Furnaces + internal buffering

Post your ideas and suggestions how to improve the game.

Moderator: ickputzdirwech

Post Reply
jonatkins
Fast Inserter
Fast Inserter
Posts: 155
Joined: Wed Sep 30, 2015 7:29 pm
Contact:

Furnaces + internal buffering

Post by jonatkins »

Is there a good reason that furnaces will continue to accept raw resources from inserters when they have several items in ther output slot not being taken? Assemblers stop getting ingredients inserted from belts/chests once a small number of items are in the output slot, so why the different behaviour for furnaces?

Why is it an issue? It's hard to see how a large furnace array is coping with supply/demand when there's all this internal buffering going on. e.g. you need more output, add a bunch more furnaces, and suddenly the input is sucked dry. However, wait some time, and the furnaces output buffers wil fill and the supply will start building up again.

You would need to make sure the very early gameplay doesn't suffer - where burner miners directly insert into furnaces. I'd think this can be solved by having the miners force the resources into the imput slot - similar to how manually inserting ingredients into an assembler will cause it to assemble beyond it's usual output buffer size. As this is how direct miner->assembler already works it shouldn't be hard to have the same behaviour for furnaces.

Only one early-game setup I can think of would break with a change like this, but it's gotta be rarely used - inserters feeding resources into furnaces, but the player expecting furnaces to buffer up internally. I've never used such a setup though - whenever I've got to the point of belts/inserters loading furnaces, I'm also unloading onto belts too.

I've noticed that chemical plants also have the same issue - but only when fed by fluids only (e.g. solid fuel). Seems there's some general logic to stop input inserters feeding in resources when output is backing up, but this doesn't apply to fluid inputs. This has never been a huge issue in my factories, but could also be changed so fluid behaviour is the same as input inserters.

Thoughts?

nuhll
Filter Inserter
Filter Inserter
Posts: 922
Joined: Mon Apr 04, 2016 9:48 pm
Contact:

Re: Furnaces + internal buffering

Post by nuhll »

I like buffering. And when you build arrays where it matters, its endgame and you have everything enought, no big deal.

jonatkins
Fast Inserter
Fast Inserter
Posts: 155
Joined: Wed Sep 30, 2015 7:29 pm
Contact:

Re: Furnaces + internal buffering

Post by jonatkins »

nuhll wrote:I like buffering. And when you build arrays where it matters, its endgame and you have everything enought, no big deal.
It normally becomes an issue for me mid-game, rather than end-game - when I'm starting to really ramp up production to keep blue science going well and working to the next stage. Things like thinking I don't have enough mines the new furnaces mean the ore unloading is running out, only to find later a queue of ore trains backed up out of the queue space blocking lines and causing havock for other resources.

If you want buffering, explicitly add it with chests where needed, or (as I often do) use trains to transfer iron/copper from a smelting array to the main base - as this adds natural buffering with the train loading/unloading.

Maybe there's an argument for configurable buffering - in furnaces/assemblers/chemical plants, etc - constant values, or even controlled by the circuit network - but that's not what I intend to discuss here, rather the simple buffering inconsistency between furnaces and assemblers as things stand now.

nuhll
Filter Inserter
Filter Inserter
Posts: 922
Joined: Mon Apr 04, 2016 9:48 pm
Contact:

Re: Furnaces + internal buffering

Post by nuhll »

Maybe sometime you can attach a cable from inserter to furnace to control how many get inserted.. :)

User avatar
ssilk
Global Moderator
Global Moderator
Posts: 12888
Joined: Tue Apr 16, 2013 10:35 pm
Contact:

Re: Furnaces + internal buffering

Post by ssilk »

jonatkins wrote:Thoughts?
There are hundreds of smelting-designs, that cope with it and have no problem to handle that well.
And there are circuits. For example: You can count incoming and outcoming items and stop incoming if the furnace would begin to buffer.
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...

Aeternus
Filter Inserter
Filter Inserter
Posts: 835
Joined: Wed Mar 29, 2017 2:10 am
Contact:

Re: Furnaces + internal buffering

Post by Aeternus »

Personally I'd rather see the reverse applied: That all production facilities manufacture up to a full stack of whatever they are producing at the factory output, even when output is backed up. That every facility functions like smelters do.

jonatkins
Fast Inserter
Fast Inserter
Posts: 155
Joined: Wed Sep 30, 2015 7:29 pm
Contact:

Re: Furnaces + internal buffering

Post by jonatkins »

As I suspeced, different people want different behaviour. However, while it's trivial to add additional buffering with chests where desired, it's far from trivial to stop it when the furnace has it built in. (especially when it's not a 1:1 ratio of input to output - e.g. bricks, stone, or productivity modules)

I wonder if there's a mod that changes furnace behaviour? I know, for example, Bob's Mods additional furnaces are actually implemented as assemblers with furnace graphics - you have to select the smelting recipe with them.

I wouldn't mind having to select a smelting recipe (could even be useful sometimes) - might look into writing such a mod if none already exists...

User avatar
ssilk
Global Moderator
Global Moderator
Posts: 12888
Joined: Tue Apr 16, 2013 10:35 pm
Contact:

Re: Furnaces + internal buffering

Post by ssilk »

Aeternus wrote:Personally I'd rather see the reverse applied: That all production facilities manufacture up to a full stack of whatever they are producing at the factory output, even when output is backed up. That every facility functions like smelters do.
Not a good idea. Think to how many resources are 200 advanced circuits. :)
jonatkins wrote:As I suspeced, different people want different behaviour. However, while it's trivial to add additional buffering with chests where desired, it's far from trivial to stop it when the furnace has it built in. (especially when it's not a 1:1 ratio of input to output - e.g. bricks, stone, or productivity modules)
There are artithmetic combinators which are able to divide...

Sorry, this is just not an issue, it's a challenge to handle the furnaces in a way, that it is working for your factory. :)
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...

agentgoblin
Burner Inserter
Burner Inserter
Posts: 13
Joined: Tue Jun 28, 2016 5:29 pm
Contact:

Re: Furnaces + internal buffering

Post by agentgoblin »

Personally I'm want logic circuit connection for furnaces/assemblers for controlling inner buffer size. With default furnaces buffer size equal stack and assemblers buffer size equal one item.

And turn on/off furnaces/assemblers, read contents, but it's another question. :roll:

Roxor128
Fast Inserter
Fast Inserter
Posts: 163
Joined: Sun Oct 02, 2016 9:48 am
Contact:

Re: Furnaces + internal buffering

Post by Roxor128 »

agentgoblin wrote:Personally I'm want logic circuit connection for furnaces/assemblers for controlling inner buffer size. With default furnaces buffer size equal stack and assemblers buffer size equal one item.
Why stop at the circuit network? Just copy the Stack Size Override feature we got with the inserters in v0.15 and apply it to output buffers on everything.

Post Reply

Return to “Ideas and Suggestions”