Science & Industry

For an extraction planet: if two t1-facs are routed into 1 t2-fac and a t1-fac cant deliver/doesn't have a cycle due to lack of r0, the other t1 will still try to send it's ressources, if the input is full will the ressources be destroyed, requiring routing over the launchpad/silos?

For a factory planet: is there any problem with direct routing, considering constant production?

On a side note: Are there any plans to improve routing? (like giving the option for facs to auto-gather required ressources from pads/silos or better visualization for existing/creating routes)

Greetings

I assume that means "yes" to the first question - material would get destroyed. Why would i do that for a factory planet?

You always should be routing

(extractor/processor) -> storage (launchpad) -> processor -> storage

Now, unless they've changed it (I haven't done PI in a while), you have to do more than just "route over" the storage structures -- that is, you literally have to tell the extractor (or processor) to dump to storage, and the next processor to pick up from that storage.

The reason you do this is because you will practically never have the ticks right, and so the storage buffer keeps you from losing the PI goo when one processor finishes before the one that will be using the stuff. Doesn't matter if it's an extraction planet or a factory planet, no buffer means lost product, and lost profits.


Assuming a 5/5 factory, you'd have a setup like this. (there are better ways to cluster - see youtube - but that's hard to draw )

1 -------\.........../------- 6
2 --------\........./-------- 7
3 ---------- LP --------- 8
4 ---------/.....\---------- 9
5 --------/........\--------- 10

Well there is always possibility of timing glitch, lower tier factory pushing into higher one before higher one has completed its cycle. Unless you stagger the factories for couple seconds/minutes when setting up and always keeping it running. I personally just route through pads instead.

Yes, the resources will be destroyed. The idiot workers will just push stuff out the airlock if there is no room. In the real world, this is called malicious compliance, or "work to rule." Always route stuff through a storage buffer. Sometimes the idiot is between the keyboard and chair. I speak from experience on this.

Well, sort of... but because each factory is a "different thing" they can get weird

Say you have a 3 stage factory planet

Basic Factories 1, 2, 3, and 4

Intermediate 5,6

Advanced A


1,2 feed 5
3,4 feed 6

5+6 feed A


Now, "on paper" it looks like what you're thinking:

0.00 (nothing outputs, 1,2,3,4 pull inputs)
0.30 (1,2,3,4) push to (5,6)
1.00 (1,2,3,4) push to (5,6), which start
1.30 (1,2,3,4) push to (5,6), which are halfway through their cycle (and have their buffer filled)
2.00 (1,2,3,4) push to (5,6), which finished and pushed to A <-- it falls apart right here

Thing is, there is no guarantee that at hour 2, your factories (5,6) will dump their product (and then clear their input buffer) before factories (1,2,3,4) finish and dump their product to the (still full) buffers.

So, using the launchpad, at hour 2, the ticks might look like this

1.59.59 --> (1,2,3,4) have 1 sec left, (5,6) have 1 sec left, each has a full buffer (2 units of P1 input)
2.00.01 --> (1,2,3,4) finish and dump product to the silo, there are now 4 P1 in the silo, (5,6) have 1 sec left
2.00.02 --> (5,6) finish and dump 2 units of P2 (A) immediately grabs. (5,6) clear buffer, then each pull 2x P1 (total 4 P1) from the silo.

Current status is then:
(1,2,3,4) -> running, full buffer
(5,6) -> running, full buffer
(A) -> running, empty buffer.

Or maybe (5,6) finish at 2.00.01 and then (1,2,3,4) finish at 2.00.02 -- but there is no way you can, with 100% certainty say that it will occur one way or the other. So dump it all to buffer, and let the factories keep themselves supplied from somewhere that won't destroy the materials.

Or maybe (3,4) and (5) finish at 2.00.01 ... since (3,4) are only feeding (6) their materials are destroyed. When (1,2) finish at 2.00.02 factory (5) gets the buffer filled halfway, leaving you always having a half-hour lag between 5 and 6. Now that 5 and 6 are out of sync, their feed of factory (A) gets out of sync and ends up wasting even more materials.

The real question is why after asking the question, and getting the best possible answer (hedge your bets and always route to storage) that you keep fighting to do it your way? If you want to do it your way, do it your way and find out if you lose anything over time.

Always route to storage or launchpad first. Anything else is either a beginner mistake, or overthinking it...