Space platform "weight/number of engines/fuel needed" equation?
-
- Manual Inserter
- Posts: 1
- Joined: Thu Oct 31, 2024 9:04 am
- Contact:
Space platform "weight/number of engines/fuel needed" equation?
Hi,
I tried search for some guidelines regarding this problem, and did not find anything. My intuition tells me - the heavier your platform is, the more engines you need to get it moving, and the more fuel you need to get it through the journey to another planet. On the other hand, if I add engines, I add weight, resulting in more fuel needed, and if I add fuel, I add weight, resulting in more fuel needed again. I don't suppose Factorio properly simulates the rocket equation, but still, I don't want to move an overkill amount of engines and fuel around space, seems wasteful.
Is there a consensus on a rough guideline (for example, 2 engines and a full tank of fuel + ox for every 100 t of platform)? Or better, is there an equation available?
Thanks,
Michal.don
I tried search for some guidelines regarding this problem, and did not find anything. My intuition tells me - the heavier your platform is, the more engines you need to get it moving, and the more fuel you need to get it through the journey to another planet. On the other hand, if I add engines, I add weight, resulting in more fuel needed, and if I add fuel, I add weight, resulting in more fuel needed again. I don't suppose Factorio properly simulates the rocket equation, but still, I don't want to move an overkill amount of engines and fuel around space, seems wasteful.
Is there a consensus on a rough guideline (for example, 2 engines and a full tank of fuel + ox for every 100 t of platform)? Or better, is there an equation available?
Thanks,
Michal.don
Re: Space platform "weight/number of engines/fuel needed" equation?
There is this thread that has some discussion about it: viewtopic.php?p=626961#p626961
There DragoonGXG tested that more thrusters have diminishing return on speed.
mmmPI posted some formula for acceleration in there albeit I cannot make much of it (I tried to run it through excel to make graphs to see what effect each value has, but it only gave weird results, probably because I don't know what exact units I need to enter into the formula).
This thread by MBas also deals with fuel economy: viewtopic.php?f=193&t=118009
There are some interesting discussions in there. it is not anywhere fuel efficient to run your thrusters at 100%. You might want to throttle them on purpose using circuit network to make the most out of your fuel, which in return will reduce the chance your platform runs out of fuel midflight but also reduce the downtime you face in orbits to restock fuel. ^^
What is also clear after both threads and my own tests is that the width of the platform also plays a big role in how well your platform behaves. Wider platforms are terribly inefficient. It is a factor in the formula that mmmPI posted, but someone also posted a graph in the thread of MBas thread showing that increasing width of the platform reduces the top-speed asymptotically. So you definietely don't want wide ships that travel. They should be as narrow as is possible per design (which also means you limit the numbers of engines) and it is better to make them just longer instead.
Anyway from my observations (without too much infinite asteroid productivity techs) more thrusters are not always preferable; they just burn more fuel than your platform might be able to sustain early on and it will then result in slowdonws midflight and/or that your platform needs to recover in orbits for long periods.
From personal ships I used... 3 thrusters was okay for small platforms to get a little feel for it.
Then I started using 5 thrusters for medium sized ones which I use to deliver science packs from the Fulgora, Vulcanus & Gleba back to Nauvis. From a design perspective 5 engines is at least as many as you can fit in the back of any reasonable design.
After that it is hard to tell because beyond that I already had research advanced asteroid processing and such. And also some asteroid productivity. That all increased resource conversion ratios like crazy. So the big ship I built for travel to Aquilo last night also only has 5 thrusters and it went there just fine (but using MBas' circuit logic to throttle fuel consumption). It is so freaking efficient that it can actually go around the solar system forever without ever needing to pause in any orbit to recover. The fuel tanks are permanently at 25k. The later reprocessing & advanced processing techs are just that efficient.
But in any case, "real life" rocket physics does not apply in Factorio. The mass of all the stuff you have built on the platform and even all of the payload contents of the hub, and all the fuel you stored into tanks have no effect on the total mass of the ship. It is not factored in (I thought it would but they don't). So forget about it for Factorio.
It means you can totally add more engines if you want to. The limiting factor will be if you can sustain it fuel-wise for the application you want to do.
