FFF Speculations

[raven bridge]

or when looking at it (for the window level)

[pale pumice]

Or when breaking/placing pipes

[raven bridge]

Yea but that's rare occurrence

[pale pumice]

yes, but it adds a constraint that we must have a way to do this

even if it is not stateful

Introducing the new algorithm made it immediately 30% slower.

this is exactly what I imagined when I was envisioning mine

am still debating the segment merging, as it is not that simple and it would cost the simulation some detail. It is a low priority at this point compared to other parts of the update time.

This is something Harkonnen tried a long time ago, to reduce indirection in the update of the fluidboxes. Essentially, instead of each entity updating their own fluidbox, all fluidboxes in a segment would be kept in a singular part of memory, and then the simulation could be updated much faster. Initial experiments showed a performance increase of 30-50% updating all fluidboxes. However this would not address any of the other issues, and would add a significant amount of complexity to the currently quite simple handling of the fluidboxes, which we decided isn't worth the price.
It seems they tried a data oriented design like I envisioned, and got a 30-50% speedup, which is also about what I expected.

[pure siren]

the line above it:
The overhauls and optimisations in FFF-271 cut the update time by some 50% and up to 10x on some high-end CPUs.
the bit in FFF-271 that is referenced:

1. Move all the fluid logic into a separate system.
2. Merge straight sections of pipes into segments.
3. Tweak the fluid flow logic, which will not be an optimisation, but a gameplay mechanics improvement (FFF-260).

Dominik has just finished stage 1, and it has been merged into 0.17 [...]```
so some of it has been implemented

[raven bridge]

Yea some nice improvements were definitely had

[pale pumice]

the thing is, it mentions the 'throughput loss over distance' problem a lot

One needs to argue whether this is desirable behavior or not

[raven bridge]

With my system, it can be handled gracefully

As you spread the gradient over the whole pipe

That said, adding splits will completely destroy this

Maybe it could find pipe cliques and work from there

[pale pumice]

realistically, most pipes are junction heavy anyways

[raven bridge]

Or yeah, consider the whole system holistically somehow

I can't see how we can sidestep the issue of "high-low-high-low-..."

[pale pumice]

wdym?

[raven bridge]

without multiple nodes (cost), or a single node (bad simulation)

[pale pumice]

I'm still not following

[raven bridge]

Consider something like this

Pipe alternate betwen being "high" and being "low" in fluid height

[pale pumice]

yes, why is this a problem?

[raven bridge]

If it's a single node, it'll look wrong to be averaged.
If it's multiple nodes (like today), it will be computationally expensive.

[pale pumice]

well ideally you use some sort of interpolation

the thing is, if we use a node per junction here, we have about as many pipes as we had before.

the number of pipes saved is not significant

I am not interested in optimizations like these for it because they are so much more complex

I think it is a better path forward to see if we can be more clever with the cell updates

[raven bridge]

First lets move all the fluid calculations to its own thread/s

[pale pumice]

that's already been done haha

I think it may even use one thread per fluid network currently!

Do you know?

[raven bridge]

I don't know. But it makes sense

[pale pumice]

problem is that the game has memory bandwidth issues

so you often can't just 'throw more cores at the problem'

[raven bridge]

Damn memory

[pale pumice]

That's why I think data oriented principles are really where any real performance gains will be with factorio

their OOP memory model is not cache optimized, and that's likely what's needed for more performance

Note that I am applying some level of speculation here based on what I can infer externally of the codebase

The thing is, as far as I can tell, the fluid boxes are already laid out in a single place in a memory-efficient way.

So they're not even a candidate for this kind of performance improvement.

[pure siren]

you mean that?

[pale pumice]

yup

So I think all things considered, the only real 'fluid improvement' can be to first pre-check the amounts of fluids that can move, then move them

this requires a second pass, but will make the fluids behave predictably and symetrically

and will probably have a 2x performance cost...

Which is why I bring up the question to people asking about for a fluid system change: what exactly do you want?

Here's posila basically echoing what I was saying

Without access to the source code, it's hard for me to say much, but I'm pretty sure there's just not really much ground to be gained with fluids. If you want it to be better, you may have to sacrifice performance.

[pure siren]

I haven't re-read FFF-260 and 274, but from what I remember (and learnt today), the UPS improving steps have been implemented. It's still an issue for megabasing.

While the fancy new algorithm would slow things down a bit, but make fluid flow more predictable in "edge cases".

[pale pumice]

yup

well, not the full fancy one

the full fancy one with flow has some weird behaviors of it's own

but there's a simpler two-pass one that can elimiate the edge cases at some performance cost.

But again, what else are we wanting out of a fluid overhaul?

[maiden magnet]

im not sure if this was already said here but i just popped in, What is the part in factorio's fluid system that like sucks?

or eats performance

or whatever

[pale pumice]

When fluids hit a fork, the way they go is determined by placement order

eating performance is caused simply by having to go and update every fluid box

[maiden magnet]

and the whole chithat above this is about how that could be improved?

[pale pumice]

Well, more how it cannot

[pure siren]

most of the chitchat aboud the proposed fluid overhaul is just wishing it into existance

[pale pumice]

All of the optimizations I have mentioned are already existant

[maiden magnet]

i wish the devs could just do some dark magic and pop it into existance

[pale pumice]

what would it even be?

[maiden magnet]

more performant & better ofcourse

[pure siren]

Good luck finding something more performant than what we have now.

[short roost]

Maybe just reduce fluid units

Wouldnt that help a little at least

And check how many pipes are in line until an intersection and teleport the fluid after the time it wouldve taken it to move to said intersection

Or something idk im not a dev

[small bear]

feature prediction: programmable speakers reading texts

[tired fossil]

Maybe also some radiosignal transmitter because of all the disconnected power grids

[cursive reef]

100% gas chamber

[short roost]

or vacuum?

[tired fossil]

Or ufo 😅

[short roost]

huh?

[tired fossil]

Vulcanos makes it easier to produce low density structures, fulgora easier way for processors/rocketcontrol modules, so i expect any easier way for rocket fuel on the third planet

[desert lake]

Are any of you considering the newly teased order of the buildings, this one potentially being last

I found it weird at first too, but I see it as a possibility

Maybe it's an advanced lab, and the window glows when researching, and it's the only thing that can process the final science pack

[cursive reef]

Order? You should be able to choose which planet you go to

[desert lake]

One notable exception

[cursive reef]

And after the normal ones there's last one

[desert lake]

Ooh, I just realized... If my lab theory is correct, I may or may not have accidentally predicted something

Do you see that tube on it

It could be just visual, but what if there's an actual fluid input involved

So who knows...

[bold shard]

this place has been quieter ever since we got our new channels lol

keeping in line with the usual FFF pacing (content next), I think we're overdue for more space platform mechanics

[lusty ledge]

I kind of want to see what they'll do to fluid mechanics, waiting on a FFF for something like that if they end up changing it

[pale pumice]

no

[uncut moon]

I think this week might be finalised graphics on entities that were previously in concept

[candid pawn]

i just dont really get why people want new fluid mechanics, besides it being a meme

[sweet wind]

I mean we're getting more robust pipe filtering in in the prototype? Does that count enough as a rework?

[candid pawn]

like if they said "we made pipe mechanics make more sense" I'd kind of shrug my shoulders and say "okay." it doesnt really affect my game. unless it fixed some really weird fluid issues i never knew they had

[pure siren]

The last fluid overhaul mentioned in FFF-260, 271, 274 only got halfway implemented.
(the performance part, but not the "fluid make more sense" part)
Until last week, I thought all of it was in the game. Fluids are working good enough for me, I see no need for a fluid overhaul. I'm just sad all this work got lost/wasted.

[desert lake]

The only things I would care about in a fluid rework are: UPS cost improvement, and pipe throughput increase

Otherwise I don't really think it's that needed

[sweet wind]

I think most of people's issues with the fluid mechanics is that they're pretty opaque and unintuitive. You can't really know what your throughput is going to be without just, building it a seeing

which isn't really an issue broadly, as I expect that's a generally popular way to play anyway, just build as much as you think you need and change/expand it as you need to

but it definitely kicks up frustrations from some of the more detail-oriented enthusiasts.

[fresh grotto]

If they can't find a way to optimise the existing fluid system, I was kind of hoping that they would add a separate "one-way" pipe system.

It could be under the guise of "pressurized pipes" with the additional benefit of higher throughout, but it has a strict flow direction - allowing it to function more like belts and splitters with the end result being more optimized.

[small bear]

u mean barrels on stacked belts?

