Factorio Forums

www.factorio.com
https://test.forums.factorio.com/

Nuclear Power Plant from the future

https://test.forums.factorio.com/viewtopic.php?t=113966

Page 3 of 6

Re: [FFF416]Nuclear Power Plant from the future (RANT)

Posted: Mon Jun 24, 2024 1:09 am
by mmmPI
FuryoftheStars wrote: Sun Jun 23, 2024 11:21 pm This does not compare. Direct insertion is from a single or very small number of machines to another single to very small number of machines, and not flexible/able to change/what have you.
That's why i called this a strawman, instead of actually arguing with argument that one state as its own one put some bad imaginary argument in the mouth of a strawman and one argue against the strawman that the made up argument are bad. The strawman is the only one who said it compare.
FuryoftheStars wrote: Sun Jun 23, 2024 11:21 pm See my last paragraph just above for part of the reason why I feel that way. The other obviously being the instant transmission, though compared to a full pipe under any system, that is moot. Though again, I could be wrong, but what they've said so far doesn't leave me with much hope.
I understand you say it's similar to electric network, and have read you previous comment, but it didn't convinced me, it is your opinion, not hard cold facts like those we'd get next week, i can see there are also players that feel this way as a positive news, you i understand as a negative news, to me it is premature in both case to say it will be like the electric network, it's not how it look like in the videos from the FFF to me but i may be wrong too. I designed the power plant as if pump were still meaning something for throughput, which may be incorrect understanding, i thought this post would help me clarify the FFF annoucement but it's not going this way atm ^^

Re: [FFF416]Nuclear Power Plant from the future (RANT)

Posted: Mon Jun 24, 2024 1:51 am
by FuryoftheStars
mmmPI wrote: Mon Jun 24, 2024 1:09 am I designed the power plant as if pump were still meaning something for throughput, which may be incorrect understanding, i thought this post would help me clarify the FFF annoucement but it's not going this way atm ^^
Well, to my understanding, pumps connected directly to tanks are supposed to be able to transfer faster, as well as break apart the segments, and of course have a max rate themselves that they'll transfer fluids between segments. So based on my understanding, yeah, you're doing it right. Don't know if it's optimal, but I believe it would work the way you are thinking with it....

Re: [FFF416]Nuclear Power Plant from the future (RANT)

Posted: Mon Jun 24, 2024 2:14 am
by KillingTimeItself
i'm sort of indifferent to the overall change. I really like that i no longer have to spend as much time thinking about moving water into the goddamned nuclear plant, because that was always a nightmare. (though ironically bringing in water via train is concerningly viable.

I also do like how the fluid flows through the pipes as would sort of be expected. But i also hate the very real UPS cost in that calculation chain happening. I wonder if it would be possible to combine both in some sort of manner that would allow reasonably "realistic" feeling pipes, while bumping down the UPS cost, and allowing for more than a pitiful maybe 1000 units of fluid p/s which is rather restrictive.

I'm not a huge fan of landfill printing nuclear plants as i find it removes some of the nature of it, as well as making it more tedious, though not significantly. I think the base game would benefit from having a well pump research of some sort, you could easily mod it in, but it makes sense in terms of progression, and is one of the little QOL things that makes the end game much nicer. I also imagine this change is going to be very advantageous for flamethrower based defenses, with massive loops of pipes running oil everywhere. That alone can be quite the UPS magnet just based on the volume of them. But then again i also like watching the flamethrowers light up in succession as they each get their own oil supplied to them.

IDK, i suspect this will probably be rolled back in the 2.0 update, or implemented slightly differently in order to preserve some of the expected behavior of pipes.

Although, if wube implemented steam expansion properly (WUBE PLS FIX IM BEGGING YOU) none of this would be a significant issue, because in the current game water is consumed 1:1 with steam, which i find rather silly and obtuse considering how steam power works to begin with. I think the commonly reported figure of expansion is 1:2000 going from water to steam, in a traditional steam engine design, couldn't tell you whether that's superheated or not, but compared to 1:1 literally any amount of expansion would be significantly improved.

Re: [FFF416]Nuclear Power Plant from the future (RANT)

Posted: Mon Jun 24, 2024 2:22 am
by mmmPI
KillingTimeItself wrote: Mon Jun 24, 2024 2:14 am i'm sort of indifferent to the overall change.
Although, if wube implemented steam expansion properly (WUBE PLS FIX IM BEGGING YOU)
I'd recommand you share your feedback about the FFF on the thread for the FFF feedback, for it to have more chance to be read by those you address it rather than on this thread which is just a personnal creation about a personnal nuclear power plant that you do not adress at all ;)

Re: [FFF416]Nuclear Power Plant from the future (RANT)

Posted: Mon Jun 24, 2024 2:26 am
by mmmPI
FuryoftheStars wrote: Mon Jun 24, 2024 1:51 am Well, to my understanding, pumps connected directly to tanks are supposed to be able to transfer faster, as well as break apart the segments, and of course have a max rate themselves that they'll transfer fluids between segments. So based on my understanding, yeah, you're doing it right. Don't know if it's optimal, but I believe it would work the way you are thinking with it....
I read in the FFF that the new system was compared to the electrical network, or rather than "the nuclear-electrical network option" was choosen. But i think this is conceptually describing the algo choosen compared to other algo rather than stating "made to be like the electrical network already existing". It act closer to electrical network than other system in the game, but just the existence of pumps having an impact on throughput makes it different. (maybe ? )

