Factorio Friday Facts

[@disskuss4268]

So, three FFFs left, and Space Age is out. Crazy!

[@taunomoldri9494]

Need to finish my satisfactory phase 5 by then!

[@ComradeQuestions]

Yeah you can still run long pipelines but the flow will only work in one direction now. Before you could do that but have multiple source/sinks along the line and it would self balance.

[@Carlos-ux7gv]

I think that is an issue. Probably you will need some circuit pumps along the way: if a>b, use pump A, else use pump B.

[@ComradeQuestions]

@@Carlos-ux7gv yeah you could have tanks at the boundaries with pumps running in/out but you'd probs end up with a situation where your pumps are constantly moving fluid in a circle. I think the only realistic solution is to run two seperate pipelines like how you would with belts.

[@sghuisman]

You could even make a ‘ring’ just pumping oil around in circles but you can grab everywhere…

[@thoriandragontamer1411]

The other thing is that pumps require power. So you'd have to run power infrastructure as well as pipes.

[@soberTrezviy]

@@Carlos-ux7gv what if even without circuit? just 2 pumps. the speed of a->b is higher when a amount is higher and vise versa, so it balances two blocks. would be funny to establish oscillations and waves

[@fakecubed]

Don't feel bad about having trouble understanding some of the stuff in this FFF. I got very confused by a number of things they wrote in it, and based on comments I've read in various places, a lot of other people are confused much worse than I was at first. I think this one could've gone through a bit more editing before they posted it. And it could've done with a reminder of the changes they were already making, rather than assuming the reader has all of those memorized and knows exactly what a change means from the prior changes.

[@nnelg8139]

If 14 seconds isn't fast enough to unload a fluid train, why not just add more fluid stops? Or use a longer train? That's what you do for bulk materials, after all.

[@TimDrogin]

14 seconds for basically double of fluids(see the capacity buf) so basically this translate to seven seconds vs vanilla 0.5. With legendary pumps it will be like two seconds

[@walksanator]

Better question is what are you doing to consume a entire tank in 2 seconds

[@TimDrogin]

@@walksanator Exactly

[@MarkMichon7]

@@walksanator serious megabases. Kovarex said a while back his goal was to break 1m SPM on his SA playthrough before release. Even if he's only reached 10-20% of that, that is some serious throughput to achieve that. Bases of that size for the sake of UPS conservation will demand the highest performance setups throughout your base.

[@magnus0017]

I feel like the "ok for vanilla mega bases" was maybe in regards to 1.1 factorio mega bases, as you wouldn't have access to quality. Like, the pumps are slightly nerfed from base game, but it should be fine? Dunno. I feel like there is going to be another round or two iterating on this, but it's continuing to go in the right direction.

[@Xterminator]

Yeah I'm thinking the same thing. The production (and therefore consumption) rates in the expansion are absolutely insane compared to current vanilla 1.1

[@KaneYork]

1200/s was about what you would be getting in 1.1 anyways if you didn't use rigorously spaced pumps and just casually added one occasionally

[@SimonWoodburyForget]

"Output rate used to be unlimited, but is now inversely proportional to the fullness ratio of the sinks." -- The fuller the fluid system (sink). The slower you can pump into it (output ratej). This is just like the fuller the fluid system, the faster you can pump from it (input rate). I assume this is just a balancing mechanic, and not intended to directly mimic any kind of real world system. I'm not totally sure what it'll affect in a real game.

[@Carlos-ux7gv]

Basically liquids are now handled in 250x250 squares instead of per-pipe basis.

[@AbsoluteHuman]

It is kinda true to pumping gases tho... And a natural flow system like a river.

[@Ghorda9]

just think of it like the tank on the back of a toilet, the water comes in from the bottom and slows down as the pressure builds up until it becomes too much and stops

[@soberTrezviy]

irl you need more pressure to push liquid higher into the tank, so with same pump power it will push slower. and vise versa, liquid column pressure helps to push out the liquid. same with compressed gas

[@insentia8424]

This mimics the behaviour of fluids in real world. Fluids move from higher pressure to lower pressure, simplifying pressure into fullness ratio. Also, them using the word output is just causing confusion. Whether it is an input rate or output rate depends on perspective. If A pumps into B, then A's output rate is B's input rate. Now the first sentence talks about an output rate being inversely proportional to the fulness ratio of the sink. But it *never* specifies whose output ratio it's talking about. The sentence only talks about the sink, so the most intuitive assumption would be to think it talk's about the sink's output rate... but that's not what it's happening, it's talking about the fluid being harder to pump into the sink the fuller the sink is, aka, the source's output rate. So you basically need to know what the sentence is meant to say, or guess the right thing, to understand it. They should have either specified whose output rate they are talking about, or used input rate instead, since the sink was the only thing being talked about.