[fresh grotto]

Hahaha, if only... but that doesn't help with optimisation.
You would need an additional assembler for every entity that outputs fluids, plus all of the inserters needed to place them onto belts.
And that's assuming that "direct insertion" of fluids from one entity to another functions any differently than the existing pipes - which I assume it doesn't 😕

[small bear]

aha! left is getting water from steam

right is getting water from ice

[bold mirage]

very good eye

[maiden magnet]

that would actually be good

[pure siren]

with the end result being more optimized.
that sounds a lot like a way to optimise the existing fluid system

[sweet wind]

#friday-facts message

🚀🌩️♻️🚮🐊😥

new lore

[pure siren]

Doesn't the pipe system have more than enough througput for 95% of the players?
The other 5% are masochists that need it to be quirky to be fun for them.

[sweet wind]

says otherwise

Nah pipes are fine rn. A little quirky in some cases but good otherwise.

[pure siren]

Launch a rocket, fly to Fulgora. Use the lightning, unlock recycling. Something about waste separation. What's the crocodile about? Oil Swamps? Use the Blue Fries to launch a rocket and go back to Nauvis.

Nauvis bit doesn't make sense, but it's probably to not give a hint which planet is optimal after Fulgora.

[bold mirage]

the flamethrower spam in fulgora is going to be insane

infinite oil basically everywhere

[topaz wing]

They will probably be unnescessary though as lightning seems to be the main enemy on Fulgora

[bold mirage]

unless the crocodile represents an enemy but yeah

[topaz wing]

good point, that crocodile is suspicious

[bold mirage]

I pretty much expect it to be barren just that message threw me off a bit

someone said it may be the recycler lmao

♻️ 🚮 recycling scrap by hand vs 🐊 using the recycler maybe?

[topaz wing]

Walkable oily puddle swamp does seem to be the perfect spot to place a crocodile that ambushes careless players

[raven bridge]

The animation is a bit crocodilish

[bold mirage]

Can't wait to recycle a perfectly good to make more bullets

[raven bridge]

Bullets are mostly and

[bold mirage]

wait to I can just export uranium bullets from nauvis

nvm

[primal nacelle]

What if fulgora has oil monsters like Armus? (The oil monster from Star Trek TNG)

[topaz wing]

an enemy that stays in the oil would also be nice because it keeps you on your toes when walking between islands but doesn't require automated defenses when you're away

[bold mirage]

wouldn't lightning take care of them automatically tho?

[primal nacelle]

Not necessarily

[topaz wing]

That would also be just the thing to not reveal before release for maximum damage

[fresh grotto]

Well, it would be a replacement/alternative to the existing system
But the end result is optimisation, yes 😄

[primal nacelle]

Ohh, that would be nice

[sweet wind]

Current theory from #friday-facts is that the recycler looks like a munching crocodile https://cdn.factorio.com/assets/img/blog/fff-375-recycler.mp4

[bold mirage]

still what's with the crying emoji at the end?

[topaz wing]

he forgot to bring enough belts

[pure siren]

JG is sad to leave Fulgora?

[small bear]

Bullets are probably useless on

[raven bridge]

Probably. But we don't know what's sitting over the "extremely dense scrap piles"

[small bear]

I’m expecting just a pile of ruins

or some fallen skyscrapers

or giant cliffs

[bold mirage]

a still active military base that has long outlived its creators would be cool

like how our factory technically keeps running after the engineer leaves nauvis at the end

[small bear]

the previous civilization doesn’t need to defend any enemies other than lightnings tho

I’m still on the no enemy theory

[pure siren]

Wube is trying to make every planet different from each other. Doing military defense and offense like on Nauvis 5 thymes is unlikely.

[small bear]

the thing is if u need to defend enemies on it will be just place turrets and walls on every island’s perimeter. Its not that interesting and is repetitive as hell. U already need lightning rods everywhere

what even worse is that u will probably just place offshore pumps to get heavy oil for flame turrets which is a trivial logistics puzzle but repetitive and tedious

and there’s no coal on so no explosives, good luck to deal with enemies with no artillery shells

[bold mirage]

that makes raiding a base the more interesting option, no long range artillery cheese, no turret creep since you can't build on the oil swamps, and no need to worry about defense, only offense

I think that would be unique enough

lightning rods seem more akin to placing power poles than turrets tbh, it's a set and forget thing

[pure siren]

Constant lightning also means Fulgora is very hostile to life.

[small bear]

week # of predicting and on

[pure siren]

I believe we won't see enemies on every new planet. There hasn't been any info about SA enemies though, so it's 100% speculation.

[bold mirage]

I feel fits betten on aquilo being endgame and all

[pure siren]

I hope will be unlocked sooner. I want to actually use it instead of finishing the game and then abandoning the save. 😅

[small bear]

“most” means not all

[primal nacelle]

I hope that Wube avoids the problem spidertron has in SE, where it comes into play too late

[pure siren]

that means some of them have the same military targets

[small bear]

from gives a strong reason to go there first

[pure siren]

or none. language is weird

[small bear]

having the same military target is just boring and repetitive

I don’t expect that to happen

[bold mirage]

having spidertron in bacchus is maybe too strong, I can see it needing science from all three planets so you can use it on aquilo

[pure siren]

me neither

[small bear]

tbh having spidertron just removes the need to personally visiting planets that u have already been to. They’re not that OP as they’re also quite expensive

its a very nice QoL to not need to visit planets personally, but I wouldn’t say its OP

as u can already do alot with remote views

[pure siren]

objectively right, but I think most players will get a bit of tunnel vision when they see the spidertron being unlocked on one of the 3 mid-game planets

[small bear]

maybe not always if is unlocked on

also is on

[raven bridge]

might not require Aquilo, but it will likely require more than one middle science

[pure siren]

yeah, if Wube can divide the players equally among the 3 planets then it's ok lol

[small bear]

tbh if u plan your play through effectively, u might find that spiders are not that vital to get early

it’s up to playstyle I think

gives u flexible management over multiple planets, gives u the best quality upgrade route, probably boost the overall productivity most

I don’t think any one of them is objectively better than the others

[sweet wind]

✋ I volunteer as someone who almost never uses spidertrons

Them being on Bacchus is not enough incentive for me to go there first

That might change with the new logistics groups though

[small bear]

doesn’t have just spidertrons th o

[sweet wind]

will need to play with them to know

[topaz wing]

I think the importance of will also be for a big part determined by how long actual travel distances will be between planets for a small player used platform

[small bear]

aha found it

this is why I think its not on the last planet

[solemn cosmos]

cliff explosives 🚫
nukes ✅

[topaz wing]

We also haven't seen all new technologies which might even the odds between planets. Who knows, there might be something new on par with the spidertron

[sweet wind]

With elevated rail and new cliff generation I'm significantly less inclined to feel cliff explosives are a necessity

[small bear]

and iirc they did mention that rushing cliff explosives can be still faster than researching nuke as the science cost is high

[sweet wind]

Also nukes are expensive themselves 💰

especially when you could be turning it into instead

[solemn cosmos]

but has da big boom!

[sweet wind]

[small bear]

iirc nuke is unlocked by space science

specifically after u get space-science-in-space

and u get some very useful techs from just satellite space science

[solemn cosmos]

i hope we get pickles in the update

[pale pumice]

To be fair, pipe throughput is pretty easy to measure using the chart on the wiki. Anyone who is curious can quickly find an answer.

[sweet wind]

While true, It's not great to have to leave a game to learn hidden mechanics in order to understand something

Again, I think the fluid/pipe mechanics as is are perfectly serviceable and do their job well

But at the same time, it is Wube, so maybe they can come up with something better?

I sure can't

[pure siren]

Most players don't have to consult a chart because they stay well below the throughput limit.

[pure siren]

Dominik has come up with something better. But he left Wube and his work is lost.

lost as in, nobody else has the insight into the proposed fluid mechanics

[small bear]

I hope we get logi chest priorities

[raven bridge]

Imagine if that...

We got priorities for train stations, why not for chests? 🙂

[pale pumice]

Hmm, would this even be expensive?

I think it could be?

[raven bridge]

That's for Wube to figure out

[pale pumice]

I mean performance

[raven bridge]

That's for Wube to figure out

[candid pawn]

I was thinking this the other day when playing Ultracube!

It would be a really nice QOL feature for bots

Belts already have it, trains are getting it, and it would be pretty easy to set priority in the requestor chest UI

[bold mirage]

maybe i'm missing something but how are there molten metals on fulgora?

you can only get those from lava or ore right?

[frail kestrel]

That looks like light oil/heavy oil/petroleum. No molten ore.

[bold mirage]

oh i figured one was light oil I didn't realize the rest were petroleum stuff too lmao

so there's a recipe that takes petroleum gas, heavy oil and stone

weird stuff

[topaz wing]

That recipe is the electrolyte

[pale pumice]

This is #1177597309572366396. If the idea has serious performance issues, we can know they won't be adding it.