I also noticed on the video of the chemical plants, there are 2 pumps as output for the petroleum gas, which is what hinted me to also use several pump in parallel, but in the FFF there are 2 pumps and then 2 paralel pipes for output, which i didnt do, my setup look more silly, 1pipe =>2 pump in paralel => 1 pipe. Maybe i missed something there. ( that's kind of my hope )

Re: [FFF416]Nuclear Power Plant from the future (RANT)

Posted: Mon Jun 24, 2024 3:23 am
by FuryoftheStars
mmmPI wrote: Mon Jun 24, 2024 2:26 am
FuryoftheStars wrote: Mon Jun 24, 2024 1:51 am Well, to my understanding, pumps connected directly to tanks are supposed to be able to transfer faster, as well as break apart the segments, and of course have a max rate themselves that they'll transfer fluids between segments. So based on my understanding, yeah, you're doing it right. Don't know if it's optimal, but I believe it would work the way you are thinking with it....
I read in the FFF that the new system was compared to the electrical network, or rather than "the nuclear-electrical network option" was choosen. But i think this is conceptually describing the algo choosen compared to other algo rather than stating "made to be like the electrical network already existing". It act closer to electrical network than other system in the game, but just the existence of pumps having an impact on throughput makes it different. (maybe ? )

I also noticed on the video of the chemical plants, there are 2 pumps as output for the petroleum gas, which is what hinted me to also use several pump in parallel, but in the FFF there are 2 pumps and then 2 paralel pipes for output, which i didnt do, my setup look more silly, 1pipe =>2 pump in paralel => 1 pipe. Maybe i missed something there. ( that's kind of my hope )
Hmm, honestly, it looked to me like they were just showing "this setup doesn't work under 1.1, but the identical one does under 2.0". I'm not sure there is much of a mechanical reason behind it, unless the fully beaconed chemical plants are capable of producing more fluid than 1 pump's transfer rate can move. But it may also be good to have pumps between your producers and your consumers, at least like with your nuclear plant. If the water level of those tanks starts to get low enough for any reason, without the pumps to break up the segments, it would start to cause input starvation on the heat exchangers (and actually, I wonder if that design would have been better with the pumps connected directly to the tanks?).

----------------------------------------

There was another reply here from someone else that seems to have disappeared. No moderator note, however I won't name names or include the direct quote, but I do want to note two things:
  • What is trivial to one, or what one derives fun from, is not necessarily the same to another.
  • Using a mod, or rather a set of mods, that were each meant to push a game's mechanics to their breaking point and then were used in combo with each other to push it even further is not justification for calling a system in the game annoying. This was the intention of the mods (and those who started using them together) for those that wanted that extra level of challenge. If one finds that system, when pushed to that extreme, annoying, then the fault is not with the base game or its mechanics.
That said, I do not disagree that the old system has faults. I've said it plenty of times and have wanted change. But I have also said every time that this (2.0 change) was not what I wanted and would have much rather stayed with the way it was if there was no other way.

Re: [FFF416]Nuclear Power Plant from the future (RANT)

Posted: Mon Jun 24, 2024 4:25 am
by functional
FuryoftheStars wrote: Mon Jun 24, 2024 3:23 am [*]Using a mod, or rather a set of mods, that were each meant to push a game's mechanics to their breaking point and then were used in combo with each other to push it even further is not justification for calling a system in the game annoying. This was the intention of the mods (and those who started using them together) for those that wanted that extra level of challenge. If one finds that system, when pushed to that extreme, annoying, then the fault is not with the base game or its mechanics.
[/list]

That said, I do not disagree that the old system has faults. I've said it plenty of times and have wanted change. But I have also said every time that this (2.0 change) was not what I wanted and would have much rather stayed with the way it was if there was no other way.
I do not know why message disappeared and mods did not DM me either. Don't know whats up with that, I guess the tone was wrong. But I digress;

OP has been insisting that this is some kind of matter where Wube is appeasing noobs by simplifying liquids completely. And part of OP's argument has constantly been the implication that nobody who understands how fluids work would desire for a change like this. OP is also not alone, with this comment in particular:
Meanwhile, the pros will quit the game and look elsewhere.


(by another person; I thought it was you but it wasn't, for that I apologize)

So maybe you can see why I bring up B&A; because I have done something that requires you to understand fluids completely*. You're going to quickly fail if you'd try to set up B&A base in scale. Be it artificial water, be it flotation, be it dealing with tons of oxygen and hydrogen (if you don't just simply burn it all off), be it bunch of other stuff: you deal with a lot of it. I'm no stranger to how fluids work and I've been able to manage them.

And besides me, Discord & Reddit are also full of experienced people who are happy for the changes (mostly thinking that it's least worst alternative, nobody is that excited in and of itself, more so the fact that it replaces a system they consider to be worse). It is simply not true that this is a change that only people who dont know how to play the game would want. And besides that, Wube has made previously changes that I also objected in order to help beginners; such was the case with oil refinery change which is the only time before couple weeks that I participated here. I disliked it, but now I've nearly forgotten about it and just accept it as it were.

*Arguably you can finish it even if you dont understand it. But you can't scale up a lot, else it starts to become a problem.

