Factorio Forums

I ran my 600SPM "mini-mega-base" on nuclear with zero UPS issues. The thought of running around placing enough solar panels to compete was horrifying.

I would guess that despite the image this forum projects, relatively few people get to the stage where UPS becomes the limiting factor. And in the "early late game", I'd say nuclear is pretty well balanced. Perhaps the solution would be some "late late" game nuclear upgrade that allows certain layouts to act as a single entity (just like solar does) to reduce UPS?

Without knowing exactly why nuclear is UPS-heavy, I wonder if a simplification that "turns off" fluid flow mechanics (in favor of instantaneous flow) through pipes for steam under certain circumstances would help?

Selvek wrote: Thu Jul 11, 2019 10:08 pm I ran my 600SPM "mini-mega-base" on nuclear with zero UPS issues. The thought of running around placing enough solar panels to compete was horrifying.

How is that possible? Do you have PC Gamer Magazine's large pixel collider?

Selvek wrote: Thu Jul 11, 2019 10:08 pmI would guess that despite the image this forum projects, relatively few people get to the stage where UPS becomes the limiting factor. And in the "early late game", I'd say nuclear is pretty well balanced. Perhaps the solution would be some "late late" game nuclear upgrade that allows certain layouts to act as a single entity (just like solar does) to reduce UPS?

Late game is late game. Splitting it up further than that really makes no difference and just adds to the confusion. The ultimate authority on your nuclear setup is power demand anyway.

As for the claim that "relatively few people get to the stage where UPS becomes the limiting factor", there's multiple issues with that:

• Different people have different hardware setups. In many cases, Factorio is run on basic workstation/laptop PC's that have very limited capabilities. I'm sure for someone with a better computer, UPS only becomes an issue late game. However, for someone with a toaster, issues could hit before they even get past initial oil setup.
• Although the majority of players probably don't get to megabases, we should still accommodate those who do. Most people who have more than 100 hours in-game have probably built a megabase by now, or at least gotten up until the point when it becomes feasible. Using the argument "most people don't" as an excuse is just not considerate of the hardcore players.
• UPS should ALWAYS be a concern from an efficiency and system resource cost standpoint. Unless you can demonstrate that your code is so well optimized that it is technologically impossible to optimize it any further on every potential hardware configuration, it really shouldn't be ignored. That being said, Factorio is extremely well optimized and the devs work hard to keep it that way after every update.

Overall, there's no good reason to really have this stance on UPS.

Selvek wrote: Thu Jul 11, 2019 10:08 pmWithout knowing exactly why nuclear is UPS-heavy, I wonder if a simplification that "turns off" fluid flow mechanics (in favor of instantaneous flow) through pipes for steam under certain circumstances would help?

Nuclear is inherently UPS-heavy because of all the calculations being done, not only by the power plant itself, but also mining and processing of uranium. Only considering UPS, solar with accumulators is a no-brainer.

KoblerMan wrote: Fri Jul 12, 2019 1:56 am

Selvek wrote: Thu Jul 11, 2019 10:08 pm I ran my 600SPM "mini-mega-base" on nuclear with zero UPS issues. The thought of running around placing enough solar panels to compete was horrifying.

How is that possible? Do you have PC Gamer Magazine's large pixel collider?

Or maybe he has just followed "best practices" where UPS is concerned ?
Or maybe Factorio is just well optimized ?

KoblerMan wrote: Fri Jul 12, 2019 1:56 am Different people have different hardware setups. In many cases, Factorio is run on basic workstation/laptop PC's that have very limited capabilities. I'm sure for someone with a better computer, UPS only becomes an issue late game. However, for someone with a toaster, issues could hit before they even get past initial oil setup.

(emphasis mine)
With a literal toaster, maybe? Do you have any concrete examples? What you suggest is borderline insulting for Wube...

KoblerMan wrote: Fri Jul 12, 2019 1:56 am Although the majority of players probably don't get to megabases, we should still accommodate those who do. Most people who have more than 100 hours in-game have probably built a megabase by now, or at least gotten up until the point when it becomes feasible. Using the argument "most people don't" as an excuse is just not considerate of the hardcore players.

Even defining "megabase" as anything you do past the end game, rather than something that taxes your computer so that you're unable to average 60 UPS, here are the following numbers :
- median play hours : 46
https://steamdb.info/app/427520/graphs/
- researched oil processing without using mods : 58%
- finished the game without using mods : 14%
https://steamcommunity.com/stats/427520 ... ce=SteamDB
So your statement seems unlikely.
(I have probably several thousand hours by now, and I never made it to the actual end game : launching a rocket with a satellite. But I'm probably not representative of the "hardcore" players either...)