Everything discussed here is 'for wube to figure out'

[raven bridge]

What I mean, is that it's an implementation detail

[candid pawn]

tbh I'm sick of people binning cool features because of performance issues that they dont even know exist or not

People binned multiple surfaces because of imaginary performance issues too

The game is 10 years old now, hardware has changed and the game's been well optimized

[pure siren]

People binning features has no effect unless they're employed by Wube.

[short patio]

Which is weird because multiple surfaces seem like they could improve performance

[pale pumice]

Yes but multiple surfaces doesn’t even seem like it would have theoretical performance issues.

Something like ‘bot priority’ is a simple question of algorithms

much like fluid overhaul. We can be more sure we aren’t getting one due to the failure to come up with a vastly better algorithm

[hybrid prism]

it would be cool if surfaces ran on different threads for better performance

[raven bridge]

The problem is memory transfers.

[hybrid prism]

why is it a problem

[raven bridge]

Because it's the bottleneck, rather than cpu utilization

[hybrid prism]

hmm

ok

that makes sense

[orchid onyx]

i predict some kind of method to transfer 1 type of item into another type

[pale pumice]

wdym type?

[bold shard]

transmutation maybe? like how in K2SE (don't know the exact mod) you can convert stuff into matter and convert it into like 10-20 other things

wouldn't be a fan of it personally

[sweet wind]

I always find the transmutation stuff really cheesy

[tough ginkgo]

Prediction: non-hr graphics will be abolished. Older, potatoesque systems will downscale the hr versions to save vram.

[raven bridge]

Prediction: Driving improvements

[candid pawn]

that would be nice

[elfin zenith]

O shit wrong channel XD

[candid pawn]

it could be, but given that all of the other buildings in that image have been unique buildings on a planet, I think this will be the unique building for bwohuo

no its a good channel for it

[elfin zenith]

Oh

[hasty rapids]

The final fff will be 534. I can feel it.

[uncut moon]

Ever?

[slender glen]

introducing: clutch

[uncut moon]

5th gear straight into reverse when you see a biter nest

[small bear]

we need car drifting

[desert lake]

We need train drifting

Multi-track drifting

[raven bridge]

Well the greatest driving improvement is the better cliff collision boxes

[ashen herald]

I predict, spider car, legs instead of wheels

[west wing]

I predict the fall of Wube

[ashen herald]

That's going to happen with the eventual heat death of the universe

[hasty rapids]

Not ever. Fff 534 will announce SA

[small bear]

SA is already announced

fff 534 is like, more than two years from now!

FFF434 sounds like a reasonable time for SA release

or even sooner

[hasty rapids]

Mark my words. Maybe that's when the world will end.

I get my special dates confused sometimes

[raven bridge]

We're not gonna get weekly updates after SA

[bold shard]

with 1.0 I think they continued until 1.1, or a little after it before going dark

[elder raptor]

I wonder how many free updates they'll do post-2.0

and will they go into making 3.0 lol

[candid pawn]

unclear. i doubt they know themselves. most of wube's history was moving the game from early access to 1.0 and then 1.1. they dont really know how they want to cary themselves in a post-release world

[bold shard]

whatever kovarex's vision is, I like to hope there is more to come

from the player's end, the interest is certainly there

[raven bridge]

I'm sure they will want to move on to other projects at a certain time in the future

I predict it will be a year or two after SA is released and stabilized

[pure siren]

I hope not. I like Factorio and want it to be supported until Windows 57.

[bold shard]

I want to be in my 90s and teaching my grandchildren on Factorio 10.1 about the metas I used to do back in olden times

[raven bridge]

"Heh... remember when quality was considered complex?"

"In my days, full Py's was just 1000 hours... so short"

[small bear]

Factorio 3D

[slender glen]

We will skip right into 4D

[small bear]

tbh I feel like it’s nightmare level difficulty to make good graphics in a 3D grid based factory game

shapez2 did it quite well tho

[raven bridge]

You have E thinking in 12D already

[primal nacelle]

Factorio with multiverse time travel? 🤔

[slender glen]

Shapez2 is a 2d game with 3D graphics engine

Unless I'm missing something

[slender glen]

Hard to make a multiverse time travel game that is not turn based

[primal nacelle]

I'd say it is hard in any case

[slender glen]

Isn't managing multiple saves at a time a factorio with multiverse time travel?

Or how would it be any different

[raven bridge]

Shapez2 has bridges

But it's just 2 levels, so it's basically 2D(x2)

[slender glen]

Are bridges any different then underground belts?

[raven bridge]

Actually x3

[slender glen]

I see

[raven bridge]

It's 3 layers, rather than a single layer

https://images.gamewatcherstatic.com/image/file/9/5f/125189/ss_4405d7fba61786df8a9a1fa8056760e799e763c6.1920x1080.jpg

So it's not fully 3d but rather 2d with several layers

[slender glen]

At least it looks readable

Full 3d factory game would be belts going in all directions including up? 😄

[raven bridge]

In Satisfactory, all belts are virtual. They can go through each other, which is kinda unsatifying

[slender glen]

What

[minor spruce]

Please make it happen!

[raven bridge]

Belts aren't checked very much for collisions, they can go through one another

[pure siren]

what?

[slender glen]

Why

[raven bridge]

I guess performance issues? Or just 3d being difficult. You'll have to ask them.

[small bear]

well belts in Satisfactory can even phase through buildings

[rocky plover]

It really shouldn't take much performance to check for belt collisions. You only need to do it when they are built and never again since they don't move.

[desert lake]

Here's to hoping the Bacchus building will make pipes and engines

[final dove]

the most interesting aspect of their layer system is that you can build any factory component at any level. and at least one of the components has multiple inputs on different layers.

if I'm honest tho they managed to strike the worst possible balance of builds being each too different to be able to save effort with copy paste but also not different enough to be interesting

[raven bridge]

Haven't played with Shapez2 yet, but in 1 I used a lot of copy paste

I really like their idea of "one machine - one job", similar to how Unix commands were initially meant to be

Now, all the target shapes are very different, so you need to design different factories for them. A big plus in my eyes.

I had a lot of direct insertion (rather than bus) in Shapez1

[final dove]

its really fun while you're still learning the ropes. I got board pretty quick once the shackles were off

[raven bridge]

Haven't played S1 beyond the original design (18 levels)

All the random shapes things

[final dove]

oh yeah, there's no reason at all to do anything but totally insulated builds

[raven bridge]

For that you design a universal machine and call it a day

[final dove]

and its not even that hard :P

just big

[raven bridge]

I wonder if I'll do more generic designs for Factorio now that we have param BPs

[final dove]

yeah I'm worried that it'll follow the same curve

super fun to design at first but then be too optimal and no longer fun

[raven bridge]

BTW we have shapez like layers in Factorio too... https://www.youtube.com/watch?v=-xP1FVh8yzk

I personally didn't like the idea of how megabases require a lot of traffic for intermediates, so instead I just designed a monolith "ores -> labs" and call it a day
https://storage.googleapis.com/soulburn-mapshot/mapshot/death/index.html?x=336.3&y=-177.1&z=2.0

[final dove]

I don't know what I'm looking at but I'm a sucker for parallax so I'm happy

[raven bridge]

It's a mod, and it actually works (slowly)

[final dove]

impressive

[raven bridge]

It's 250eSPM btw

[final dove]

250x10^spm

[raven bridge]

Yea 250 with SPM zeros after it

[final dove]

how does your cpu handle so much science!?

[raven bridge]

Wube are Wizards

[final dove]

I love scrolling around in those mapshots. hope you don't mind me being nosy

[raven bridge]

That's why it's there

[final dove]

that poor old speed mod build. abandoned

[raven bridge]

Yea it was replaced by the factory in the south desert

[final dove]

how's the earlygame mall work out for you with just the one yellowbelt of iron? I always get impatient without having a whole belt on top of an extra one for just the gears + chips

[raven bridge]

Totally fine. That's what speedrunners do

I did upgrade it to red and later blue though

I totally forgot about the "extra" iron belt on the right hahaha

[final dove]

huh, what am I doing wrong then lol

[raven bridge]

Do you buffer too many items?

