Item Recycling/Recycler/Scrap Yard (merged topics)

[MF-]

and cause stacking trouble (items with different raw resource value would be technically different)

[kovarex]

The recycler would have customized recipes for all items.

[Arakasi]

and cause stacking trouble (items with different raw resource value would be technically different)

depends on considering ID only or also some metadata in the container. I think its not necessary to differ two "same" items depending on their construction process.

[MF-]

and cause stacking trouble (items with different raw resource value would be technically different)

depends on considering ID only or also some metadata in the container. I think its not necessary to differ two "same" items depending on their construction process.

Exactly, there should be no difference.
AFAIK stack = one item data + count (therefore the items must be exactly the same).

Keeping separate data for each item in the stack would defeat the codewise purpose of stacking and have negative network game bandwidth impact.

[rlerner]

I came up with an idea of a dense coal mod that allows an assembler to combine 10 coal into one "dense coal" with a fuel value of 10, into stacks of up to 256. It allows much more compaction and even better storage.

A coworker of mine had finished the mod as I wasn't getting it working.

But what I propose here is the reverse -- A way to get raw materials out of a product.

Since not all items are directly replaceable (i.e. Fast Underground Belts do not require a normal Underground Belt -- Small Powerlines are not required for Medium / Large Power lines, etc), I often tend to box these up and destroy them as my factory grows.

The scrap yard would allow reversal of non-smelted (i.e. crafted or assembled) items.

Small Power Line = Wood + Copper Wires

Copper Wires = Copper Plates

It would allow a way to recycle old materials, or re-purpose excess production. It should be a technology that can be researched (like Solar Panels, i.e. Green Technology) and should only cost electricity to split into the parts.

Thoughts?

[FreeER]

Not to be rude but this has been mentioned a few times (try searching the forum for 'recycle'). However, if I am not wrong I believe the devs have said this will be implemented, eventually (it's not a priority).

[rlerner]

If so, my mistake. My work blocks this site (not Factorio.com, however), so it's rare that I can get on here and read up (and see all of the ideas), and I've always had issues with phpBB's (gag) search.

If a mod would like to merge the topics, or just append my post as a vote++, that would be fine.

[DexterNemrod]

IMHO a good idea ...
but it should work with slight losses (after all, that´s what you normally have in recycling as well)

Maybe it could work like some kind of reverse factory ... you enter which recipe your scrapyard factory should produce and you enter the products and afterwards get the raw matedrials out of the output:
Example:
Recipe: Scrapping Copper Cables:
Needed: 5 Copper cables
Output: 2 Copper plates

(normally you would need only 4 copper cables, the 1 cable more should reflect the losses due to not being able to recycle with 100% efficiency)

[MF-]

Note that items are allowed to have multiple recipes.

[DexterNemrod]

Which is no problem if the assumption made by you in another post above is correct, that is:
Stack = ItemID + Count

With ItemID being the same for an item regardless of the way it was produced.

To reflect the different ways of construction the item could just have different recipes for this item in the scrapyard/recycling facility,
so that you can decide for yourself whether the item/s should be broken down into w items of resource A and x items of Resource B
or into y items of Resouce B and z items of resource C.

(but of course the availability of multiple ways of construction makes it necessary to carefully balance the recipes of deconstruction, so that, in no way, you will gain more resources than you invested by constructing and afterwards deconstructing an item ... but rather have a slight loss)

[ssilk]

I think we should avoid to maintain a second "recycle tree". This just makes much more work. Instead we should think about what recycling really is. When you look at the process of recycling, you see, that it is a juggling with probabilities. I would like to have that also. Recycling has two components: the input material and time. And the output depends very much on random.

