#Pin Miner with overflow system

1 messages · Page 1 of 1 (latest)

turbid steppe
#

For v. 4.0 onward, linked below:

  • produces 75% of P-P-P-P- or P-P-P-P-P-P-(hex) of the total mining output, works for any number of extensions, adapts autonomously
  • compatible with any mode and difficulty (hex, insane)
  • 12 outputs, distributes equally across all 12 belts for any number of extensions
  • integrated overflow system
  • wireless

Requires a full shape as primer, third floor and overflow splitters unlocked, slow start-up time.
It's a combination of the following versions, thank you for sharing them:

Update 16.04.26: Major changes are coming in 1.1 regarding shape miners. The concept of a pin miner will presumably not work for 1.1 and beyond

📎 PM_3_4_W.spz2bp 📎 PM_3_4.spz2bp Screenshot_331.png
subtle vector
#

just wondering, is this also the new method FatcatX made (Sikerow's Pin Maker)?

turbid steppe
# subtle vector just wondering, is this also the new method FatcatX made (Sikerow's Pin Maker)?

All the methods, including FatcatX', use the same principle. They just reshuffle the position and layout of buildings. The biggest distinction, if even noteworthy, is how each version deals with scaffolds, shapes that are recycled.

Some versions can accumulate scaffolds faster and have a unique detection system locking the filters. Too few scaffolds in the cycle and it's too slow, to many and it can stall, also losing efficiency. The faster these can accrue, the better the start-up time.

cerulean lotus
#

I had always hoped that integrating overflow was possible. I'll have to look at this later to see how it works with any number of extensions too (that's just amazing)

turbid steppe
small pikeBOT
turbid steppe
small pikeBOT
# turbid steppe New 3.0 version! 👇

Blueprint Infos :

Version : 1122 / 0.1.1, Blueprint type : Platform, Blueprint cost : 2,060, Platform unit cost : 2
Building count : 355, Building size : 12x14x3, Building tiles : 437
Platform count : 1, Platform size : 1x1x1, Platform tiles : 1
Icons : layout.Layout_ShapeMiner, building.PinPusherDefaultVariant, <empty>, <empty>

cerulean lotus
turbid steppe
#

Edited the top message, it now adapts dynamically, no more clearing belts required 😎

cerulean lotus
#

also something that is very easy to overlook: your miner (at least the version I tried) generates a warning ("conflict" overlay). this may or may not matter to you. some players find it extremely annoying
(I turned them off at about the 100 hour mark, but they are on by default)

#

under what scenarios does it need to have 2 belt readers?

turbid steppe
small pikeBOT
# turbid steppe Easy fix: no warning signs should appear now

Blueprint Infos :

Version : 1122 / 0.1.1, Blueprint type : Platform, Blueprint cost : 2,067, Platform unit cost : 2
Building count : 356, Building size : 12x14x3, Building tiles : 438
Platform count : 1, Platform size : 1x1x1, Platform tiles : 1
Icons : ShapeExtractionGroup, building.PinPusherDefaultVariant, <empty>, <empty>

cerulean lotus
#

you don't even need the signal producer, it can just be an empty tile, right?

turbid steppe
turbid steppe
cerulean lotus
#

Are these different for a reason?

image.png
#

How long should it take to stabilize on full throughput after pasting in? I'm still getting gaps in floors 1 and 3 after like 10 min

#

I pasted it with 3 extensions already present

turbid steppe
cerulean lotus
#

itty bity gaps (like 99% throughput) are unfortunately acceptable due to the wierdness of loops in the simulation, but these gaps are like missing whole shapes and a half

turbid steppe
cerulean lotus
#

this overflow splitter is diverting shapes occasionally (which is just a symptom, not a full problem)

image.png
#

do you have a spread out model of this whole thing?

turbid steppe
# cerulean lotus this overflow splitter is diverting shapes occasionally (which is just a symptom...