Maybe you switch to too early

[final dove]

I don't think so. but the red belt that downgrades after the gears is a nice touch

[raven bridge]

going to eat ttyl... feel free to write more

[final dove]

see ya!

enjoy your food

[small bear]

usually I need a bit more than one yellow belt of iron for early mall. like, 1.5 belts or 2. I like to have one belts of iron next to half belt gear halfbelt chips

[final dove]

^^ that's what I do

I can't be trusted to upgrade it later more than just redbelt it so it has to be able to still function once all the steel and oil stuff gets added too

[small bear]

I almost never automate refineries and chem plants until I get

for some items that are not used in bulk I usually just handcraft

which saved me some buffered items

[final dove]

I see your point and raise you: if you automate it then the construction bots will just handle it

(and yeah I know you can hand craft and then plop it in a storage chest)

[small bear]

the thing is u can handcraft and then hand feed into red/yellow chests

well, having bots to handle them all the way is still handy, which is why I switch to eventually

[final dove]

in the rare times I actually have logibots I'm a fan of using the trash slots to hand items off to the network

[small bear]

after I have I almost don’t need handcrafting anymore

there are rare times where I rely on + handcrafting

using personal requests

I act as the logistics hub myself

[final dove]

handcrafting is so slow. I find I basically only handcraft the items that take steel before I get them added to the mall

[small bear]

but usually at the time I mass produce I already have unlocked

[final dove]

interesting

I can't live without construction bots once they're in sight

[small bear]

depends on what u craft. If it’s your first 10 train stations I’d say its fine to handcraft

[vocal rock]

stuff like assemblers take 0.5s with all intermediates - totally fine to handcraft

[small bear]

and are different things

[final dove]

I'm in the habit of adding those to the mall well before I end up using them

stations that is

[final dove]

whoops you right

[small bear]

I tend to not do this because I find it will eat my intermediates for something I don’t need now

[final dove]

that's why they go alllllll the way at the end of the mall

[small bear]

and the materials never goes there because I use other mall items too fast

[final dove]

good problem

as I get faster at the earlygame I've started building a second long inserter assembler and redesigned a lot of my builds to bias the red bois. I'm liking it

[small bear]

iirc handcrafting speed is 1x, speed is 0.75x

so it’s actually faster

[final dove]

that it is

if you have all the intermediates

[small bear]

iirc I needed less than 20 chem plants to get infrastructure to unlock

[final dove]

I think I handcraft something like 7 of them? and that lasts me until I get around to putting them in the mall

[small bear]

yup

I also handcrafted my first 4 nuclear reactor cores

[final dove]

I go very oil light to start

[small bear]

they last quite a while before I need another nuclear plant

[final dove]

and let bots quadruple it once I've got construction bots automated

[final dove]

I've been putting off redoing my nuclear build...

so I regret to say I do a LOT of steam engines before slapping together an ugly nuclear build that I'm sure has so many flaws

do you have a nuclear build that you like?

[small bear]

yup

I like direct feeding steam from exchanger to steam turbines in a 1:2 ratio. It’s not the optimal ratio but it looks neat

[final dove]

those builds are certainly aesthetic

[final dove]

that's nice!

this is the kind of mess I usually make

just one loooong line

that tiles

[small bear]

this design seems to need a lot of landfill

[final dove]

it does indeed 😅

in exchange I can pack the water pipes and not have to think about routing them

[raven bridge]

I put my offshores near the reactor, so they can be built over rivers rather than needing lakes

[final dove]

I lived this life one time. never again...

[raven bridge]

? I don't use pumps

[quasi crest]

why not?

[final dove]

if memory serves you can fit 5 offshores into a single pipe if you use pumps like this

[raven bridge]

For my nuclear reactors I mean

See here https://storage.googleapis.com/soulburn-mapshot/mapshot/death/index.html?x=-240.1&y=-193.0&z=4.6

I'm working on a better version, but that idea is the same there too

[final dove]

I build the line long enough that by the end it needs pumps and so I just do it thruout to not need to think about it

yeah, instead of splitting the pipe you can use pumps

[raven bridge]

Well you don't need pumps for water because the exchangers reduce the pressure quickly

This specific design is built around 5 high lines so it can be easily tilable

But it's more efficient in space to make it 10 high, as it allows stacking exchangers and turbines 3 high

[small bear]

I did need pumps between my exchangers based on testing, but I think it’s a quite rare design

I was feeding 24 exchangers using a single pipe, and there’s an underground jump in the middle

[raven bridge]

Yea that would do it.

[final dove]

ok, real fff speculation time. if we assume the fusion reactors are going to be a thing how do we think they'll work?

[bold mirage]

Asteroid mining for helium-3 with the space platforms could be interesting

[final dove]

what do you do with he-3 ?

[bold mirage]

Iirc its a good candidate for nuclear fusion fuel irl

[final dove]

now that's cool

what form is it in in space? solid chunks or particles or something?

[bold mirage]

I hope we get some space exclusive resources (apart from space science of course) that are useful on the surface

[final dove]

me too

[bold mirage]

Im planning on shipping down iron for quality given that its a freebie

[final dove]

reasons for platforms to wander around besides doing a train impression would be a great addition

[final dove]

me too!

and you know that if the platform is made mostly of iron that that platform will be self building for sure

[bold mirage]

Even things like steel or engines, thats only iron too

[final dove]

[small bear]

tbh u will probably not run out of ore patches once u have big miners

[bold mirage]

also (and on fulgora too) are free, so space mall can be a thing lmao

[bold mirage]

true, but until you get to that stage a small free source of iron can help a lot (as well as ice on the more water limited planets)

in the endgame it doesen't matter anymore

[small bear]

u get big miners quite early

before that stage space platform are just expensive to setup

[bold mirage]

that depends on the order you choose to visit the planets

if you go vulcanus first, sure, that my be better

[small bear]

I mean even without big miners, ore patches lasts quite long. If u go to Fulgora first u can get quality miners sooner, which also helps slow down patch depletion

[bold mirage]

still I think there is some value to having some production without "stealing" resources from the rest of the factory

[final dove]

and no matter how expensive, I'll take an iron source that I can just copy paste over having to setup another outpost

[small bear]

depends on the infrastructure cost

if it’s more expensive than mining prod research

[bold mirage]

that and how much you are willing to wait for ROI

[small bear]

getting free resource from space will probably take more IRL time than setting up miners

[final dove]

at worst its a resource sink for when nauvis would be idle otherwise

or any other planet for that matter

[small bear]

infinite research is resource sink already

there are many infinite research that only requires Nauvis science packs

[final dove]

tempting

[quasi crest]

is it not tritium and deuterium thats used for fusion?

[bold mirage]

yes

idk if he-3 is actually used or is a more theoretical thing

but I've heard many people mention that one of the best reasons to make a mining outpost on the moon is because of its richness in helium

but given that's not coming anytime soon SA is the best next thing

[small bear]

I hope tomorrow will be a space platform FFF

[pure siren]

oh, it's the day before FFF already
this week zoomed by :)

[pale pumice]

that makes sense actually. engines especially

but it seems kinda weird if it's a chemplant equivalent...

[small bear]

tbh engines and motors will probably stay as assembler recipes so assemblers remain useful for intermediates

[pale pumice]

I mean, there are still plenty of assembler uniques

but I agree

[small bear]

tbh there isn’t a lot intermediates left for assemblers once circuit related stuff are taken by EMP. Concrete, gears and LDS are taken by foundry

[pale pumice]

Would you mind listing the ones left?

I feel it is still sizable

(I can't - am at work)

[small bear]

electric motor and bot frame

that’s all

oh and Uranium fuel cell

ah I totally forgot Nauvis science

they’re all made in assemblers

oh I assumed that will have its specialized machine here

and RCU is removed

[pale pumice]

I suppose that much is true

yeah, that's pretty much all known intermediates

[bold mirage]

there's not much room for what the 4th building could be

[small bear]

the 4th building doesn’t have to buff Nauvis intermediate

[raven bridge]

Are assuming the 3rd building is a / ?

[small bear]

yup

[raven bridge]

The 4th building could be something completely different

[small bear]

like the centrifuge on Nauvis

[raven bridge]

Yes.

It could be a lab, a weapon, a roboport, a train thing, a miner of sorts, or something completely different like the lightning rods

[small bear]

I still expect it to be an assembler with a list of special recipes

[raven bridge]

Particle accelerator turning lead into gold

[short patio]

I think it would make sense for the foundry to make pipes, maybe as a recipe that needs to be researched

[ashen herald]

How about pipes that make sense to be hot without melting, I mean if they can have molten lava piped through, couldn't they also be impervious to flame turret fire?

I suspect that they will make me happy though

[raven bridge]

Pollution overhaul when?

[candid pawn]

do you have any ideas on how that would work?

i like how there's been so many QoL FFF that we're feeling that every system in the game is going to be reworked. (Not that I would be opposed to a pollution overhaul but)

[ashen herald]

Pollution might just be turned off on planets with no enemies

Or in fulgora where it's probably already polluted to the max

[vagrant pumice]

Vulcanus is already polluted too, there is sulfur gas and lava everywhere

[maiden magnet]

I dont understand what half of that means, but more vram would be neat

[pure siren]

Vulcanus' pollution could be renamed to vibrations or so and that wakes the Balrogs up.

[tough ginkgo]

Factorio started off with low-res graphics. Then 'high-res' files (starting 'hr-') of twice the resolution were introduced for people who could use them. See for yourself in e.g. data/base/graphics/entity/assembling-machine-3. VRAM is video RAM. Rationale for the prediction is a) it simplifies prototype definitions and now is the time for breaking changes, b) factorio already supports CPU and GPU downscaling, so people who can only run low-res would only be disadvantaged by slightly longer startup time and slightly suboptimal graphics.

