Bus or not to Bus

[Breith]

Yeah, this is what I meant. But I'm not sure to understand your point.

You cant really over product: if your belt is full, then the assembly machine won't product more (cannot export the good from the assembly machine). It's not related to keep some free space to easily increase the production later, when you need to.
You also can use your robot network as a buffer as well as a good way to monitor your production. Even if you don't use robot, add a passive (or active) provider chest at the end of your production line, link the belt to the inserter: if the belt is not moving (i.e. you're producing more than used by the factory), fill the chest and if the product reach a given amount, then do not export the product from the assembly line (or block the motion of a belt. Or do nothing, everything is full anyway).

If you do not constraint the vertical development of your base, then you cannot have a walking path north close to your processing area. But actually... does not really matter. I do not need a walking path above my processing area: the only thing I have north is my wall and a (vertical) line of gun turrets. I have a walking path here (about 100 tiles north from my bus. Whatever the distance you use it's easy to move a turret line with robots if you need more place. Way more than moving a processing area. I moved my wall twice already, when I had to increase the Copper Wire production column).

[Northgate]

I will only use logistic bots for transfering buildings to my own inventory.

Because I can't craft anything manually and have to automate everything(I even have an assembly for iron sticks) I need a frequent access to my resources on the bus and I feel I have to sacrifice vertical expansion space in order to gain quicker access to the belt. Otherwise the bus would likely become really messy. Sure I could do the whole 32 lane bus in the middle and then have the processing areas north and south of it, but then the access to the resources requires a lot more underground belt and belting off the resources every 4 or 5 tiles would probably result in a mess. And if I think I need more space and can just leave space for an additional column instead of expanding vertically.

Like Light said, in case I need to expand my production I have 2 temporary support bus lines which can go zig zag.

[Light]

I'm currently also experimenting with a ore bus to bring the raw ore to the refining states and the refined ore to the sorters and the sorted to the smelting. But I haven't decided on a setup yet. Anyone having a good setup?

I may as well share what I've come up with.

Sorting

I came up with this sort of design after a good couple hours of tinkering with it from the start of my game. I haven't updated it since then, thus it's rusty around the edges and I see things I'd want to improve, but since I love trains they transport the goods to each respective area. I do use robots for the short trip to the train warehouses, as I couldn't come up with a decent way to do it at the time. I may revisit the design in the future, but it functions perfectly fine.

The trains then go to the smelting area, which cost so much stone that I spent more time waiting than building it. It's still not finished but I like where it's going so far.

Smelting

It's no surprise I built it right next to the large coal deposit, you're going to need it.

Just note that these setups are not ideal, as I don't look up pre-designed material since that would ruin the uniqueness of making a factory. But hopefully it gives you some ideas to build off of.

[Northgate]

That looks really beautiful. Very clean setup. The only thing that would bother me is the amount of empty space. I think I would've put the Leeching plants vertically next to the sorters.

And if I remember correctly the ratios in the smelting ares are slightly wrong. I think it's: 2-1-6-10-15.

I personally never bother refining the Saphirite because it doesn't provide anything that you can't get from the others and I appreciate that extra bit of iron.

[Light]

That looks really beautiful. Very clean setup. The only thing that would bother me is the amount of empty space. I think I would've put the Leeching plants vertically next to the sorters.

And if I remember correctly the ratios in the smelting ares are slightly wrong. I think it's: 2-1-6-10-15.

I personally never bother refining the Saphirite because it doesn't provide anything that you can't get from the others and I appreciate that extra bit of iron.

A large reason for the empty space was because I wasn't sure if expanding was required at the time, so I ensured there was room to do so if needed. After days of using it to satisfaction, I agree it can be compressed, but with the sorters placed horizontally instead.

The ratios aren't exact, but the different Mk being used corrects it to a degree. Production of plates has thus far been more than satisfactory so I didn't bother to really improve the prototype design that's used now.

Actually, copper is my major enemy since it seems I can never get rid of the stuff fast enough. I imagine if I used solar power that wouldn't be the case, but it's leading me to come up with an idea to correct the issue.

