Factorio Forums

I finally got a .13 factory to "end-game" tech levels, ie blue belts, ass3, and level 3 modules. I wanted to build a green circuit plant with maximum productivity+ speed (beacon) bonus.

Since with prod3 modules cable:circuit is about 1:1, the setup is simple:


Every plant has 4 productivity modules and 6 beacons for speed 4.25 (+240%) and productivity 1.4.

This should produce 11 circuits/s using 9.5 copper and 7.9 iron, with 23.8 cables/s as intermediate (see below). From a simple test setup, a stack inserter can move 26.5 items/s between containers (so the cable is fine with a single inserter), 12.5 items/s from container to belt, and 11 items/s from belt to container (probably slowed by the jittering in picking up items from a moving belt). So, in theory a single stack inserter should be enough for all input and output.

What I see, however, is that circuit production is occasionally stalled by lack of iron, while the cable production runs fine. This is strange, as iron requirements are lower than copper requirements (since the circuit plant runs a bit below max).

So, what I think is the problem is that the iron inserter only starts inserting when the iron input buffer < 2, and before the inserter gathers the 12 iron and puts them in the assembler, it runs out of iron. What I don't understand, however, is why the copper input doesn't stall in the same fashion (the screenshot shows two inserters from the copper belt, but it runs fine with a single inserter as well). Does the output buffer come into the calculation at any point?

A workaround is to put a chest between the iron belt and the assembler. This allows the stack inserter to work more quickly as it no longer needs to gather material, and the assemblers seem to work more or less full time:



So, my questions are:
1) Why can't a stack inserter keep up with iron input for the circuit production?
2) Is there a better way to deal with this than the buffer chest? (two stack inserters from a belt doesn't seem to help)
3) What is *your* design for end-game circuit production without bots?
4) Are there any "scientifically valid" numbers on stack inserter speed from/to belts/containers?
Edit: Taking the chest workaround and tweaking layout a bit, this is a design to fully saturate a blue belt with 4 circuit assemblers (each producing 11 circuits/s, blue belt has capacity 40/s). Input should be 28 iron and 35 copper plate per second. The copper cable is divided over two belts, so you can place them side by side and share the copper belt. Note that current design has a shortage of modules, I'm working on it
I didn't quite like how flat this design is, so here's a more vertical (but somewhat less compact) equivalent. This saturates a blue belt from one side:
All comments/feedback welcome!

The problem is that the inserters wait for the input buffer (inside the assembler) to drain to a certain level. I think it's 5 iron plates in this example.

Only then the inserter starts gathering up iron plates from the belt in order to insert it into the machine. Unfortunately, the gathering process takes too long, and by the time the inserter decides it has finally had enough and inserts the plates into the machine, the machine has already run dry.

That's why you get the frequent stalling. The solution is to use 2 Fast Inserters instead of a single Stack Inserter. I don't know another way to work around this issue.

We had discussed it in another thread a while ago.

Right, that's what I thought, but what I don't understand is why the copper cable plant is not affected the same way, as it has identical input and speed?

Also, I guess my chest workaround is another way around which probably has higher throughput and only requires a single inserter facing the assembler, but it takes a bit more space around it.

BTW, can you link the thread you mention? I can't seem to find it...

In my setup even 3 fast inserter caused stalling (but way less frequent than 1 stack inserter). I needed bufferchests everywhere and then it worked^^ I posted it as a bug but the reply was: We know about it, not a bug. Use Bufferchests

I agree that it's a bug or at least inconvenient, I would suggest increasing the input buffer input to max(recipe_input*2, inserter_stack_size). Can you link your bug report?

viewtopic.php?f=6&t=30368

I would think that if a single stack inserter can pull enough off the belt to keep the buffer chest full, then the inserter should be able to do the job itself, without the chest. At that point its just a timing issue that could be fixed in the code.

when the inserter takes on the belt, it waits a few tenth of a second if the stack is not complete or if it can not be taken.
When he takes a factory or a safety he does not expect the next item on the belt and takes all he can in one shot, it is more stable

I put a screen again later, but I use chests as buffers

vanatteveldt wrote:3) What is *your* design for end-game circuit production without bots?
4) Are there any "scientifically valid" numbers on stack inserter speed from/to belts/containers?