KoblerMan wrote: Fri Jul 12, 2019 1:56 am UPS should ALWAYS be a concern from an efficiency and system resource cost standpoint. Unless you can demonstrate that your code is so well optimized that it is technologically impossible to optimize it any further on every potential hardware configuration, it really shouldn't be ignored. That being said, Factorio is extremely well optimized and the devs work hard to keep it that way after every update.
[/list]
Overall, there's no good reason to really have this stance on UPS.

Selvek wrote: Thu Jul 11, 2019 10:08 pmWithout knowing exactly why nuclear is UPS-heavy, I wonder if a simplification that "turns off" fluid flow mechanics (in favor of instantaneous flow) through pipes for steam under certain circumstances would help?

Nuclear is inherently UPS-heavy because of all the calculations being done, not only by the power plant itself, but also mining and processing of uranium. Only considering UPS, solar with accumulators is a no-brainer.

While Wube's dedication is commendable, I don't think that they should spend too much time on issues past the end game, that affect only a minority of players.
Everything is a tradeoff : Nuclear is, by definition, not viable as power for the megabase-that-is-so-big-that-you-run-into-UPS-issues.
Optimizing stuff comes at the cost to gameplay : pushing this logic to the extreme : turning Nuclear into another solar would certainly "optimize[] that it is technologically impossible to optimize it any further", but what's the point ?

Using IRL numbers for solar and nuclear is not a good idea, as gameplay > realism.
The realism of the game is pretty good as it goes, and mechanics (like the fluid simulation) > numbers.
(And if one wanted to do that, one would have to start first anyway with the energy in coal and the calorific capacity of water - solar and nuclear only later.)

And btw, the high consumption of water by nuclear reactors is realistic :
For instance, for the Fukushima power plant, they razed the 35m cliff that it would have been located on, so that the costs of pumping seawater are lessened !


And last summer, we had to temporarily shut off some nuclear power plants, as they would have made the river too hot with their waste water!
(Only a tiny fraction of water actually ends up evaporated.)

(and this is IMHO good for gameplay, as the player has now to deal with with logistics of water, which is good for a late-game mechanic like nuclear.)

Rebuttal of bullshit "facts":

• My "toaster" runs 1GW/60SPM at 1.15Ghz at 60UPS. And it's a totally chaotic factory with zero UPS optimizations. Your random assumptions only make your argument weaker.
• Tuning game mechanics for the hardcore players is "not considerate" of the 99% of normal players.
• In a world where you have infinite resources and infinite development time and developers with infinite IQ available you can try to optimize for "every potential hardware configuration". In reality you can not. Having to spell out tautologies makes me sad...

KoblerMan wrote: Thu Jul 11, 2019 9:44 pm

eradicator wrote: Thu Jul 11, 2019 8:58 pm

• (500-15) * 0.2kJ * 60 == 5820MJ, production of turbine
• (165-15) * 0.2kJ * 30 == 900MJ, production of engine

Alright, I see my math is incorrect. I didn't have time to proofread when I posted. I think actually that the discrepancy (if there is one) is the MJ value of the fuel type against total water consumed, and the case to be made is that the water cost of nuclear is just way too damned much compared to regular steam.

Apparently you also didn't have time to understand my post, because it says right there that nuclear uses *less* water per MW produced because hotter steam has more energy. Maybe you should try to build a 1GW+ coal power plant for comparision...

Btw, if the fluid-optimizations ever happen, then nuclear will profit too.

They already did :
https://factorio.com/blog/post/fff-271
(And I remember someone saying that upgrading his 0.16 megabase to 0.17 made him get back to 60 UPS again ?)

IIRC the plan forward is to make the fluid system less optimized, but easier for new players to understand ?
(With an updated wave equation algorithm, and, for instance, making fluids spread more evenly at intersections, while keeping traveling wave oscillations at 0.16.51/0.17.0 levels...)

Huh, you're right. Some of it seems to be in, i thought it was all postponed. (Had to look it up int he annotaed roadmap thread.)

FFF 283 wrote: So the fluid mixing prevention and fluid update optimisations are in 0.17, but the new algorithm was put aside for further research.

