or with minor heating oil loss over time, whichever works)

The oil and water pass through the heat exchangers through separate pipes separate lines and both fluids are allowed to flow freely, basically as a 3x1 or 3x3 boiler that uses no fuel, and as they remain in the same heat exchanger their temperatures slowly equalize. With this, two fluid cycles exist instead of one: one for oil gaining heat in solar panels, losing it in heat exchanges, and returning the the panels and the offshore pump --> heat source --> steam engine chain we all know and love.

The core idea with heating oil and heat exchangers however, is that in order to heat up a given amount of water, you only need a fraction as much oil, meaning that the oil travels slower. If your fluid is traveling a third as fast, it's spending three times as long solar panel, and thus each panel is three times as efficient.*

Now there are two ways to do this: Either the oil can get much hotter much easier, or it can have a higher specific heat than water. (read: the recipe for the heat exchanger only drops it one degree for every three degrees it heats up the water) I favor water having the higher specific heat and oil having the higher temperature, because means that a countercurrent exchange, while it is naturally more efficient, is not strictly necessary, instead being a neat trick players reward themselves for finding. Which is a theme here: This entire setup with solar-thermal is the game finding a dozen ways to say "Hey, you don't have to do this, and it is a bigger problem/harder puzzle/small extra thing to consider... but it's better." and teaching players to chase that voice.

The fact that oil getting hotter but having a lower specific heat is more realistic... is just icing on the cake really. That said, this specific heat stuff isn't actually necessary and if it's not worth changing the code, there are workarounds.

In any event, you've set up your solar plant with heat exchangers and heating oil and now you're doing great: You've got a setup where you've got a bunch of of solar panels setup to run, three times as efficient as before.


Solar-thermal: Step three: Optimization.

Finally, and this is the insidious part, eventually you're going to hit a peak: Pumping capacity.

With inline pumps, fluid can only travel so fast and so far. About 750 tiles for a single pump working at 30 units/second but that length decreases massively as you add more speed.

Now I haven't actually tested this, but from everything I know it makes sense that out in the reaches, you can just send oil through your pipes in parallel. Up back at the base and for longer steam engine lines, where your mixing all those fluid lines together at the central power hub, you need to pump differently. And thus the player is encouraged to learn the mechanics of high speed pumping.

In the end, the greater the throughput of the heating oil, the bigger the solar farm can be, and the more engines it can power. And once you hit a peak, or are happy with what you want, you just copy the pieces of the setup into your blueprints, and paste it down again when you need it anew.

Final notes:

All of this is based on a few very simple and knowable principles:

1: Solar panels and heat exchangers, like boilers, don't pump fluid but merely heat up or equalize the temperature of the fluid(s) as it remains within the object and the object has a source of energy.

2: Steam engines burn through a set amount of water per second to produce power. More steam engines need more hot water.

3: Fluids in Factorio have a speed limit, based on distance, which can be overcome via adding more pumps, more frequently.

So you go from step one! "Hey! Free boilers by daylight!" To step two: "Oil and heat exhangers triple the efficiency!" To step three: "Pumping mechanics help make my system bigger!"

So this is why I proposed solar-thermal: it's three new items, a solar boiler, a fluid that gets hotter than usual (and it can even just be light oil, heavy oil, or petroleum gas with some code changed), and a boiler that equalizes the temperatures of different fluids with a bias based on specific heat, but it gets this complex. And each machine is pretty simple and behaves with very consistent principles.

And it's still only ever as complex as you let it get. It's a patient puzzle that challenges the player in their own time.

Footnote:

*Why specifically three times as efficient? Three reasons: First and most importantly it's a big number that rewards the most complicated part of the setup process. Secondly, using oil as a working fluid needs to be at least twice as efficient as using water as a working fluid, in order to match the flow rate of offshore pumps to the flow rate of small pumps and make the third stage meaningful. Finally, in the real world water has a specific heat of ~4 and boils at 100 degrees Celsius. Oils have a specific heat of ~2 (a little more for petroleum based oils, a little less for vegetable oils) and the peak temperature of steam engine based solar thermal energy is around 600 degrees Celsius. So realism as a tool to inform gameplay.

Replacing solar electric with solar thermal could be a neat thing. There are some pressing issues however.

1) Closed liquid loops don't function very well. I've seen the nuclear mod attempt it and it's not pretty as the fluid system seems to need air bubbles to work properly. That's a major reason why steam engines consume their water in the first place, so players don't have to concern themselves with loops.

2) Lack of sensors to supplement the system. This system makes it very important that cold liquids get excluded from storage containers and heat exchangers. Otherwise the player is moving cold oil, which is pointless, or storing up cold water, which will devastate the system with every sundown.

3) Map starts with only starting water become completely invalidated. There is an absolute limit to the player's water supply, after which they can no longer generate energy using steam power, after which they reach total peak energy. They will need another energy source.

4) You can already store supplemental energy in the form of hot water tanks. It's relatively pointless because regular fuel boilers can cook on the fly just fine, and aren't a huge issue to have.

What if both systems exist? If the player still has electric solar, then they'll just like use that and not care about thermal solar. Sure there may be a price point where one thing excels over the other, so at the best it becomes a matter of personal choice. But if the player chooses electric, nothing changes.