Well a stack inserter from a chest to a belt does 766 units per minute. That is scientifically proven, I've measured it in game with a combinator contraption. Chest->Chest is easy, it's 2.31 * stack_size units per second.

For end-game production you could use this (doesn't stall, does about 2950 EC/minute), no bots: Somebody on Reddit has recently posted a better setup even (slightly more energy efficient): https://www.reddit.com/r/factorio/comme ... build_013/

But for actual end-game production I prefer to make the electronic circuits together with the Advanced Circuits and not put them on a bus. That does make more sense to me since that's where you need almost all of them (and for making PU, also crafted in-place).

I actually put together one of these builds back in 0.12:



That one ran at the speed of the copper wire assembler (i.e. fast inserter throughput in 0.12 was not the limit), I haven't tried my hand at redesigning for 0.13 but it could certainly be a little more compact, certainly the long-handed inserter to chests wouldn't be needed at all.

my incredible template to produce more without bots (Factorio v0.13.8+)
in theorical, this can produce more than 4k EC/m in theorical, this can produce 4 "Module of production 3" per minute my calculation (my MOD helmod on factorio MOD site)

siggboy wrote:Somebody on Reddit has recently posted a better setup even (slightly more energy efficient): https://www.reddit.com/r/factorio/comme ... build_013/

The 8x8 layout is the best for minimizing module costs. But I prefer this version of it.

@Dave, thanks, that looks nice and compact (and certainly neater than mine )

@Helfima, your mod looks fantastic, I'll check it out!

DaveMcW wrote:The 8x8 layout is the best for minimizing module costs. But I prefer this version of it.

Really nice. Certainly better than the thing shown on Reddit. I'm pretty sure it's enough to use 2x Fast Inserter for the inputs from the belts. Going to test that now.

BTW, Dave, have you made new versions of your modules for making Red and Blue Circuits that you've used way back for the "1 Minute Rocket Defense" build? When you've published it you've shown the blueprints for these modules, and I really like them.

So I'd be interested in your versions of these for 0.13.

I'm still happy with my high-volume red circuit design. Maybe you could clean it up a bit with stack inserters, but inserters were never the limiting factor.

I would probably not use the blue circuit design now, simply because the endgame does not require that many blue circuits anymore.

DaveMcW wrote:I'm still happy with my high-volume red circuit design. Maybe you could clean it up a bit with stack inserters, but inserters were never the limiting factor.

Yeah, that's the one I was talking about. I've cleaned it up a little, also since I don't want to craft Plastic Bars on the spot (and, of course, to get rid of the interleaved belts at the output; I don't want to get visited by the Belt Police, I'm sure you've got them in your pocket so probably not an issue for you ).

What about this for a simple vertical layout? It is tileable so each beacon should affect 8 plants (4 on each side) and each plant has 8 beacons. I didn't do any real calculations, but a single copper->circuit + copper should be able to supply 8*1.4=11.2 plants, probaly just 11 because the circuit is not at 100%. All plants are at 5.5 crafting speed (4 prod 8 speed), so this should produce (1/8)*11*1.4*5.5=10.5 red/s consuming 15 plastic and green and 30 copper (so I guess I need to upgrade these belts to blue to get 100% production, but I first need to upgrade my oil refining to get more plastic...).

That layout looks great, I have to try it out on my sandbox map. By the way, you should install Creative Mode, then you get item providers and consumers, makes it a lot easier to create these kinds of test setups. I've also made a small map and pimped it in the map editor, putting concrete everywhere and removing the trees.

@siggyboy: Good tip, I'll have a look at the mod!

I really like that mod

New screenshot: It doesn't quite get to theoretical potential of 635 red/minute, but gets close (611 red). The problem is copper cable, I divert some from the top assembler to help the circuit production, but now the cable runs out (otherwise the circuits do, I think). Anyway, it's good enough for me, and I like that it is a compact rack that can easily be replicated to both sides. It 'wastes' two beacons by moving the copper cable down to get room for the extra inserter, maybe something smart can be done there to save one row and align the assemblers to the beacons again.

I made a version of the green circuit inspired by DaveMcW's setup above. It loses some efficiency but it can still mostly saturate a blue belt without using belt braiding:
viewtopic.php?f=202&t=33355

[Koub] Topic moved to Show your creations.
Now that the designs are complete, they deserve to be in "Show your creations"