But that said, it only really matters much during early game. Late game... screw it. You get so many chunks that you never run out of fuel and can sustain lots of thrusters if you wanted to. ^^
There DragoonGXG tested that more thrusters have diminishing return on speed.
mmmPI posted some formula for acceleration in there albeit I cannot make much of it (I tried to run it through excel to make graphs to see what effect each value has, but it only gave weird results, probably because I don't know what exact units I need to enter into the formula).
This thread by MBas also deals with fuel economy: viewtopic.php?f=193&t=118009
There are some interesting discussions in there. it is not anywhere fuel efficient to run your thrusters at 100%. You might want to throttle them on purpose using circuit network to make the most out of your fuel, which in return will reduce the chance your platform runs out of fuel midflight but also reduce the downtime you face in orbits to restock fuel. ^^
What is also clear after both threads and my own tests is that the width of the platform also plays a big role in how well your platform behaves. Wider platforms are terribly inefficient. It is a factor in the formula that mmmPI posted, but someone also posted a graph in the thread of MBas thread showing that increasing width of the platform reduces the top-speed asymptotically. So you definietely don't want wide ships that travel. They should be as narrow as is possible per design (which also means you limit the numbers of engines) and it is better to make them just longer instead.
Anyway from my observations (without too much infinite asteroid productivity techs) more thrusters are not always preferable; they just burn more fuel than your platform might be able to sustain early on and it will then result in slowdonws midflight and/or that your platform needs to recover in orbits for long periods.
From personal ships I used... 3 thrusters was okay for small platforms to get a little feel for it.
Then I started using 5 thrusters for medium sized ones which I use to deliver science packs from the Fulgora, Vulcanus & Gleba back to Nauvis. From a design perspective 5 engines is at least as many as you can fit in the back of any reasonable design.
After that it is hard to tell because beyond that I already had research advanced asteroid processing and such. And also some asteroid productivity. That all increased resource conversion ratios like crazy. So the big ship I built for travel to Aquilo last night also only has 5 thrusters and it went there just fine (but using MBas' circuit logic to throttle fuel consumption). It is so freaking efficient that it can actually go around the solar system forever without ever needing to pause in any orbit to recover. The fuel tanks are permanently at 25k. The later reprocessing & advanced processing techs are just that efficient.
But in any case, "real life" rocket physics does not apply in Factorio. The mass of all the stuff you have built on the platform and even all of the payload contents of the hub, and all the fuel you stored into tanks have no effect on the total mass of the ship. It is not factored in (I thought it would but they don't). So forget about it for Factorio.
It means you can totally add more engines if you want to. The limiting factor will be if you can sustain it fuel-wise for the application you want to do.
But that said, it only really matters much during early game. Late game... screw it. You get so many chunks that you never run out of fuel and can sustain lots of thrusters if you wanted to. ^^
Re: Space platform "weight/number of engines/fuel needed" equation?
The formula that i'm aware of posted in the other thread with units left as a mystery is this one :
space_platform_acceleration_expression = "(thrust / (1 + weight / 10000000) - ((1500 * speed * speed + 1500 * abs(speed)) * (width * 0.5) + 10000) * sign(speed)) / weight / 60"
Thrust is in Newton most likely,
weight in kilograms or grams,
speed in kilometer per second ,or meter per second, or per tick ( 1/60th second )
width most likely in tile where 1 tile = 1 meter
There is an additionnal thing, when you are "leaving" a place you have to fight an additionnal negative 10 km/s speed, and when you are "arriving" in a place you have a boost of around 10km/s, this allows out of fuel platform to not be completly stranded in space but instead drift slowly toward their goal or destination if they are more than 50 of the way to it. This is quite significant at slow speed !
@ Medusalem I suppose to use it properly you need to compute the thrust produced by your thrusters given their fuel level, and calculate the integral to know your actual speed over time. Desmos and Wofram alpha are handy tools for those but i'm too busy playing with combinators to try something x)
Re: Space platform "weight/number of engines/fuel needed" equation?
Sure, one could do nasty integrals. But it doesn't necessarily have to be me, because after doing some googling it seems like there were already others at it:mmmPI wrote: ↑Sat Nov 02, 2024 9:22 am @ Medusalem I suppose to use it properly you need to compute the thrust produced by your thrusters given their fuel level, and calculate the integral to know your actual speed over time. Desmos and Wofram alpha are handy tools for those but i'm too busy playing with combinators to try something x)
https://www.reddit.com/r/factorio/comme ... thrusters/
The best one so far (most accurate):
https://www.reddit.com/r/factorio/comme ... top_speed/
I am always surprised how fast people are with stuff like that.