That said, I do not disagree that the old system has faults.
I think there's a consensus that the new system is not ideal at all. Nobody really thinks its better. It's just currently the least worst known. So if we start from the premise that we need a new system becauase of SA introducing new problems that old system cannot really facilitate that well, then we really need something new.

Currently the only alternative that I know of is the "increase volume" one and that obviously is a bandaid solution. As far as I know, there's no feasible alternative that anyone has ever given that isn't increase volume one. Fluids Must Flow I guess could be a third alterantive, but it also doesn't solve issues related to input starvation (and there might come mechanics that employ input starvation) + it probably has severe UPS overhead, although that is just a guess.

So yes, I will conclude that this isn't exactly a system that I want. But it is something I want over the old system, and I can't also come up with a better system either. It solves a lot of issues that can arise of old system and doesn't bring any new ones. The FFF also states about new potential, but I'm not ultimately sure what they mean.

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Mon Jun 24, 2024 8:47 am
by mmmPI
Please do talk about your past failure of using fluid with mods in a post in which it won't be off topic
this topic is about the consequences of the changes on the vanilla fluid system taking the example of designing a nuclear power plant, not on why you like the change based on different train setup or whaetever else was tried in the past and the degree of satisfaction, please do that in the FFF it will be welcome there whereas here it is off-topic and irrelevant, (in case it wasn't clear already)

Re: Nuclear Power Plant from the future (FFF 416 RANT)

Posted: Mon Jun 24, 2024 10:56 am
by MeduSalem
mmmPI wrote: Sun Jun 23, 2024 5:54 pmI have to compete against myself to not argue about everything to death x)
Same here, but Anakin would be glad to hear that it is a power you can learn. ^^
mmmPI wrote: Sun Jun 23, 2024 5:54 pmIn my now moderated story i added a little drama piece where i took as example someone asking for a way in the game to make 3 divisable into 2 equal integers.
Well you can do integer division using remainder. For some people it may just not be a "pretty" solution, but it is a legitimate solution. :>
At least nothing wrong about it.

mmmPI wrote: Sun Jun 23, 2024 5:54 pmI can follow this reasonning, like if the nuclear power plant has 432 turbines or only 120 they could have their fluid increasing at different rate when suddenly connecting to the steam tanks based on pressure. But i think , and from reading the other thread, that there won't be a limit on the number of machine per segment as that could be arbitrary and not under player control where a segment decide to form. ( // huge mess) Maybe i'm not getting it.
For the idea I suggested yesterday...

I am going to use an example with Heat Exchangers, which can output 103 units of steam per second.
And lets say the upper limit pipe segments can handle is let's say 6000 units of fluid per second (up to balancing of course)

Then you could connect for example 58 heat exchangers (= 58*103 = 5974 steam/sec < 6000) to it.
But if you do 59 heat exchangers (= 59*103 = 6077 steam/sec > 6000) then a random spot of the pipe segment just breaks and fluid/steam starts leaking out and possibly brings the fluid transfer to a halt.

If you don't want a pipe break mechanic in the game you could instead start scaling down the active output of the heat exchangers proportionally to what the maximum pipe-segments can handle. (It would act like a throttle system on the machines on the producing side)
So if you hook up 59 heat exchangers... it would be 6000 / 59 = 101.7 Steam/sec per heat exchanger.
That would be lower than the 103 steam/sec each Heat Exchanger could possibly do at maximum.
Then no pipes break, but you are not using each heat exchanger as optimally either.

Obviously with different producing machines connected to the pipe-segment the proportion each of the machines can insert into the pipe-segment would be a little bit more complex, it would be a weighted proportion (similar to the weighted proportion they use on the consumer side, just for the producer side).


So hypothetically, you could connect any arbitrary number of machines.
But if there are too many producing machines connected it might break the pipes. (That is also UPS friendly)
And if no pipe-breaking mechanic would be used there they would only output at a proportionally lower output rate. (Potentially less UPS friendly since you might need to calculate the proportion for each entity that wants to input into the pipe-segment pretty much every time)
And as player you could easily control that by simply not connecting too many producing machines to a pipe-segment OR using circuit network to shut off producing machines as necessary to stay within the limit.


That said, it would only put an upper limit on the Fluid PRODUCING side. You would just not be able insert more fluid/second to the segment than the upper limit allows.

That would NOT limit how much you can draw per second from the total fluid that is already stored inside the pipe-segment.
There you can still draw as much per second as you wanted to. So you can even hook up 200 steam turbines even if 100 would be optimal for the 6000 steam/sec.

Obviously for pulsed systems (like nuclear plants with a steam storage buffer that have varied steam consumption depending on power demand)... you can possibly see higher burst consumption than the example 6000 steam/second, since as said, there is no limit on how much you can draw from a segment.
But you would NOT be able to permanently maintain >6000 steam/second because eventually the tanks/pipes will drain eventually and then all the heat exchangers will kick in and output a maximum of 6000 steam/sec into the pipe-segment and cannot do more than that because the pipe-segment just says "nope".

As said, 100 steam turbines would be optimal for the given number, but if you need to run more steam turbines permanently than that you would simply have to make a new separate pipe-segment with additional heat exchangers for potentially another 6000 steam/sec and another 100 steam turbines.


