Design of Extreme Industry Sets - An Analysis of FIRS and XIS
What does FIRS do right, that XIS does wrong? And how do I use these design principles in my own industry set?
When you're moving freight at all in OpenTTD, the industry set is what truly gives your playthrough identity. There are many different industry sets available. I'm personally interested in the design of extreme industry sets, as I am currently working on my own. "Extreme" here means having a very large amount of cargoes and industries. The most popular "extreme" industry sets are XIS and FIRS 4. In essence, FIRS is multiple industry sets at once, and the only ones that I would classify as extreme are Steeltown and In A Hot Country. XIS is built upon the code of FIRS, thus inheriting a lot of features from FIRS, and also making it a lot easier to compare the two. XIS is generally considered worse than FIRS, supposedly all over the place, whereas FIRS is generally considered to be well designed. In this essay I will be looking into what FIRS and XIS do differently, and how to learn from this.
XIS is often cited as lacking a general direction and theme. It inherited this from FIRS 3 Extreme, which was being deprecated for that reason around the time of the development of XIS. The benefit of this is that it allows many different parallel chains to coexist. FIRS 4 Steeltown is widely considered a golden standard because of its extreme directionality; everything points towards the ultimate goal of producing vehicles.
As I said before, I am currently working on my own extreme industry set, the DPRK Industry Set v2. When it's finished I want all the industries to lead to a single ultimate goal, much like Steeltown. In the process I want to make it so you can tie up loose ends and have many smaller chains. That way you have somewhere for your steel to go before you connect up all the industries needed to produce vehicles. I want to use this analysis to critically examine my own, and other people's industry sets, and see how I can improve my own design, and what I should avoid. I hope it will also be useful to others looking to make their own industry sets.
I heard someone say that XIS is the prime example for "less is more". This raises today's question: is less really more? I'm sceptical.
The numbers
Code: Select all
Industry Set| XIS| Steeltown | In A Hot Country----------------------------------------+-------+-----------+------------------Cargoes| 62| 48 | 29Industries| 82| 30 | 32XIS is clearly far bigger than FIRS. It's more than twice as big as In A Hot Country, and it even dwarfs Steeltown. Interesting to note here is how Steeltown has slightly fewer industries than In A Hot Country, while having one and a half times as many cargoes.
OpenTTD has a cargo limit set at 64 different cargoes, so XIS has space for only 2 more cargoes. When designing a large industry set, this limit can really creep up on you, so it's important to properly deal with this limit.
If all you want is to have as many trains as possible, then we can clearly see that XIS is the winner here. Let's take a closer look at XIS.
Everything goes to everywhere
So I started up a quick game with XIS. Vehicles seem like a logical goal to aim for, as many different industries come together to eventually produce vehicles. To make vehicles, we need to supply the assembly plant with vehicle bodies and parts. Vehicle bodies require steel, which requires pig iron. I can obtain pig iron by supplying the blast furnace with iron ore, limestone and co*ke. co*ke is provided by the co*ke oven if you supply coal. A decent chain, as can be expected of XIS. So I build my tracks and I send my iron ore train to the blast furnace and prepare the trains for the limestone quarry, and then I see that iron ore alone was enough. This was quite disappointing to be sure. At first I thought this was just because iron ore was the core component, but even providing limestone was enough.
As it turns out, all FIRSids have this problem where you can just supply any of the required cargoes and the secondary industry will go brr just fine. I personally find that somewhat problematic. After all having everything all come together at the end is the beauty of things like that.
Another problem XIS has is the parts cargo. I've been informed since starting this essay that this exists because the original FIRS code is somewhat difficult to work with, which is why XIS has only 62 cargoes. Regardless, this parts cargo kinda ruins XIS in some fundamental ways. You can obtain vehicles by simply sending wool from a farm to a textile mill, and then sending the parts you get from that to an assembly plant. 6 very different types of parts; textiles, glass, nuts & bolts, tyres, electrical components and plastics, go to 6 very different secondary industries; factories for furniture, books, vehicles, vehicles again, goods and a port. This can make it somewhat difficult to figure out what to do with those parts. Because if everything goes everywhere, then where does it go? This is the thing I like least about XIS.
To fix this, XIS would preferably have to get rid of some industries, cargoes or links. XIS isn't necessarily beyond saving, but a lot would need to change to fix this problem.
I wouldn't say this is necessarily an example of less is more, but rather an example of how technical limitations might force us to make decisions that in hindsight might not have been the best decisions.
So what about FIRS? How does it improve on this?
What’s this feeling of warmth?
Second in the list is the smaller of the three; In A Hot Country. Unlike Steeltown and XIS, it doesn't have a cargo like vehicles that would require a large chain, so I just started wherever. Just like in the base game, you just combine whatever industry you fancy. It's quite relaxed, and reminds me a bunch of toyland, which in my opinion is the best economy out of the 4 base game economies.
