assembler barrel intake

[Veklim]

Except how do you handle empty barrels?

Ahhh, bots are my answer to that. You can exert pretty damned precise control via bots and logistic chests. Use a requester chest every now and then which requests the empty barrels and move them from there to the train station/belt bus/etc which you are using for transfer. My preferred method is to use a simple system whereby half of a train car has empty barrel slots and the other half has full ones, that way you wire up the loader and unloader inserters in pairs which only remove a barrel from the train when an empty one is put in. make sure you have a few spare barrels floating in buffer chests amongst the system at either or both ends and you're absolutely fine.

If using belts the same basic principle can easily be applied, merely run parallel tracks, one in either direction, or better yet a loop, since inserters always place to the far side of a belt, you can place full barrels on the nearside of the unbarrellers and the empties will always go on the other side of the belt.

[mrvn]

Except how do you handle empty barrels?

Ahhh, bots are my answer to that. You can exert pretty damned precise control via bots and logistic chests. Use a requester chest every now and then which requests the empty barrels and move them from there to the train station/belt bus/etc which you are using for transfer. My preferred method is to use a simple system whereby half of a train car has empty barrel slots and the other half has full ones, that way you wire up the loader and unloader inserters in pairs which only remove a barrel from the train when an empty one is put in. make sure you have a few spare barrels floating in buffer chests amongst the system at either or both ends and you're absolutely fine.

If using belts the same basic principle can easily be applied, merely run parallel tracks, one in either direction, or better yet a loop, since inserters always place to the far side of a belt, you can place full barrels on the nearside of the unbarrellers and the empties will always go on the other side of the belt.

A single loop works. But if you have a branch then you get a problem. Sure you can loop it but then all barrels first go down one branch and only what's left go down the other. What if you want half the crude oil going one way and the other the other way? You can't split it half way along the loop. Now you need two loops and have to split before barreling.

[Jap2.0]

Honestly I fail to see why anyone would say barrels are not useful.....

...

Barrels are awesome because they are compact, logistically sanguine and offer a whole host of logistically intricate and unusual arrangements. Just lately a friend of mine made a 32 core nuke plant which can only function at the efficiency it achieves due to the space reduction allowed by water barrels instead of pipes.

...

My point is this really, barrels just add more options to the table, they have a more complex set of reasons to be used and often occupy the extreme ends of the number crunching game (i.e. a fully compressed blue belt of barrels represents, afaik, the highest possible fluid throughput in the vanilla game (for a single tile wide strip of any given length, certainly)). If you want to calculate precise ratios for bot transfer of fluids to a series of production nodes (especially if you're circuit controlling them en masse, etc) then you can calculate the throughput of one assembler unbarrelling fluids and figure out precisely how many processes/second that can achieve, ergo how many assemblers you are hard-piping in a nice little box, etc. In my eperience, the further into automation you go, the more appealing the barrel becomes.

...

3) Some people use barrels and logistics for getting lube to engine and blue belt plants. I could see that being helpful if a) you need lube in a bunch of different areas, or b) your factory is so dense that routing pipes is a pain. In most cases, routing pipes is pretty easy, but then, barreling and using logistics is pretty easy too. So, again, it really doesn't matter which way you pick.

Conclusion - no, you don't need barrels. But they're not useless. They could provide some tiny benefits to some people in some circumstances.

Many of you seem to be missing he main use for barrels. As Slevek and Veklim said, barrels are mainly useful in dense, bot-based factories where you can just plop down a requester chest and assemblers and not have to worry about long pipes which may not even fit in some very dense bot factories. When used this way they can have insane throughput as well- one assembler can produce 150 liquid per second (actually 112.5 for level 2 assemblers or 187.5 for level 3 assemblers). When you're using sulfuric acid, that's enough to supply 300 processing unit assemblers or 37.5 battery assemblers. That is easily supportably by a moderate size logistics network. This would just make bots too OP, but that's just my opinion.

Favorite use for barrels: Heat steam to 500C. Put into steel barrel. Flight at high speed through cold, high altitude air. Drop barrel at steam turbine setup. Extract steam, which is somehow still at 500C.

I agree .

[mrvn]

Favorite use for barrels: Heat steam to 500C. Put into steel barrel. Flight at high speed through cold, high altitude air. Drop barrel at steam turbine setup. Extract steam, which is somehow still at 500C.

I agree .

You forgot: Store in storage chest for 3 days in case of a cold night.

[Jap2.0]

Favorite use for barrels: Heat steam to 500C. Put into steel barrel. Flight at high speed through cold, high altitude air. Drop barrel at steam turbine setup. Extract steam, which is somehow still at 500C.

I agree .

You forgot: Store in storage chest for 3 days in case of a cold night.

Also put some in a sealed vault in a very cold area like the Norwegian seed vault that was in the news just over a month ago in case the biters overrun your base, so you can get a head-start on power.

[buggy123]

Barrels are a little clunky and don't get a lot of love. So, I think they should be improved, and a good way to do this would be to allow machines to natively support barrels.

What do I mean by "natively"? Well...

Lets look at a assembler:

09-27-2024, 17-03-11.png (73.95 KiB) Viewed 431 times

We want to make blue circuits, and we have everything we need. Circuits of all sorts, belts in and out, and sulfuric acid... in barrels. Right.

So what do we need to do? All of this:

09-27-2024, 17-06-27.png (120.05 KiB) Viewed 431 times

We need a second whole assembler, and two more belts, and we need to handle getting rid of the extra barrels. And sure, this isn't too bad, because of course one unbarrelling assembler can usually feed multiple machines. But at the same time, you could just do this:

09-27-2024, 17-08-52.png (53.63 KiB) Viewed 431 times

And it just works. No transporting empty barrels around, no playing about with the total number of barrels in the system... it just works.
Yeah.
It's not hard to see why barrels aren't very popular.



So, what if instead of all that, we just do... this?

09-27-2024, 17-10-51.png (57.08 KiB) Viewed 431 times

"But wait!", you say, "That won't work, it has to be unbarrelled by a dedicated assembler."

Ah, but what if it did just work?

BarrelMockup1.png (30.59 KiB) Viewed 431 times

What if you put the barrels it, and it just automagically empties them into it's own internal buffer, and gives you back the empty barrels?

Or, if instead you have a more complicated machine that creates and uses fluids:

BarrelMockup2.png (39.07 KiB) Viewed 431 times

Then you have a couple of slots, one for each type of fluid.


One complication that does emerge is if a machine both needs and produces empty barrels. I don't think this is a especially difficult design problem to solve; one solution that comes to mind is to treat barrel slots like chest slots when appropriate, allowing inserters to both add and remove barrels. On top of this, barrels could avoid being filled until the moment a inserter actually pulls them out of the machine, which would help reduce clogging issues. Another solution could be to have separate input and output slots for fluids that occur more than once in a recipe, and only fill the output slot if the input slot is full.


On the whole, this means that barrels become a alternative instead of excessively situational/a downgrade. Instead of pipes, you have the choice to supply barrels via belts. You still have to remove empty barrels, but on a whole its simpler. Complicated fluid recipes would present a player with a choice between complex piping and a inserter filtering/belt sorting problem.

Both a challenge to solve, but of different sorts.

09-27-2024, 18-02-42.png (310.77 KiB) Viewed 431 times

[Koub]

[Koub] Merged into an older thread with the same suggestion.