Sushi thread

[hardy fable]

because if you have 1 belt then remove the item you'll have voids

[stuck scaffold]

https://cdn.discordapp.com/attachments/139677590393716737/1281468702839279688/image.png?ex=67299cbb&is=67284b3b&hm=91eb2f74e8f70c6b17edcb9edcaecf4af472a4d2aadb6174dd98697618e1b393&

[hardy fable]

oh yeah ok I see

[stuck scaffold]

Oh yeah building and using are different

[hardy fable]

on the last splitter there are 2 input belts so it works

[crude quest]

wtf am I looking at lol

[stuck scaffold]

1:10 mix lol

[hardy fable]

lemme guess, for ?

[stuck scaffold]

https://cdn.discordapp.com/attachments/139677590393716737/1283443827733958710/image.png?ex=672a34b5&is=6728e335&hm=07a663e9fe46e61cd110e56fb6dedb35ca618677a333f1de61f6fe72ea6d285d& 1:2;3 ratio

@hardy fable This is why my answer to your question was so poorly worded lol

[crude quest]

Something like this?

[stuck scaffold]

That's 1:1:1:1

[crude quest]

Whats the reason to go out of that ratio?

[stuck scaffold]

Uhh, so like, I tried to explain "This is how you get any compression ratio you want."

[crude quest]

In some of my loops recently I've gone to this 3 splitter setup that is more non-blocking. Is this sane?

[stuck scaffold]

I think there's a missing concept if you're using more than 1 for filter and resupply

[crude quest]

I think the missing thing is that circuit you had that stops overflow of certain items

[stuck scaffold]

Yeah that little circuit is a nifty little thing. "Everything = 9" means the belt doesn't move unless all specified input items are fully present, and you have a combo controller to turn it on and off as well as specify what you want

I would strongly advise it or something like it, for compression mixers. If you can find a more optimal solution I'll use that instead

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

[crude quest]

Cheers

Thanks for all the help

My experiments:

[stuck scaffold]

idr how to sync sides

Theoretically there's something there I just don't remember it

[crude quest]

sync sides?

One key thing I'd love to solve is: Having a massive backlog of various quality of an item, and "washing" through it all in a loop to get out the lower chance qualities without blocking when it gets a large string of higher chance qualities

[stuck scaffold]

Conceptually, prioritize the higher quality/lesser count

[stuck scaffold]

Side sync is that it's possible to have different ratios with the same items on different sides. I think if it's different items you don't need to worry about it but I'm dredging up old memories so let me double check

[crude quest]

Nar it'd be 1 input belt of mixed quality

[stuck scaffold]

Yeah so I was right about sides at least

[crude quest]

Basically I think I need a FILO - First in last out - to wash it all?

[stuck scaffold]

idk I havent' played with quality at all yet

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
6x2 to 1

[crude quest]

Well forget quality - Being able to wash out rarer items from a massive backlog

[stuck scaffold]

Looks like ratio

[crude quest]

Fuckin tasty man

Though I think it needs to loop otherwise it'll break on non full science research right?

[stuck scaffold]

Yeah, you know recycling tho, I don't need to demo that for you every time so I put a chest

[hardy fable]

there you go champ

I'll do a stacked version one day

[crude quest]

Oh I'll check it out tomorrow! Thanks. By stacked so you mean stack inserters?

[hardy fable]

yup

that's beautiful

that's good for 1200 spm, crazy

shit, even with turbo belts a single stack inserter is not enough

it works for recycling because it is fed stacked items

but I guess the input also has to be stacked to some degree

tho I always have chests there so it'll be ok

heck yeah let's just do that

let's go, I don't even need more than fast belts on the input

[normal gulch]

thats insane

I do have one question.. are inserters guarenteed to take a full stack off the sushi belt? Guess it doesnt matter in ur design

[hardy fable]

yeah it doesn't matter at all

even if the first inserter doesn't take a full stack, the next one will take the remaining

after all, 1200 spm is a lot of labs

just 3 fast inserters are enough

and I'm not even inserting in a chest like a lab is supposed to behave

OH

that's why

they added one more step in SA to match the stack size

explains why 2 fast inserters are enough

what's crazy is that both are working 100% of the time but no item is passing through

[crude quest]

Hrm so my factory already ships multiple qualities of science. I wonder if the splitters can filter for any quality of a type

Otherwise I'm in trouble

[hardy fable]

there's this

[crude quest]

Okay great

[hardy fable]

available in both inserters and splitters

[normal gulch]

you can do basically anything you want

[hardy fable]

I love this game

[crude quest]

You finished it yet? I refuse to hit aquilo until I setup good factories elsewhere first

[hardy fable]

lmao no

not at all

[crude quest]

About 100 hrs into this save

[hardy fable]

I ran away from nauvis like a coward cuz biters were getting annoying (also the lack of resources and map generation)

then did all researches on vulcanus

I just lack artillery and LDS productivity researches iirc

and rn I'm on gleba

designing the factory is one thing, but building it is another

so much landfill required 😭

[crude quest]

Hilarious your not in fulgora given your sushi proclivity

[hardy fable]

I'm not even sure I'll do sushi there

yeah I don't think so

I'll just sort

and then I'll recycle using conditions similar to petroleum cracking

like to get iron plates I'll recycle what I have the most at the moment between gears and iron sticks

[crude quest]

You kind of just recycle it down until there is nothing left.

[normal gulch]

I did sushi to start but im already in throughput issues with red belts and its making like 10 science a minute. next iteration will be sorted and much higher throughput

[hardy fable]

tbh I tried fulgora a tiny bit in my save (before reloading it) I did electric generation with the thunder and accumulators and that's about it

[crude quest]

Main issue IMO is actually the space to build

[hardy fable]

sounds kinda inefficient

[normal gulch]

Its easy; accumaltors + lightning + backup steam

[slim wave]

I thought I was gonna do sushi on fulgora but it turns out it really doesn’t call for it

[crude quest]

It is but the mine patches start at like 40m. So you didn't need to scrimp

[slim wave]

The overflow and throughput to consider makes sushi much more challenging than designated belts

[hardy fable]

I almost did sushi on gleba tho, to fit spoilage, nutrients and bioflux on the same belt

but ended up using 2, long handed inserters aren't that expensive

[slim wave]

Gleba makes more sense for sushi because the production chains aren’t so intertwined

Means there’s not as much complicated programming to do

[stuck scaffold]

I mean, you can do funny things with SushDI

https://cdn.discordapp.com/attachments/139677590393716737/1287843886626115685/image.png?ex=6729bbd5&is=67286a55&hm=8f27b82dbf2ec667650518c82d39f50bf7d5fca5d00b8d7453c1b290991abacc& Sushi's kinda good for intertwined chains specifically I think, if you can get all the inputs on one belt, then you only need one belt

[umbral gorge]

Cursed

Also my main problem with doing things this way is that if you are low on a resource then all of the ratios get messed up

[stuck scaffold]

Yeh, that's why I recommend the cutoff switch

[stuck scaffold]

Wire and a constant, not so bad

[tacit cedar]

https://factoriobin.com/post/obzjy1
latest one

simple circuitless rate limiters
though circuits seem to be baseline now (even if you still have to research combinators) so probably obsolete

[stuck scaffold]

Platforms are tight for space

[tacit cedar]

yeah these aren't great for platforms either

I do not know if they have a usecase anymore

[stuck scaffold]

At this point it's a "when the only tool you have is a hammer" problem, with sushi I think. People are forced to sushi and only have circuits

Mechanical sushi was always a very dark art

[hearty relic]

minus mine

[crude quest]

Man. There seems to just be no way to avoid the deadlock where when you get a recipe with more than 1 ingredient and you're trying to feed in qualities - Eventually one of the item types deadlocks the whole thing if a long strip of a quality comes though:

[hardy fable]

yup, I knew it would happen

since it is statistically possibly it will happen (aka Murphy's law)

[hardy fable]

wait, it's bound to happen if you don't have the same amount of quality "steps", what the hell have you done?

put the same amount, tier and quality of quality modules in ore and plates production

and then do not put quality in the gear production, else you'll unbalance everything

[tacit cedar]

this color scheme is really nice

[hardy fable]

I have yet to find a science pack color combination that doesn't look good 😄

[tacit cedar]

this is so sad why can't I add multiple filters to splitters

[hardy fable]

one splitter per item isn't that much bigger tbh

especially in comparison the the size of the rest 😅

[tacit cedar]

it's not looking good

[hardy fable]

this layout is a bit smaller

even tho I'm using 2 splitters per item from filtering and priority

couldn't we use one btw 🤔

I'll try that when I get back from work

[tacit cedar]

alright glanced at ur idea and made something from it

[hardy fable]

that works yeah

[crude quest]

That completely ruins the scale of things though - At every step you want to try to increase quality

[hardy fable]

yeah but what do you do if you don't have enough high quality copper plates to go with your high quality gears?

and if you just increase plate production, what do you do with you excess of regular copper plates?

if it's to send them to recyclers then putting prod modules instead of quality modules is more efficient

trust me I tried #1215078107334057984 message

[slim wave]

So you’re doing this flow chart to find optimal proportions for quality and non quality products?

[hardy fable]

yeah it was to compare the ore consumption per item produced using different factory structures

[crude quest]

I'm not sure how to read those images?

And you're right - What I'm trying to do seems to be impossible to make deadlock free :/

The plan however was to always have enough regular machines to process regular outputs

But I'm sadly finding it being a colossal grind since how each step needs 5 factories

[hardy fable]

the explanation is on the first post of the chain, and the value are just unspecified amounts (what's important is the ratio between them)

what's really important is what I wrote below the pics, you can't get a higher quality ratio than the shortest recipe chain has

and the best way to do that is to put prod after quality, within the same chain

[crude quest]

So you're saying not to increase the quality within production chains? IE Green chips have quality in the ore/plates but prod in the circuits=>chips?