That said, I only used 6000 units of fluid/second as maximum throughput for a pipe-segment as an arbitrary number. It could be balanced to any other number. Whatever the devs would feel is right. And then the number of machines would scale accordingly. I am sure with some proper numbers you can get preeeetty close to the number of machines you can possibly maintain in the current fluid system.


In either case it would not be much different than the belt system works. You also cannot fill more into a belt-segment than what it allows for items/second. But there is no limit on how many items you can draw from the belt. And if you have buffer storage somewhere in between, then it will eventually run dry.



You could... but that is something I am personally not too fond of... also make the upper limit of throughput per second be dependent on the number of pipe elements that make up that pipe-segment. The longer the pipe-segment is, the more individual pipe elements it contains (tanks weighing worse than just normal pipe), the lower the maximum throughput could be. It would be like a "how compact/inert is the pipe-segment"-factor depending on the size of the pipe-segment. That would be pretty much similar to how it works currently. And it would eventually require you to place annoying pumps every other few meters again to split it up into more smaller pipe-segments so each can maintain a higher throughput.

Then there would still be no real flow calculation: The fluid would still transfer from A to B instantaneously within a pipe-segment, the fill level would still be equally proportional at each point of the pipe-segment, and branches/mergers inside the pipe-segment are a non-issue... as said... no real simulation.
BUT you would have at least almost all the other characteristic throughput limits of the current system for gameplay/challenge reasons.

mmmPI wrote: Sun Jun 23, 2024 5:54 pmI do not often "design around limitations", that would be the case initially, to test those, but mostly then i design for a purpose which is different in many games
Well, I also don't always design around limitations. There definitely is a difference between "what is the maximum possible" and "what do I actually need practically". Depending on the game and sub-feature those two things might not always be the same.

But if you do a mega base in Factorio then very often you get to the point of "what is the maximum possible" because there is little reason not to push it to the limits if you actually need it.

Re: Nuclear Power Plant from the future (FFF 416 RANT)

Posted: Mon Jun 24, 2024 5:16 pm
by mmmPI
MeduSalem wrote: Mon Jun 24, 2024 10:56 am Well you can do integer division using remainder. For some people it may just not be a "pretty" solution, but it is a legitimate solution. :>
At least nothing wrong about it.
No you can't divide "3" into 2 different integer that are the same size. 3/2 is 1.5 and it's not an integer, there is no way to add 2 integer to make 3. That's what i meant. I will maybe make the syntax clearer in the drama piece ^^.
MeduSalem wrote: Mon Jun 24, 2024 10:56 am That said, it would only put an upper limit on the Fluid PRODUCING side.
That would NOT limit how much you can draw per second from the total fluid that is already stored inside the pipe-segment.
From the FFF it doesn't seem like this is the case for the system that is being implemented to replace the current one :
Each segment inherits its volume from the fluid boxes that comprise it and can hold one fluid.
Machines can push fluid into a segment at an unlimited rate, and can pull from a segment at a rate proportional to how full the segment is. In other words, if a segment is half full, then the pulling rate is half of the maximum.
I think you meant to describe your idea with more details, but i am still not convinced "segment" of hardcoded capped lengh would be optimal ( unless they have a size of 1 :D ). As it would be a new thing for player to monitor, that is even less "realistic" (imo) than the previous system where some arbitrary boundaries occuring even in straight pipes would be the norm and players would have to juggle with that.

MeduSalem wrote: Mon Jun 24, 2024 10:56 am You could... but that is something I am personally not too fond of... also make the upper limit of throughput per second be dependent on the number of pipe elements that make up that pipe-segment. The longer the pipe-segment is, the more individual pipe elements it contains (tanks weighing worse than just normal pipe), the lower the maximum throughput could be. It would be like a "how compact/inert is the pipe-segment"-factor depending on the size of the pipe-segment. That would be pretty much similar to how it works currently. And it would eventually require you to place annoying pumps every other few meters again to split it up into more smaller pipe-segments so each can maintain a higher throughput.
Then there would still be no real flow calculation: The fluid would still transfer from A to B instantaneously within a pipe-segment, the fill level would still be equally proportional at each point of the pipe-segment, and branches/mergers inside the pipe-segment are a non-issue... as said... no real simulation.
BUT you would have at least almost all the other characteristic throughput limits of the current system for gameplay/challenge reasons.
That sound more like the system described in the FFF, i'm afraid for you / happy for me :D, ( which is why i made this post to compare how the different understanding of the FFF translate in game and roughly clarifiy with the newer informations that we are given ).


but the sentence is :
There is no longer a realistic fluid "flow" through pipes; fluid pushed to a segment will be immediately available at any point along a segment. This is the "nuclear electric-network type solution" that was discussed in previous FFFs. The result is that pipes "just work" and you almost never have to worry about throughput.
To me i read that as there is no more "flow calculations" , yet there is still a little something that cap throughput and need be tweaked, but you ALMOST never will be bottleneck, (unless one system really is bad and in such case one were already complaining and will continue do so).

MeduSalem wrote: Mon Jun 24, 2024 10:56 am Well, I also don't always design around limitations. There definitely is a difference between "what is the maximum possible" and "what do I actually need practically". Depending on the game and sub-feature those two things might not always be the same.
But if you do a mega base in Factorio then very often you get to the point of "what is the maximum possible" because there is little reason not to push it to the limits if you actually need it.
I meant that i design trying to achieve something in particular, not megabasig necessarily, if i want my base to be symetric and without robots, or without central electricity facility, or in a map that as no water, or in ribbon worlds. I do need to know the "limitations of the system" so that i know what i'm allowed to do which gives me margin for creativity but the process of designing is not necessarily always trying to reach that limit ( as it would be the case in megabasing).