[@MikeKrasnenkov]

So basically we get something very like 1.1 fluid behaviour, but singular "fluid box" is now 250 tile stretch of pipes and tanks, while pumps do the role of what previously were actual pipe segments, and limited to 1 way only. Theoretically I think this is a great compromise and I think I might like it more than both 1.1 and previous 2.0 god mode pipes. I think putting 2 pumps between pipeline stretches will make things work in both directions? Especially with decreasing flow rate when it approaches fullness. Not sure how useful it will be though. Overall I am once again very happy with how devs came up with great solution. I did not like previous 2.0 mechanics, but accepted it as necessary compromise. But this update looks really great, and they once again did something completely unexpected. What a team!

[@TimDrogin]

It’s a nice balance change. And, to be honest, having a main pipeline with occasional(not every 100 tiles) nicely placed pumps is not bad at all

[@Desette]

For the 256 tiles versus 250 question it might be how you need to include the pump and the pipe coming out of the ground for each chunk to have all of the components in the same area. Meaning 1 pump is two tiles with the pipe coming into or out of it would be 1 tile. 3 tiles total. On each side of the chunk would take 250 to be 256. So everything fits neatly inside a chunk. I think there was something about how chunks process the information in the game within a chunk as well might be another reason.

[@Snoogen11]

13:42 if that is how it works, that's actually somewhat realistic to how pressure works in real life. The more water you put in a barrel for example, the less space there is for the loose oxygen inside, which means the oxygen gets pressed closer and closer, meaning if you add more water, that water has to PUSH all that air out of the way. It's the same with how charging batteries work. It's much easier to fill a barrel/battery from 0-80%, than it is 81-100%

[@DeronMeranda]

I still do a lot of barreling because I find the challenge so satisfying, as well as all the animated action going on. Besides trains full of barrels, fluid wagons are fun too. I'll even run a fluid train for short distances, just because I can. Pipes are my least favorite because once placed they just sit there and don't have any "interaction" other than being a trip hazard.

[@kryzethx]

It's really weird that they defined it by area, you would think it be simpler to just define it as length, based on the farthest source to the farthest endpoint. Maybe even make different tiers of pipes to go further (since I think quality pipes with varying lengths would be annoying to deal with)

[@DeannaEarley]

Re the block size. If it stopped at chunk boundary, there would be the chance of a two pipe long segment that crosses that boundary, requiring a pump.

[@aidanallen3187]

I like the pump changes (though agree 256 tiles is better), and in practice it resembles Satisfactory's vertical pressure requirements. I think the main reason this change was made is to make fluid trains more viable for long distance, as long pipelines (especially if more than one pipe is being run in parallel, which may be needed given the nerf to pump throughput) now have an energy cost associated with them (n*30kW per 250 tiles). Realistically I only see this as a concern on Vulcanus for molten metal/sulfuric acid transport or on Nauvis for oil

[@tobydods4515]

I think "the special exception for storage tank flow has been removed" at 13:48 refers to the oddity in 1.x that if you had Fluid Wagon -> Pump -> Storage Tank it would empty in less than a second but if you had Fluid Wagon -> Pump -> Pipe -> Storage Tank it would take ten times longer to empty the wagon. If so, I am happy this oddity has been removed in 2,0 as it never made any sense to me.

[@AbsoluteHuman]

No, it wasn't an "exception" it was a consequence of how fluids worked. The exception they are talking about is something that was in place in 2.0 to mimic these rules I suppose.

[@Ghorda9]

the reason why the later example was much slower is because the pipe segment bottlenecked the flow with it's lower throughput.

[@peterschmidt1900]

My main question is: did they fix the heatpipes being so UPS inefficient?

[@mennovanlavieren3885]

Yes, nuclear mega bases!

[@DVDmatt]

13:34, my understanding of this is that if a 'sink' (i.e. destination) is 0% full it gets100% flow rate, if it's 50% full it's 50%, and if it's 90% full it's 10%. This means it's quick to fill a container part way, but quite slow to fill up the last 10%

[@Carlos-ux7gv]

Just like cell phone batteries then.

[@timewarper9575]

WUBE is cookin

[@adavidhenry]

I don’t think it’s based on where the map is just the max extent of your fluid system

[@nnelg8139]

I think his point was that if you have large chunk-aligned blueprints (e.g. 256-wide city blocks), you won't be able to simply put one pump at the edge to transfer fluids to the next block.