BlueTemplar wrote: Fri Jul 12, 2019 7:24 am

KoblerMan wrote: Fri Jul 12, 2019 1:56 am Different people have different hardware setups. In many cases, Factorio is run on basic workstation/laptop PC's that have very limited capabilities. I'm sure for someone with a better computer, UPS only becomes an issue late game. However, for someone with a toaster, issues could hit before they even get past initial oil setup.

(emphasis mine)
With a literal toaster, maybe? Do you have any concrete examples? What you suggest is borderline insulting for Wube...

No, it isn't. I'm also not insulting Wube with this, nor intending to even mention them by any means. Prior to 0.17.x, UPS was a muuuuch bigger problem than it is now. The optimizations with fluid mechanics alone for 0.17 pretty much invalidate the last part of my statement anyway. Look, the point is, UPS drops still happen, and it's more likely to be with users who unfortunately are not running the best hardware.

BlueTemplar wrote: Fri Jul 12, 2019 7:24 am

KoblerMan wrote: Fri Jul 12, 2019 1:56 am Although the majority of players probably don't get to megabases, we should still accommodate those who do. Most people who have more than 100 hours in-game have probably built a megabase by now, or at least gotten up until the point when it becomes feasible. Using the argument "most people don't" as an excuse is just not considerate of the hardcore players.

Even defining "megabase" as anything you do past the end game, rather than something that taxes your computer so that you're unable to average 60 UPS, here are the following numbers :
- median play hours : 46
https://steamdb.info/app/427520/graphs/
- researched oil processing without using mods : 58%
- finished the game without using mods : 14%
https://steamcommunity.com/stats/427520 ... ce=SteamDB
So your statement seems unlikely.
(I have probably several thousand hours by now, and I never made it to the actual end game : launching a rocket with a satellite. But I'm probably not representative of the "hardcore" players either...)

Your statistics (and anecdote that you've never launched the rocket) only further prove my point that "the majority of players probably don't get to megabases" considering total number of players having even launched the rocket in vanilla is just 14%.

Changing the definition of "megabase" is just going to lead down another path of pointless semantics. When I typically define the word "megabase", I mean a base with post-rocket launch where overall production is mentioned in science per minute rather than any other unit of measurement. This is the definition that I decided on for myself since when people post about their "megabases", this is the trend that is displayed most of the time. Your definition of "megabase" may differ, but just know that when I use the word, that is the context that I have chosen to use it with.

BlueTemplar wrote: Fri Jul 12, 2019 7:24 am

KoblerMan wrote: Fri Jul 12, 2019 1:56 am UPS should ALWAYS be a concern from an efficiency and system resource cost standpoint. Unless you can demonstrate that your code is so well optimized that it is technologically impossible to optimize it any further on every potential hardware configuration, it really shouldn't be ignored. That being said, Factorio is extremely well optimized and the devs work hard to keep it that way after every update.
[/list]
Overall, there's no good reason to really have this stance on UPS.

Selvek wrote: Thu Jul 11, 2019 10:08 pmWithout knowing exactly why nuclear is UPS-heavy, I wonder if a simplification that "turns off" fluid flow mechanics (in favor of instantaneous flow) through pipes for steam under certain circumstances would help?

Nuclear is inherently UPS-heavy because of all the calculations being done, not only by the power plant itself, but also mining and processing of uranium. Only considering UPS, solar with accumulators is a no-brainer.

While Wube's dedication is commendable, I don't think that they should spend too much time on issues past the end game, that affect only a minority of players.
Everything is a tradeoff : Nuclear is, by definition, not viable as power for the megabase-that-is-so-big-that-you-run-into-UPS-issues.
Optimizing stuff comes at the cost to gameplay : pushing this logic to the extreme : turning Nuclear into another solar would certainly "optimize[] that it is technologically impossible to optimize it any further", but what's the point ?

Why shouldn't Wube spend time on "issues past the end game"? Is it completely unreasonable for them to focus on a whole 14% of their player base? Ideally, developers should want to satisfy as many subcategories as possible while also keeping the players as a whole satisfied. Sure, maybe they should wait on endgame stuff until post-1.0 launch after most of the bugs are squashed. But the biggest criticism I've ever heard of Factorio is that "once you launch the rocket, there's no point to playing anymore because you have f***-all to do". Why wouldn't Wube expand on that?

To say that optimizing comes at the cost to gameplay just does not make any sense whatsoever. Optimizing nuclear will not turn it into "another solar" and in fact will do the opposite, giving players a choice for once to go nuclear instead of always needing solar.

