0.16.7 Belt Compression... Whats the big deal?

Circuit-free solutions of basic factory-design to achieve optimal item-throughput.
Involving: Belts (balancers, crossings), Inserters, Chests, Furnaces, Assembling Devices ...
Optimized production chains. Compact design.
Please provide blueprints!
Forum rules
Circuit-free solutions of basic factory-design to achieve optimal item-throughput
Post Reply
Footy
Burner Inserter
Burner Inserter
Posts: 9
Joined: Mon May 01, 2017 11:19 am
Contact:

0.16.7 Belt Compression... Whats the big deal?

Post by Footy »

So I've read a lot about belt compression and seen various solutions...
But it seems to me (unless I've missed something big) that there is a really simple solution.

Old belt compression method used the following methods, the underground and splitter inserter (not shown) used a speed work around and the side loading was loading into a belt without sufficient space.
Image

These methods always have felt like small exploits but things we have done because we are lazy...

0.16 has a few new/old method(s) to ensure compression, all we need to do is modify the designs...
Image
This I think will become the default output method to build your new designs around, it's simple and effective.
I will post a few designs below for various assembler setups and even an early game smelter setup. (They are all tilable)

Image
here we have (left to right) 1in/1out, 2in/1out, 2^2in/1out.
The Red Science and yellow Inserter are capable of filling the output belt to full density, the Assembler 2's is not (due to lack of input resources).

Smelter Below
Image

They are all built around a simple concept of output onto an isolated belt then split it into the main output.

Train Stations....
I have a couple of options they both fill 4 belts to full compression without need for too many upgrades.
Image
They essentially have the same goal, send 2 belts with more ore than 1 can handle into a splitter then pull 1 output. end result is a fully compressed belt.

These solutions are not complicated or require lots of tech or circuits.

So given we can mine, smelt, assemble, unload from trains without compression issues, can I ask what the problem with compression is?

Footy

spaghetti335
Burner Inserter
Burner Inserter
Posts: 6
Joined: Fri Sep 08, 2017 4:59 am
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by spaghetti335 »

You jerk. You're really going to enlist my help in making these and you're not going to give me credit?

Also, train unloader #1 is all my design :) #2 was all Footy though.

Jap2.0
Smart Inserter
Smart Inserter
Posts: 2339
Joined: Tue Jun 20, 2017 12:02 am
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Jap2.0 »

(Almost) Everyone agrees that underground belts were cheaty. Sideloading is somewhat controvertial, but in many cases it makes sense and the community agrees that it should work. There are several other topics on this, notably:

viewtopic.php?f=38&t=54945
viewtopic.php?f=38&t=55565
viewtopic.php?f=5&t=55005
viewtopic.php?f=7&t=54627

So I doubt this will add much to the discussion. Perhaps it would fit better in show your creations?
There are 10 types of people: those who get this joke and those who don't.

Koub
Global Moderator
Global Moderator
Posts: 7175
Joined: Fri May 30, 2014 8:54 am
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Koub »

[Koub] Moved to Show your Creations > Mechanical Throughput Magic (circuit-free)
Be aware that splitters do not always compress perfectly (in fact, they randomly lower compression on fully compressed belts), and the devs are still working on the belt mechanics, so all this might be useful for just a few days.
Koub - Please consider English is not my native language.

Dragony
Long Handed Inserter
Long Handed Inserter
Posts: 77
Joined: Sun Jul 09, 2017 8:13 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Dragony »

Keep your fucking Rube Goldberg machines for yourself. If THAT is what my factory is supposed to look like, I will just stop playing the game once and for all.

Actually, I am here to plan and build an optimized, elegant and good looking(!) Factory. If I want to build stupid Rube Goldberg machines, I can just play Crazy Machines.

Don't you even REALIZE anymore that the simple task "Put product on the belt" needs even more space than the task "Produce product"??

Koub
Global Moderator
Global Moderator
Posts: 7175
Joined: Fri May 30, 2014 8:54 am
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Koub »

[Koub] Please stay civil. You may agree on the fact that you disagree, but no aggressivity, no personal attacks. The last post on this topic is really borderline.
Koub - Please consider English is not my native language.

Footy
Burner Inserter
Burner Inserter
Posts: 9
Joined: Mon May 01, 2017 11:19 am
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Footy »

I took it as a joke Koub, I'm sure everyone appreciates brilliant designs.

Dragony
Long Handed Inserter
Long Handed Inserter
Posts: 77
Joined: Sun Jul 09, 2017 8:13 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Dragony »

Footy wrote:I took it as a joke Koub, I'm sure everyone appreciates brilliant designs.
You must be a very happy being....

User avatar
5thHorseman
Smart Inserter
Smart Inserter
Posts: 1193
Joined: Fri Jun 10, 2016 11:21 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by 5thHorseman »

Hopefully this won't anger anybody like every other post in this thread has :)

I don't put a lot of thought into my designs, and I borrow things from others like we all do, but I came up with this modification to what I consider a simple block of 48 furnaces (which is what a single red belt can supply with ore with no bonuses). My original design was to just run the ore and coal down the sides and feed the plates into the middle, using underground belts to compact. With those gone, I reworked it to instead split the 48 furnaces into 2 sets of 24, and have each half-fill a belt that meets its sister belt in the middle in a splitter. As you can see, it does a pretty good job of compressing that belt.