That reminded me of something i should have written somewhere earlier :

DISCLAIMER

The current power plant design may not also work because of too many heat exchanger in a row for heat transfer to reach the lasts ones. That would have been kind of a limit to have in mind when doing the BP. I didn't really care about it, i'm pretty sure in vanilla current condition you can't make a tileable power plant with the heat exhanger only one side of the nuclear power plant due to that, at least i don't know of any and my attempts didn't work, so i'm really surprised i'm not getting more comment on that, i guess everyone just want to give its opinion on the FFF and not on the power plant :twisted: :twisted:

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Mon Jun 24, 2024 5:54 pm
by Koub
Some time ago, I designed a tileable 6nx2 nuclear plant and from what I remember, it works reliably at its maximum theoretical output, so fluids were not an obstacle to building it. And I didn't even need fluid pumps.
So 2.0 fluids will not trivialise nuclear power plants, which are already pretty straightforward.

Seed :


Tile :


Feel free to test it (couldn't include the landfill, because it would get my post above the char limit).

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Mon Jun 24, 2024 6:30 pm
by functional
Arguably even OP's nuclear power plant is too complex. Nuclear power is so extremely efficient that once you learn kovarex enrichment, there's no real point to add that whole buffer because having the nuclear reactor 24/7 on barely consumes a resource that is practically infinite anyway. Only in very limited circumstances does it make much sense to not run them 24/7, as each reactor consumes 1,8 u235 per hour. OP's setup consumes 28,8 per hour.. so even a very tiny deposit would last for a long, looong time, while being burned 24/7.

Funnily enough, the new system actually incentivizes it slightly more from my perspective since you don't have to do millions of pumps. The cost to benefit ratio iis more on par; low effort for low gains.

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Mon Jun 24, 2024 6:41 pm
by mmmPI
Koub wrote: Mon Jun 24, 2024 5:54 pm Some time ago, I designed a tileable 6nx2 nuclear plant and from what I remember, it works reliably at its maximum theoretical output, so fluids were not an obstacle to building it. And I didn't even need fluid pumps.
So 2.0 fluids will not trivialise nuclear power plants, which are already pretty straightforward.
Feel free to test it (couldn't include the landfill, because it would get my post above the char limit).
The underlined part i agree with, it is really part of my point, when i say it wasn't broken before, despite what i called "unfounded critics" which would be "it was required to put pumps everywhere", or "fluid behave in ways that are impossible to use at large scale in vanilla".
I think It is a nice power plant you posted there and it demonstrate well how to use the current fluid system while having the limitation in mind and make nice things x).


From what i understand of the math you used : There are 12 heat exchangers per offshore pump, because offshore pump produce 1200 water per second, and heat exchanger consume 103 per second, they will "underperform" a little, but 1200 fluid can easily propagate each second in the array, this part will be unchanged in the future system i suppose because it is already "good".

In my design it doesn't matter at, there are "pumps" used to write a text, all connected by a single pipe, feeding into the nuclear power plant, i'm not saying it is trivialized, you still need to count the number of exchanger and pump in ration and exchanger and core in ratio. But you can see how my design wouldn't require finding a lake or a proper place to build the power plant whereas yours, i had to,( really i had to use the editor to put water in precise spots it was less fun than for my design where i could goof around with the pipes ), that was also an argument from aka13 which is valid in my opinion, that the new system would simplify that. Although maybe not to the extreme shown in the early pictures, it is still one of the stated goal of the change.

You have used many simple small network of every fluid, there is no 3 ways junctions in your design, although there are S-bends, the overall small size of each fluid segment makes it not a problem for throughput to have extra added pipes that cover less distance than the superior undergroud pipe. So you built a system that works. There are many pipes in parallel for each section of 20 turbines, which consume 1200 fluid per second, not as much as the heat exchanger could produce at full speed, because that would be 12 * 103, but exactly what the pump produce.

This means in full power draw, this is stable, and if you consume less you will have buffer that can handle peak.

The maximum throughput expected of any sort of fluid thought any 1 pipe is 1200/second in your design Koub.

In my design it is all the water for all the power plant no matter how many tiles there are that would go through a single pipe. ( around 25 000/s )

I think it highlight well the differences between "old" and "what i think may be the new based on incomplete information about throughput numbers".

Not everything from the change is represented though, because in Koub's design there are no junctions that would not work properly, as there are no (real ) junctions at all x). There are junctions in the fluid mechanic sense of the term because the heat exchanger output their steam not in line so they are technically a junction for when it comes to try and predict fluid flow, but given the fluid system in which they output is properly dimensionned in terms of lengh and consumer/producer, it is not a problem.

Those are the kind of things that i find allow to have a rationnal discussions because it allows me to compare "things that were done properly before" and "things done properly after the change". To highlight were the differences lies, and maybe shows the area of uncertainty concerning the understanding of the FFF.