Also someone put in the effort and made a table on Wiki already... ^^
https://wiki.factorio.com/Thruster
So there we have the consumption values for any given fuel level without having to look at the graph.
The only values that are missing in the table is the direct thrust values, but those are also specified in the thruster:
Q1: 9.9MN-102MN
Q2: 12.9MN-132MN
Q3: 15.9MN-163MN
Q4: 18.9MN-193MN
Q5: 24.9MN-254MN
So together with the relative % thrust values from the table above you can determine the actual thrust for any fuel level.
I think with the table above one can make pretty nice assumption already.
From the thread of MBas and his circuit logic that can limit the fuel consumption, a value of C (consumption rate) around 40-50 is probably the most optimal. You get decent speed for the amount of consumption and I have been running these values for quite some time on my platforms. If you have lots of fuel available and can't consume it all then you can totally go higher as well, but if you are fuel starved then around 40 seems a sweet spot.
Also that one Demos calc is pretty accurate because it calculates exactly the top-speed values I have in-game for the consumption rate I limited my thrusters to for the mass & thruster config. ^^
Re: Space platform "weight/number of engines/fuel needed" equation?
MeduSalem wrote: ↑Sat Nov 02, 2024 12:28 pm I am always surprised how fast people are with stuff like that.
....
a value of C (consumption rate) around 40-50 is probably the most optimal. You get decent speed for the amount of consumption and I have been running these values for quite some time on my platforms. If you have lots of fuel available and can't consume it all then you can totally go higher as well, but if you are fuel starved then around 40 seems a sweet spot.
Yeah attention to hastly conclusion though x), to me it seem there is no particular sweet spot at 40 50 % this is just a matter of preference right ?, the place where there IS a sweet spot however i would say is 75% , because that's when you reach max thrust which is the important stat for speed, which is the important stat for how much time the platform take to move from point A to point B which is the important stat if you want to have your lab running and your science not stopping, you need to dimension the ship to carry the amount of science pack your factory consume during the time it takes to bring them roughly.
And as going above 75% is only going to waste fuel for no thrust up to 80%, but then between 80 and 100% it's the same waste, i think one should aim for 75%.
When considering how many thrusters, you need the mass and width of the platform to know the speed you will reach when having them at full thrust so that you can know the time it would take to travel. That's how it will be eventually, even if in early game you are scare in ressources and you throttle the fuel additionnaly.
Fuel efficiency is only when you are begining space logistic and you are not sure your ship can make it for me, but then you can start as low as 10% or 20% or 30% imo
I have played a bit with the formula and i think sometimes people are just mis-interpreting the curves crossing on the factoriopedia as something significant but it's just the scale and the curve on the same graph that gives this impression, it is not something that is a mathematical answer to an optimization question, although that could be a personnal "rule of thumb", unlike the "75%" which may be a "bad" rule of thumb for early game, but that has a real mathematical signification as "optimal".
Re: Space platform "weight/number of engines/fuel needed" equation?
No. It is not really a "hasty" conclusion but one I found to be suitable after toying a little bit with other values that only made it worse.mmmPI wrote: ↑Sat Nov 02, 2024 3:01 pm Yeah attention to hastly conclusion though x), to me it seem there is no particular sweet spot at 40 50 % this is just a matter of preference right ?, the place where there IS a sweet spot however i would say is 75% , because that's when you reach max thrust which is the important stat for speed, which is the important stat for how much time the platform take to move from point A to point B which is the important stat if you want to have your lab running and your science not stopping, you need to dimension the ship to carry the amount of science pack your factory consume during the time it takes to bring them roughly.
And as going above 75% is only going to waste fuel for no thrust up to 80%, but then between 80 and 100% it's the same waste, i think one should aim for 75%.
When considering how many thrusters, you need the mass and width of the platform to know the speed you will reach when having them at full thrust so that you can know the time it would take to travel. That's how it will be eventually, even if in early game you are scare in ressources and you throttle the fuel additionnaly.