Also, when I originally said THIS...

KoblerMan wrote: Fri Jul 12, 2019 1:56 am Unless you can demonstrate that your code is so well optimized that it is technologically impossible to optimize it any further on every potential hardware configuration, it really shouldn't be ignored.

...I was basically putting out the sentiment that no particular hardware configuration should be forgotten when considering optimization. My exact wording of "it really shouldn't be ignored" is not a hard literal translation for "always keep optimizing with no regard to anything else". This really has nothing to do with anything particular in the game, either, but rather a philosophy towards optimization and UPS as a whole.

BlueTemplar wrote: Fri Jul 12, 2019 7:24 am Using IRL numbers for solar and nuclear is not a good idea, as gameplay > realism.
The realism of the game is pretty good as it goes, and mechanics (like the fluid simulation) > numbers.
(And if one wanted to do that, one would have to start first anyway with the energy in coal and the calorific capacity of water - solar and nuclear only later.)

And btw, the high consumption of water by nuclear reactors is realistic :
For instance, for the Fukushima power plant, they razed the 35m cliff that it would have been located on, so that the costs of pumping seawater are lessened !


And last summer, we had to temporarily shut off some nuclear power plants, as they would have made the river too hot with their waste water!
(Only a tiny fraction of water actually ends up evaporated.)

(and this is IMHO good for gameplay, as the player has now to deal with with logistics of water, which is good for a late-game mechanic like nuclear.)

What the engineers did for Fukushima is really just the Factorio equivalent of moving your nuclear plant closer to a large body of water to reduce the distance water is pumped through pipes. It really has not much to do with the consumption.

"Only a tiny fraction of water actually ends up evaporated" may be true for real life, but in Factorio, the steam quite literally vaporizes once it's used. There is no method in Factorio for condensing used steam back into water to limit water consumption. This actually makes it less realistic than real life.

As for temporarily shutting off some nuclear plants for making the rivers too hot, what in all honesty does that have to do with the price of cheese on the moon? Because it really doesn't relate to Factorio at all.

eradicator wrote: Fri Jul 12, 2019 10:40 am My "toaster" runs 1GW/60SPM at 1.15Ghz at 60UPS. And it's a totally chaotic factory with zero UPS optimizations. Your random assumptions only make your argument weaker.

UPS is not optimized by the player, it's optimized in the code. You can't sit here and tell me that you can manipulate UPS to any reasonable degree on a software basis. Sure, you can design your factory to be more efficient which will decrease UPS cost, but unless you have double the structures being half as efficient I really quite honestly don't believe you.

eradicator wrote: Fri Jul 12, 2019 10:40 am Tuning game mechanics for the hardcore players is "not considerate" of the 99% of normal players.

Sure it is. You're incentivizing normal players to broaden their horizons and try something new. Normal players are in focus almost all of the time anyway. That's like saying it's not fair to the 99% of people who get cake to give cake to the 1% of people who almost never get any.

eradicator wrote: Fri Jul 12, 2019 10:40 am In a world where you have infinite resources and infinite development time and developers with infinite IQ available you can try to optimize for "every potential hardware configuration". In reality you can not. Having to spell out tautologies makes me sad...

It's a hypothetical, as stated above. Of course it's blatantly unrealistic. Again, my point here is not that this should be the case, but rather that further optimization should always be considered. And also again, Wube does a fantastic job at this and they have my respect for it. I'm talking about all game devs in general.

eradicator wrote: Fri Jul 12, 2019 10:40 am Apparently you also didn't have time to understand my post, because it says right there that nuclear uses *less* water per MW produced because hotter steam has more energy. Maybe you should try to build a 1GW+ coal power plant for comparision...

Ultimately, whether or not my math is correct, my point is that nuclear guzzles water faster than offshore pumps can even dish it out. No amount of optimization will change that. It's a balance issue.

Hmm, have we entered a Tassadar / Overmind mutual destruction mode ?

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm No, it isn't. I'm also not insulting Wube with this, nor intending to even mention them by any means. Prior to 0.17.x, UPS was a muuuuch bigger problem than it is now. The optimizations with fluid mechanics alone for 0.17 pretty much invalidate the last part of my statement anyway. Look, the point is, UPS drops still happen, and it's more likely to be with users who unfortunately are not running the best hardware.

0.17 isn't stable yet. I've also never has any issues, nor see anyone complain about UPS before setting up first oil (specifically due to fluid calculations).