The major design flaw of my sorting area that is constantly on my mind is that the ores are sent to a main bus, and as such once the row is maxed out the sorting area starts to shut down as it clogs up. I'm now in the process of finding a way to deal with the overflow using circuit networks to keep the other ores flowing while excess ores are stored away until the demand does increase for them. I just haven't had the time to sit down and test the design ideas I've come up with yet.

I do appreciate the clean compliment. This is my third factory and my first Angels+Bobs one. Keeping things clean in vanilla is straightforward, but with many Angels byproducts and Bobs expanded tech, it has been quite the experience trying to come up with ways to maintain that clean look. A lot of my game time can be attributed to learning the new ways and then designing around it. My fourth factory onward will likely be Angels+Bob again, it's just too much fun to design for.

[ukezi]

your solution would be mixed sorting that gives one type of ore only. so copper can clog up and nothing else stops.

[Breith]

I'm moving out of the sorting that gives several types of ores to the one that gives only one. It might be less efficient on the amount of ore output but it's way easier to manage the ore supply to the melting area.

[fractalman]

Actually, copper is my major enemy since it seems I can never get rid of the stuff fast enough. I imagine if I used solar power that wouldn't be the case, but it's leading me to come up with an idea to correct the issue.

Turn it into a liquid and void it in a steam engine. Warning: this method will break in the next big update.

Personally, I use the fancy processing for my iron ores but not my copper. The ratios seem to be working out pretty well as long as I'm getting my iron+copper from a mix of slag processing (mostly from dirt water electrolysis) and saphirite/jovialte(sp?).


I also shove all the ores from sorting into a warehouse and then filter-inserter from there, so there's lots of space for everything.

[Exasperation]

What I do is set up a primary sorting line with the multi-output sorting methods, a secondary sorting line with the single-output sorting methods, and side-load the output of the secondary line onto the output of the primary line. That way, if I'm using (for example) iron and copper in the same ratio as I'm getting them from sorting, the primary line handles everything, but if I'm not using enough copper and it backs up and slows down the primary line the secondary line will kick in to fill the gaps in the iron line. I also throw in buffers between the sorting and the side-loading merge to smooth over temporary shifts in consumption without having to fall back to the secondary sorting line.

[Breith]

What I do is set up a primary sorting line with the multi-output sorting methods, a secondary sorting line with the single-output sorting methods, and side-load the output of the secondary line onto the output of the primary line. That way, if I'm using (for example) iron and copper in the same ratio as I'm getting them from sorting, the primary line handles everything, but if I'm not using enough copper and it backs up and slows down the primary line the secondary line will kick in to fill the gaps in the iron line. I also throw in buffers between the sorting and the side-loading merge to smooth over temporary shifts in consumption without having to fall back to the secondary sorting line.

The issue is not the loading from another line, it's more if you line out of the multi-output is full of one element (i.e. the one from the sorting machine to the bus). At this point, the whole multi output line stops (my cobalite line is plug like that since hours since you almost use no cobalite at all in the early game).

It's a good way in the early game to have some ore, but definitively I think that this multi-ouput method should be abandonned in the middle/late game.

[Northgate]

What I do is set up a primary sorting line with the multi-output sorting methods, a secondary sorting line with the single-output sorting methods, and side-load the output of the secondary line onto the output of the primary line. That way, if I'm using (for example) iron and copper in the same ratio as I'm getting them from sorting, the primary line handles everything, but if I'm not using enough copper and it backs up and slows down the primary line the secondary line will kick in to fill the gaps in the iron line. I also throw in buffers between the sorting and the side-loading merge to smooth over temporary shifts in consumption without having to fall back to the secondary sorting line.

The issue is not the loading from another line, it's more if you line out of the multi-output is full of one element (i.e. the one from the sorting machine to the bus). At this point, the whole multi output line stops (my cobalite line is plug like that since hours since you almost use no cobalite at all in the early game).

It's a good way in the early game to have some ore, but definitively I think that this multi-ouput method should be abandonned in the middle/late game.

I would do it the other way around. I'm usually just sorting Saphirite in the early game and everything else get's crushed and then meldet immediately. Once the game progresses and the need for the higher level ores come up, I start to float and leach Bobmonium and Rubyte to make sure that I don't produce stuff I don't need yet.