Fuel efficiency is only when you are begining space logistic and you are not sure your ship can make it for me, but then you can start as low as 10% or 20% or 30% imo
So it sure depends on the situation, but that is why I wrote "if you are fuel starved" and want to make the most out of it, which applies the most to early platforms.
My early platforms (the ones without advanced asteroid processing) were somewhat fuel starved with higher fuel consumption. Those worked way better after I set C~40 or the likes after adding the fuel consumption circuit logic. There I had the highest engine efficiencies I have seen on my platforms and barely used any fuel from the tanks anymore after a trip.
When I set C to higher values the thruster efficiency started dropping. Sure the platform arrived faster, but the downtime in orbit to restock fuel was much more atrocious as well. And in Nauvis orbit with the little amount of chunks you get... that is annoying.
The thing however is, my platform's travel times are not a bottleneck. They spend more time waiting for science packs in orbit than they spend going to Nauvis, unloading, restocking fuel & traveling back. So I didn't care that they were traveling somewhat slower with a lower C value. ^^
Anyway for the platform I built yesterday which can go to Aquilo, also using all advanced processing stuff obviously, I set the C value to way higher already because of the ridiculous amount of chunks in the region. Like anyone, I cannot get rid of all the chunks and needed to limit the grabbers from inputting more.
But in any case I agree that it definitely is a matter of preference and what you aim for and which game phase you are in.
Yea, as said that one calculator above already does a fine job with that. ^^mmmPI wrote: ↑Sat Nov 02, 2024 3:01 pm When considering how many thrusters, you need the mass and width of the platform to know the speed you will reach when having them at full thrust so that you can know the time it would take to travel. That's how it will be eventually, even if in early game you are scare in ressources and you throttle the fuel additionnaly.
But that said, from toying with the formula I also noticed that the width of the platform has way more ill effects than the mass of the platform. ^^
The mass of the platform barely effects top speed at all.
So the actual compromise I predict people to design around eventually in the future is... thrusters vs platform width.
There probably is a point where more thrusters would make the platform so wide that it barely gives any more top speed anymore.
Re: Space platform "weight/number of engines/fuel needed" equation?
Even before reaching that point you'd want to make proper use of the width of your platform by putting thrusters that are not gimped at 40% thrust i believe.
Maybe your priorities will change in the future and you will say that it was a hastly conclusion without all the information available
I can understand that higher efficiency would be more interesting early game, but as low as 40 to 50 seems to be too low for general rule of thumb or so called optimal, i think it's more something that keeps changing depending on the ship and travel you want to give it and size of platform and purpose and quality of materials. With the defined spot at 10% and 75% for "optimal".
Re: Space platform "weight/number of engines/fuel needed" equation?
For early the early game platforms the 40% was the right call I think. The platforms are so small then still, you have limited machines on them; maybe not even a quality item either. So fuel production was definitely slow and one cannot process the chunks fast enough (I mean those gathered midflight).mmmPI wrote: ↑Sat Nov 02, 2024 7:14 pmEven before reaching that point you'd want to make proper use of the width of your platform by putting thrusters that are not gimped at 40% thrust i believe.
Maybe your priorities will change in the future and you will say that it was a hastly conclusion without all the information available
I can understand that higher efficiency would be more interesting early game, but as low as 40 to 50 seems to be too low for general rule of thumb or so called optimal, i think it's more something that keeps changing depending on the ship and travel you want to give it and size of platform and purpose and quality of materials. With the defined spot at 10% and 75% for "optimal".
I mean sure, you can build a somewhat bigger, over-engineered platform with several machines with more solar panels, more circuit logic and whatnot. Then it is definitely easier to run higher thruster fuel levels sustainably, but that just takes more time to get ready & launch into orbit.
I mean in the end how long does it even take to travel those 15000km distances on average. A minute or whatever? (Haven't timed it) What does it matter if it can do it in 40 sec or whatever when the platform then idles in orbit for multiple times longer because it burned through all the fuel reserves. ^^
For later platforms that can make use of all the advanced asteroid processing methods... I agree however. There you can ramp it up all the way to the point where you get the 100% thrust (aka 75% fuel level). Because you easily have enough fuel & chunks to sustain that permanently without any downtimes to restock fuel.
Re: Space platform "weight/number of engines/fuel needed" equation?