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm Your statistics (and anecdote that you've never launched the rocket) only further prove my point that "the majority of players probably don't get to megabases" considering total number of players having even launched the rocket in vanilla is just 14%.

Changing the definition of "megabase" is just going to lead down another path of pointless semantics. When I typically define the word "megabase", I mean a base with post-rocket launch where overall production is mentioned in science per minute rather than any other unit of measurement. This is the definition that I decided on for myself since when people post about their "megabases", this is the trend that is displayed most of the time. Your definition of "megabase" may differ, but just know that when I use the word, that is the context that I have chosen to use it with.

My point is that there is no meaningful definition of "megabase" (lol, wrote "MAGAbase" on first draft...). People start to talk about SPM after the launch of the first rocket because that's the most obvious thing to do, especially with the infinite sciences. But that SPM might vary by orders of magnitude (with corrsponding UPS impact).

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm Why shouldn't Wube spend time on "issues past the end game"? Is it completely unreasonable for them to focus on a whole 14% of their player base? Ideally, developers should want to satisfy as many subcategories as possible while also keeping the players as a whole satisfied. Sure, maybe they should wait on endgame stuff until post-1.0 launch after most of the bugs are squashed. But the biggest criticism I've ever heard of Factorio is that "once you launch the rocket, there's no point to playing anymore because you have f***-all to do". Why wouldn't Wube expand on that?

To say that optimizing comes at the cost to gameplay just does not make any sense whatsoever. Optimizing nuclear will not turn it into "another solar" and in fact will do the opposite, giving players a choice for once to go nuclear instead of always needing solar.

I said "not too much time". Because trying to optimize for the bases that are so big that they end up with UPS issues cannot be solved by definition. And at some point, solar will always win.
It's not worth if for Wube to expand on the gameplay post the rocket (at least before 1.0) because the huge mod community does it for them. I doubt that all (or even most) of the players are going to keep playing the same game once they've launched a rocket (sat requirement was removed in 0.17 I see...). (There's obviously the players already with mods for which it's harder to know, but they are somewhat of a minority too.)

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm "Only a tiny fraction of water actually ends up evaporated" may be true for real life, but in Factorio, the steam quite literally vaporizes once it's used. There is no method in Factorio for condensing used steam back into water to limit water consumption. This actually makes it less realistic than real life.

Cooling towers were initially planned for vanilla :
https://factorio.com/blog/post/fff-164
(As well as nuclear meltdowns due to overheat.)

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm As for temporarily shutting off some nuclear plants for making the rivers too hot, what in all honesty does that have to do with the price of cheese on the moon? Because it really doesn't relate to Factorio at all.

Just to show how much water can be used by nuclear reactors IRL...

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm Ultimately, whether or not my math is correct, my point is that nuclear guzzles water faster than offshore pumps can even dish it out. No amount of optimization will change that. It's a balance issue.

Concrete proof, please. At which point you are unable to fit enough pipes to deliver water from offshore pumps ?
(Blueprint/Save.)

BlueTemplar wrote: Fri Jul 12, 2019 8:46 pm Hmm, have we entered a Tassadar / Overmind mutual destruction mode ?

Only if you read all of my posts in this thread in The Overmind's voice...

"Serve the hive...
I control the groove!
Serve the hive...
I control the way you move!

"How'dya like my groove, Daggoth?"
*FWOOSH*
"WELL DONE!"

Yeah, yeeah! Yeah, yeeah! Yeah, yeeah...!

En Taro Adun!
Hi...

An immortal classic !

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm Ultimately, whether or not my math is correct, my point is that nuclear guzzles water faster than offshore pumps can even dish it out.

Every time i disprove one of your lies, you fall back to "meh, but i don't like it!", it's quite boring. I could link you to tons of threads by people who "miracilously" managed to build enough pumps to use nuclear (which i happen to be one of). But as you don't care about facts why should i bother? Great job at wasting everyones time Mr. Troll.

Have fun bathing in your imaginary "victory". Bye.

eradicator wrote: Sat Jul 13, 2019 9:23 am

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm Ultimately, whether or not my math is correct, my point is that nuclear guzzles water faster than offshore pumps can even dish it out.

Every time i disprove one of your lies, you fall back to "meh, but i don't like it!", it's quite boring. I could link you to tons of threads by people who "miracilously" managed to build enough pumps to use nuclear (which i happen to be one of). But as you don't care about facts why should i bother? Great job at wasting everyones time Mr. Troll.

Have fun bathing in your imaginary "victory". Bye.