It does seem rather directionless without any end goal to work towards; you don't really feel like you're getting closer to something, but the chains are all fairly small, so you can easily chain together a bunch of industries, which is rather satisfying. On top of this, In A Hot Country is rather interconnected, so adding a new industry to your network feels rather satisfying. This gives In A Hot Country a rather fuzzy feeling
Like all FIRSids, in In A Hot Country, only one of the cargotypes is required for a secondary industry to start producing, and it's not very picky. This means chemicals alone are enough to produce copper. Sure by supplying everything you get more, but I kinda dislike doing things this way, and it's not really intuitive.
In A Hot Country also has a bunch of cargoes that go straight into the trading post. This trading post produces supplies whether you chuck diamonds into it or not, but it is somewhat of a shame, as it does make for rather boring gameplay.
Like a magic
I was really looking forward to Steeltown until I learned that you can just send sand to get turned into engines which you can then send to an assembly plant to make vehicles. Just take a pile of sand, chuck a license plate on it and boom; it's a vehicle. The pioneers used to ride these babies for miles. It is very disappointing, as I had imagined it was a lot like factorio, where you eventually get to your final industry where you assemble everything you've worked towards, and then you win.
But sadly, because FIRS requires players to supply any, and not all cargoes, the excitement of getting to work towards a goal kinda just dies. If you ignore this issue, and impose the requirement onto yourself, then Steeltown provides a nice and deep experience, which is quite enjoyable.
Another thing Steeltown does is keep related industries close together. With the large interconnected web of industries in Steeltown, you get to condense the chains somewhat, and this also makes it a bunch easier to get started. I think this is a fairly nice addition, which helps tie everything together.
The thing I dislike most about FIRS is magical ports. In A Hot Country also has these, but in Steeltown it is so much worse. These magical ports just magically cause cargoes to appear out of thin air without you having to do anything to obtain them. In Steeltown there are 2 magical ports, each producing 5 different cargotypes. 7 cargoes in Steeltown exist purely to be sunk into these ports, and 8 cargoes can exclusively be obtained through the magicians that are harboured by the ports. I like the idea of being able to export some cargoes that you otherwise wouldn't use, in order to get new ones, but that's not how this works. It just appears out of nowhere, and you can supply stuff to make it appear out of nowhere faster.
What is the purpose of these magic ports then, and why does Andy like them so much? Probably because Andy prefers smaller maps, and this way, he didn't have to add a bunch more industries to the map, which already quite packed. Alternatively he's just trying to avoid making extra sprites. Reasonable tbf.
XIS also has these ports, but they're never the only way to obtain these cargoes, so it is a very welcome improvement. This way you don't end up like potash, which exists solely to be chucked into the sea. At least it isn't as boring as the power plant in the base game.
Is less more?
Steeltown and In A Hot Country have roughly the same amount of industries, although Steeltown has many more cargoes. Discounting the magic port cargoes puts them both at roughly the same size either way. Regardless, Steeltown manages to make the larger amount of cargoes work really quite well by weaving the entire economy together in one large web, which gives you a much more coherent network, and as such is a bunch more satisfying.
Meanwhile XIS is truly too big for its own good. The omnicargo "parts" means it's not ideal for a game where you just want to move cargo from A to B, and it performs much better as an industry set for eyecandy games. I am currently playing a game with XIS that is all about eyecandy, and I must say that XIS is just fantastic for it.
From this all I can say is that if you're mindful of your limitations, more is simply more. In A Hot Country shows that it doesn't have to be one interconnected whole to be fun, while Steeltown really highlights what that can do to a game.
I believe that industry sets larger than Steeltown could definitely work, but you'd need to work with your limitations, and not against it.
How can I do better?
The first version of the DPRK Industry set v2 won't be much bigger than v1, but I do have ambitions to make it much larger than even Steeltown. My current cargo flow draft has 55 cargoes and 61 industries. Many cargoes are used in both much larger and much shorter chains. With this I hope to be able to make it satisfying to add small parts to your network while you're working towards a bigger goal, and I'm hoping to make it so at the end everything will all come together, when all the pieces will fall into place.
Another benefit of doing it like this is that it allows me to release updates in smaller parts, without having to rip out some older connections. I can also make it so the player can have a smaller set of industries if they prefer.
Something Steeltown teaches me is that I might do well to consider making it so some similar industries spawn close to each other. You wouldn't want to have to drag 55 different cargoes halfway across the map.
One thing I do realise now is that if there's too much going on at a single industry, it might be nice to split it up into multiple smaller industries, but besides that I'm not seeing any obvious flaws. The main issues I had with FIRS and XIS were things I was already going to do differently. Perhaps this made it so I couldn't properly assess these industry sets, but who knows. I suppose the only way to see if my ideas are truly better is to put it all together and find out.
So what do you think?
No pics no clicks. Seriously.