Unlike example of things "not done properly before the change" because there is an infinite amount of those and it wouldn't me meaningfull to take every single mistake that can be made as a source of speculation (imo and for this thread) and discuss about what the person who made the mistake didnt understand from the previous system as this reasonnings/learning could have been done earlier than the annoucement of the FFF and i find is becoming useless after the FFF announce the fluid rework to start discussing about the mistake players made with the old system.
( those i am receiving them in PM if someone is interested )

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Mon Jun 24, 2024 6:52 pm
by mmmPI
functional wrote: Mon Jun 24, 2024 6:30 pm Arguably even OP's nuclear power plant is too complex. Nuclear power is so extremely efficient that once you learn kovarex enrichment, there's no real point to add that whole buffer because having the nuclear reactor 24/7 on barely consumes a resource that is practically infinite anyway. Only in very limited circumstances does it make much sense to not run them 24/7, as each reactor consumes 1,8 u235 per hour. OP's setup consumes 28,8 per hour.. so even a very tiny deposit would last for a long, looong time, while being burned 24/7.

Funnily enough, the new system actually incentivizes it slightly more from my perspective since you don't have to do millions of pumps. The cost to benefit ratio iis more on par; low effort for low gains.
Hey ! this is a much more interesting prospective take than earlier imo, i appreciate it ! I do not share your reasonning, but it made me think of it, i disagree but i do not think it's off-topic apart maybe from this part where you say "do millions pump" that i feel is more based on how YOU did and the picture you sent me, more than how it was necessary to do as demonstrated by Koub's setup where there is no pump at all.

To me with the "old" system there was no incentive to do large scale nuclear because it was not UPS efficient, kovarex enrichment becomes "better" if nuclear is more "viable" in UPS.

So i would build "less solar".

And to me "nuclear" was a side power , because solar was the main.

Therefore it was acting on/off regularly, when the battery charge level of the solar needed it or when spike.

It made no sense for me to run nuclear 24/7, i have even designed system to reduce fuel consumption to not waste uranium feeding nuclear power plant for no reason. I found it a very fun challenge, to me that correspond to what the game has to offer as fun time of problem solving, and learning how to make a nuclear plant and why, the relation with the larger base.

With the change, if nuclear UPS viablity is better, then there is more incentive to me to start avoid solar and do more nuclear, and less incentive to do nuclear fuel control setups.

It's not that bad as more nuclear is better than no nuclear and full solar, but if nuclear "becomes like electric-network" that is negating the added puzzleness. (imo)

Re: Nuclear Power Plant from the future (FFF 416 RANT)

Posted: Mon Jun 24, 2024 7:23 pm
by MeduSalem
mmmPI wrote: Mon Jun 24, 2024 5:16 pmNo you can't divide "3" into 2 different integer that are the same size. 3/2 is 1.5 and it's not an integer, there is no way to add 2 integer to make 3. That's what i meant. I will maybe make the syntax clearer in the drama piece ^^.
And I say you can. The result is 2 * 1 plus 1 remainder. All integers. xD
It is not pretty and not what you want, but a viable solution. :P

mmmPI wrote: Mon Jun 24, 2024 5:16 pm From the FFF it doesn't seem like this is the case for the system that is being implemented to replace the current one :
Yea, obviously it isn't the case. That is because it is an "idea" of what could be done to improve upon Fluid 2.0. ^^

If fluid 2.0 would already work at least like I suggested, people worried about "infinite throughput" would likely complain less because they would be calmed to know that there is an upper limit on throughput as a design challenge to work around and not an infinite throughput.

mmmPI wrote: Mon Jun 24, 2024 5:16 pm
There is no longer a realistic fluid "flow" through pipes; fluid pushed to a segment will be immediately available at any point along a segment. This is the "nuclear electric-network type solution" that was discussed in previous FFFs. The result is that pipes "just work" and you almost never have to worry about throughput.
To me i read that as there is no more "flow calculations" , yet there is still a little something that cap throughput and need be tweaked, but you ALMOST never will be bottleneck, (unless one system really is bad and in such case one were already complaining and will continue do so).
Well whatever they mean with "almost never have to worry about throughput" is at best... vague. ^^
It still implies near to infinite throughput.

I am looking forward to the next part of Fluid 2.0 FFF already. And I wonder if they will change something in terms of infinite throughput because it is the most common concern.

mmmPI wrote: Mon Jun 24, 2024 5:16 pmThe current power plant design may not also work because of too many heat exchanger in a row for heat transfer to reach the lasts ones.
And as funny as it is, I hope they rewrite the heat transfer system one day as well. ^^
Not that is really that bad... it works.
But similar as current pipes... also has some quirks because of build order as well... hence why symmetric builds never result in symmetric temperature spread. It is always somewhat off... and that can make a design that would hypothetically work perfectly, not work as perfectly. xD

Re: Nuclear Power Plant from the future (FFF 416 RANT)

Posted: Mon Jun 24, 2024 8:00 pm
by mmmPI
MeduSalem wrote: Mon Jun 24, 2024 7:23 pm And I say you can. The result is 2 * 1 plus 1 remainder. All integers. xD
It is not pretty and not what you want, but a viable solution. :P
I meant to illustrate something that make no sense or is impossible that could be understood as such. I will think of something else where it would make no sense there is a remainder if it's needed to illustrate why i don't want to hear people on this topic explaning what they think the system could have been because it would be off topic if not in relation with a practical example like the power plant, as it would easily diverge in endless speculation from unequally informed individual with ranging degree of self-awareness, as opposed to practical consideration about the consequences of the change announced in the FFF that leaded to the power plant presented being designed :D