[@SimonWoodburyForget]

@@nnelg8139 I don't see any real problem with placing two of them or aligning them to the 256-32 tiles. The size of a big pole was only a problem because it was lower than the minimum size of a chunk, so you needed 2 to 4 per chunk to cover all directions and now you only need 1.

[@lascivio]

Radar is 224 range so just build smaller city blocks. but i do agree either 256 or 224 would be best

[@Big-E-TheBouncer]

1.1 fluid has been a headache of mine wince I've started playing factorio. Hoping their changes solve all fluid issues.

[@HoberMallow-tg3ph]

Trying to imagine how the whole pipe segment works now, is the whole 250x250 pipe segment limited to filling at a rate of 6000 per second or can i fill it up from a lot of different positions and break that limit? Also are the fluid users also able to collectively drain the pipe segment faster than 6000 per second? I guess I'm just curious because the article didn't really cover edge cases like this.

[@JudgeNicodemus]

Dunno if I missed something but I wonder what the Fulgora enemies will be.

[@Ghorda9]

i don't think there will be any, we already have the lightning storms

[@pedroluz7724]

The tanks in real life actually take longer to fill the more full they are because many times the feed to the tank is from the bottom and the more fluid you have the more pressure you have against you Of course if they feed the tank from the top this doesnt apply but many times they feed from the bottom

[@HaibaneKuu]

I mean in real life the whole fluid "network" isn't a single container either. A tank is a tank that's filled and emptied, and pipes are usually completely full and pressurized. At least from my understanding.

[@pedroluz7724]

@@HaibaneKuu yes, i think this inverse filling speed is only aplied to tanks no? From what i got from the FF

[@HaibaneKuu]

@@pedroluz7724 The basis of the whole 2.0 fluid network is that the whole network is a single fluid entity that's filled and emptied as one. Pumps break it of course.

[@pedroluz7724]

@@HaibaneKuu then i probably misunderstood it That is a bit more annoying because that means we will never have the pipes full when were feeding the assemblers

[@Ghorda9]

@@HaibaneKuu in real life toilet tanks work like this

[@lolmao500]

A mod I like is the mod where it puts a red-yellow-green light on factories to show how the factory is going... its very useful to see quickly where problems are. Anything on that?

[@Xterminator]

Highly unlikely that will be implemented due to performance concerns

[@bearmike_360KB]

New players are using long pipelines to get oil in 1.1

[@namshimaru]

Pumps can be legendary too though

[@t1nytim]

I don't use trains. So will pump fluids a lot. So I'm happy things are being simplified. Realism is nice, but not necessary even remotely for me, so definitely not wishing they would head towards more realistic.

[@PianoMeister3]

11:01 I wonder if they're going to nerf offshore pumps down to keep the ratios the same. 200 boilers from one offshore pump seems a little broken.

[@GeneralBolas]

Probably not. The number of offshore pumps you need for things other than boilers/heat exchanges has not changed. So nerfing them just because they changed the heat capacity of steam could cause real problems.

[@AbsoluteHuman]

​@@GeneralBolas now you can finally use the ratio of 34 (I think?) boilers to one coal belt!

[@Carlos-ux7gv]

Will liquid tungsten be used against enemies on the flamethrower turret?

[@mennovanlavieren3885]

Liquid tungsten? WCl6 maybe.

[@Ghorda9]

it's not a fuel so no

[@Carlos-ux7gv]

@@Ghorda9 you didn't understand my question :-)

[@datman6266]

I like this

[@virtual_GaRy]

The size limitations which need pumps feel so artificial. The game will even tell you where to put the pump. I think they should come up with some better system, possibly limiting throughput again in some less funky way as in 1.0

[@TlalocTemporal]

Reducing the flow cap of fluid operations that extend farther than 250 tiles might work. It's less arbitrary and still encourages fluid operation within 250 tiles if that was for performance reasons.

[@virtual_GaRy]

@@TlalocTemporal is just very difficult to define what's a source and a drain with machines coming on and off and complex looping trees of pipes. But I'm sure something like it can be done somehow

[@AbsoluteHuman]

​@@virtual_GaRy how the new system defines the point of placing a pump then? Where's it? Does it break your system as 50-250 or 250-50 or 150-150…

[@Ghorda9]

@@virtual_GaRy you're overthinking it, just because a machine isn't operating right now doesn't mean it magically switches from being a drain to being a source.

