parts or end pruduct factorys ?

[Kewada]

is it better to build a specifit factory for every thing i want to automate like or to building.
or build multiple smaller factorys that just produce a part like iron plates and then distrebute them to the locations where i need them ?

[josmith7]

Both approaches work.

My first game I went with multiple smaller factories each building a specific item or component.

Then post-blueprints I went, in part just to switch things up, to the all-inclusive factory for any high production rate items (rockets, solar sails, proliferator, fuel rods, warpers, science, etc.). However all low production rate (basically buildings) did use very small factories building each component because my building production lines spent the vast majority of their item idle; and so it didn't make sense to me to give them full dedicated factories.
Also, my personal preference, was to keep general smelting of basic materials separate from the all-inclusive factories -- so I centralized smelting and treated things like iron, copper, titanium ingots, refined silicon, graphite, glass, and stone bricks as universally available building blocks.

But on those high production rate items, it meant that for example I was producing processor as the end product of an all-inclusive factory (for use in purple science), but completely independently was also producing processors as a subcomponent within the all-inclusive factory making quantum chips end (for use it green science).

Going multiple smaller factories lets excess capacity get taped for low production rate items like making buildings - and the factories are simpler to design and easier to stamp down. However with multiple production chains drawing on the same products they can sometimes be a bit slower to chase down the ultimate bottleneck when production isn't keeping up with your expectations.


But like I said (and despite spending more text on the later options) both can work really well -- it's all a mater of what kind of play style you prefer.

[SmoothCrimnl]

one big detriment to building individual parts lines is that you can't determine how/where your end product is pulling it's components from.

for example, my processor build is 90P/S - 3 output lines of 30P/S feeding into a IPL. That will require 180P/S blue chips - 2x builds at 90P/S each (i use the same 3 output lines at 30P/S each). The processor build will run one of the blue chip builds all the way down before it starts grabbing from the second one. most of the time it doesn't matter, but there are occasions when it can cause backlogs on production.

[Ghevd]

I'm pretty new myself. Only 106 hours. I first went with a polar bus. It was beautiful.
The bus did manage to get me to some yellow science but, unfortunately, it didn't last once I unlocked interplanetary logistics.

Instead of scrapping my starter world I just moved the other planet in my solar system I had not used. It had titanium but no silicon. The planet was nice and close though seeing as we both orbited the same gas giant. All I left my starter world with was about 8 interplanetary stations, some fuel, and enough solar panels to make a belt of them around the equator. Had to make a few trips back and forth occasionally but it was like 0.10 AU away.

So now I have a interplanetary logistics for everything. Iron ore in, iron ingots out. Magnets. Copper. Silicon. Everything comes in to a single station designated to make one item. Then the same station acts as the storage for the end product. Each station proliferates the starting product if it is freshly mined like ore and coal and water etc and of course it proliferates the finished product to avoid having to do it later.

Once a product is in a station I can request it basically anywhere I need it. Even my science is built using this method and then sent to where my research labs are.

When I start over once combat is realized I think I will just go with a setup that is more easily transitioned to the logistics drones to avoid moving. They are an inevitability.

I recently finished green science and finally attained warpers but unfortunately I kinda lost interest. Which is a shame because I haven't even attempted to make sphere parts yet.

Hopefully once there is a way to fail, i.e. combat, my interest will be rekindled. As the game stands now there is no fail state so, which ever way you decide to design your base doesn't really matter because worst case is you simply tear it up and rebuild.

[aphfid]

the beauty of dyson sphere program is that you can try those options and many more.

i've played a number of games now, probably up to around my tenth game. this game i'm playing making everything from raws. not because I've decided its the best way, but because it's an approach i haven't taken yet, and I am totally loving it.

my response would be "what would you like to do?". do that. you can always do something different next game.

[Kewada]

now that i unlocked the logistic station i try a parts factory. so i deliver what ever i need to the factory process it and then deliver it every where its needed. saves lots of belt work.

[yes]

while both approaches are valid, its still best to always only build as much as you need. what i mean with that in terms of "build mulitple small factories that produce just one intermediary product" is that its better to not preplan large factories (say build 100 iron/s, while you will actually only consume 50 - totally random numbers btw), but only scale up as you go along, this way you will always stay on top of demand without experiencing sudden shortages (because you think, well i have plenty of iron anyways. while it may actually just be that towers are full, and you only realize you would have actually needed 120/s way too late and everything grinds to a halt). ofc you can counter that with speadsheets, but thats kinda too sophisticated for my personal taste ^^
its actually pretty simlar to "complete factories which produce from raw materials" just that components/intermediaries are centralized.

personally i usually end up using a mixture of both. the very basic materials are centralized, everything else produced on-site.

producing everything on site is a bit tricky with h2 and h3 in early and midgame, but manageable. here the centralized approach - i usually have a dedicated planet - is, in my opinion, far superior since you can use priorty belting. its also the only thing i have more complex belt networks - but maybe thats just habit and i could do it differently.

but ofc ultimatively there is no right way to play DSP. as long as you enjoy the game and yourself you are doing the right thing, no worries ...