[hardy fable]

no I'm saying you should have quality modules in mining drills and furnaces, but not in the => assembler

because it would cause a mismatch of the quality ratio between the ingredients of the circuit assembler

same goes for , you should use quality to make ore and plates but the gears must use something else otherwise you'll have a mismatch

and now I wonder if you can't combine the two 🤔

if you put quality modules everywhere then making circuits will generate too much low quality iron plates
and making science packs will generate too much low quality copper plates
so you have a bunch of both but that doesn't mean they can be further processed
I mean you can still use them to make low quality circuits and/or red science packs but putting quality modules in those assemblers will not work

[crude quest]

That's what I'm saying. Ideally every step is qualitied. The ratios are indeed a problem and will deadlock eventually. But maybe its more efficient if you eat the backbuffers through qualitied recycling to keep it moving

[hardy fable]

I'd need to crunch the numbers in order to get a definitive answer

uh, that would only make matters worse

if the issue is that you have too many high quality items then quality recycling them will make a lot of even higher quality items

but we're getting quite offtopic now

ping me in #quality if you have further questions

[shut current]

is...this a 1:2:3?

Sorry, a bit late to reply

holy fugg its compact.

[stuck scaffold]

ikr? I do like the layout of the 123

[shut current]

Just trying to wrap my head around the prio-resplit-supply from the recycling end of the sushi belt, but in terms of a primary mechanical "gearbox", its beautiful

[stuck scaffold]

With some of those, you don't even need to recycle. Mechanical ratios in particular, you put the right ratio on the belt. Anything that comes back around has partners

[shut current]

Oh lol, you just wrap it around to the main input as it is. I...have tried this in the past and always had deadlock problems if a particular input is lacking. I guess it does need some signalling to ensure all ingredients are present.

[stuck scaffold]

Yes, it will do that. The cutoff circuit is only wires though, that's a nifty little bit of defense

"out belt does not move unless all ingredients fully present on the belt"

[shut current]

Oh, I would have gone for:

Sulfur Input checks if Engine Input is > 5
Engine Input checks if Red Chip Input is > 5
Red Chip Input checks if Sulfur Input is > 5.

...

Wait, should it be
A checks B
B checks C
C checks A

or the other way around?
ie. Predecessor Checks Successor or Successor checks Predecessor?

[stuck scaffold]

That's probably fine, it ensure you have at least one item on the second lane, but it would trigger as soon as an item entered that lane, so I'm not 100% sure

The way I do it, you put a 1 on each ingredient you need, on the combo, and just hold read all input belts. Everything must be exactly 9

If you shut the combo off, the mixer shuts off, that's also nice, but primarily it makes sure you don't pass with an empty input

You will get a full belt into your mixer if it moves

https://cdn.discordapp.com/attachments/139677590393716737/1290040314387628063/image.png?ex=672dc4ea&is=672c736a&hm=56b91fffbfc2ed54bf1eef3bd01f5b48db1a9efb3727690470ddbf2553c409ea& I put an output cutoff on mine sometimes too

[shut current]

9 wouldn't work unfortunately. A whole belt holds 6 items total, and ideally you have 6 on the 1/2 belt, 4 on the 1/3 belt, and 2 on the 1/6.
Since you can't ensure both lanes are filled equally at the same time unless the belt is compressed, youd need to make sure at least 4 is on each belt, plus the ratio variance.

6 (3+3 because we need 3 pairs) for red chips

• 5 (3+2 because we need 2 pairs) for engines
• 4 (3+1 because we need 1 pair) for sulfur

Everything =< 15

[stuck scaffold]

Oh it's on the input side

You check inputs, not outputs

[shut current]

I do mean inputs.

[stuck scaffold]

If the input dries up, that's when the problem is happening, immediately shut it down

Inputs are pure, should be 8/belt?

[shut current]

3.5 per lane if you really wanna split hairs.

Though I might be wrong and conflating with item rates per lane.

[stuck scaffold]

You might be

That's a working example of =9 lol

It's hold read, just items present on belt, and a full belt is 8 items

[shut current]

Using a loader sure.

I'd like to disable-throttle a loader with RNG to simulate shaky input and see how a non-CC wire-only total condition performs when individual lanes aren't nicely matching each other, ie. a couple one lane, a few more on others, but not lined up.

[stuck scaffold]

I did this in testing. It stutters if supply < demand. It'll wait until the belt has 8 and release back under

https://cdn.discordapp.com/attachments/1277595827547799552/1281488364234866749/image.png?ex=672da38a&is=672c520a&hm=a306db9a22020dc399dc79a5fd18ab8b87d08331a32c485fb1b832683485dc34& Kinda hard to tell in this image but there are lines b/c show-gaps is on

Here, this is more clear

Because it's being cut off the instant the inputs start to dry up, the mixer itself is preserved.

The whole thing operates in compression so the instant you stop the out belt, all the ins stop too

[ruby quail]

how viable would starter sushi mall be with the fact that i also try to get lazy bastard and first nest kill with arty, also with space age (so i have ways to go)?

[red obsidian]

sushi is a bit expensive at the very start (a whole extra belt going around with extra inserters, vs direct insertion) and you don't see the benefits for a while (making a wide variety of buildings easily; you only have like... 4 main products to start off with)

Still viable, not sure it's worth it before at least green science

[ruby quail]

i mean it is completed already, surprisingly easy to just leave it be and come back to few stockpiled items

and yeah there is this problem that items need to be unclogged, but it's not that bad

[maiden jacinth]

hey guys, how do I pop this into just one decider combinator pls?

[red obsidian]

Option 1. < 10 => ❌ (1)

Option 2. < 10 => ❌ (1)

Option 3.
< 10
OR < 10
OR < 10
OR < 10
=> ❌ (1)

not sure which you're trying to accomplish

[maiden jacinth]

I am trying sushi for the first time ... I want to stop when x of each are on there

I have tried option 3

no worky for me, unless I misunderstand

[red obsidian]

why do you want one condition for all of them combined?

[maiden jacinth]

just to save space for my ship, lol

[red obsidian]

wouldn't you want to stop each particular input separately?

[maiden jacinth]

I figured it would be easier to add things to this list too

[stuck scaffold]

That looks like an each <10 output X(1)

Constant to init values

[maiden jacinth]

which part is wrong please? I would like to do the OR list

[red obsidian]

mmm. mall.

[elder plover]

if you want to have individual limits for different item types, it's much easier to have a combinator that checks EACH [green] < EACH [red] => EACH: 1
Where [green] is the contents of the sushi belt, and [red] is the contents of a constant combinator that contains the amount of each item you want on the belt

The output of the combinator can then be used to set filters, or add conditions on the inserters

[maiden jacinth]

nice idea, thanks

[slim wave]

I’m doing new things with space platform sushi where a sushi belt’s loopback might be done on a separate belt or a network of belts

I use sushi loops that can void materials as voids for other loops

[stuck scaffold]

That's legal. I've been able to input regulate so far but I haven't seen much

Simple target count. Hold read is very nice, I just don't pick up materials if I don't need them.

SushDi made it into space, too. That belt space is valuable and this is seemingly sufficient for the job rn

[stuck scaffold]

Idk how long the principles will last, but for now it's working

There's a possible edge case where I run out of fuel with a full ammo belt? As long as I'm consuming ammo I should be fine so I don't think I need to defend it

[warm vine]

are you using a mod ? the colors look much more vibrant than on my game

[stuck scaffold]

nope, I do have contrast cranked iirc

[hardy fable]

you can tune the colors a lot in the game settings

[chilly flicker]

I'm assuming this makes sure to only input more of a specific science if the existing one has been consumed, right ? What about if there's no input of new science ? Does that mean that'll leave holes in the belt ?

[tacit cedar]

yeah it leaves holes

[chilly flicker]

Aw. Wait, so those weird splitter contraptions, they're .. how's it called .. rate limiters, or something ? Turn a belt into 1/6th of a belt ?

I wonder if it's possible to make a circuit based rate limiter where, say, a signal comes in, with a value of n, and it limits the belt to 1/n-th. If so, i could have a fully compressed belt when I have only a few science types, and as soon as 1 more science type comes in, the signal changes, and suddenly all my 1/3rd belts become 1/4th, leaving room for the 4th science type.

[hardy fable]

then you may run into the same issue as the not rate limiting splitter sushi

cuz if each third is full they're going to need buffer space to store what they must lose to get from 1/3 to 1/4 (1/12 exactly)

[chilly flicker]

Ah, but only until those get consumed by the labs, right ? .. Hm, I wonder if it's worth it in the end. Even on the worst case, if I have only 1 science pack, red, having a belt 1/12th filled, if i don't have too many labs, I can surely still feed the last one before the first one needs more science, right ? (I'm gonna use red belts for this, they're twice as good as yellow and you can get them fairly early, and they're not too expensive. Blue belts are just too expensive for me, and not THAT much better than red ones)

Also it's gonna be 1/12th and not 1/6th filled with 1 science because I want to save the 2nd lane of the belt to get all the rotten agricultural science out. (Or maybe that situation will never happen and I shouldn't worry about it ? Although i'm fairly sure I can get into a situation where some of that science rots INSIDE the lab before it's used)

[stuck scaffold]

If you're worried about spoilage, loopers probably ain't it.