My point is that you get all the math and curves telling you there's nothing magical about 40% x)MeduSalem wrote: ↑Sat Nov 02, 2024 7:46 pm For early the early game platforms the 40% was the right call I think The platforms are so small then still, you have limited machines on them; maybe not even a quality item either. So fuel production was definitely slow and one cannot process the chunks fast enough.
Some people build ship that do loop around all the planet, and add a planet when they unlock it, other people make ships that do only 1 planet and add ship when they unlock planets.
The speed you need/ want and the size of your ship(s) can vary a LOT, it just happened to work for you given the particular ship you have but change the level of physical damage research and you need a different amount of iron versus ice in your ship, and wether or not you use reprocessing, and wether or not you have asteroid productivity research those also matter , same if you use foundries or modules or red ammos on your platform, all of those have an impact on the ressources you have that are at least as much important as the fuel efficiency imo regarding uptime/downtime in early game.
( it's also low enough that it won't fail apart from maybe the worst of ships )
The number of thruster is a little different to me, it is what will set the constraint on the ship. If there was a classification for ships it would be by number of thrusters then weight then purpose or carrying capacity if it matters.
But really when you build a ship, you don't know for sure the final weight, you know however the number of thruster, that's coming way before in the process of making the ship, if you don't know the number of thruster you can't dimension your fuel production, and your crushers and so on, you need to have an idea quite early in the process.
Knowing the number of thrusters allows you a maximum weight to still be able to travel on time for spoilables. Otherwise you need to account for the couple weight/capacity. Because a slower ship due to more weight may not be problematic if it can carry more material. To maintain the same throughput overall.
Regarding "fuel tanks", i think they are worse than "water tank" + iron ore and carbon regarding density of storage. And even water tank maybe worse than ice. I think in early game it may be necessary to buffer fuel because it also allow to buffer the energy needed to produce it, so you can produce fuel and ammo at different time , when travelling / when in orbit transfering and thus need less solar pannel/smaller ship/faster / more efficient.
But later it becomes more convenient to just dimension the fuel production so that you don't need to buffer fuel, particularly when the advanced fuel recipe is unlocked and you are swimming in asteroid chunks. And i feel like people are going to reach that stage faster and faster as the technique to deal with the puzzle in different planets becomes familiar. I expect ships with lots of time invested to be that stage, because you can go thru the early mid game while still exploring and making test on your ships, or even quite fast, to the point where "the late game begins" and you start to dig in ship making a bit more, with advanced tech.
Re: Space platform "weight/number of engines/fuel needed" equation?
Not saying it is "magical". But it seems like it is the area the devs balanced the "100% base line" for fuel consumption around.
It is the point where you have about 100% fuel consumption, but get about 75% thrust.
Just looked into the game and the exact value is actually 43% fuel capacity and not 40%. 43% is where you get 100% fuel consumption and 75% thrust. It seems to be where the orange and the green graph intersect in-game.
With 75% fuel capacity, sure you may get 100% thrust, but you also increase the fuel consumption all the way up to 186%.
Sounds menacingly inefficient, but as said, if you have sooo much fuel that you don't care, well then screw it. Fill the thrusters up to 75% capacity.
I think another good trade-off spot is actually 50% capacity. That is where you get 84% thrust and 119% fuel consumption. Kinda somewhat where both values are almost "equally off" of an optimal 100% and might be also something to settle on if you value both thrust and fuel consumption equally.
Hypothetically to maximize fuel efficiency but still get lots of power out of it you could for example when deliveirng SPs from Gleba to Nauvis ramp it all up to 75% or even 100%, and on the way from Nauvis back to Gleba... let it recover in "eco" mode because it will probably have to wait for packs to be ready to be shipped anyway ^^
Anyway an interesting topic that will probably keep people busy in the long run. ^^
Yea, different people like different playstyles. Me for example I like point-to-point connections for the science packs. Less waiting for stuff to be delivered. ^^
Well, the various chunk processing techniques and whatnot only change how much fuel you can produce for a given amount of chunks; and thereby how many thrusters you can sustain in a reasonable fashion as a side effect. ^^mmmPI wrote: ↑Sat Nov 02, 2024 8:29 pm The speed you need/ want and the size of your ship(s) can vary a LOT, it just happened to work for you given the particular ship you have but change the level of physical damage research and you need a different amount of iron versus ice in your ship, and wether or not you use reprocessing, and wether or not you have asteroid productivity research those also matter , same if you use foundries or modules or red ammos on your platform, all of those have an impact on the ressources you have that are at least as much important as the fuel efficiency imo regarding uptime/downtime in early game.