[vagrant pumice]

I predict a train fff

[quasi crest]

what more can they add to trains tho?

[pure siren]

smart trains powered by AI

[quasi crest]

so... better interrupts?

[desert lake]

I'll tell you what they could add

The fabled multi requester

Or more likely some way to rig one up

The simplest option would probably be the ability to change a station's name with signals

[desert lake]

I know about that

That's what gave me the particular idea in the first place

[small bear]

the generic interrupt response was a lot more positive

[desert lake]

Oh, I'm not saying it's likely to be added

I mean I personally probably wouldn't even use it

I was just replying to the question of "what more could they even add to trains"

[small bear]

I’m hoping we get a space platform FFF today

I want to see more gameplay showcases

and more tech tree

[desert lake]

I'm personally super curious about space platform mechanics

How crazy can we get with these things

Because we know there will be "more advanced crusher recipes"

The implications are so exciting

Space trains are cool, but what I'm really interested in is making the huge stationary ones

I hope there's at least some incentive to make those besides just white science

[pure siren]

We'd need some way to get Copper. Without deincentivising ground bases.

[desert lake]

The asteroids are just called "metallic asteroids", not specifying iron, and they look like this

I'm no space expert, I very well could be completely wrong, but I don't think iron would oxidize and change colour in space

So that orange stuff could potentially be copper, if we really stretch it???

Ah, wait, the iron doesn't oxidize, apparently it's formed as oxide to begin with

It makes a little more sense this way, as the asteroids looked a little too orange for something players would first and foremost gather iron from

[pure siren]

The asteroid resources have been mentioned, Copper was not one of them. They could be part of the more advanced crusher recipes that have been hinted at though.

[small bear]

pretty sure we will be able to get copper from asteroids later (perhaps with the advanced crushing)

and even some oil sources

[quasi crest]

oil would be impossible as is requires dead animals and dinosaurs to be buried for like a million years

[fluid lichen]

could be in a solid from and be used to only craft things that use plastic

[pure siren]

Coal Liquefaction to the rescue. Although I think we shouldn't be able to craft everything in space.

[fluid lichen]

space hub

[small bear]

if a platform can craft its own components mid flight I think it mean it has access to most Nauvis resources

carbon liquefaction perhaps

or alternative chemical production routes from carbon

[pure siren]

research enough belt stacking that you can grab some barrels on the platform

[small bear]

technically with ice and carbon we can rearrange them into other chemical products

[fluid lichen]

and again it could just be something they were trying out but did not work how they wanted to because it ends with a ?

[small bear]

I’m quite confident that they already have something that works well

otherwise they wouldn’t say “more about that some other day…”

[desert lake]

Well, duh

Also, one thing I potentially see happening is space not having access to any form of oil, but carbon being able to make plastic

And potentially being able to get sulfur out of oxide asteroids

I don't think heavy and light oil would see much use up there

[raven bridge]

I didn't expect radar circuits, but it's welcome

[small bear]

yeah I thought we would have a new entity for that and able to set multiple channels

radar as a single channel connection is quite an elegant solution

hum... so the default speed is 10 crafts per second?

4.33 = 260/60

then if we assume its still 0.5s per gear, then the crafting speed of foundry is 5

[bold mirage]

20 molten iron in the foundry means 10 per gear craft so 5 per iron plate right?

[small bear]

and according to the input number, each gear take 10 molten iron

[small bear]

it could be also 10 per iron plate

so casting gear would be always more efficient

compared to casting plate -> assembler crafting gear

[bold mirage]

hm, i guess

[small bear]

5 is a weird fluid input

usually its at least 10

[bold mirage]

i figured the extra efficiency of the recipe would come from the +50% prod

[small bear]

casting plate and casting gear has the same amount of built in prod. but if u cast plate and craft gear in assemblers u get an extra prod step

[bold mirage]

oh I see

i thought casting plate didn't have prod at all, like barreling/unbarreling

and you got the extra prod from smelting

ok so yeah 10 seems reasonable

[small bear]

and I sus that steel will be 20

[bold mirage]

Something to consider is that a wagon full of plates is 4k items, a fluid wagon full of molten iron could be 5k to 2.5k depending on the plate to fluid value

Im honestly hoping its the former since molten metal trains sound fun

[small bear]

don't forget the built-in prod

x1.5

even if u don't use prod modules in foundry, a wagon of molten iron for plates is already 3.75k items

[bold mirage]

The thing is fluid trains are more versatile, a molten iron train could work as a plate, gear, or steel train

At the very least im planning on a fluid only bus on vulcanus

[small bear]

if u use for casting, then fluid train is more dense than plate train

[fluid lichen]

I personally think its a way of remoting the player into a planet without actually traveling there. but there will probably be limitations like inventory size and needs to be within radar range

[small bear]

what is that

[fluid lichen]

not part of the foundry

[small bear]

imagine if casting steel only takes 10 molten iron

I'd say 20 is more reasonable tho

[bold mirage]

And then productivity tech on top of that

[small bear]

yeah. The thing is that prod tech may apply to both the casting and smelting steel recipe

[bold mirage]

Why casting? Isnt that molten iron?

[small bear]

but its still a recipe that makes steel

[bold mirage]

Ah sorry i misread

I thought you said melting the iron

[small bear]

ah

[bold mirage]

Yeah it should apply to both processes

[small bear]

I think it makes sense that 10 molten iron for both plate and gear. As the maximum amount of extra prod u can get by casting plate -> assembling gear is 4x which is +100% (2x)

if we compare different routes of steel, in the most extreme case where u maxed out recipe prod bonus (so its +300% prod for both casting and smelting steel), the difference between casting steel vs casting plate -> smelting steel is that the latter has an extra prod step in foundry when casting the plate (4x and bultin prod) which is +150% (2.5x).

so it makes sense that iron efficiency for casting steel is 2.5 better, which means instead of 5x cost compared to plate, it should be 5/2.5 = 2x the cost of plate, which means 20 molten iron

for steel assume max recipe prod research

[half mountain]

“THE CLAW”

[bold mirage]

with 10 per gear and accounting for +50% prod we are looking at around 1700 fluid units per second

so two pumps in most cases

wait how is it getting enough fluid, it consumes 40 per tick but pumps can at most fill the machine once per tick, so 20 per

[small bear]

perhaps the assumption is wrong

machines could be filled more than once per tick

[desert lake]

Could the claw belong to the thumbnail cryptid from fff 389?

Could someone do pixel measurments or something

[small bear]

u can see in the mp4 its moving

behind the machine UI

and its not shown in the machine UI so its definitely not part of the foundry

[desert lake]

I mean, thanks for pointing out the obvious, I guess

I just realized, the thumbnail cryptid has some dead trees near it

That could potentially be used for height reference

[fluid crow]

who noticed the space balls reference in the recent fff?

[small bear]

u mean the ender pearl

[solemn cosmos]

you guys are too smart

[raven bridge]

They obviously wanted to leak this

[solemn cosmos]

you dont say

[ashen herald]

I'm going to say that it's an addon to the engineer, or an automaton that is like the engineer, pre-spider remote control unit.

[vagrant pumice]

so it is the robot-like unit from the other fff thumbnail

[ashen herald]

Which fff?

[hybrid prism]

389

[fluid lichen]

this image

[short patio]

Last Friday's mention of videos taking a long time to make is probably implying that they need to spend a good amount of time going over every video with a fine comb to make sure only the intended information is leaked

Because just filming driving around in the car would be pretty straightforward

[small bear]

quick, someone calculate the base power of foundry

hum, if my math is correct then each foundry is about 2.5 MW at default

[bold shard]

certainly not trivial, but its to be expected

[tough ginkgo]

if its max power is 1MW then via a few assumptions I get 37.5MW (ignoring unknown min power) but still just short of 37.6MW so probably wrong (edit: foundry is legendary)

[short patio]

The foundry being legendary shouldn't affect the power draw. You can just calculate the energy increase from the number of speed modules affecting it

[tough ginkgo]

i threw in an additive 150% energy cost for the foundry being legendary; dunno if this is correct

[short patio]

Legendary quality items don't spend more energy per craft

It's explicitly a bonus for them on space platforms, that they do more with less power

[small bear]

quality machine doesn’t increase power consumption

[small bear]

I just checked in 1.1 that all machines has their idle power = 3.33333% of working consumption

[tough ginkgo]

i think it's generally 1/32

do you have a source for quality machines not increasing power consumption?