Looped mixers. But what you're asking for is certainly possible. At the least, delay decay is viable, where you calculate the tick delay till next pair of items and release them then

I think a 12:1 is probably preferable though, if rate limiting is on the table? hard to say

If you swapped ratio before you ran out of the older science's techs you could maybe do it but it'd be tough

[umbral gorge]

if you're worried about spoilage, burn it

[hardy fable]

I have seen a weird behavior with rate limiters and spoiling items, the items that are looping will continue looping infinitely

I noticed that when I made a sushi belt in sandbox mode where I was feeding and destroying the entire belt all the time, and spoilage managed to appear in its rate limiter, which shouldn't happen since I'm always renewing the items and the spoil time is 1 hour

factorio determinism :D

[chilly flicker]

If only the splitters weren't blocking.
Imagine a splitter with a filter that sends every item of a certain type "right", but instead of sending every thing else "left", it still SPLITS everything else. We could then add filters to the sides of the splitter to keep our rate limiters the way they are but just tell each splitter to redirect spoilage a specific way as to get it out of the rate limiter asap

In that case, the filter would change from being "I want only this item on this side" to "I want this item on this side only"

maybe a toggle option as to not break existing designs ?

Is it worth suggesting ?

[stuck scaffold]

I fail to see the difference between the individual functionality that already exists, and that? Particularly for this use case, rate limiters, I'm fairly sure you need extra splitters either way. The rate limiters must do very specific things based on input material count and can't be randomly removing portions

For a rate limiter to do its job, it can't be also filtering out, I think. But even then, filtering generally messes with splitting and vice versa.

[hardy fable]

imo it's fine as it is rn

it has its quirks for sure but that's part of the fun

[stuck scaffold]

Conceptually, I understand how the recycle loop works for mixed balancers, but it still fucks my head up when I see a recycle belt loaded with one item type and it works

It's an item stack, it's just weird timing alignments. If you remove X from the mix every time, and you ask for an item off the "Stack of correct mix", then it's correct to return that item. Still does my head in to see 100% of one item as "mix"

[hardy fable]

yeah ikr

you see them turn around without ever leaving, feels so weird but I'm sure the math checks out

[stuck scaffold]

Math checks out, I've ignored the loops with no repercussions so far.... brain still hates it

[dapper robin]

Is there a blueprint for 2.0 sushi science?

[mental spruce]

just make its its not hard

[stuck scaffold]

@hearty relic Okay so here's the million dollar question: Is it possible to get transformational sushi going on the belt or na? I feel like there's a hack but I can't quite put my finger on it

The problem as I see it is that these recipes decompress but there's gotta be a thing there

I don't see much of a way to prevent overloading something either

Starvation sushi could prolly work

I feel like rock cracking has to be circuited too, and is recommendable.

I don't see a fully mechanical solution besides having multiple belts?

[hearty relic]

Transformational can work, but you would need to guarantee the asteroids dont exceed more than a certain amount.

Otherwise, you could oversaturate with other asteroid types

[stuck scaffold]

Right, that's the big problem, you can oversaturate many things

You're replacing like 10:1. It's possible to output block but.... idk how

[hearty relic]

So yeah, not a full mechanical solution

[stuck scaffold]

If I were processing metallic asteroids to ore immediately after peak iron plates, I think that would be a proper output block

[hearty relic]

You could theoretically do asteroid conversions to waste excess materials

[stuck scaffold]

Yeah, that, I don't see a way around circuiting

Maybe cracking could be directly inserted

I think I'd rather take lazy machines over circuits?

I can't figure out a way to get them to block each other, either. How to stop an iron dump when insufficient ice

[hearty relic]

Why not just convert it?

You have asteroid recycling

[stuck scaffold]

Yeah, w/ circuits

It just postpones the problem anyhow, which is carbon/ice/plates balance

[hardy fable]

#930147588160782436 message

I don't have the blueprint of the stacked version but it's an easy modification

[lament dew]

circuitless rainbow sushi
didn't like how other designs needed a dedicated placeholder item or bottlenecked each science to maintain a fixed ratio (which fails to compress the belt if any packs are missing), so I figured why not use the science itself as placeholder items? excess science fills in any gaps, and once the missing pack is restored the placeholders go into overflow chests, without the need to bootstrap the placeholders manually (i think you only need one overflow chest but one pair of inserters isn't fast enough)

[umbral gorge]

read the belt, put a logic condition on each inserter that it only inserts x product on the belt if < certain amount

[stuck scaffold]

Erm.. transformational sushi has, as part of the design philosophy, forbidden this explicitly

[umbral gorge]

i misunderstood the question then

[stuck scaffold]

https://cdn.discordapp.com/attachments/139677590393716737/1279104782300348509/image.png?ex=6740a728&is=673f55a8&hm=c97f238e8b318dbde78318951693a18127fdbfa29b965ea55c94c78c1eeed0d6& This is uncircuited "transformational sushi"

The idea is that if you pick up an item off the belt, you have room to place items on the belt

It's regulated by positioning on the belt. I won't claim to fully understand how it works, but I can get it to work mostly? XD

Ostensibly, if you remove a plate and a gear, you have room for a red sci. etc.

The harder part is also getting inserters and belts to not overload and kill it 😉

@hearty relic is really the guy to talk to about transformational sushi. But speaking on that topic, I may have discovered some sauce. I'm using a very full belt to move iron ore occasionally. Regarding the topic from the original response there, I think that might be how to get transformational to work with this weird expansion

So the reason transformational works at all normally, is that most recipes compress. An inserter takes 1 plate, 1 gear, 1 circuit, so you're going to have that much room while it operates. if it gets starved, when it begins operations again, it picks something up, and so has room to output. Now, my secret behind getting too many trains on a single track, and more than 1 belt worth of items on a single belt, is to not use the entire belt/track for every item, each item only uses a stretch

Like there, some stuff gets through, but for the most part the iron ore is only on the belt for a short stretch

It's possible the expansion could be mitigated that way somehow? I'm not sure.

[stuck scaffold]

Okay the order is super fucked up @hearty relic , it's crusher, consumer, asteroid collector LOL. Asteroid collection prevented by crushed asteroids, I don't know how to stop crushing though. Single item per lane I can do now.

[hearty relic]

Are you putting more stuff on the lane than one type of asteroid?

[stuck scaffold]

Idk if that works with more than 1 but it works with 1, I think

[hearty relic]

i mean, it probably wont, but i was just trying to figure out if there any missing points. Kinda hard to see the full loop in what you've got and compared to last time, may have underwent significant changes

[hearty relic]

latest concoction. Blue science @ 30 SPM;

running stress testing now

[hearty relic]

seems to be stable. Deploys transformational and standard belt reading.

[clear wedge]

Blueprint String

[item=chemical-science-pack]x30 Mk.1

(Space Age is not supported yet, sorry!)

[hearty relic]

very nice footprint overall.

funny thing ,the merger of the circuit line with the blue science line blocks green circuit oversaturation when coal is starved because engines are briefly put on the line and take priority over the circuit lane

this creates a small but significant buffer that doesnt compress the green circuits fully allowing it to recover when coal is restored

but essentially all lanes are sushi.

[half phoenix]

this seems to help https://mods.factorio.com/mod/read-belt-contents-hold-all-belts-read-belt-count?from=search

maintain desired item per belt segment ratio!~

(without counting manually, since undergrounds)

[hearty relic]

Actually, im seeing peculiar behavior where it reads past undergrounds

[hearty relic]

@half phoenix note the bottom loop here. I know I explicitly made two all belt readers here. Note the loop that cuts through the engines. Theres no all belt reader on any of those belts so the only conclusion i can reach is its reading the undergrounds as well

[half phoenix]

not my mod, try opening a discussion thread! they seem responsive

🙂

fwiw, both of these read B10

[hearty relic]

No no, im saying that what i shared is vanilla

[half phoenix]

okay, it seemed like you were talking about the mod I linked, so I guess I lost the plot

[stuck scaffold]

Thyme's =9 mixer guard no longer functions. I can't show it anymore cause half the belt is stacked and it's only the one spot that it happened, RC mix, but it died twice with 9 items on the belt instead of 8, which reads 10

I had to change it to >=9 cuz it died twice

Acutally I can see it being kinda weird, there's no excuse for this to be pulsing but it is

Both plastic and wires seem to not read at least 8 sometimes, they should be

[hearty relic]

Just to make sure, we are still talking about vanilla right?

[hearty relic]

space foundry sushi

[stuck scaffold]

Yeah, still vanil. The stop state it was one extra item

WELL Spage, so technically not vanil I guess

[tacit cedar]

that's in space? where are you getting the oil from?

[hearty relic]

Blueprinr error in snadbkx

[hearty relic]

It could be in space if its swapped for coal liq

[stuck scaffold]

You can do a whole coal-liq chain, can't you?

Then liq the coal

[tacit cedar]

yeah simple coal liquefaction works

but his build had normal oil refining

which is why I was confused

[stuck scaffold]

Steam's not so hard to come by in space either but yeh even simple liq

You can barrel liquids and put them into space, I def had a setup that can handle that

Actually this should handle sushiliquids in general, but there is an input and output side to the machine, so you can't have a switching machine that does both, unfortunately

All multiinputs on chemplants can have water on that side 😛

[hearty relic]

@stuck scaffold Whats the sushi called where you are analogically treating belts like train routes?

theres a particular circuit flavor sushi ive used where it involves only releasing items onto a belt when its respective item isnt detected elsewhere. Its not quite like a memory cell setup, or an all belt reader, but just checks if the item is at the given position to release or not release the item.

[hardy fable]

wasn't it a blood belt or something like that? 🤔

[dim rose]

blood belt with hormones?