It's actually the cause, because it shouldn't divert any after the cycle is saturated. The same issue is also present in the old wireless version, it boils down to overflow splitters acting very aggressively.

The 2.0 version shouldn't be affected by this, since there are no overflow splitters like here. Not sure this can be remedied while keeping it adaptive, I've tried all sorts of methods to mitigate rogue shapes on the wireless setup

cerulean lotus
#

I would be trying to get the 3 extension case working as flawless as possible.

cerulean lotus
turbid steppe
# cerulean lotus itty bity gaps (like 99% throughput) are unfortunately acceptable due to the wie...

Did they occur previously? Or is there a noticeable and consistent difference between 2.0 and 3.0?

While testing many different pin miners, including all the ones linked, gaps also appeared when in combination with fully saturated space belts, and they are all the same in terms of size and frequency. I'm yet to find a solution that is almost spotless. These gaps don't appear to follow any clear patterns, and more interestingly, almost completely dissappear when not using space belts.

I don't believe this to be only a quirk of the simulation (99%), but rather due to some innate property that all pin miners share coupled with how space belts handle and balance load.

All pin miners have one big loop of stacking, swapping and cutting. With no space belts, it looks like it can find a stable equilibrium to keep the cycle constant, while with space belts, it slightly disrupts it which leads to micro-outages.

cerulean lotus
# turbid steppe Did they occur previously? Or is there a noticeable and consistent difference be...

My primary testing environment consisted of this:
bypassing the splitters on the output (replacing the first split with a straight belt) then having the 3 output belts go straight for the length of a 1x1 then into trashes.
I have not done any sort of testing with space belts yet. Pin miners are very complex so I'm taking it very slow.

I have very heavily modified your miner (it doesn't matter which version because it's so different) into a new version that appears very reliable (I haven't tested space belts or trains at all yet). My new version only works with 3 extensions (actually it may sorta work with 2... I'd have to test that) and also it outputs evenly to all lanes simply because I had the space for it. (It has integrated overflow protection)

turbid steppe
cerulean lotus
turbid steppe
#

I have very heavily modified your miner (it doesn't matter which version because it's so different) into a new version that appears very reliable

Since you didn't use space belts, at least from version 2.0 onward you shouldn't encounter gaps, I tested this at 1x speed and it's pretty much perfect. It doesn't surprise me your modified version works too, but what does is why you claim this test failed on your end.

The gaps I am referring to are the ones that appear when using fully saturated space belts, anything else should be fine.

Screenshot_335.png
cerulean lotus
#

I may have simply not waited long enough

#

I was staring at it the entire time so it probably wasn't that long

#

I don't remember

turbid steppe
#

But was it 2 hours? I doubt that 😅 Give it at least 5 minutes, it's ludicrously long but then we know it's only a matter of start-up time.

cerulean lotus
#

yeah it was 5-10 minutes

#

yeah I have re-created it. it appears to be steady at full throughput but then there are large gaps. I assume it needs a very long time to stabilize

turbid steppe
#

So in essence, we know for certain that your version gets to 180/m more quickly. Yeah as I said before the start-up time is much slower, due to how little room we have to kickstart the thing. That's not a problem, just a property. But now I'm curious, mind sharing yours?

cerulean lotus
#

quick question, what are these for? I'll share mine right now if they are needed otherwise I will tweak it to remove them then share it

image.png
turbid steppe
cerulean lotus
#

I noticed a back up that slowly grew over time, which those would prevent so I kept them

#

(over a few hours)

#

I can't call this flawless yet but here's the WIP:
(3 extensions only)
(untested in hex)

image.png 📎 message.txt
small pikeBOT
turbid steppe
# cerulean lotus I can't call this flawless yet but here's the WIP: (3 extensions only) (untested...

