Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[umiman]

I'm even more lazy. I don't even set up circuits or depots or anything.

I just put everything on a 30 second timer and overload the network so nothing will ever be empty. It's unefficient but rocket fuel is super cheap so who cares. All the lines are at max compression and there's no headaches.

It just goes like this.

Code: [Select]

Unloading Station > Copper Smelters A
                  > Copper Smelters B
                  > Copper Smelters C

1 train for each. Problem solved.

[n9103]

I'm even more lazy. I don't even set up circuits or depots or anything.

I just put everything on a 30 second timer and overload the network so nothing will ever be empty. It's unefficient but rocket fuel is super cheap so who cares. All the lines are at max compression and there's no headaches.

It just goes like this.

Code: [Select]

Unloading Station > Copper Smelters A
                  > Copper Smelters B
                  > Copper Smelters C

1 train for each. Problem solved.

Pretty much exactly my approach.
If I notice a train comes back short on resources, then it means it's time to expand the rail line further out to the next patch.

[umiman]

Yup. It served me more than well enough for my 1,000,000 green circuits per hour factory.

I had more problems expanding the bus to be big enough to unload all the plates rather than the trains not delivering enough. Which is why I will be using motorbitch's blueprints in the future, though I might tweak it a little. I'll probably make it 12 inserters per carriage and reduce some splitters. It just won't look as pretty and neat.

[Putnam]

What do y'all bus, anyway? Usually I only bus iron/copper, but in my latest game I bussed circuits too and it was, uh, actually pretty helpful, so I should probably do that more often.

Eventually I'm probably going to give into the bus and start bussing stone bricks or something.

[umiman]

You don't really need a bus if you're making average "launch one rocket" factories. This is for several reasons:

1. You don't have the proper tech to do it justice for most of the game.
2. It's ultra annoying to upgrade if you start with regular belts. This is the worst part.
3. It's way too expensive for you to plan for both materially and strategically early on (unless you have no biters, though it'll still be expensive).
4. You don't use enough resources to justify a bus. Maybe a really small one.

Buses are better for when you've already basically beaten the game and want to make uber factories. Also way better then since you don't have to manually lay down all the belts yourself.

Then you're looking at many, many belts for copper and iron. Some for steel. Some for each of the chips. That's basically it.

You don't want to bus things like metal gears or copper wire as this is extremely inefficient. Belts don't move things fast enough even at blue belts and more importantly, they don't move enough of them in a single stack. This is extremely problematic for those two items as they get used in massive quantities. So when it comes to items like metal gears and copper wire, you don't want any belts involved whatsoever if you can avoid it and have everything moved directly from assembler to assembler. This is also the reason why everyone designs red and green chip blueprints to direct feed them.

For other stuff like plastic or modules... Eh. Depends on you. If you want to bus it you can. I personally wouldn't and would just use a train or make it on site. I also would rather use trains instead of buses if I could, but I'm not creative enough to figure out how to directly feed a thousand assemblers through trains without going through a conveyor belt. Actually there's an idea... I could manually filter each cargo carriage to carry specific things and have the trains move around supplying the assemblers in a big line somehow. Maybe train stops every single section? 1 second timers? Or one ginormous train? Unloading into buffer chests that unload directly into assembly machines with no belts involved.

Oooo....

[motorbitch]

Yup. It served me more than well enough for my 1,000,000 green circuits per hour factory.

I had more problems expanding the bus to be big enough to unload all the plates rather than the trains not delivering enough. Which is why I will be using motorbitch's blueprints in the future, though I might tweak it a little. I'll probably make it 12 inserters per carriage and reduce some splitters. It just won't look as pretty and neat.

12 inserters will only speed up the unloading of the train *if* the chests are empty. but it wont put more ore on the belts, as 8 are enough to saturate the belts. so the chests wont be empty. the long and short of it is, it wont speed up anything.
also, you can use the empty spot to put a requester chest there with an inserter that loads *into* the train. have it request ore and put ore on auto trash so the stuff will be neatly inserted into your factory.

[PTTG??]

I ended up doing copper, iron, and steel for a simple bus (one lane each of a level 2 belt, so barely a bus at all). It was actually pretty cool. To be honest though, I think my next approach may be train based cargo shuttling.

[motorbitch]

yeah, i found busses dont realy work very well for copper and iron, at least not with expensive recieps. when i first tried it i read about that this to do it and that you just rebalance the belt after you take out stuff and gucci. but it doesnt work that way. no matter how hard you rebalance a belt, you wont magically saturate a belt that has half its contents taken out, and a belt will only support its items/s taken out of it on its whole lenght. so what i do now is yes, i do have a main bus of sorts, but that i only use for factories with low demand.  lines with high demand, like green electronics get their own dedicated lines.
another thing expensive reciepts told me that its a smart thing to do to have your iron smelter lines interrupted by gear wheel factories. they will use up a lot of iron, thus its smart to have this drain before you even start to bus. of course, the belt limit then means it will be hard to get enough ore to your smelter/gear line to saturate both the gear and the iron lines.
still, if you want to feed a very hungry factory, like a gun turret fac, one fully saturated belt is MUCH better then two belts that are half empty.

[milo christiansen]

You need to run dedicated lines for green circuits at a minimum. other than that, a bus should be enough. If not, then run a few more feeders to merge into the bus after it thins out too much.

I keep wanting to try a train base, but I never quite get around to it.

[MagmaMcFry]

Busses don't need to be balanced, and they don't need to be full-width all the way. If you need half a belt of iron at the first fork, and half a belt at the next, then you can make your iron bus narrower by one after that, it's pointless to keep the bus at full width afterwards.

[milo christiansen]

Whoa! The latest FFF really set the fox in the hen house!

One of the devs was talking about how he wishes he could go back and remove logistics bots, you should hear the megabase types on the factorio forum! You know the kind I mean, max modules everything, no belts, all bots. Boring.

One guy even tried to say that belts are "too easy, just drop more blueprints". I thing he got belts and bots confused

[umiman]

It could be something worth investigating though.

Those all bots factories are lame as hell. I think logistics bots should honestly just be limited to delivering things to you and vice versa.

When people make all bot factories they remove all the challenge of logistics. This seems to be a big deal for the dev team as when they were asked about adding z-levels they mentioned their concern about it.

Here: https://www.reddit.com/r/factorio/comments/6e6tkw/im_the_founder_of_factorio_kovarex_ama/di83rf7/

Real levels, like underground, would be easy to do, but it would introduce new problems. As you don't see both levels at the same time, it is much harder to understand the factory layout compared to simple 2d view. It would also diminish the important part of the logistic challenge.

Maybe instead have a dedicated bot moving thing that has to be on the ground. Like a giant walking robot with a big bag.

[motorbitch]

dunno, i think they are nice to fill a niche, like delivering high value low quantity stuff like red or blue circuits. maybe just make them much more expensive to use, so there is a real tradeoff for not building a belt.

[MagmaMcFry]

Bots are very expensive to use though, they require massive amounts of power in comparison to belts and trains. Also bots are actually pretty low-capacity: If you have a strip of perfectly packed roboports, then this strip can carry only 5 times as many items as an equally wide strip of blue belts, and that's even assuming that robots have infinite speed. That's basically peanuts compared to train throughput.

[umiman]

You know what would be fun?

If we could build war bots. Like in Star Wars. Just have factories building armies of them, and factories building the weapons and vehicles for them.

It'd give us a nice end game goal to aspire to other than make 1000 science per minute.

We'd need a more capable opponent though.