MeduSalem wrote: Mon Jun 24, 2024 7:23 pm Yea, obviously it isn't the case. That is because it is an "idea" of what could be done to improve upon Fluid 2.0. ^^
If fluid 2.0 would already work at least like I suggested, people worried about "infinite throughput" would likely complain less because they would be calmed to know that there is an upper limit on throughput as a design challenge to work around and not an infinite throughput.
I understand there is tight limit between what i consider off topic and not and that may seem contradictive sometimes x), i understand your idea and why you mention it, and i will try to make it more about the current power plant so that it stays on topic so as to illustrate :D (sorry)

What you suggest means on the picture with the pump, people complaining before will still do saying that the segment being created require pumps at arbitrary distance difficult to count (in my opinion). I think the current FFF already hint that thoughput will not be illimited somehow, but i have read it several time and i still don't get how it would work or what does it mean, to me "thoughput through 1 pipe" will be seemingly infinite in some cases like in the sections without pumps on my power plant where around 25 000 water can flow per second though a single pipe, then split into 2 parralel pump and 1 pipe again. It is not the case on the video of the FFF where there is petroleum gas as output and 2 pumps are required for the ouput lane. I wonder why.
MeduSalem wrote: Mon Jun 24, 2024 7:23 pm Well whatever they mean with "almost never have to worry about throughput" is at best... vague. ^^
It still implies near to infinite throughput.
What do you mean in terms of power plant ? i mean in practice what would be the limiting mechanism that makes it "near" or "almost". How does that translate in game design, how would look like a blueprint according to this ? How could there be a rules simple to understand that leads you to design this way ?

This is what i questionned myself with, when making the power plant, that's why i put 3 pumps on the side thinking it's still doable in gameplay to tell player to read the throughput of a pump and the game would be to manage how many connexions between segments, somehow, but i have no idea if this is what was meant by the FFF or if someone can come up with something that makes more sense than my speculation that would convinced me to re-evaluate what i think i have understood from the FFF^^
MeduSalem wrote: Mon Jun 24, 2024 7:23 pm I am looking forward to the next part of Fluid 2.0 FFF already. And I wonder if they will change something in terms of infinite throughput because it is the most common concern.
fluid teleporting was "shown" in the videos to me, not infinite throughput TOWARD THE INSIDE AN ENTITY, ( teleport + infinite pull ) rather infinite throughput PER PIPE , ( you need 2 pump out that are pushing fluid onto the next segment, but that fluid is coming from a single pipe ? )
like in my design, that is the thing i find silly. There are nuances of "infinite throughput" :)
MeduSalem wrote: Mon Jun 24, 2024 7:23 pm And as funny as it is, I hope they rewrite the heat transfer system one day as well. ^^
Ah that's another topic i have seen less people complaining about x)

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Tue Jun 25, 2024 3:00 am
by functional
mmmPI wrote: Mon Jun 24, 2024 6:52 pm
functional wrote: Mon Jun 24, 2024 6:30 pm Arguably even OP's nuclear power plant is too complex. Nuclear power is so extremely efficient that once you learn kovarex enrichment, there's no real point to add that whole buffer because having the nuclear reactor 24/7 on barely consumes a resource that is practically infinite anyway. Only in very limited circumstances does it make much sense to not run them 24/7, as each reactor consumes 1,8 u235 per hour. OP's setup consumes 28,8 per hour.. so even a very tiny deposit would last for a long, looong time, while being burned 24/7.

Funnily enough, the new system actually incentivizes it slightly more from my perspective since you don't have to do millions of pumps. The cost to benefit ratio iis more on par; low effort for low gains.
Hey ! this is a much more interesting prospective take than earlier imo, i appreciate it ! I do not share your reasonning, but it made me think of it, i disagree but i do not think it's off-topic apart maybe from this part where you say "do millions pump" that i feel is more based on how YOU did and the picture you sent me, more than how it was necessary to do as demonstrated by Koub's setup where there is no pump at all.

To me with the "old" system there was no incentive to do large scale nuclear because it was not UPS efficient, kovarex enrichment becomes "better" if nuclear is more "viable" in UPS.

So i would build "less solar".

And to me "nuclear" was a side power , because solar was the main.

Therefore it was acting on/off regularly, when the battery charge level of the solar needed it or when spike.

It made no sense for me to run nuclear 24/7, i have even designed system to reduce fuel consumption to not waste uranium feeding nuclear power plant for no reason. I found it a very fun challenge, to me that correspond to what the game has to offer as fun time of problem solving, and learning how to make a nuclear plant and why, the relation with the larger base.

With the change, if nuclear UPS viablity is better, then there is more incentive to me to start avoid solar and do more nuclear, and less incentive to do nuclear fuel control setups.

It's not that bad as more nuclear is better than no nuclear and full solar, but if nuclear "becomes like electric-network" that is negating the added puzzleness. (imo)
Nuclear is still not good for UPS versus solar. For upkeep it:

- requires centrifuges for refining and enrichmentx
- requires logistics (belds, trains, bots, inserters)
- requires mining (with sulfuric acid)
- requires heat flow calculations
- requires bunch of unique entities

All of these add just a tiny bit of overhead, but its hard to compete with virtually free (which solar is).