[stuck scaffold]

Sounds like something in the blood belt range, yeah, there's a couple variations inside that even. Delay decay is just "Whenever you see an item, add X ticks to the delay" and include own additions. That way if you get a pulse of some extras, it "remembers" that and delays for longer. This helps with keeping even belt spacing, but max's method of delay reset is even more so. It doesn't remember multiple items, just that whenever it sees an item, it resets completely and begins waiting again.

[stuck scaffold]

Oh wait, belt chunks is what you meant, isn't it? The iron ore I transfer into the hub only has to go on this short stretch (mostly) while the carbon originating in the southeast corner has to circle the entire belt around before being picked up again. They do compete but only in the iron section, this type of behavior can be thought of like trains

It's not so much a "sushi type" as it is a way to understand sushi I think

Like with trains, if you have many sections of high throughput that don't compete with each other, they can be on the same rail network (belt) and exchange some trains (items) but for the most part you're able to get far more than one track(belt) worth of throughput with locality. That's true of other belt stuff too not just sushi

It's not sushi but it's the best example I have on hand of that behavior I think, the u235 on the top belt. It's also exchanging lanes but immediately being brought back to one lane so at least it's more clear because of that I think. The "traffic" is between the output and input inserter, if I had their positions flipped, it'd actually look very much the same here, but with sushi or in other circumstances like more than one machine doing input and output, it would not because that "traffic" has to circle the entire belt.

Just like in trains, minimizing travel distance is good.

[hearty relic]

I could have sworn it was a way to approach doing sushi, if you go by the definition of "a belt having more than 1 dedicated item per lane"

[slim wave]

Putting 2 beacons like this will save power

You only need these 2 beacons for equivalent effect

[stuck scaffold]

If it didn't take 2 seconds to fix I'd leave it lol

[slim wave]

Just hate to see someone use extra modules like that lol

[stuck scaffold]

lol I'm switching over to legendaries rn

[stuck scaffold]

Yeah that's effectively what it is. If most of the consumption is done in individual chunks like this, you can get more than one belt's worth on it. Same way I did the steel in the 20k and got like 6 belts onto one belt

You can also share between however, but similar to trains, you can use think of it in parts rather than the whole. If you're producing 60% of a belt's worth of iron plates, but consuming most of those plates immediately, then it's not 60% of a belt for the whole belt, it's just for that chunk. The extras slip out and around but in practical use cases most of it gets picked up

[mental spruce]

so how are you enjoying SPAGE ichi

[stuck scaffold]

It's p good

I've never had so many things where I was like "actually bots tho?"

[stuck scaffold]

I'm doing a bit of quality sushi but honestly it's been overproducing copper so hard that I think this was a mistake. I have so much quality copper sitting

[dim rose]

iirc someone wrote some code to generate rate limiters (with Python?, maybe JavaScript) but I can't find it anymore. It's probably not more than a year ago. Do any of you know where it is? I thought it was posted to this thread.

[slim wave]

If you’d like to try and study it there’s a good conversation in here that @tacit cedar contributed most to where they find a very simple way to convert repeating binary fractions to rate limiters. I’ll see if I can find it real quick

#930147588160782436 message @dim rose

[tacit cedar]

if the output is 50% or more

[tacit cedar]

I made this method by having maxreader tell me how to do it
I didn't use code though

[slim wave]

But it would suggest there would be a simple way to make a program for this

[half phoenix]

so who in here is working on interplanetary sushi logistics for platforms? is it possible?

[slim wave]

My thoughts on it were to have a demand code of items sent from the requester planet to the platform; then the platform would go get those items from the provider planet

Each request code item would represent 50 or 200 of itself

Or you could just keep a stock of a whole lot of items

[neat idol]

https://katiska.cc/temp/factorio/sushi/sushi-science-yellow-turbo.mp4

It's just so simple to setup and use 🙂

[hardy fable]

I recently figured that putting agricultural science pack on the science sushi isn't a good idea

else the max spm is limited proportionally to its freshness

if they are all at 10% freshness the whole belt is at 1/10 of the max spm

but if you put it on another belt you'll have around 12 times more agri science than any other so you can handle that case

[foggy junco]

question, what if all the labs are full with spoilage (due to e.g. some time without research)?
The sushi belt will remain full (since no packs are picked up since the labs aren't researching), so there will be no space to output the spoilage, which would just lock the entire system

[neat idol]

That sushi belt is filled only to 75% capacity so there's always room for spoilage. Turbo belt is 4 yellow belts, and that setup uses just 3 belts

We could add one extra set of yellow belts, or use a clock, if we wanted to fully saturate it

[neat idol]

I noticed the same thing, but I just got around it by designing my throughput and number of labs needed around expected freshness

I thought about feeding extra agri science to account for lost freshness, but since that should be only couple of percents anyway (depends on production and platform speed) 90~99% is easily achievable

[hardy fable]

90% maybe, 99% definitely not

first, the agricultural science pack can only be made on gleba (making it on site from bioflux would be better since it spoils 2 times slower) so you know you have 1 hour to work with

the time you remove from that is all it takes to get the science pack from the biochambers to the rockets on gleba (let's assume it's instantaneous), then the time it takes to get to space (I believe it's at least 20 seconds) then the time it takes to get to where your labs are (likely nauvis, so around a minute at 250km/s), then how much time it takes to get down (probably the same as the rocket launch, maybe less) and then the time it gets to the labs (because of the sushi belt it might take at least few dozen seconds)

but there's something more to consider here

you're not going to ship a single science pack at once

you'll likely ship at least a full thousand, since that's the rocket capacity of the science packs

but you'll also likely not be able to make a thousand science packs instantaneously

so it will take time to fill that rocket

and during that time the packs will slowly spoil

assuming a constant production, the average freshness is just the average of the most spoiled one (the first crafted) and the least one (the last crafted)

at the end, 6 minutes for all of that sounds realistic but 36 seconds definitely isn't imo

[neat idol]

Ok, 99% is not easily achievable nor practical, but it's achievable. (I designed my biolab columns around 90%) (and by 90~99% I meant start conditions before labs, not at labs)

In my test setup I had:

• 2000 SPM and I was transporting 1000 freshest
• 500 km/s platform

Transport accounted for 0.83% spoilage (15000/500/3600) and average spoil from biochambers to rocket was around 15 seconds (60*1000/2000) -> 0.42% spoilage, and rocket and landing pod animations took maybe another 15 seconds

With this setup we got around 1.6% spoilage before labs. We could get it to <1% with faster platform and more overproduction, but I don't think that's very practical.

Reasonable expectation is around 3% before labs.

[hardy fable]

oh damn, that's a fast boi

[tacit cedar]

how does blood belt work? this backed up

the belt on the top left that the inputs are connected to is set to pulse
the read all belts is just for the total on belt thing

[stuck scaffold]

Blood belt is supposed to be a sampling kind of thing iirc

[tacit cedar]

I tried doing exponential moving average of what passed over the one belt

[stuck scaffold]

I think EMA was not quite blood belt? Similar concept though. Blood's not as useful anymore with the measure whole belt, I think

[tacit cedar]

oh yeah absolutely theres likely no benefit in making this

[stuck scaffold]

There's a couple ways to sample, if you want to keep stuff spread out instead of clumped, which I believe was part of it

[tacit cedar]

yeah spread out makes more sense

[stuck scaffold]

Delay since item has been seen, I think you need to read the input belts too, or else it dumps? What you've got sounds more like delay decay

https://cdn.discordapp.com/attachments/1215073493142474833/1348280402820731001/image.png?ex=67d96f79&is=67d81df9&hm=727ae8452408f70f3c3b483fda9498084920b7325331b213b43d3232a0b72196&

It would've dumped 2x more if not for the stuff already on the belt

[tacit cedar]

I don't really get it tbh

[stuck scaffold]

For every item you see, count however long you want the delay to be. I think I did 1000 for precision reasons, and then every tick I subtracted like 100 something from iron

If you want one item every second, add 60, subtract 1 per tick, then the input belt allows items if count is low, and also pulse reads to acknowledge it has put items on

That's not blood belt tho

Blood belt, iirc, measures a small area of a large belt, and adds based on that. so if a stretch of like 5-10 belts doesn't have that item on it, or that item count is low, then add

[tacit cedar]

this seems pretty good

[stuck scaffold]

Aha, doing a sushi demo?

[stuck scaffold]

Seems well enough spaced and close enough to the ratio to function

[tacit cedar]

ya more or less

lmk if you have more types

im also not 100% sure how to categorize

sub-categories being "you can apply these techniques to any of the above ones"

[stuck scaffold]

Delay decay has two forms, one where you reset and one where you keep total count.

#space-age message This should have purple sci in it, that one is done up delay decay style so you can at least see an example circuit

It really is just a memory cell, a constant decrement, and an arithmetic that multiplies all pulses by a constant

Delay reset will always reset to whatever consistent delay, so it'll be say 20 ticks after the last time you saw that item. That one will keep track of batches, so if it sees 5 items, it waits for 100

You've also got Jouster's transformational sushi, I can only sortof explain the concepts behind that, should really ping him for better details

Hmmm what else

[tacit cedar]

so like
you start with 0, subtract 1 every tick, +20 whenever you see an item? and add more whenever it's less than 0 (and count the ones you add)?