But it changes nothing about the fact that working thrusters with higher fill levels makes them consume dis-proportionally more fuel in the first place only for a diminishing amount more of thrust.
mmmPI wrote: ↑Sat Nov 02, 2024 8:29 pm But really when you build a ship, you don't know for sure the final weight, you know however the number of thruster, that's coming way before in the process of making the ship, if you don't know the number of thruster you can't dimension your fuel production, and your crushers and so on, you need to have an idea quite early in the process.
[...]
Knowing the number of thrusters allows you a maximum weight to still be able to travel on time for spoilables. Otherwise you need to account for the couple weight/capacity. Because a slower ship due to more weight may not be problematic if it can carry more material. To maintain the same throughput overall.
But judging from the calculations that people implemented in desmos, the weight of a platform pretty much does not matter at all for the top speed it can reach.
It is just involved in the acceleration phase and determines how long it takes the platform to the top speed. But from changing the mass in the desmos calcs I get the feeling that a platform with 5 thrusters that weighs 200 tons reaches the same top speed eventually as a platform with 500tons. (given they have the same width).
However the acceleration phase can become a significant part of the trip though; that probably takes some more investigation how long it actually takes to get to another planet with a certain speed and how much of it is dedicated to accelerating the platform to that speed. Because if it becomes a significant part of the travel time then it is better to add a couple more thrusters to get it up to top speed faster.
[edit 1]
While writing I did some more tests. And yea can confirm. The desmos calculations are correct. This one: https://www.desmos.com/calculator/eykhbatbn6
Mass does not matter for the max speed at all.
Just added several tons more platform foundations onto one of my ships, accelerated and it got to the same top speed as it did without that additional mass. Probably just took a tiny amount more time to accelerate, but otherwise same speed. ^^
For the acceleration calculations the mass is part of the denominator and a value of 10000 gets summed to it. So a platform of 1000 tons would still only barely make the platform accelerate slower than a platform with 200 tons. It would only really start get noticeable if you try to approach the 10000 tons where it would start to become a major part of the denominator. ^^
[/edit 1]
[edit 2]
So now that I had some more time I did some more testing with the calculator, entering lots of different thruster configurations since it seems to be pretty accurately reflect the numbers I also get in-game.
My observation from toying with the calculation is... funnily enough... you cannot really increase the top speed of platforms by all that much. Only a single digits km/s with every thruster.
That is unless you use quality thrusters. Those are the only ones that allow you to get even higher top speeds. So if you want to decrease the travel time of your platforms there is no way around crafting higher quality thrusters. (so much for "quality is optional" xD)
The formula seems to develop that way because for every thruster you add, you have to increase the total width of the platform by another 4 tiles since otherwise you cannot fit them (I saw some ways where some people did multiple rows already, but I will leave those out for now).
And those 4 additional tiles of width for each thruster reduce the top speed thanks to how they are factored into the formula. It pretty much almost cancels out the additional top speed gain from the thruster you place on it. As said it barely increases the top speed by a few km/s with with every thruster.
I think they did the formula that way on purpose to give incentive to craft higher quality thrusters.
But the final straw and why going for more thrusters is not really worth it is that the few km/s in top speed you get, are bought through the ridiculous amount of higher fuel consumption from the additional amount of thrusters that you would need to sustain.
It also means that in return is you will have to place dis-proportionally more chem plants to refine even more fuel to make up for it as well and that just takes up even more space on the platform. And all that for no good reason since you don't get anything out of it anyway.
In other words... from my tests I feel like it never pays off to place more thrusters beyond a certain amount:
(I assume that the thrusters are placed directly next to each other without gaps and 1 tile to the left most and 1 tile to right most thruster that are necessary to get the pipes for fuel in, so that is the "+2" with the tiles.)
1 to 2 thrusters (at least 8+2 tiles width respectively) is definitely much too less. There you are still far off the possible top-speed "limit".
3 to 4 thrusters (at least 12+2 or 16+2 tiles width) already gets pretty close to the "wall" of speed increase.
With 5 thrusters (at least 20+2 tiles width) you definitely hit the wall in speed increase already.
