I like production optimization games, what can I say?

Foundry is a voxel based production optimization game, a sort of weird hybrid of Minecraft and Factorio. It had a sneak peek demo of sorts last October, and I’ve been somewhat intrigued since. I like both of those games, and the dev company is a relatively small Canadian one, so why not give it a go.

At their core, this genre of game is Excel with graphics. The goals are about making numbers go up, which feed into other numbers, who also need to go up. And then somewhere in the middle realizing “what a sec, I need WAY more numbers” and truly appreciating the scale of the games.

One important part about Excel (or math in general) is that every new variable added can exponentially increase complexity. So the pace of added variables has to be one that’s controlled and mathematically sound. You can’t have one recipe need 5x and the next one need 100x. It’s also bad form to add so many variables and the inability to balance them – which is mostly found through a starter bus. In these games you need to produce everything (foundations, belts, smelters, etc..) and the number of variables is directly related to the ability to ingest. e.g. If a machine can only accept 2 materials, you can only have 2 materials.

So what does Foundry bring to the table?

• Voxels, which is a fancy word for grid-based exploration in 3D. Think Minecraft.
• Tons and tons of decorations, all from the start. Neat.
• Procedurally generated worlds. Every world is different, so there should be inherently less complex construction as the world is variable.
• Multi-player. While I haven’t done this, I can see tremendous value.
• Exhaustible resource nodes. You can’t just mine the same spot forever.
• Foundations to power everything. You need factory floors, which is a very annoying bit as your character can only jump 1 voxel.
• Loaders. These are initially limited to 1 space, then within an hour 2. The 3 space loader takes a while to unlock.
• Many machine types. Smelters, miners, assemblers, casters, pumps, etc…Many. Frankly, too many.
• Storage and the ability to limit. Used for buffering. This is great!
• Farming. Sort of. It takes more energy than it produces, not sure of the point yet.
• Research cores and stations. This is how you move along the tech tree to unlock more things to do.
• Low voltage and High voltage power. I dislike this a lot, more on it later.
• Subterranean mining. This is is interesting.
• Tasks. Not much for now, but an attempt at world-building.

Looking at the above, you can see a ton in common with Factorio, Techtonica and Dyson Sphere Program. There isn’t much in common with Satisfactory here.

Voxels – This is more related to world construction & destruction. You’re given a hand drill and you get to drill (or blow up) blocks to make open spaces. It’s also used to clear trees and rocks, that block foundations. In 99% of cases, your lack of vertical mobility means you need to go through things rather than above them (e.g. mountains). It’s interesting but finicky, in particular when laying down belts.

Procedurally Generated World – The map is infinite in size. “Seed hunting” make little sense. There’s no long-term travel options (e.g. trains or others), which means belts everywhere!

Foundations – You need a LOT. And when you think you have enough, you need more.

Loaders – They initially only grab 1 space away, then you get a 2 space loader which is essential for a main bus. The 3 space loader is a long ways out, and generally not needed.

Research Cores – I really like this model. You can self-direct the path of advancement. The research tree needs some serious QA work. One panel for crafting, one panel for mech, one panel for decorations would be ideal. Research is also a major bottleneck near mid-game, where the scale of later (tier 3+) items require a TON of material.

Machines – All machines have a config panel that is placed in such a way that if you put loaders in front, you cannot read it. They do what you think they do. It would be nice to have throughput data more obvious than 4pt text. Machine ratios can be a bit weird.

Low Voltage – High Voltage – This is how you power things. Most basic things need low voltage, which is simple enough to manage. You can also convert high voltage to low, which replaces early tech. Cool. You cannot convert low to high, and the only way to transport high is through power poles & wires, which have a 15 square limit. So like 50 power poles to get to from a generator to an extractor, which is 2-3x more than it should be. I’m sure this will be tweaked. Oh, and a platform can only take 1 transformer at a time, which is 25MW, which means “pods” of power. It breaks up foundations for a purpose I have yet to grasp. Oh, and power pole can’t clip, so your best bet is to dig them underground. I guess this is the ‘new to the genre’ part.

Belts – This is a weird thing to me. Straight belts are fine. Want to turn, you need to manually select. If you want to go up or down though, you need a different type of belt and cannot click-drag them in place. If you want to split or merge, you need a specific tool. It’s finicky. And you need a ton of them.

Advanced Pieces – The concept of a main bus lasts about 1 hour, until you need to build mk 2 belts due to insufficient plate throughput. You’ll quickly need to build those upgraded belts, but to actually do so needs some ninja belt pasta because it’s an offshoot of the main bus. Once you unlock steel production, you’ll simply conclude that a main bus doesn’t work here in the traditional sense, and you’ll have “pods” of construction. This is advanced logistics, and while I do expect it, it comes a little too early here.

Elevators – I think this is a really cool part of the game, where conceptually you dig a giant hole, put an elevator, then mine below ground. I say conceptually. Digging that hole requires explosives, and the field of view means you’re likely to fall into said hole with no way out (Jetpack is the only practical option). Then actually putting an elevator in needs substantially more room than you would think as it will be blocked by seemingly random things. And finally, once you do have an elevator in the hole…now you need a cave to mine and power within. The finicky parts of this are just ooooozing potential.

I’m curious as to how this game as an EA will move forward. It really doesn’t do enough to distinguish itself from its peers, where I’d argue Techtonica is the closest comparison and is a few years ahead. On the very positive side – it doesn’t crash, is fairly well optimized, has good art style, is fluid, and has generally “good bones”. If my gripes are all about quality of life, I think that’s an amazing achievement.