[stuck scaffold]

Yeah, that's the concept in a nutshell

You can fine tune the numbers, I think mine was like +1000, -160 or something like that, I needed it 1 item between 6 and 7 ticks

[tacit cedar]

i dont see combinators in there

[stuck scaffold]

Hmmmmmmm

[tacit cedar]

except the one constant ofc

[stuck scaffold]

I could've sworn that's the BP I pasted on map

[tacit cedar]

this one is still neat though
I like what the constant combinator is doing

[stuck scaffold]

lol the on/off control switch is nice, yeah

Circuits there on the purple

[stuck scaffold]

This should be the important bits

And some unimportant bits lol

[tacit cedar]

just playing around with the idea
I figure I should give it a maximum amount

first combinator is each * 50 from the belt on pulse
second is each - 1 and outputs into its own input

(the input belts on the left are yellow, so no single input can clog the rest)

[stuck scaffold]

omg you can decrement on thing with the new circuits can't you

That's not behaving quite like it should? The timer is set based on how often you want them to be fed

Delay decay specifically spreads them out the most

[tacit cedar]

maybe i have the numbers off idk

turning the input belts to read-pulse helped

[stuck scaffold]

lol it's supposed to be nice and spread out

[tacit cedar]

this is what it looks like with the input belts being yellow

so maybe it's just clogged

[stuck scaffold]

Oh I understand why it's dumping like that, I think. The spacing between measurement and addition is wayyy different

It's supposed to pulse a lot >.>

Measuring before addition might be necessary too? Or at the latest where you add

It's supposed to end up mega spread out

[tacit cedar]

maybe "50 per item, -1 per tick, 7 items" is not the right ratio idk

yours is 1000 per item, and -135 per tick (for iron)

[stuck scaffold]

7.4 ticks between items, though it ends up being ~15 ticks for 2

so 8/s approximately?

Little more

[tacit cedar]

this is 1000 per item, -200/tick

input belt is enable <=0 and read:pulse, top is just read:pulse

[stuck scaffold]

That's weird, idk why it's dumping. But I'm pretty sure read on main belt should be before additions

You don't want to add more if there's an item there already

I'll whip up a test case, sec

[obsidian roost]

Error

2.0 / Space Age Blueprints are not yet supported.
Please be patient until BP Radar has been updated.

If you want to help with the update and have experience with Rust feel free to reach out to my developer here.

[stuck scaffold]

bpbot, not bpradar

[stuck scaffold]

I don't have EE or nixie tubes but I can get you the core 😛

[tacit cedar]

i noticed that delay reset can get really high or low numbers which can take a while to auto-correct

when idle

(nothing outputting -> high #, nothing inputting -> low #)

oh wait i figured out the problem

[stuck scaffold]

-100 on four over 1000 is enough to overload, so they have to be slightly lower

[stuck scaffold]

Belt has to keep moving?

[tacit cedar]

the read on the left wasnt actually going to the input

[stuck scaffold]

This is -90 over 1000

[obsidian roost]

Error

2.0 / Space Age Blueprints are not yet supported.
Please be patient until BP Radar has been updated.

If you want to help with the update and have experience with Rust feel free to reach out to my developer here.

[stuck scaffold]

ffs I just pwnt myself

[tacit cedar]

hope this helps

[stuck scaffold]

It'd probably be a bit more clear with a gap between control and belt, but timing things right means the gap needs to be the same space

[tacit cedar]

just did +1000, -90 on my design with 4 inputs and it seemed to work more or less

[stuck scaffold]

omg blue belts pulse twice

Red belts only allow one through at a time, not two, it spreads out so much nicer

Also it lets me know -90 is still too much lol

[tacit cedar]

-50 seems fine for this

it overfilled with red belt tho

[stuck scaffold]

Hrm, so, I think if you want to do 7, it'd probably be south of 50 but maybe north of 45

[tacit cedar]

-40 might be too much too, but it seems to have fallen into some kind of balance where when it tries to add more, it's blocked

[stuck scaffold]

It's wild that the 1 tick delay on bluebelts is too much lol

Sec I can change a wire color and adjust a tick, that should be enough

[tacit cedar]

oh ya i have no idea if my wires are correct

[stuck scaffold]

lol where you measure makes a huge difference, and yeah adjusting that tick caused it to not double drop

[tacit cedar]

-30 fills very slowly but seems kinda nice

[stuck scaffold]

So Idk if science is a great use case for it

[tacit cedar]

oh yeah science probably not

plus this has no output

[stuck scaffold]

Yeh, the point's kindof to maintain a given rate on the belt

Maybe 5% is returning on the loop, maybe 30%, who knows, it needs to be at 60% though so get it to 60% of the items per second are that thing

[tacit cedar]

i still think mine are the most aesthetic

[stuck scaffold]

lol if it wasn't so balancer heavy I'd use them more frequently

[tacit cedar]

fuck combinators
my ups belongs only to splitters

[stuck scaffold]

Rate limiting to combine back onto one belt though

[tacit cedar]

this solution is also very clean

[stuck scaffold]

I'm a huge fan of compression sushi tbh

[tacit cedar]

this is probably peak ups for sushi

[stuck scaffold]

lol belt rate, yer

[tacit cedar]

plus the separate sides

so inserters dont have to choose between left or right

[stuck scaffold]

https://cdn.discordapp.com/attachments/930147588160782436/1279875941807558686/image.png?ex=67d9c05b&is=67d86edb&hm=0190feed0ef2ee5eb6f58202b834c718c530d8f2da8ce72c466b72154220510d& Belt rate sushi is snasty

[tacit cedar]

ok compression sushi is actually kinda neat

wait my top one is not my latest RLs hold on

[stuck scaffold]

https://cdn.discordapp.com/attachments/930147588160782436/1279871626346758166/image.png?ex=67d9bc56&is=67d86ad6&hm=60f95e1bb6959aee5d4fa6d2626a596b37b5cdfdc4f35016b36ae3d46e895cdf&

[tacit cedar]

there we go

[stuck scaffold]

lol yeah, compression is my normal go to for rate limited single belt now

It's always gonna be rate limiting for something else

I'd like a cleaner way to get multiple outputs out, when that's possible, but even 2 is a mess most times

[stuck scaffold]

One of the neat things about belt rate limiters is that quite often, you can end up with multiple outputs for specific situations, this is a nice one because it's a 2:1:1 ratio, so I need a quarter of a redbelt of green chips and plastic, and a half a belt of wire. So I don't rate limit one belt to 25%, four times, I just split it into fourths lol

Your rate limiters do that kindof thing too if you know where to chop em

[tacit cedar]

my rate limiters are more "it just works" puzzle pieces

man mixers are hard

[stuck scaffold]

Ya, the problem is that compression mixers rely on backpressure

So multi-output can't ever have free flowing inputs

[slim wave]

That’s not necessarily true

They created pressure in the middle but not at the end

[stuck scaffold]

It's been some months but I seem to remember backpressure being essential to each splitter's functionality, so if one doesn't have backpressure it doesn't do the mixing

[stuck scaffold]

Proving me wrong would be lovely, I could do a lot of work if that's not true lol

[tacit cedar]

I wish there was a way to fix this

i guess i could run it through part of a balancer

not very clean tho

3 tiles high my beloved

[tacit cedar]

my method scales pretty easily if you have the space

I guess a 16:16 balancer works for this case tho

wait how does no loop work

[stuck scaffold]

It only works for certain recipes iirc

And it needs a refeed chest at the end, and it still scary lol

I'm not wholly convinced that style is viable but in some cases it prolly can work

[lament dew]

hope to god the ratio is perfect so there's no clogs

[hardy fable]

the most basic theory says it works because of determinism

a more accurate theory would tell you otherwise tho, because the grabbing pattern of inserters is pseudorandom and thus might cause a lower entropy somewhere on the belt

and if none of the 8 items is the one that the last machines need, they're stuck

one way I found in order to solve this was to make a partial loop, aka bringing back the last items to the middle of the feed belt

https://cdn.discordapp.com/attachments/139677590393716737/1250558034615205898/image.png?ex=67dbdbad&is=67da8a2d&hm=265accaad261a0289156b4491e6db4e9a3d2a2628e252f6e03c41670b786df0d&

you can see that being implemented here on the top and bottom lanes

middle lane is 100% safe because of the most needed item (electronic circuit) has its own dedicated half-belt

[stuck scaffold]

The design I remember working, was an LDS one, where copper was one lane, and the plastic/steel was sushi'd, and it still needed a guard chest on the end. When plastic or steel was 4 on the belt, it had to pick up that item and put it into a chest, which got fed back into the last machine

[warm gyro]

Maybe controlling inserter filters or something? Im not sure

[hearty relic]

Evolution of transformational sushi taken to an extreme.

[warm gyro]

I didnt know you were trynna solve the steel and plastic stuff lol

[hearty relic]

The crux of the matter as ich pointed out is that everything on the belt needs to be consumed in order to properly upgrade. Very risky stuff tbh, but easily corrected with robots

[stuck scaffold]

lol yeah I mean, I'm not actively trying to solve that, it's functional for some specific builds from what I hear but I've never tried it personally. I don't like the concept, having to monitor a belt end

[warm gyro]

I wonder if DI copper di to lds is possible, would make it so there is probably more space for belts so my 12 beacon setup wont be needed

[stuck scaffold]

Ofc it's possible. You could alternate copper and steel

Though steel is lowest, and it's also possible to just pour LDS, I do think you get more out of your ore by taking the extra prod

[stuck scaffold]

I have no clue what kind of ratio this is close to but you can snake about 4-5 beacons on any machine

[warm gyro]

Oh yeah lol spage

Wait, lds doesnt need prod cuz of its research

[stuck scaffold]

Bingo

[warm gyro]

So this is kinda viable

[stuck scaffold]

Yeah it's not totally awful, I think it's within reason for a couple beacons but who knows

[warm gyro]

Anyways this isnt sushi

[stuck scaffold]

true

[warm gyro]

corrected with robots why?