This is great but doesn't bring anything new to the table. If we only care about optimizing start-up time and an even distribution across all 12 belts (as oppose to just floors), then maybe this can work (altough it doesn't do the latter yet), but it loses every other perk as a consequence. (to clarify, reliability isn't affected, we've established it eventually works flawlessly for any version)

It doesn't work for 0 and 2 extensions, it doesn't adapt and doesn't ensure it distributes equally across floors for every case. A simple fix won't suffice here. It's just backpaddling to a more reactive but less flexible variant.

(and it's not like 10x faster when reaching 180/m either, altough hard to quantify, I have found it to be only marginally faster, maybe 1.3x - 1.5x when freshly pasted onto an asteroid)

Trust me the longest time was spent trying to meet all those requirements 😂 while keeping it reliable ofc, but the start-up period was the last of my worries...

Older Pin Miners are much faster than this, because their layout allows for scaffolds to be accumulated very quickly. So if I had to only build one with 3 extensions that saturates belts quicker, I would scrap the current version entirely, because it wasn't designed for this in mind.

turbid steppe
#

New cheaper version, less convoluted, it should stabilize faster compared to 3.0:

Screenshot_336.png 📎 PM_3.1.spz2bp
small pikeBOT
silver robin
#

Does this require the 3 layer upgrade to work?

cerulean lotus
turbid steppe
# silver robin Does this require the 3 layer upgrade to work?

Oh wow 😂 , unfortunately it does, I'll go out on a limb and say that this is currently impossible to achieve without, a 2 floor BP needs at least another 1x1 if we wish to preserve all functionality, but I'm happy to be proven wrong.

Alternatively, it could be done but with very limited throughput...

silver robin
#

All good I just got the game this week, working towards milestone 8 and trying to set up something more permanent, ty

turbid steppe
turbid steppe
#

New v. 4.0

  • wireless with optimal reliability (thoroughly tested)
  • faster start-up period
  • even distribution across all 12 belts (allows for faster space belt saturation)

In terms of building count, it's about as expensive as v. 3.1, and contains only about 20 buildings more than the very first iteration (1.0). This is likely the last major improvement for this miner specifically.

Direct Pin Makers, 2-floor mode, compatibility with every shape permutation etc. could be new ways to explore and improve pin production, but as for the miner, I believe this offers a very solid package 🙂

📎 PM_4.0.spz2bp Screenshot_337.png
small pikeBOT
sage verge
#

Very cool

modest glen
#

Am i blind or do these go nowhere? I swear I quadruple checked

IMG_1780.png
cerulean lotus
turbid steppe
turbid steppe
#

Tidied it up, final version:
Sorry for the frequent updates recently, this will be the very last iteration of this miner. Feel free to adjust to your needs 🙂

📎 PM_4.0.spz2bp
small pikeBOT
glass anvil
#

this is a masterpiece of engineering, absolutely love the design

sage verge
#

Yeah, a lot of us just use these and treat pins as the fifth shape

turbid steppe
#

Alternative version only for 1.0 and above:
On 1.0 mining tiles on shape miners are unlocked differently, which made routing and distributing easier. It contains fewer buildings and is more symmetrical 🙂

image.png 📎 PM_5.0.spz2bp
small pikeBOT
subtle vector
#

Alternative version only for 1.0 and above:
On 1.0 mining tiles on shape miners are unlocked differently,
May i ask what is different then ?

subtle vector
#

owwww.... now i remember the 1.0 change yes you are right

turbid steppe
modest glen
#

Why would it become obselete

sage verge
#

I forgot to post this here. It gets smaller when you remove the miner constraint. But it still is roughly the same size as any 4:3 pin maker layout

2026-03-01_at_23.14.18.webp
turbid steppe
raven furnace
#

I haven't followed very closely but afaik the feature in question will be an optional upgrade, the current miners will still exist

turbid steppe
raven furnace
#

maybe you should ask to be in the manufacture hard mode beta (if you're interested in providing feedback ofc), since the discussions about that feature happened there