[small bear]

I think its common knowledge but u can probably find it in the quality and/or space platform FFF

[small bear]

depends on how u calculate it. it is 1/31 of max power, or 1/30 of the working consumption

[tough ginkgo]

it's not in the quality one. do you mean "Higher quality chemical plants produce fuel faster while consuming the same amount of power" from the space platform one?

[small bear]

quality only give u benefits

[tough ginkgo]

ok ok but what is your source?

[small bear]

quality FFF?

All higher quality entities have more health, but apart from that, they get unique bonuses based on their type:

[tough ginkgo]

the statement i posted above "Higher quality chemical plants produce fuel faster while consuming the same amount of power" is from a FFF and is useful, are there others?

[small bear]

no

but thats sufficient evidence

[raven bridge]

Both of you have default red discord profile, which makes this conversation confusing

[tough ginkgo]

so here's my point.... if energy consumption goes up by 150% and crafting speed goes up 150% then your chemplant will consume the same amount of power for the job, i.e. per craft

[fluid lichen]

wait theres 2 people?

[small bear]

my point is that quality wouldn't be that powerful on space platform as the mentioned if it also increase power consumption

[tough ginkgo]

but it wouldn't per unit craft if energy consumption went up with quality bonuses - this is one interpretation which i think is valid enough to consider

in the absence of any other hard facts

[small bear]

also I'm 70% sure that u can find some devs in discord confirming that quality doesn't increase power

its basically a common agreement at this point iirc

[raven bridge]

Of course it won't increase power consumption

[desert lake]

There’s also the fact that quality increases the bonuses of modules, but not their debuffs

I think it’s pretty obvious quality is supposed to be a 100% positive thing, different from modules

[raven bridge]

I do wonder though, if inserters will take more power. By default, if you increase their speed, their power increases because it's "power per swing". In JQ I had to specifically reduce power usage so the final result is the same power

Also, use less power

[tough ginkgo]

do you get my point that increasing power consumption at the same rate at crafting speed keeps the power consumption per craft constant?

[raven bridge]

Yes, but they never said anything about that.

[desert lake]

Right

[raven bridge]

Power in machines is "when active"

[desert lake]

I mean I guess it could potentially play a part in those weird “ammo manufacturing pollution cost” calculations people do to prove that red ammo secretly sucks

But I don’t see it as something particularly significant

[small bear]

its energy, not power if u calculate it for per craft. Unless u said 'per craft per unit time'

[desert lake]

It’s trippy to see you two converse, it’s hard tell who’s who 😭😭😭

[tough ginkgo]

don't tell my physics teacher

[raven bridge]

Because they are both

[small bear]

tbh the value of red ammo is more for offensive attacks instead of for defense because fast killing with less turrets matters more in that case

[small bear]

did this make it better?

[raven bridge]

Perfect

[small bear]

[final dove]

you have become Pupperbox

[small bear]

[final dove]

[small bear]

week 3 of hoping a space platform fff

Factorio 2.0 makes me unable to play Factorio 1.1, and Factorio 1.1 makes me unable to play other games

[bold shard]

so keeping in mind the pattern- new content/something big next week?

[raven bridge]

Next will be more features. New planet will next month

[small bear]

hopefully space platform fff

[raven bridge]

What about biter dating sim?

[raven bridge]

Prediction: Shovel-guy will be unveiled in 2-3 weeks.

Prediction: Shovel-guy can't enter vehicles, but you press Enter on it, it transforms into a car

[desert lake]

I'm still on team early spidertron, and believe shovel guy is just an armor suit

But the fact that he was teased a whopping two whole times may indicate some greater significance, so who knows really

[fluid lichen]

i think shovel guy is a remote in armor suit

[candid pawn]

transformers!

[vagrant pumice]

omg I'm dying for some transformer or mecha style robot

[candid pawn]

hopefully there's some hardcore transhumanism stuff where we need to leave our humanity behind and replace our flesh with metal

[vagrant pumice]

another mod to the graveyard (mechanicus engineer)

I'm sure the robot or armor suit would make sense when we get the Bacchus FFF

[candid pawn]

Bacchus has poison gas that melts biological material but machines are unaffected

[small bear]

we will probably have a fff dedicated for spidertron remote

[raven bridge]

Radars connect "per surface". This, unfortunately points that we won't have a multi-surface planet.

[short patio]

What would you do with multiple surfaces on a planet?

[bold shard]

if there were underground layers you could travel to, you could have big parts of your production/processing there, substantially shrinking your above-ground footprint

[short patio]

It does make sense for radar to not penetrate under the ground

[bold shard]

well there are ground penetrating radars IRL, but I don't think they can scan beyond some tens of feet square.... very megabase unfriendly

[small bear]

if there were underground layers they would have added the underground train instead of elevated ones

[pure siren]

could connect two (adjacent) surface levels with a pair of radars at the same coordinates

[raven bridge]

That's a novel idea

[ashen herald]

I wouldn't rule out a subterranean expansion in the future, hopefully it would build upon SA though.

The radar circuits are more of a convenience thing, not something revolutionary. What would be revolutionary is if we could connect radar groups so they only listen to ones within their radar group.

And also I think it points to long range transmission like a big radar for interplanetary

I think it will be cumbersome without it unless the platforms can innately be smart enough to know what resources need to go where and when.

[small bear]

the landing pad will probably take the big radar role

[topaz wing]

Yeah that makes sense. it's limited to one per planet as a hub and it also has a radar on it

[ashen herald]

Interesting idea

[raven bridge]

It does indeed have a radar on it!

That said, you want to connect it to wires to request stuff, so I would not think it connects to other radars, other than the platform

[ashen herald]

That's what I was thinking, that it could act as a more advanced train station, but actually transmitting information through that wouldn't go beyond resource requests.

[elder raptor]

I'd love that as a, say, 3.0 update. I don't think it's super likely to happen on its own, but with other stuff like interstellar travel where you want to make the most out of a small space (again)

There's many buildings and natural resources that would limit subterranean expansion; the rocket silo, and... well, it's the only one I can think of rn

Having to put regular support pillars, etc.

[small bear]

trees

[elder raptor]

Yeah, basically all natural resources will

[desert lake]

I've thought about this subterranean idea a lot recently, trying to come up with things for it

The thing with it is, it's really hard to find any roles for it besides "put more factory in less space"

Then again, maybe that's enough for some people

Unless we moved a significant chunk of mining to the underground layer, or something else drastic like that

Subways would be neat I guess, but there's the question of how you would unload those underground trains

[elder raptor]

I assume you could just build stations underground

[pure siren]

There're some underground mods that add another surface just like the "surface".

[ashen herald]

The main focus of subterranean layers I think would be resource distribution, where it would be more dynamic in finding different resources with richness. Dangers might also make it interesting, like having to build more supports going deeper, and having to deal with gases, or fluids of different temperatures that could slow down or hurt your machinery if not properly handled.

I just would love if prospecting were a thing, where you don't know how big an ore vein is just based on the surface, or might not even be visible at all from the surface. Maybe ores are more mixed underground leading to increased strategy for filtering.

[fresh grotto]

I could only see a subterranean layer working if it involved improving efficiency to make up for the added complexity and management.

Perhaps incorporating some elements of the new planet mechanics into one location - i.e. liquid metals, "recycling" mixed ore into large quantities of useful raw materials

[solemn cosmos]

i think we need more artillery shells. like we should be able to make poison capsule shells. crafting recipe would be the same but instead of explosives in the shells, you put poison capsules

[primal nacelle]

Nuclear artillery

[uncut jay]

orbital artillery would be the next step

[tired fossil]

Also need some stuff for asteroids bigger than "medium"

[desert lake]

More artillery shells are always a neat idea on paper, but in practice it's just
a) It's weaker than vanilla shells, and is never used
b) It's better than vanilla shells, so it completely replaces them

Because artillery shells are very simple things

They make an enemy base disappear, which then provokes an attack on your stationary defenses

There's not a whole lot of room for nuance, they either do their job well, do it poorly, or do it too well

[candid pawn]

there's nothing wrong with an item completely replacing a worse item. for example, the assembler mk2 completely recplaces the assembler mk1

[desert lake]

Well, of course, I didn't say there was anything wrong with it

Just that there isn't room for as much variation as some people imagine

But also there is a specific problem with artillery shells, that being their balancing of triggering instant attacks

If a shell was a little too destructive, enough to wipe out enemies around its target, that would start dipping into OP territory

And considering it isn't really meaningful to increase any other stat of an artillery shell other than its explosion radius, that's probably one of the only results that could be achieved by an upgraded artillery shell

Unless they made it sort of like a reverse poison capsule, that only destroys nests and worms but not enemies

[candid pawn]

that seems like a fun idea. what about a artillery that delivers a poison capsule that only kills biters and spitters, but not worms and nests?