[@AdamL_18]

when i saw this FFF, my first tough was, Oh no here we go, they messed it up. performance wise, the first FFF itteration was way better.

[@derekcook8201]

I'm confused as hell

[@SekGuy]

The only true option is to use barrels!

[@Xterminator]

@@SekGuy if only they hadn't been nerfed to hell years ago...

[@mqb3gofjzkko7nzx38]

Legendary Barrels?

[@SekGuy]

@@Xterminator it was clearly a joke! haha

[@lucidmoses]

They are not doing exactly what I outlined in my design doc I sent. Seems close but if close enough you can accomplish the goal a better way with zero UPS overhead. Simply limit the individual pipe length to 250 (or what ever number). That can be entirely calculated at placement time and would be so fast humans wouldn't be able to measure it. This would even make more sense as a long running line to get water may need pumps while short run that go over bounders would not.

[@JeamesBooth-n3p]

John Dale

[@samserious1575]

The pump nerf is fine with on pipes, but in terms of loading and unloading trains, I feel this is an indirect nerf on fluid trains. 2 pumps before is enough to completely fill/unfill a fluid wagon in slightly more than a second. A nerf like this will make such operation occure a bit more than 10 seconds. Me personally I prefer trains to be quickly in and out of their stations on full load and unloads quickly.

[@GeneralBolas]

I can understand this thinking, but the loading/unloading speed is quite comparable to inserters. For a 200 stack item, it takes (now) bulk inserters about 22 seconds to load or unload a wagon. The new stack inserter will be 16.5. Base quality pumps working optimally will be 14 seconds. So not particularly bad; certainly lower than the near instantaneousness of 1.1, but I don't think it's really going to affect much. What *will* be affected is what an optimal loading/unloading setup looks like. To keep your loading pumps working at max speed, you need to keep the fluid segment full. To do that optimally, you need to break things into 2 pieces: the part that generates and buffers the fluid, and the pipes directly connected to the pumps that load the wagons. Between the two need to be enough pumps to keep that small fluid segment full at all times, which will likely be 2x the number of pumps that used to load the train. Something similar needs to be done on the unloading end; the drain pumps need to pump fluid out so fast that the unloading pumps always have somewhere to put the fluid.

[@Ghorda9]

the pipe throughput for long distances are handicapped by the mandatory pumps, this change was very necessary for balancing. Pumps can also only go in one direction

[@Omnifarious0]

I find this all far more confusing than fluid flow is currently.

[@KeinNiemand]

I hate this I was hoping I could do some intresting stuff with pipes which is now impossible. Also I hate the 6000 fluids/s machine limit that puts a hardcap on how much speed any recepie involving fluids can have.

[@GeneralBolas]

You may want to consider how much fluid 6000 fluid per machine input *really* is. Also, it's not per machine; it's per machine *input* (or output). A chemical plant set to make plastic has 2 fluid inputs for petroleum, so it could in theory use 12000 fluid per second. But, let's think about how much fluid 6000 per second really is. Plastic is a 1-second recipe consuming 20 petrol. In order for a chemical plant making plastic to cross that threshold, it would need an effective crafting speed of 300. Given a chemical plant at "Legendary" quality (crafting speed 2.5), that means you need to get a 120x multiplier. So that would be an accumulated speed bonus of +12,000% speed. I know that high-quality modules and beacons are strong, but I don't see them giving a 3x3 machine that level of speedup. And even if you could, that chemical plant would have to consume 300 coal per second *and* spit out 600 plastic per second. Even with the new stack inserters at high quality and fully-stacked green belts, that's just not likely for a 3x3 machine. Advanced oil processing is more likely to run into this problem, but only because it's a 5x5 machine and thus can have more beacons. But even then, beacon scaling makes achieving +12,000% speed unlikely. I could really only see this limitation affecting modded play. Like Krastorio 2's "Advanced" chemical plants that have a crafting speed of 8 along with regular beacons and singularity beacons. Throwing quality into that could hit that limitation.

[@Ghorda9]

you can still use multiple pumps per area

[@miendust]

This doesn't make remotely sense.... the previous version was better...

[@billweasley1382]

Sounds like they had a fix that worked GREAT, and now screwed it up again because a tiny minority of crybabies didn't think that the fix was "realistic" enough.

[@AbsoluteHuman]

No, it's clearly a their own own design preference... They wanted the fluids to be at least SOME kind of challenge, not just another version of electricity. The question is if it's the right type.

[@Ghorda9]

the previous version was completely broken and made fluid wagons pointless

[@orlikforcebalancer7495]

first