But my point originally was that that "nuclear fuel control setup" is nearly entirely pointless thing to do and yet you did it anyway in your caricature build. You could have simplified it so much by removing circuits and all the steam containers from it, yet you didn't.

You kind of demonstrate in practice that you don't even go for the simplest possible setup even while trying to make a point about how simple the setup can get after the change. I think you mostly accidentally demonstrated that you're not even going for the theoretically simplest setup even in a situation where you should, yet you are concerned about less simple design being poossible after these changes.

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Tue Jun 25, 2024 3:40 am
by FuryoftheStars
functional wrote: Tue Jun 25, 2024 3:00 am
mmmPI wrote: Mon Jun 24, 2024 6:52 pm
functional wrote: Mon Jun 24, 2024 6:30 pm Arguably even OP's nuclear power plant is too complex. Nuclear power is so extremely efficient that once you learn kovarex enrichment, there's no real point to add that whole buffer because having the nuclear reactor 24/7 on barely consumes a resource that is practically infinite anyway. Only in very limited circumstances does it make much sense to not run them 24/7, as each reactor consumes 1,8 u235 per hour. OP's setup consumes 28,8 per hour.. so even a very tiny deposit would last for a long, looong time, while being burned 24/7.

Funnily enough, the new system actually incentivizes it slightly more from my perspective since you don't have to do millions of pumps. The cost to benefit ratio iis more on par; low effort for low gains.
Hey ! this is a much more interesting prospective take than earlier imo, i appreciate it ! I do not share your reasonning, but it made me think of it, i disagree but i do not think it's off-topic apart maybe from this part where you say "do millions pump" that i feel is more based on how YOU did and the picture you sent me, more than how it was necessary to do as demonstrated by Koub's setup where there is no pump at all.

To me with the "old" system there was no incentive to do large scale nuclear because it was not UPS efficient, kovarex enrichment becomes "better" if nuclear is more "viable" in UPS.

So i would build "less solar".

And to me "nuclear" was a side power , because solar was the main.

Therefore it was acting on/off regularly, when the battery charge level of the solar needed it or when spike.

It made no sense for me to run nuclear 24/7, i have even designed system to reduce fuel consumption to not waste uranium feeding nuclear power plant for no reason. I found it a very fun challenge, to me that correspond to what the game has to offer as fun time of problem solving, and learning how to make a nuclear plant and why, the relation with the larger base.

With the change, if nuclear UPS viablity is better, then there is more incentive to me to start avoid solar and do more nuclear, and less incentive to do nuclear fuel control setups.

It's not that bad as more nuclear is better than no nuclear and full solar, but if nuclear "becomes like electric-network" that is negating the added puzzleness. (imo)
Nuclear is still not good for UPS versus solar. For upkeep it:

- requires centrifuges for refining and enrichmentx
- requires logistics (belds, trains, bots, inserters)
- requires mining (with sulfuric acid)
- requires heat flow calculations
- requires bunch of unique entities

All of these add just a tiny bit of overhead, but its hard to compete with virtually free (which solar is).

But my point originally was that that "nuclear fuel control setup" is nearly entirely pointless thing to do and yet you did it anyway in your caricature build. You could have simplified it so much by removing circuits and all the steam containers from it, yet you didn't.

You kind of demonstrate in practice that you don't even go for the simplest possible setup even while trying to make a point about how simple the setup can get after the change. I think you mostly accidentally demonstrated that you're not even going for the theoretically simplest setup even in a situation where you should, yet you are concerned about less simple design being poossible after these changes.
Does it even matter? Even if mmmPi were to remove all the circuits and extra tanks and whatever other superfluous items, that doesn't change the overall point of the design demonstration.

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Tue Jun 25, 2024 4:44 am
by functional
FuryoftheStars wrote: Tue Jun 25, 2024 3:40 am Does it even matter? Even if mmmPi were to remove all the circuits and extra tanks and whatever other superfluous items, that doesn't change the overall point of the design demonstration.
Makes it more difficult to take as a serious concern that people can do a simple design when most people (including me) overcomplicate their designs virtually by nature. While OP hasn't stated it explicitly as far as I can recall, lot of people have complained to the effect "No more fluid trains. What's the point anymore?" as if people actually opted for simplest possible designs. The amount of people sticking circuit networks in nuclear reactors is a testament to the contrary. And by the way, I do it too despite knowing how absolutely futile and pointless it is, if I would use nuclear reactors.

Re: Nuclear Power Plant from the future (FFF416/RANT)

Posted: Tue Jun 25, 2024 5:19 am
by functional
Also, I'd be curious what you folks think about this reactor. Dunno about ratios and all since this is B&A and I can't be bothered to switch modlists.

Image

There's no circuits anywhere. Temperature of reactors stays sub 1000 celsius (so all is consumed) and there's not a single pipe in the whole design (besides heat pipes). Nothing I've done here that isn't possible in vanilla.

This reactor design is an order of magnitude simpler than OP's and it's possible right now. In fact, something similar to this (but with actual ratios) is what I design if I go nuclear. Only thing required in vanilla is enough water and landfill. If you need to pump in liquid, it's also trivial. Just need millions of pumps.
All times are UTC
Page 3 of 6

Powered by phpBB® Forum Software © phpBB Limited