[ashen herald]

I like when the item that completely replaces the item uses said item in the recipe

Well pre-recyclying it feels necessary, but in 2.0 I'll just recycle obsolete items

[raven bridge]

Telescopic recipes have their merits, but there's something nice in non-telescopic recipes. IR2/3 does a lot of different tier items with different recipes. Solved by adding a recycler 🙂

[frail kestrel]

Hopefully the enemies on other planets will have different resistances than biters. Say you had an enemy building that is super resistant to explosives so normal shells won't be efficient, but is vulnerable to a different damage type.

[raven bridge]

Interestingly, in vanilla, armors are not telescopic. But that's just to save players from eating their grids

[desert lake]

expect vulcanus enemies to be fireproof, and we already know asteroids are laser resistant

[frail kestrel]

Did they say they actually laser resistant? I thought the problem was lasers are power hogs which isn't great in space.

[desert lake]

Past a certain size, asteroids become practically laser proof

iirc

I'll dig up the exact info

[small bear]

I almost always eat my grids in SE when upgrading the armor 😅

[frail kestrel]

So both reasons. I wonder what larger asteroids will need. Artillery against moving targets seems impractical. Flamers in space makes little sense. Tesla weapons make even less sense in space. So I guess something we haven't seen.

[raven bridge]

at least a gun turret

[pure siren]

a tank turret? I'd like that a lot.

[raven bridge]

My prediction is also a tank turret

[solemn cosmos]

yeah your right.

[tired fossil]

Would guess a rail-gun with supercapacitors thingy

[pure siren]

If we can deploy it on the ground, I'm all for it.

[small bear]

I wonder if we will be able to set enemy filters in turret UI

[quasi crest]

but... why??

[small bear]

heavy weapons only attacks large asteroids/big biters

something like that

laser turrets only attack spitters

enemy filter can be useful when u have a combination of different turrets

[quasi crest]

interesting

but i have flamethrowers

[small bear]

perhaps u don’t need filteres in flamethrowers, but in other turrets

or, imagine if theres fire immune enemies

[primal nacelle]

I kinda hope there is variety like that. Even on Nauvis. Even if it is an optional setting to have enemies which randomly/something have a high resistance or immunity to a single damage type.
Ofc that would further nerf spidertron, since no additional weaponry

[quasi crest]

i was just joking about how op flamers are in 1.1

[small bear]

do u know how op land mines are in 1.1?

[quasi crest]

yes

but i cant bother

i dont even have a real wall

[raven bridge]

Mines are incompatible with flamethrowers, unfortunately

That said, if you put the mines closer that fire range, it can work

[orchid onyx]

try vanilla defense with rampant. finally something to burn for the flamethrowers

[elfin zenith]

Week unknown of predicting biter pets.
Week u%$^own of predicting steam locomotives

[little seal]

My prediction for next FFF:

||Factorio releasing on Android platform||

[slender glen]

predict someting else please

[pale pumice]

'something else' is my favorite JG subsentence

[hybrid prism]

i predict we learn more about space

[bold shard]

klonan's FFFs have pretty much all been mechanics so good chance it will be that again https://factorio.com/blog/author/Klonan

and also worth noting that klonan said earlier that he wrote tomorrow's

[hybrid prism]

hmm

i see

[candid pawn]

so far ear's written every new planet writeup

[solemn cosmos]

WAIT

SPELLING ERROR

[candid pawn]

what does it mean to be a technical FFF and for how long have we had this polling feature?

[bold shard]

too late, ship set sail!

it just got announced to me a moment ago, so brand new I guess

[solemn cosmos]

i cant edit it the heck!

[solemn cosmos]

same. i saw it and wanted to test it 😭

[bold shard]

def a useful feature

[solemn cosmos]

there

[raven bridge]

QoL

[candid pawn]

pls no more

qol

[solemn cosmos]

we have had a lot of qol 😭

[pale pumice]

more major QOL would be welcomed still

[raven bridge]

Of course there will be more QoL that we haven't even though about, but now can't live without

[candid pawn]

QoL FFF's are good but emotionally cripping because

• I want to hear about SA!
• I cant use them right now so they make me miss them when playing normal factorio

[bold shard]

we "only" have 2 dozen more FFFs until september, so there really can't be too many slow/small FFFs

unless they are going to hit us with some giant ones near release

[raven bridge]

We're not going to hear about Aquilo until launch, is my guess

[bold shard]

😢

or it'll be vague/open ended like the intros for the last 2 planets

[candid pawn]

this will be a bot FFF

[sweet wind]

Gonna be going over all the statistics screens. They don't really work with multiple surfaces that well right now

[ashen herald]

I think it's going to be another QoL sneak peak, and more easter eggs as usual.

[pale pumice]

What's funny is DSP already has this nailed imo

[sweet wind]

I should really play DSP, I didn't vibe with the janky feel the game had on release. I'm sure it's gotten better with time.

[raven bridge]

A new game over screen is almost a given

[raven bridge]

The biggest thing I want overhauled/touched is pollution.

[small bear]

spider trains!

[quasi crest]

wdym by that?

[slender glen]

https://tenor.com/view/tank-engine-meme-scaring-memes-gif-19928134

[quasi crest]

WTF IS THAT!?

Who made this abomination?

[raven bridge]

That's just nightmare spiderthomas. Nothing to write home about

[primal nacelle]

A pretty mild nightmare-thomas even. There are worse ones. By a lot.

[solemn cosmos]

i guess it was a polish

[elfin zenith]

Week unk^%^^ of pre@$^ting biter pe^s.
Week u%$^own of pr@#%cting steam locom&#%$ves

[topaz wing]

I notice that we mostly have about 4 qol/technical FFF's between planet/new bigger feature FFF's, so i expect something bigger next week

[primal nacelle]

next week is number 404 though, and close to april's fools. They may meme to some extent. Or pull off shenanigans and subvert expectations. Or just pull off shenanigans

[raven bridge]

Indeed. But I feel they didn't do it with 401 and 403, so I predict they won't do it with 404 either, even though it's a much more known error code

Also, the planned release is around 420, blaze it

[bold shard]

404 will be weapons, and the title will be "making enemies disappear"

[tough ginkgo]

"Fish are slippery and don't stack well on belts - so we've added a barrelling recipe. Earendel has added Delivery Cannon support for this to the next release of Space Exploration".

[ashen herald]

lightning rod on spidertron makes sense

[raven bridge]

Lightning rod equipment??? And then a tesla charger like in K2?!

Oh man I do wish for a tesla charger like in K2.... would be worth bringing to other planets

[small bear]

I don’t see any lightning rod in the image tho

[small bear]

spidertron can be just immune to lightnings

just like trains

[small bear]

perhaps they will announce things that will not added into SA

or things they removed

and they can start the introduction by mentioning RCU

[final dove]

visible equipment on the spidertron perhaps?

[ashen herald]

It's definitely not something spidertrons have currently

[small bear]

now I wonder if there will be remote driving car and tank

and add grid to tank

[raven bridge]

What are you seeing here that I don't?

[desert lake]

That does not look like a lightning rod

[desert lake]

And trains aren't immune to lightning

[small bear]

they are

[desert lake]

We know rails are immune

Nothing was said about trains

[small bear]

#friday-facts message

JG leaked it

[desert lake]

Something was said about trains

Anyway, when I first saw the spider thing it kind of reminded me of the top of a distractor bot

[slender glen]

[small bear]

it looks like the spider is facing towards right

[desert lake]

Oh you know what that makes a lot more sense

I see it now

That's just the rocket battery part, it's nothing new 😭😭😭

[small bear]

[raven bridge]

If trains are immune to lightning, what do we use interrupt-in-interrupt for? It must be something else.

[desert lake]

This is how I saw it 😭😭😭

I thought it was some kind of new model of spidertron with a hole in its head and vertical lines around it

[topaz wing]

That also means that we probably don't need the interupt in interupt on Fulgora, but on Bwuhuo for something

[raven bridge]

Wait so literally no leaks in 403? For shame

[slender glen]

or you just don't know about them

[raven bridge]

Yes. I'm so bad at finding leaked stuff, it would be nice is someone pointed little old me at the right direction.

[slender glen]

under the sofa

[ashen herald]

Whatever object that is casting a shadow on its head looks tall and thin. Part of the object casting the shadow is brown and the edge of it is clipped by the edge as if we're not supposed to see the whole object. So why you don't see a lightning rod is because you can't see the whole object. I've compared it to the current spidertron and they don't have whatever that thing is.
(Need I remind you this image is a screenshot from a video in fff 403)
Just use your imagination a little.

[tough ginkgo]

I'm sure you've found some leaks that don't seem worth mentioning...

[raven bridge]

I'm trying to get a dev to leak something......

[tough ginkgo]

Stranger things have happened

[slender glen]

Is that called phishing?

[desert lake]

It does have it

It's nothing new

The black line is the edge of the little rocket battery on the back of the spidertron

[ashen herald]

Ok

[bold shard]

you gotta use leading questions, like so:

JG, what is your favorite feature of the endgame power armor in aquilo?