[GAMING_Alligator]

Originally posted by Kewada:

is it better to build a specifit factory for every thing i want to automate like or to building.
or build multiple smaller factorys that just produce a part like iron plates and then distrebute them to the locations where i need them ?

Build a factory for everything, resources and buildings alike (although for most buildings it may just be one single assembler). Feed raw materials and everything you build into your logistics network, no direct links between assembly lines. That way if you need to scale up one thing you can just build another line wherever and feed it into your logistics network without worrying about anything else.

[ariabenko]

My personal choice - using individual planets with suitable resources (or in suitable systems) for a particular product or critical component. The logic being to reduce the necessary shipping (but not to eliminate it entirely, so I can still use planets consistently).

E.g.:
- Green turbines can be built from scratch on lava / volcanic ash planets. Lots of iron/copper - and voila.
- Titanium Alloy factory best built on volcanic ash planet: iron, titanium, sulphur
- Proliferator I-II-III factory, IMO, best built on organic planet with tons of coal and, ideally, spiniform rare veins.
- For CPU: I look for systems with one planet with tons of copper and silicon, to deploy a huge microcrystalline factory, and another lava/volcanic ash planet for the circuit/CPU factory itself.
- I use barren desert planets for an occasional giga-factory that takes minimum logistics but consumes a ton of space: e.g. producing nanotubes.

[Bobucles]

For the typical person, single product bite sized factories work best. The main reason is thanks to blueprints and logistic towers. Design a simple part of the factory, blueprint it, and connect it to the rest of your base through towers. Expanding X production becomes very easy, just print out more X chunks as needed.

The more complex a factory chunk gets, the more difficult it becomes to maintain. For example take a blueprint that starts with ore and goes all the way to producing yellow cubes. What happens when proliferators become wide spread? The entire factory chunk needs its ratios reworked from scratch. What a mess. With simple factories, the exact ratios don't even really matter. Just build close enough, then when proliferators mix things up, just stamp down more XYZ chunks to compensate. Low IQ design, but incredibly flexible and highly useful.

[aphfid]

I'm not going to put out an opinion regarding better or worse. Just share an experience, because there are so many ways to play the game and what one person loves another might hate. I've played a lot of dyson sphere and usually I go with some variation of one planet being dedicated to producing one or a few components and shipping around the components as needed.

I'm playing my first game of building everything from raw and it is a totally different experience.

Usually my play-style is reactionary. For example. Want more white cubes - set up more white cube production. Hmm, not making enough green cubes. Make more green cubes. Need more planar filters to make more green cubes. Need more casimir to make more planar filters and so on. I would describe my usual play-style as expand and fill in what needs filling in after expanding. There are always variations. Some games I might not transport any ores at all - only ingots and refined materials.

Playing from raw is a totally different experience. I built a really shabby blueprint that takes in raws - crude oil, crushed rock, iron, copper and titanium ore, etc and uses them to make red, yellow, blue, purple and green cubes and then uses those cubes to make white cubes and then feeds those white cubes into science.

If I'm ready to produce more science, I just drop another science blueprint and put drones in the towers in the blueprint. Done. As long as I am producing enough raws, there is nothing more for me to do, so I spend the vast majority of my time creating new blueprints. If I need more raws, I just put down a few more mining machines on whatever planet is convenient to do so and it's done.

The blueprints take me a long long time to do. But I find it incredibly satisfying. The experience reminds me of those videos of people setting up long chains of dominoes and knocking over that first domino and watching all the rest fall in succession. I love it. I haven't completed my first set of blueprints yet, but can already see a lot that needs improving. I'm going to remake my science blueprint to be more efficient and I'm going to include the ray receivers in my science blueprint. The idea is that whenever I have x amount of spare capacity I will simply drop another science blueprint - all inclusive with ray receivers and artificial stars to generate energy.

Once a blueprint is dropped, as long as it doesn't have a design flaw, such as not handling hydrogen properly, it never needs to be looked at again. It just keeps working forever as long as there are raws to feed it. No chasing magnetic coils or green motors. All my raws are shared globally so any shortage is quickly and easily resolved by harvesting more of whatever raw is lacking. I just pick a planet with what is needed and set up harvesting of a bit of everything.

I really enjoyed all the previous games I played with different variations of different planets producing different components. I'm really enjoying this game, producing everything from raw, but it is a totally different experience. And I will really enjoy my next game, which will be a variation of producing everything from raw. Instead of exporting raw ingredients, I'll convert my raw ingredients to ingots, refined oil etc but still use the expanded blueprints to make complete science blueprints and all-inclusive small carrier rocket blueprints that fire the rockets they make.

I'm not going to rate one way as being better than another, just different experiences. I will probably play another game at some time in the future making different components on different planets, because that is the experience I will choose at that time.

[Kyrros]

Originally posted by Kewada:

is it better to build a specifit factory for every thing i want [...] or build multiple smaller factorys that just produce a part like iron plates and then distrebute [...]

Yes.