How*

[stuck scaffold]

what if the chest overflows? XD it's a corrective measure for in case the corrective measure doesn't work. Or you can jsut yeet it if that happens but then that breaks belt ratio, if you had perfect ratio

[warm gyro]

The good thing is LDS is needed in such small numbers for yellow sci that doing lds to science di isnt needed

[stuck scaffold]

With the chest and compression mixers, that's deterministic, so you can mostly trust it to come back around at some point and just feed it into the final machine

[warm gyro]

Ohhhh that makes sense now

[stuck scaffold]

https://cdn.discordapp.com/attachments/1277595827547799552/1278823696131297420/Unbenannt.png?ex=67dc83e0&is=67db3260&hm=a5282c806ca63cafa689773d385b1ac2ee037107947c11cecfff5c935f4bea0d&

You can see how it starts to get a bit more sorted as the belt goes on. That's b/c they're consuming at different rates, so there will be 5 plastic swings for every 2 steel swings, but not quite b/c usually only plastic actually gets a full hand

Can see pretty big stretches completely void of steel.

[warm gyro]

Oh dang

Thats an interesting way to do that

Huh

[tacit cedar]

ok yeah this seems to work

not "loopless", but at least it doesnt have the entire recycling step

the "sort and reinsert" thing

[warm gyro]

Is it max uptime it can do?

[tacit cedar]

wym?
I didn't put any thought into the amount of assemblers if that's what you're asking

i just put a bunch and added speed it would break as soon as possible

[warm gyro]

Erm alr

I meant like belt is 45/s so if max uptime i mean you have enough assemblers for 45/s to be fully used and then you see if all of them are consistently on

[tacit cedar]

rn im just playing with thyme method
i want to get a small lds ratio

idk if 3 tiles high is possible with the double loopback tho

maybe with belt weaving

[tacit cedar]

this is so ugly

[stuck scaffold]

3 high compression is tough, that vastly favors the fatty ones

[tacit cedar]

oh wait i hopped off factorio without getting bp

[stuck scaffold]

https://cdn.discordapp.com/attachments/139677590393716737/1290040314387628063/image.png?ex=67dc73aa&is=67db222a&hm=5ea5adebb529f0383fe69f8db3b541fabb7be553ba2f9be4b0e362e328421f48&

I wouldn't consider making that 3 high lol

[stuck scaffold]

That it's even doable for LDS is kinda impressive tbh

I guess theoretically they all should be?

[tacit cedar]

idk if it's doable in vanilla, since I used green belts

the weaving SEEMS doable without though

but i havent tried

[tacit cedar]

idk some of my bigger things i didnt bother, like my 7/1000 rate limiter

[stuck scaffold]

I think b/c of how compression works, you can always just isolate important subgroups

It all comes down to which 2-into-1 you do

[tacit cedar]

the funny uranium ratio

[stuck scaffold]

lmfao that's mirrored, it always seemed to me like it should be basically the same thing in compression except only one of them instead of two

[tacit cedar]

yeah it's somewhat mirrored

it splits it into a 7/1000 and 993/1000, it just swaps which one is the output and which one is recycled to the front with priority input

[tacit cedar]

well there were 3 input belts so i figured 3 high made sense to go for
you have considerably more input belts

[stuck scaffold]

lol yeah, fair enough

[tacit cedar]

on the other hand... belt weaving the inputs...

[tacit cedar]

no weaving, 4-wide

plus looks funny

plus no green belts

these are fun to make tho. here's purple juice

[warm gyro]

AHHHHHHHHHHHH rails on belt 😨

thats like 1 beaconed assembler on that belt 😭

[tacit cedar]

yeah but it's 1:1:30
that totals to 32
it's such a perfect ratio
how can I not?

like look at that

NO loopbacks

[warm gyro]

idk 1:1 is a perfect ratio to

such a perfect ratio that you dont need sushi 😨

but ig the mats for rails would be good to put on the belt

[tacit cedar]

direct insertion is objectively bad for ups
if i dont see the sushi then i close the game
and thats 0 ups

[tacit cedar]

ill make it if i know the ratio

[warm gyro]

direct insertion is objectively bad for ups you lost me on this part

[tacit cedar]

seeing the sushi is required

i do NOT put circuits on the belts

[warm gyro]

😭

[tacit cedar]

i do NOT use undergrounds (as much as possible)

[warm gyro]

Big brain

[tacit cedar]

ups is NOT for adding more science. it's for adding more splitters

[warm gyro]

oh true

checks out 💯

lol

[tacit cedar]

let me find my mining outpost sec

[warm gyro]

idk id rather have more logi bots moving items from A to B but B has an inserter into B

[tacit cedar]

custom made 7-4 lane balancer for my mining outpost

this is what factorio is all about

you can see rails on belts too

hell yeah

[warm gyro]

😨 this looks cursed

who ate the last 2 splitters

[tacit cedar]

they werent necessary, i just needed a TL 4:4

which means I could leave out the front or back 2 balancers

[warm gyro]

ah ight

also rails on belts bottom left 😨

[tacit cedar]

what are the mats for it? honestly ive never really played with big builds/beacons

i ask because i figure the material ratio changes based off of modules and beacons

[warm gyro]

stone rod and steel

DI'ing rods is also usual

[warm gyro]

true w beacons the ratio would be a bit dif

but there isnt prod for rail assemblers

nor prod modules or e furnaces

only prod for rods

[tacit cedar]

oh true

technically modules in spage

[warm gyro]

lmao

so ratios would match if everything has the same speed

[tacit cedar]

so input is efurnace, prodmod, iron (-> rod -> rail), stone (-> rail) and steel (-> rail)?

[warm gyro]

yerrr

now you needa get the ratios, maybe use

+faq calculators

[empty compass]

Could not find "Calculators" in FAQ tags. Did you mean "Calculator"?

Web: FactorioLab (Most overhaul mods supported)
Web: KirkMcDonald (No mod support)
External app: YAFC: CE (Full mod support)
External app: Foreman2 (Full mod support)
Mod: Factory Planner
Mod: Helmod
Mod: Rate Calculator

[tacit cedar]

im on factoriolab already rn

[warm gyro]

I think ichy might have some stuff abt di purple sci + sushi but i think he did prod modules and efurnaces di'd too idk

[tacit cedar]

2 prodmod : 2 efurnace : 50 steel : 15 iron : 30 stone?
am i dumb or does that feel like too much steel?
is it counting efurnace recipe? idk how to disable that tbh (im probably using this software very incorrectly)

it might be 15 steel?

without any prod

[warm gyro]

hmmmm this seems like its less dense than rails by themselves 😭

[tacit cedar]

rails are 33% less dense than the iron and stone though right?

calculators are too hard for me i just want to put more splitters

[tacit cedar]

2:2:15:15:30 adds to exactly 64 holy

nope seems like too much stone

idk whats going on with it

also heres the epic design i made

[stuck scaffold]

1:1:30 for 32 is funny, the UPS comment is for memes and is very funny lol

Sushi is not for UPS reasons, ever

1:1:30 is an excellent compression ratio and lends itself to the sushi task well, if you wanted to belt rails(which isn't great for UPS), but the big deal is that spacing on the belt. Because it's 1:1:30, when it tries to pick up an EF or prod module, it'll pick up one or maybe 2-3, if some rails have been removed. The gap between those items is large, so it'll always spend many swings to pick up a full hand

That kind of thing comes with the territory on sushi. I'm not building sushi designs for 100k

[warm gyro]

Ohhh so my lds setup wasnt the best for ups 😨

lol

[stuck scaffold]

lmao I distinctly remember trying to warn you of that at the time

It's the way inserters interact with sushibelts, if they can't grab a full hand, they swing multiple times

And they also stay active for longer trying to grab a full hand

I love sushi, and I love UPS optimization, but I know better than to spend a whole bunch of effort trying to mix them lol

[warm gyro]

but compact and ups efficient builds are cool 😨

i mean sushi can also be used for other stuff

[warm gyro]

w design btw lol

the middle belt wiggling is a chef's kiss

[warm gyro]

hmmm NUHHUH that totally didnt happen

anyways i dont think it took that much ups from me lol

[stuck scaffold]

lol it wasn't the killer, no

[warm gyro]

in fact it helped!

cuz i had everything limited to 600 spm 🧠

due to the lds bottleneck

[stuck scaffold]

"Helped"
"LDS Bottleneck"

[warm gyro]

it helped with the UPS!

but didnt help with the spm

[hearty relic]

Isnt that just DI builds?

[tacit cedar]

at least my RL sushi has one thing over thyme's compression sushi
i dont need circuitry to turn it off if an input runs dry
afaik even looped thyme sushi will overfill the belt if an input is empty, but mine just leaves empty space

[stuck scaffold]

Yeah, the combo-free switch is nice at least

Everything = 9 and it doesn't take power, just wires and a constant combo for an on/off switch, I can tolerate that

[tacit cedar]

i guess you could use a dummy item as a placeholder somehow

[stuck scaffold]

That's how we used to do it a long time ago

lol this post is 7 years old now damn

[tacit cedar]

oh this was simpler than i expected

smaller

wait are these just 6:1/7:1 balancers

thats funny

get rederived idiot

[stuck scaffold]

7:1's didn't work back then

This is them backing up b/c 1:7 provides slightly more on a belt than a 7:1 can handle

[tacit cedar]

bad game :(

[stuck scaffold]

They were 40/s belts though not 45, iirc

Super ancient, patched long ago

Looped restriction works now I think?