[raven bridge]

Or... what's the favorite graphics you did for Space Age?

[pale pumice]

I already asked that one

I didn't want any leaks for it though, it was an actual question

#friday-facts message

we have too many damn FFF channels

[scarlet bane]

I feel like remote driving should be a tech to unlock 🔓

[pale pumice]

why?

[raven bridge]

It predict it will be part of the automated trains tech

[desert lake]

This is the 3rd time I see someone say this

I still don't get it

That's like if upgrade planner a tech

[raven bridge]

Automated trains are already a tech required to set trains to automatic. That's basically the "remote control" for trains, so remote driving makes sense at that stage.

[ashen herald]

Train tech is too many techs, they are so fast to research, might as well be one research

[raven bridge]

They are there for learnability I guess. Maybe they'll be triggered (but that would be annoying)

[ashen herald]

Sorry I was dumb about the spidertron, I really wanted it to be something new.

[pale pumice]

I agree with this decision

[small bear]

I predict fish farming on

[pale pumice]

That would make sense, actually.

It at least solves the legendary spidertron issue haha

[small bear]

or fish farming on Aquilo if spidertron is unlocked there

if we can remotely drive cars/tanks and allow bot equipments in them, then we don’t need spidertrons that early

[vagrant pumice]

It isn't as game-breaking as like cows or some alien equivalent so I think is possible

[void sapphire]

I think remote driving non-spidertron vehicles is good, but adding bots to then in my opinion diminishes the value of the spidertron quite a bit. One of the main features of the spidertron is that it is the only vehicle that can carry personal bots, and unlocking that earlier for other vehicles would mean much less of an incentive to research the spidertron as it would no longer provide that bonus.

[small bear]

I thought the main thing about spidertron is that it can walk over buildings and obstacles

yeah and bot support

[void sapphire]

I’m not saying bots absolutely shouldn’t be added to other vehicles, but before they are it’s important to consider how that will impact game balance.

[small bear]

Imagine Wube breaking the pattern by having both and unlocked on

[violet willow]

Since I think counting FFF patterns allowed me to accurately predict 11-12 weeks between each MAJOR FFF (planet) I want to try something similar

pattern of Klonan posting FFFs since 367 (each number represents the fffs it took until he appears again):
5, 2, 2, 3, 4, 1, 4, 1, 6, 5, 3

[short patio]

Since spidertrons are now unlocked on another planet instead of from Nauvis tech, it makes perfect sense to me to add a mid-game tech that can be used for remote operation of your base before you get the spidertron. I've said before that trains would be nice, because it has the severe limitation of being confined to rails, but you can customize them with additional storage and they have automatic pathfinding on your rail network. It would make remote expansion on Fulgora nice, and incentivizes learning to use trains before leaving Nauvis

[candid pawn]

I would really enjoy a train that has a roboport and logistic access

is there a mod that has this feature right now?

[pure siren]

I think yes for Robotports. Lemme check.

https://mods.factorio.com/mod/RoboportWagons

[desert lake]

Personally I think a roboport locomotive might be better

Because then the bots can be powered by burning the locomotive's fuel, and cargo wagons can be attached to it for bots to take materials out of

[candid pawn]

A train with bots and enough rails and signals could path using ghost rails to a ghost station :D

[desert lake]

And if they made its speed super slow so it's inferior to regular locomotives, it could just roll and build without stopping

[void sapphire]

Maybe it’s speed could be related to bot speed research.

[sudden quest]

I want to see other applications of the asteroid claw arms. Would be a cool way to implement automated construction from trains. Make a special claw wagon that can take items out of other wagons and can place them within a certain radius (upgradable by research). Also, let us place the claws on lake edges for automated fishing

[short patio]

They could be a way to automate collection of biter drops, if they wanted to implement those again

[sweet wind]

I would be surprised if they did. Drops form enemies have extremely opaque throughput. Not exactly fun to deal with in a game where throughput quantities matters a lot.

[sudden quest]

I could see it working out. Sure you couldn't do perfect ratios, but more pollution -> more attacks -> more loot. If you need more alien guts, build a bigger factory.

[unborn lintel]

They still haven't revealed any enemies yet, you think it's a surprise for release?

Or they delayed because they're adding the trapdoor spider I suggested

[sweet wind]

My guess is they probably don't want to show placeholder graphics for enemies, and I expect enemy graphics to be finished pretty late compared to others.
Leaving them for a suprise would sure be interesting though.

[unborn lintel]

Oh yeah art departement makes sense

[raven bridge]

People keep asking for rate calculator in vanilla, and I disagree wholeheartedly because it only works with perfect ratios
Nothing is good as actually running the simulation. If there was a vanilla way to easily copy a build to a sandbox, with an easy way to set infinity belts, run it in for a bit, and give out statistics of that. Then that is something I would approve.
The blueprint GUI needs work anyway, so this is a good place to work on.
Tips & Tricks already have live simulations, so it's not out of the picture
A surface where you can design blueprints, with creative tools that don't get pasted in the world, and live simulation and statistics.
Works well with the idea of production per surface.

So this is a hopeful speculation

[primal nacelle]

Kind of built in test lab? That would be nice

[raven bridge]

Yea like Editor Extensions, but tied with the blueprint GUI

Could be even show you how the BP tiles

[pure siren]

Don't think Wube wants us to have such tools in vanilla.

[raven bridge]

The BP screen needs a rework

From how you can't zoom into it, only allows removing entities, alignment is hard to get right....

[pure siren]

Yes, that is something we need. Plenty of QoL FFFs left, there's a chance :)

Rate calculator was a confirmed no iirc though.

Something like Blueprint Labs I don't think will come.

[raven bridge]

Yea Wube don't like misleading features like Rate Calculator

[topaz wing]

yes the blueprint test lab is something i'd really like to see, and it would fit too as the current blueprint screen is kind of clunky. You can do it right now by going to another sandbox save but of course that is not smooth for gameplay.

[sweet wind]

I've tried mods like the blueprint designer lab, none of them work well enough for me to actually want to use them over just going into another save file or placing the blueprint and editing it that way.

Better blueprint QoL would be much appreciated. Being able zoom in and add to a blueprint would be really nice.

If we could test blueprints using an in-game simulation that would be sick

[raven bridge]

Try editor extensions

[pure siren]

I don't see running simulations in the blueprint "editor".

[candid pawn]

it only works with perfect ratios
what do you mean by this? it seems like it works fine with non perfect ratios

I dont think rate calculator should be in vanilla either but I dont get what you mean. sometimes it spits out a fraction of an assembler? well then round up

Oh, I got rate calculator and stuff like helmod mixed up. Yeah rate calculator is more finickey

[bold shard]

can't believe I've never heard of this before. I always just have a "test world" without ore, enemies, trees, or cliffs to build and test various bps/setups. this is much more convenient being able to do it "in game".

[raven bridge]

I always thought Editor Extensions was just for sandbox worlds, but no. There's a button that puts you in a sandbox surface, with a ton of useful "infinity-<something>" entities.

[bold shard]

yeah those are nice. I've been using creative mod for a while which has a lot of that https://mods.factorio.com/mod/creative-mod

[raven bridge]

True, but EE was made by Raiguard

[bold shard]

he makes some solid stuff. running creative experiments in-line with your running game is what sells it for me

[chilly widget]

To be fair, the EE testing lab is much more limited than i.e. Blueprint Sandboxes. You also need to be an admin to use the editor at all.

My main usecases for EE are for mod development and game development

[raven bridge]

Limited in what ways?

[chilly widget]

• You need to enter the map editor to use it
• There aren't any accommodations for Space Exploration

[raven bridge]

Would be fun when they announce the new blueprint editor to be something like Editor Extensions 🙂

[candid pawn]

how will sandbox surfaces work when specific buildings can only be placed on specific planets?

[bold shard]

surely theres some built in way to disable/bypass that

[small bear]

sandbox mode 2.0?

[sweet wind]

My assumption would be not enforcing placement restrictions on a potential blueprint surface/simulation.

[sweet wind]

I have tried EE, and it works well in a test world, but there are oddities when using the in-save sandbox surface. I'd rather just have a dedicated save file.

An extension of the simulation system to work with blueprints would be really cool, I can imagine allowing you to place "I/O" entities like infinity chests, belts and pipes that will be in the blueprint but won't be placed/previewed outside of simulation.

[elfin zenith]

Week 1 of predicting 'romance' update
Week $##$%%%% of predicting stea$ lo$%&iti%@s

[median coral]

I'm hoping for more specialized exoskeletons, like modular armor with a larger equipment grid but you can only place energy, batteries, and a singular utility item, such as exoskeletons. (I need to be able to move faster. Default speed is so slow)