[Exasperation]

What I do is set up a primary sorting line with the multi-output sorting methods, a secondary sorting line with the single-output sorting methods, and side-load the output of the secondary line onto the output of the primary line. That way, if I'm using (for example) iron and copper in the same ratio as I'm getting them from sorting, the primary line handles everything, but if I'm not using enough copper and it backs up and slows down the primary line the secondary line will kick in to fill the gaps in the iron line. I also throw in buffers between the sorting and the side-loading merge to smooth over temporary shifts in consumption without having to fall back to the secondary sorting line.

The issue is not the loading from another line, it's more if you line out of the multi-output is full of one element (i.e. the one from the sorting machine to the bus). At this point, the whole multi output line stops (my cobalite line is plug like that since hours since you almost use no cobalite at all in the early game).

It's a good way in the early game to have some ore, but definitively I think that this multi-ouput method should be abandonned in the middle/late game.

That's exactly the issue I'm addressing, though; here's a simple example setup to demonstrate the idea:

priorities

In the image, you can see that iron has backed up since it's being consumed slower than the Saphirite sorting can output it. That means that the Saphirite sorting slows down (it runs as iron is consumed, which means it can't produce copper fast enough to meet demand). That causes the copper to side-load from the copper sorting to meet the shortfall, but the iron (and some of the copper) is still being provided by the Saphirite sorting.

If we doubled the demand for iron, we would be consuming it in the exact ratio provided by Saphirite sorting, so the entire demand would be provided from the Saphirite sorting, with no need for the iron or copper sorting lines to run.

If we tripled the demand for iron, the situation would reverse; the copper line would (slowly) back up, causing Saphirite sorting to slow down to match the copper consumption instead of matching the iron consumption. All of the copper would then be provided by Saphirite sorting, while the iron sorting would start filling the shortfall in the iron line, and the copper sorting would be idle.

The point is that a setup of this sort can dynamically respond to changes in demand, using the multi-output methods when possible, and the single-output methods to cover for differences between demand and supply.

[ukezi]

I'm moving out of the sorting that gives several types of ores to the one that gives only one. It might be less efficient on the amount of ore output but it's way easier to manage the ore supply to the melting area.

you get more ore out of it actually since you get no slag. that you only get if you go up to purified with the not mixes sorting.

[Breith]

That's exactly the issue I'm addressing, though; here's a simple example setup to demonstrate the idea:

priorities

In the image, you can see that iron has backed up since it's being consumed slower than the Saphirite sorting can output it. That means that the Saphirite sorting slows down (it runs as iron is consumed, which means it can't produce copper fast enough to meet demand). That causes the copper to side-load from the copper sorting to meet the shortfall, but the iron (and some of the copper) is still being provided by the Saphirite sorting.

If we doubled the demand for iron, we would be consuming it in the exact ratio provided by Saphirite sorting, so the entire demand would be provided from the Saphirite sorting, with no need for the iron or copper sorting lines to run.

If we tripled the demand for iron, the situation would reverse; the copper line would (slowly) back up, causing Saphirite sorting to slow down to match the copper consumption instead of matching the iron consumption. All of the copper would then be provided by Saphirite sorting, while the iron sorting would start filling the shortfall in the iron line, and the copper sorting would be idle.

The point is that a setup of this sort can dynamically respond to changes in demand, using the multi-output methods when possible, and the single-output methods to cover for differences between demand and supply.

Outputting the ore to the first level or ore processing from smelting's mod is the best idea ever. I didn't think about that, and you're right, it solves a good part of the issue. It might be a little bit more annoying for advanced ore 'cause of the space usage, but it's not a really big issue.

you get more ore out of it actually since you get no slag. that you only get if you go up to purified with the not mixes sorting.

You can get ore from slag processing. In addition to the crushed stone processing, you get more ore at the end (didn't go through the math to know exactly how much more). Actually it also provides more flexibility, since you can redirect the slag processing to the type of ore you need the most at an instant t.

[ukezi]

You get at 0.4 ore per slag out of slag processing. Also 0.4 ore out of 25 crushed.