I looked at more thrusters, but it became obvious that the Factorio Gods frown if you make the platform any wider than those 5 thrusters. You get almost no speed increase in return even if you would place more thrusters. More than that are pointless. You are just burning fuel like there is no tomorrow. You are only battling against the width of the platform at that point and you could avoid that altogether if you just don't make the platform wider than 5 thrusters.
If one wants to go faster than that, then swapping them all out for quality ones is the only way.
Alternatively if one does not care about top-speed the quality thrusters would allow one to make wider platforms.
And obviously if you go further out in the solar system where you can make lots of fuel it might also not matter at all. ^^
[edit 2]
Re: Space platform "weight/number of engines/fuel needed" equation?
That's my point it seem a coincidence due to the scale choosen and the choice of putting all the curves on the same graph and using %. The fuel consumption represented goes from 0% to 200% as the max, the efficency from 0 to 100%. It's an arbitrary choice, it could have been written on 0 to 100% for the max fuel consumption and the curve would have intersected differently yet everything in game would be the same.
it's just wording x) you can just say it's 93% of max consumption and you still get 100% of thrust waow 7% saved.
I think it's misleading to consider things this way, i think diminishing return apply for things that you can increase indefinitly, like the width or number of thrusters. However for thrust per thurster, there is a breakpoint, it's not forever diminshing return, it stop having ANY return at some point, when they have 75% fuel, they reach their max thrust. Then there is no return. Before that, imo you are paying 4 tile of width, for only 70% of the benefit of your thruster (because you want to save an infinite ressource).
And you would remove more than a third of this speed-that-is-not-easy-to-increase-due-to-the-wall by not allowing full fuel in thrusters ? I can't believe that Especially after seeing how costly it is to add a thruster because it adds some width and your observation worded as "mass doesn't matter". To me that leaves only width, number of thrusters and .... the % of their thrust you get from them , to me this needs to be maxed out because as said it can !!!
If you want to convince me of anything regarding fuel efficiency, you would need data showing the spawn of asteroids , because otherwise there is nothing to back up any reasonning imo. Only technical solutions to "potential-problems". And not only those in the factoriopedia, those that explain wether or not you get more asteroid if the ship is wider, and the % of asteroid spawned that one ship is able to transform. Because it means little to me to consume half or double the amount of fuel regarding the ressources if you don't know how much ressources you have. It only mean something regarding the footprint of the chemplants and the energy cost of making the fuel for the power of the platform and its footprint. This is measureable , and imo it's not much investement to build for full consumption , the footprint can be made small and energy cost cheap once you have techs and research. And before that you're still in starter base so it's not necessary to optimize as much imo, it's temporary and things will get upgraded anyway.
Re: Space platform "weight/number of engines/fuel needed" equation?
I already wrote "if you have sooo much fuel that you don't care, well then screw it. Fill the thrusters up to 75% capacity for 100% thrust."
I am doing the same for Aquilo & beyond as well because advanced processing and bigger asteroids give more chunks than you could ever process; I am also not running "eco mode" there. ^^
From my experience with it I can't say the same applies to early game and the inner planets though (especially Nauvis with the handful chunks that trickle in) when you still don't have the resources or techs to make bigger platforms capable of (efficiently) processing more chunks yet. So there is a phase where you might not want to use fuel as aggressively to minimize downtimes.
For now I have nothing more to add to the "fuel efficiency of thrusters" part of the discussion because it would only go around in circles.
I only hovered a couple times over the graphs for the routes in the solar map that connect the planets that show the types of asteroids that spawn on the way. However I haven't looked at them in a too detailed fashion yet and can't say how the data is packaged and whether anything "useful" can be read from it in regards of actual spawn numbers or whether it just tells the relative percent chance that is not much valuable without knowing hard number for spawn rates.
I would look at it a little bit but I don't have the time to play for a while (vacation is over, now busy with other real life stuff). ^^
Anyway even if I don't get to look at it in foreseeable time I am sure someone will put a table with spawn rates on the wiki eventually as well. There are always people even more freakier about it. I am sure someone already discussed exactly that in other places like discord/reddit or wherever and that someone is doing exactly that research right now.
But otherwise I agree, there sure is a formula involved how the spawn rates of asteroids relates to the current speed and the dimensions of the platform. It only is a matter of time until someone figures it out. ^^