[tacit cedar]

ok im willing to try it
I don't really know what you DI and what you don't

[warm gyro]

try DIing EVERYHING

[tacit cedar]

i can hardly make a sushi belt if I'm starting with liquid ore, though

[warm gyro]

i mean like everything on screen in machines

mb

[tacit cedar]

so just those 6 inputs?

no idea what tech level to go with either
pain

[warm gyro]

yee

[tacit cedar]

cooking in progress

[stuck scaffold]

The highest rates benefit the most from DI. With purple, it's 30 rails to 1 of the other 2 ingredients, so for every purple craft, directly inserting rails instead of belting is -1 insertion on 30 rails

https://cdn.discordapp.com/attachments/1215073493142474833/1348285147266547713/image.png?ex=67e2ae65&is=67e15ce5&hm=aedef7e266984c73b6f0d3af0fe5d69ce0366c8a645bfe964105ea74a0da2144& SushDI is fun tho, materials for the last tier of crafts all on a sushibelt, then last craft all DI

Output back to sushibelt

Sushi and DI play reasonably well together, they both have the benefit of compaction. Multiple belts in one for sushi, so it plays well with DI's 'try to fit everything together,' it would be harder to run a different belt for every material in that DI build

Every item you move DI that was sushi, is more room on the sushibelt to do other stuff, too. I think they have synergy

[tacit cedar]

well heres what i have so far
I prefer ratio sushi so the belt is usually full

[stuck scaffold]

Yeah, I put speed modules in the module maker it looks like, and eff in the furnaces, I think they were either t1 or t2 prod

[tacit cedar]

there will probably be holes in the sushi belt for output, if it's constantly on
but if it backs up then I may be screwed

[stuck scaffold]

Yeah, you can only fully load on input, after that it has to have holes

And being able to reboot from full pack is def a thing as well

https://cdn.discordapp.com/attachments/1277595827547799552/1279838123542446222/image.png?ex=67e22ee2&is=67e0dd62&hm=37b97d8659ad6cdcecf8483cd2bf2210ec1f045d1a89c6903087489ce1eab5a8& lmao I tried to ratio them this way at one point ig

[tacit cedar]

I have no idea what I'm doing.

I have the building ratio here

[stuck scaffold]

Prod and speed changes ratios

It's 20:5:4 with this moduling https://cdn.discordapp.com/attachments/139677590393716737/1279982709913026661/image.png?ex=67e2b58a&is=67e1640a&hm=26474135644a1875d86d937571b53fef15d0a683988e00a877aa3653a23a85eb&

Plus DI needs backpressure to keep it healthy, there's a lot of fine tuning that goes into them to make sure it works right

>= : ✅
< : ❌

[tacit cedar]

I didn't have speed
I wanted to keep it simple with just prod purple

since it also doesnt change input ratios

[stuck scaffold]

I remember fucking with that moduling a lot to get that ratio, so if it's different b/c no speed, a diff build is probably appropriate

[tacit cedar]

I have a simple per-minute ratio already

[stuck scaffold]

Oh sick, flat 8:4 is nice

[tacit cedar]

though im not sure if blue belts are fast enough for this

i dont think so

i could look into splitting it into two belts

[stuck scaffold]

Steel's a lot there, as purple is known for

[tacit cedar]

im not COMPLETELY sure how to get two equal belts, i thought i figured this out already

does the splitter fix it? i cant tell

[stuck scaffold]

That's what I was talking about the other day, compression needs backpressure on inputs to do the thing I think

If it doesn't have backpressure you get free flow

[tacit cedar]

seems to work

[stuck scaffold]

I think it's just backpressure that's the dividing line, it mixes evenly with just a single bluebelt to choke the rate down to backpressure. -1 YB effectively

[warm gyro]

whats w the wood on the belt

[stuck scaffold]

wood=recycle

[warm gyro]

filler?

[tacit cedar]

yeah wood is filler for ratio

it'll be recycled later

[warm gyro]

oh

[warm gyro]

what is this voodo 😨

[stuck scaffold]

Backpressure is what causes the mixing, I'm like 97% sure

[warm gyro]

why does it not mix when no back pressure?

[stuck scaffold]

Splitter mechanics, when they're both in free fall it just acts like a passthrough

[warm gyro]

oh

[stuck scaffold]

It doesn't activate the "take from one, take from the other" thing

[warm gyro]

yeah okay

huh

[tacit cedar]

honestly im not completely sure how to recycle this

[warm gyro]

so a splitter is not a 2 belt balancer when all belts full demmand

thats kinda funny

[stuck scaffold]

Once you've compressed the mixer it's a lot easier, you can always get down to 1 input per item at some point

https://cdn.discordapp.com/attachments/139677590393716737/1281468702839279688/image.png?ex=67e22ebb&is=67e0dd3b&hm=f90db3610f88cbcfeee863565a33effd7694ff3a9a82f65a576ee87bba5f00cb&

Just recycle and prio the recycled into that input. B/c you've got one belt out, it's always less than one belt in per item

[tacit cedar]

seems to work? idk

[stuck scaffold]

No rate choking?

[tacit cedar]

i dont know what rate choking is

[stuck scaffold]

On the final splitter

[tacit cedar]

i dont see gaps if thats what you mean

[stuck scaffold]

Is that at full throughput or is there something after it slowing down rates? both look like bluebelt rates

Oh you're doing a split in the middle to reduce? What in tarnation

[tacit cedar]

yeah i have two belts of output

[stuck scaffold]

Omg no this is genius

You split it into two pieces, both at ratio, then halved

Yeah that's legal, you aren't using a compression on the final step though

Still, if you want two compressed belts out of a compression mixer, that's better than anything else I've seen

Neat aF, but I'll stand by my statement on backpressure and compression mixers, you didn't break that 😛

Also idk what'll happen when you backup outputs, probably it clears fine but 🤷‍♂️

[tacit cedar]

the middle part stays compressed but i dont think that would break anything

[stuck scaffold]

Rates is rates, it's still moving a half belt each and final splitter is back to compression and should 50-50. I don't see a problem

I like the form of this solution a lot though b/c it works perfectly with compression mixers, that are already 2^N. Four belts out? Make four compression mixers, then mix those. If you have a 32 belt compression mixer then that's 4x8, which is very nice

You can always repeated halve a compression mixer

[tacit cedar]

this is peak sushi tbh

[stuck scaffold]

Yeah however many more belts you need, you stop that many layers early on compression, neat

[tacit cedar]

i am lost in the sauce

[warm gyro]

you made the sauce

[tacit cedar]

now that ive overcomplicated the easy part, i get to figure out how to tile machines

[warm gyro]

wait wtf thats 7k iron? just like that? 😨

dang belts are expensive

[tacit cedar]

i guess i could just have like 2-3x as many assemblers as i need

[stuck scaffold]

I mean, 16 > 8 lol

Di just cares about >, the machines shut down very gracefully, it isn't as bad as you'd think

[tacit cedar]

also this needs blue belts due to throughput : ^)

[tacit cedar]

yeah probably

[tacit cedar]

i could also save some by not using as many undergrounds

[warm gyro]

1 tile is worth 7k iron imo

[tacit cedar]

true

but I could save at least a single underground even with the same footprint, by making the underground that crosses the giant hole better

[tacit cedar]

https://factoriobin.com/post/ril5i8
I dont think there smuch difference, but the 1/50 is new iirc

[placid zenith]

btw, here is what this looks like in the upcoming blueprint bot updates

[tacit cedar]

simulate 5000 ticks so the numbers appear in the nixie tubes

[placid zenith]

I'm going to add 1234 or something

[placid zenith]

woo, numbers added!

[tacit cedar]

update: still struggling

[stuck scaffold]

What's this guy even doing

The stacker off the belt. All the purples have no input from belt it'd seem

It'd be real nice to have the rails be bulk inserters though, they're the lion's share of the volume being moved. Longs will prolly work for rates here and UPS not a concern so it's not a big deal

[tacit cedar]

this seems to function

and tileable once I remove the funny loop belts

idk if having all the science on the left side of the belt is a bad thing

[stuck scaffold]

If it fits it fits, can swap later if it's paired especially

[tacit cedar]

well I've been running it with just a normal loop (no actual recycling) and it hasnt broken so far

[tacit cedar]

thinking about it, I feel like it should break at some point
since the left loop has more purple science taking off it

[slim wave]

Mm beacon space

[tacit cedar]

simpler example

[stuck scaffold]

That second belt is really not so bad, is it

[stuck scaffold]

https://cdn.discordapp.com/attachments/930147588160782436/1279871626346758166/image.png?ex=67e44856&is=67e2f6d6&hm=020b9c0ed140af8b7c834fca7188295f994366f74c8e77e39c60a94c8bb4a892& Split and merge is actually great for a couple of ratios, anything you can get onto power of two I think. I don't know if I'd consider it another category of sushi or not, it's beltrate sushi still?

Half a red is a yellow, two yellows is a red 😄

[hearty relic]

And 2 reds is a green

Good times

[tacit cedar]

fulgora...

[slim wave]

Why is one lane all gears

[tacit cedar]

thats only a third of the inputs

which would make sense considering it's a third of the outputs

[slim wave]

Oh I see that it may be like a standing wave in your balancer

The turns in the belts would offset those big chunks of gears

[tacit cedar]

i cant tell if this version looks better (playing around with reordering the inputs to save space)

wait i forgot to do the back half

idk it still looks bad

[tacit cedar]

i dont think im getting better

[warm gyro]

what is the point of all this 😨

just use a recycler

[tacit cedar]

fulgora...

[stuck scaffold]

You're going the wrong way tho? XD

That's how we get it, idk why you'd want to remake that sushi ratio

[tacit cedar]

know thy enemy

[tacit cedar]

ok im done with this for now

squished it enough horizontally

[hearty relic]

This is the latest sushi build. Its built with gleba in mind, but its practically vanilla. 50 spm, and some abhorent sushi implementations.

iron and copper suffocate in a few places so it needs to be boosted.

[foggy junco]

why don't you just use cirucit based input limiting?

cause that looks like it can clog quite easily if some inputs are underfed, no?