Examples:

I try to begin with the simplest: An iron gear wheel is made of iron plate, this takes 10 secs ( https://forums.factorio.com/wiki/inde ... gear_wheel ).
So if you recylce it, by a probability of 1:2 you get either an iron-gear-wheel or an iron-plate. This "decision" takes 10 seconds. If it remains as iron-plate, the process is over and the plate is put out.
But if it remains as iron-gear-wheel, the process goes into the next round: By a probability of 1:2 the wheel remains a wheel. This decision takes 10 secs.
You can imagine, that after four rounds, the probability to get iron-plates is 16:1 - which means every 16th iron-wheel remains as wheel. Now 40 seconds are over or 4 "turns".

This is the limitation of the recycler: It has only a limited time or number of turns to deconstruct things!

What does this mean? In this case: by a chance of 1:16 an inputed wheel, is not recycled! Some electricity is burned, some time has been taken, but nothing happened!

Lets take for example an transport belt: consists of wheel and plate. And a wheel consists of plate. Now you have 3 paths: Either it remains as transport belt, or it gets a wheel and a plate or only plates. The probability, that it remains as belt sinks to 1:3. But the time, needed to output only as plate rises. But now, by probability of 1:3 there it remains as wheel and plate. Then - in the next turn - the wheel is tried to recycle: See above!

You see: The calculation is relatively simple:
Step 1: Look, which ways it was possible to create an item. For each way, the time of all former steps are added.
Step 2: Remove the steps, which are not possible to obtain in the time, the recycler has left (for example maximum 40 secs).
Step 3: dice, which step will remain. Subtract the needed time from the time, the recycler has left.
Step 4: By random take one item you have now and try to recycle that (Step 1)

Finish after N cycles or if the recycler has no time left to recycle any item or if the internal recycler storage is full.

In other words: This means for example a blue potion is recycled to a fast inserter, some belts, iron plates etc. Or some other combinations. This is also, what comes out from the recycler.

Of course: you can reinsert the recycled stuff. Now the fast inserter gets a small inserter and a belt and some other stuff. And so on. So - like in real life - deconstruction is as complicated as construction because it need sorting out of needed and unneeded stuff. (*)

I think to built a working recycling factory could be really, really interesting (!), because you need to think with probabilities and how to maximize probabilities to gain a special item. Because it may need to recycle something and then rebuild something other to get what's needed.

And it offers really interesting things! Think for example to an item, which is not researched yet. But there is another item, which can be constructed in 2 different ways: one way includes the not researched item, the other the ready researched. So instead of researching the first item, you can build the second and recycle it and by some probability you will get the not-researched. Could be a very cool strategy.

The rest is tuning: How much time has a recycler? Maybe it makes sense to double his speed or every recycle-step takes a fixed number of time? Will also lower the needed calculations... How much cycles will he do? Endless, till time is over? How many internal storage will he have? ... Edit: Maybe if not enough time is left, items are dropped?


(*) Filtering only the needed stuff is currently very complicated with filter inserters. So by logic, I think we need by minimum a filter inserter, where you can inverse the filter (= insert everything but this!) or - as I already suggested - a filter splitter (see the other thread).

[ssilk]

And to add that another strange idea: How about inserting not only one item, but many?

For example: If you input 3 iron plates, you may get a belt by low probability of course.

This idea is of course not thought to the end as above.

[ssilk]

Instead of loosing materials because of deletion from map, the item should be recycled in some way described above and the material should be spread in the near surrounding of the item, you deleted.

Example: removing an electric pole, the red- and green-wires are lost! Instead the wires are recycled, maybe you get wires and alectric circuit... just need to fetch is up...

[MF-]

Instead of loosing materials because of deletion from map, the item should be recycled in some way described above and the material should be spread in the near surrounding of the item, you deleted.

Example: removing an electric pole, the red- and green-wires are lost! Instead the wires are recycled, maybe you get wires and alectric circuit... just need to fetch is up...

Even extra regular copper wires are lost.
Yes, wires should be accounted for. I already suggested that: https://forums.factorio.com/forum/vie ... ?f=6&t=367