The gap is 6 wide because of 2 reasons: 1) That's how far red undergrounds stretch and 2) my train loading blueprint accepts 6 input belts, so that's the maximum I can really utilize.

The gap will allow me to stack this 6 times. I've so far stacked it twice, which is why the other belts are currently not doing anything. The 6-6 balancer (not my design, I got a blueprint book from this very forum that has all balancers from 1 to 8 belts. Very handy) allows me to quickly add more furnaces later and just have them work, now that it's all set up.

I'd appreciate constructive criticism as always :)

Image

(and yeah it's not getting any ore right now. Adding that 2nd row shifted the supply problem to the iron fields, which I now need to expand)

spaghetti335
Burner Inserter
Burner Inserter
Posts: 6
Joined: Fri Sep 08, 2017 4:59 am
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by spaghetti335 »

It's a nice design Horseman, and particularly nice for you as it does what you need. In that sense, it's perfect.
Feedback will depend on what specifically you want feedback on.

Compressed belt? I haven't tested it, but I'm pretty sure you've got that nailed. At worst, you've got ~99% compression.
Compact builds? There's room for improvement. You've got spaces between those furnaces; I'd think there'd be a way to get rid of those.
Beauty? I prefer the ones up top, but beauty is in the eye of the beholder.

It all depends on what you're going for. Regardless, it's always fun to see someone else's build :)

User avatar
5thHorseman
Smart Inserter
Smart Inserter
Posts: 1193
Joined: Fri Jun 10, 2016 11:21 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by 5thHorseman »

spaghetti335 wrote:Feedback will depend on what specifically you want feedback on.
Nothing in particular. I just find I tend to overlook obvious flaws once something is functional.

Caine
Fast Inserter
Fast Inserter
Posts: 213
Joined: Sun Dec 17, 2017 1:46 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Caine »

@5thHorseman

Personally, I am not a big fan of the sideways extraction of the furnace content like that. It is a fairly big, resource heavy footprint which does not scale easily and cleanly unless you know exactly how many rows you will build ahead of time.

A simpler variation of what you are doing can be seen in this image: https://imgur.com/a/23rUT
Source: https://www.reddit.com/r/factorio/comme ... ltersetup/
Though it also wastes resources due to extra output belt over the half length of the array.

In a recent collab with Xterminator, Katherine of Sky shows a very simple and fairly clean solution at the 16 minute mark (though she can save a bit on undergrounds):
https://youtu.be/drXOAyPZU0M?t=962

milo christiansen
Fast Inserter
Fast Inserter
Posts: 106
Joined: Thu Jul 30, 2015 7:11 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by milo christiansen »

That is actually a kinda crappy design. Here is a more compact one:

Image

Note how it runs all the belts down the center? That saves several tiles of width. Now that the underground compression exploit has been fixed I make paired rows, splitting input and merging output. Before each of these lines would have been twice as long and each would be fed independently. The only other change I made to these for .16 is the balancers on the output. Once side loading is fixed these could be removed.
Blueprint

Footy
Burner Inserter
Burner Inserter
Posts: 9
Joined: Mon May 01, 2017 11:19 am
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Footy »

how is that more compact than my design?

milo christiansen
Fast Inserter
Fast Inserter
Posts: 106
Joined: Thu Jul 30, 2015 7:11 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by milo christiansen »

It isn't more compact than your design, but it is much better than the previous two designs posted (by 5thHorseman and Caine). I should have been more clear.

Caine
Fast Inserter
Fast Inserter
Posts: 213
Joined: Sun Dec 17, 2017 1:46 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by Caine »

My point was mainly to show that compression has some very simple solutions. But indeed, your design is way better if compactness is your primary criterion. In fact, I think that only train smelting is more compact than this (with all of its own pros and cons). Your solution is a bit on the expensive side perhaps (lots of undergrounds and splitters). Do you even need those balancers at the end? Aren't both sides naturally balanced?

I tend to utilize the version shown by Katherine. Personally I detest long handed inserters and try to avoid them wherever possible. My preference is also to limit underground usage as much as possible, but sometimes you just need them to satisfy a greater goal.

You are solving belt compression in the horizontal dimension, do you also have a solution for a vertical arrangement (i.e. 48 or more furnaces in a single column)? Your setup will become very wide, especially if all your resources flow in a single direction (e.g. from south to north).

milo christiansen
Fast Inserter
Fast Inserter
Posts: 106
Joined: Thu Jul 30, 2015 7:11 pm
Contact:

Re: 0.16.7 Belt Compression... Whats the big deal?

Post by milo christiansen »

You can arrange it as one big column, but then there is a problem with compressing the output.

Possible solutions include belt braiding (could be done with yellow and red belts) or timing the inserters with a circuit.

The output balancers are there to merge both lanes into a single lane to fix side merging compression issues. You could probably merge both outputs with a simple splitter or a "demand" balancer as well. I have a large factory making belt parts and inserters, so I really don't care about cost, only footprint. In this case the footprint is the same either way since the furnace line sets the width and this is probably a bit shorter than a single even-pull demand balancer would be.

Post Reply

Return to “Mechanical Throughput Magic (circuit-free)”