[tacit cedar]

you can just disable the output if any of the inputs are empty

[half phoenix]

any of you got a sushi space casino yet?

feels like that could be a way to get it running quicker and with fewer crushers

[plain hare]

Just so you know, there have been devs talking about how they’re killing space casinos soon

Prob gonna remove quality module in reprocessing

My non casino re processor is sushi and set recipe from circuit tho and that’s cool yea

[half phoenix]

so is that a no?

[slim wave]

Yes we’ve seen those

[shut current]

This mindset sounds very " n O o O o , y o u H a V e t O p L a Y i T t H e C o R R e C t w A Y " and has never boded well for any game.

[shut current]

#combinators message

Managed to smooth it out a bit using a further 30tick Randomizer between the output and the Asteroid Grinder.

[half phoenix]

got a bp?

[half phoenix]

I'm going full editor if they kill the casino and don't buff anything else. upcycling is just a tedious grind

[stuck scaffold]

Casinos were borked, and I never touched them because I didn't need them. I'd appreciate this balance patch, this was not the intended gameplay experience and I saw that immediately

Legendary should be a little hard to get, it's still too easy to acquire without casinos imo

If wube thought they could hotfix it, they would have, which to me indicates they're being considered about this move.

They're probably taking so long because it's a player meta thing, you don't have to use it if you don't want to, the rest of the game is right there. Legendary was fun for me, I didn't experience any grind for it, and I am hyper-sensitive to grind, I cannot abide grind. The "not casino" playspace is still seeing exploration but I think they're only considering it because everyone's casinoing, the balance isn't so harmful. Legendary isn't that tough, but it is a fun puzzle https://cdn.discordapp.com/attachments/1215073493142474833/1348300752883548181/image.png?ex=681778ed&is=6816276d&hm=3d4565d61d6897502fe3e49e2d053eb7e97c4f00e10c6eb30349d5c1d7e21a32&

[plain hare]

There was talk of quality cast lds nerf too

[stuck scaffold]

I heard that and I am stoked tbh

Factorio is almost completely a game without clear optimal points

[plain hare]

I always wondered what they were thinking with those recipes xD

Anyway

I’m stoked to do some gleba and fulgora quality shenanigans

I kinda just want to make my next gleba base process all the fruit with Qual modules

That’s where I’m at with it lol

[stuck scaffold]

lmao opportunistic qual is probably decent on gleba, I think it's doable with a sushi fulgora too

That'd be fun, full sushi fulgora

[plain hare]

Yea it should be a good time and I’ll just plant more fruit and live my best life

[plain hare]

I wanna do partial sushi fulgora and gleba

I feel like full sushi never goes that well lmao

(It makes good pictures)

[stuck scaffold]

lol it's so tough to get it functional

[plain hare]

Did you see the blue sci sushi I made

[stuck scaffold]

Don't think so, I've been in the wizard cave

[plain hare]

https://media.discordapp.net/attachments/753031298368536577/1366189664549212171/image.png?ex=68174b06&is=6815f986&hm=0a7eccb90719f56f5fd5cf470cca57c1aaae0db1d56dec118b727224a2e26ce5&

Herra was trying to get their blue sci print smaller and I said ‘sushi belt’ :D.

Got to like 85% the size with sushi

[stuck scaffold]

lol sick, yeah compression's one of the things sushi allows, I got a lot more out of DI by using sushi

A lot of DI builds aren't easily beltable

https://cdn.discordapp.com/attachments/139677590393716737/1279983685981966398/image.png?ex=68177273&is=681620f3&hm=ea14581282c589437e5e854f8db852d584fbf2361912cf33e0f1f1f9a10a7b9a& This would be hell to belt

[plain hare]

Yeeep

Pretty factory

[stuck scaffold]

Very few wasted tiles though, like insanely few wasted tiles for DI

Ty

[stuck scaffold]

Some real slick tricks here, the gears and pipes are nice lane management

[plain hare]

It was sooo close to working with just a yellow belt

But the last red chip / blue sci machine would kinda go on / off so after spending an hour on it I said screw it red belt xD

[hearty relic]

no thoughts/inputs on legendary ore breeders?

[plain hare]

Like from legendary bioflux?

[hearty relic]

Yeah

[plain hare]

Oh. Well my first thought would be that you want to turn the ore into something else right ? So you probably have a step later that can only use quality modules

In which case a lot of times the goal should be lower than legendary bacteria, but epic or rare or uncommon more…

Because legendary bioflux is hard to make right

[hearty relic]

And while thats true, the appeal to me for doing legendary cultivation is you free up the entire chain going forward for straight prod.

[plain hare]

Yea of the parts of the chain that use prod, but, generally there is an end product

Idk, what are you gonna make?

[hearty relic]

Dont know, but im aware its more dependent on what im trying to produce which certified whether the bacteria cultivation is worth it or not.

[plain hare]

Well the cool part is it just costs fruit right

[hearty relic]

Ye

[plain hare]

So it’s pretty easy to make more fruit :D.

I was messing around with it and 3 step quality mash - bioflux - bacteria seemed to produce pretty cool amounts of Qual ores imo

That legendary replication almost never worked but uh , I only made a small lil setup.

If you had science size scale and only stole the high tier bioflux for bacteria it could be cool

This is what I wanna try

Make like common , uncommon agri sci and turn the rest to legendary mall

Full quality gleba base, sushi belts everywhere ofc

[shut current]

Dosh already did it, quite a good video too I might add.

[stuck scaffold]

The cost of the legendary bioflux chain, in terms of build size/complexity, is a great balancing point. The flux chain isn't even that good, it could use a bit of a buff but I'm not sure how you'd do that. I'm sure we won't see the factorio world switching to that

[hearty relic]

Personally, i still want carbon to recycle into spoilage in order to buff biochambers

[hardy fable]

any source on that? also will they kill max productivity normal to legendary LDS transmutation? it sounds more OP than space casino imo

[plain hare]

#1325697123055833131 message

#quality message

#space-age message

i wanna say there were a few other rumors but don’t remember where

[plain hare]

And yes killing quality lds casting was mentioned too

[hardy fable]

yup, found it right under your second link, thanks
#quality message

[half phoenix]

lds shuffle is kind of op but if they kill space casino and don't buff anything else, it's the editor for me

[hardy fable]

I agree, lds is very op compared to other alternatives (even tho it requires ton of research) but space casino doesn't sound this good

I mean the design sounds more challenging that just upcycling every single item out there

[haughty solar]

#930147588160782436 message
I need help understanding combinators in this image.
I suppose it enables the belts for a fraction of time proportional to values set in constant combinator, it looks pretty ,,,

[stuck scaffold]

that looks like some form of delay decay, yeah. When it sees an item, it delays so long until it sends the next one

[steep cloak]

+faq codesushi

[empty compass]

Codesushi

#bot-stuff message

[steep cloak]

@haughty solar

[haughty solar]

Thank you

[tacit cedar]

lds ratio so ugly

[foggy junco]

why do LDS from plate?

i get that with steel prod and legendary prod3 you get more LDS from ore if you craft it like that compared to casting it

but that's just worse ups wise which matters more at that point, no?

[shut current]

Because not everyone has the #space-age expansion, and this thread is a sub of #vanilla-chat .

Although....I am seeing being used...

[foggy junco]

yeah hence my question

[tacit cedar]

should work with blue belts

i just wanted the longest underneathies

[tacit cedar]

favorite sushi I've made so far (1/16 ratio)

[plain hare]

Yes

[foggy junco]

🍣

[plain hare]

Yes yes. We can talk easily here.

[hoary trail]

@plain hare so the chest method is nice in that you don't need as much wire

[gritty cradle]

I did

[foggy junco]

wait you can create threads

i'm envious

[hoary trail]

but it's not so nice in that you need to centralize the sushi kitchen

[plain hare]

You actually don’t need to centralize it!

Multiple sushi kitchens can feed the same belt.

[hoary trail]

but then you would need multiple loop points?

[plain hare]

Yes, you would. But that’s… common for me now

The sushi loops are in all my train blueprints

[gritty cradle]

If you have a mechanism reading items per second on the belt I wouldn't call that natural but I haven't seen the reading mechanism either

[plain hare]

I have special powers muahahahah

[foggy junco]

now i can finally automate biter destruction faster

[hoary trail]

so then you'd need to filter out whatever item in the master sushi belt that needs to be looped to its own chest at the insertion point?

[plain hare]

That would be active measurement. I don’t read the belt at all, I just input item / minute requests that I desire.

[hoary trail]

that seems no longer very compact

[gritty cradle]

But how does it know when to stop inputting!!?!?

[plain hare]

It never stops.

Everything on the belt is looped back to the input, so the belt never buffers

[gritty cradle]

Okay so ammo can eventually clog your belt??

[foggy junco]

No

[plain hare]

No xD . Here let me log into discord on my computer and I can show you in voice

[foggy junco]

The input is limited

[plain hare]

With streaming

[gritty cradle]

How is it input limited if you have something that constantly puts something on the belt

?

[foggy junco]

Input priorit splitters

[gritty cradle]

I just need an illustration of what is going on

[hoary trail]

how do you calculate the chest limit?

[plain hare]

here hop in voice

im gonna stream

[hoary trail]

I can't do that right now 😦

[gritty cradle]

I can't I'm at work lol

[hoary trail]

^

[plain hare]

oh lol. well.

[hoary trail]

say I want 93.75 red circuits per minute on my sushi belt

[plain hare]

i can take photos now tha show

[hoary trail]

what does that translate to in terms of chest buffer size?

[plain hare]

hmm I think id need to see your system. are you making red circuits and placing it right on the sushi belt or is there a dedicated red circuit belt?