It's not a "victory" until nuclear is balanced better. I'm not trolling if I'm legitimately concerned about the viability of one of Factorio's most awesome features late-game. I would show you a save but I don't have one on hand. I'll work towards one and update the thread.

KoblerMan wrote: Sat Jul 13, 2019 5:57 pm

eradicator wrote: Sat Jul 13, 2019 9:23 am

KoblerMan wrote: Fri Jul 12, 2019 8:05 pm Ultimately, whether or not my math is correct, my point is that nuclear guzzles water faster than offshore pumps can even dish it out.

Every time i disprove one of your lies, you fall back to "meh, but i don't like it!", it's quite boring. I could link you to tons of threads by people who "miracilously" managed to build enough pumps to use nuclear (which i happen to be one of). But as you don't care about facts why should i bother? Great job at wasting everyones time Mr. Troll.

Have fun bathing in your imaginary "victory". Bye.

It's not a "victory" until nuclear is balanced better. I'm not trolling if I'm legitimately concerned about the viability of one of Factorio's most awesome features late-game. I would show you a save but I don't have one on hand. I'll work towards one and update the thread.

It will never be "balanced better". No matter how you optimize it something that uses O(n) time will never compete with something that uses (O1) time. That's the whole definition of O() notation. Solar + accumulator will always be free power (UPS wise) and nothing you can do to nuclear power will ever compete with that. Just alone that inserters need to insert fuel and remove spend fuel cells means an O(n) cost. So even if you optimize nuclear power down to 0 CPU time otherwise the inserters will still kill you compared to solar + accumulators.

KoblerMan wrote: Fri Jul 12, 2019 1:56 am

Selvek wrote: Thu Jul 11, 2019 10:08 pm I ran my 600SPM "mini-mega-base" on nuclear with zero UPS issues. The thought of running around placing enough solar panels to compete was horrifying.

How is that possible? Do you have PC Gamer Magazine's large pixel collider?

Nope. I have a stock Dell that I bought maybe 4 years ago for around $700.

As for why I had no UPS issues... well, I have no idea. I know there are plenty of people with bases much "mega-er" than 600SPS, but I had an 8-reactor nuclear blueprint that I kept stamping down (at least a dozen times, although I haven't opened that save in a while), and it always worked fine.

Selvek wrote: Tue Jul 16, 2019 4:00 pm

KoblerMan wrote: Fri Jul 12, 2019 1:56 am

Selvek wrote: Thu Jul 11, 2019 10:08 pm I ran my 600SPM "mini-mega-base" on nuclear with zero UPS issues. The thought of running around placing enough solar panels to compete was horrifying.

How is that possible? Do you have PC Gamer Magazine's large pixel collider?

Nope. I have a stock Dell that I bought maybe 4 years ago for around $700.

As for why I had no UPS issues... well, I have no idea. I know there are plenty of people with bases much "mega-er" than 600SPS, but I had an 8-reactor nuclear blueprint that I kept stamping down (at least a dozen times, although I haven't opened that save in a while), and it always worked fine.

Do you mean 600 science packs total per minute, 600 science packs each per minute or actual 600 science packs each per second? I don't think anyone has build a 600 SPS mega factory yet. Or at least not one that runs at 60 UPS. That would be 36000 SPM and a rocket every 3 seconds.

mrvn wrote: Thu Jul 18, 2019 9:54 am

Selvek wrote: Tue Jul 16, 2019 4:00 pm

KoblerMan wrote: Fri Jul 12, 2019 1:56 am

Selvek wrote: Thu Jul 11, 2019 10:08 pm I ran my 600SPM "mini-mega-base" on nuclear with zero UPS issues. The thought of running around placing enough solar panels to compete was horrifying.

How is that possible? Do you have PC Gamer Magazine's large pixel collider?

Nope. I have a stock Dell that I bought maybe 4 years ago for around $700.

As for why I had no UPS issues... well, I have no idea. I know there are plenty of people with bases much "mega-er" than 600SPS, but I had an 8-reactor nuclear blueprint that I kept stamping down (at least a dozen times, although I haven't opened that save in a while), and it always worked fine.

Do you mean 600 science packs total per minute, 600 science packs each per minute or actual 600 science packs each per second? I don't think anyone has build a 600 SPS mega factory yet. Or at least not one that runs at 60 UPS. That would be 36000 SPM and a rocket every 3 seconds.

Based on him using "SPM" in his original post, "SPS" was probably a slip-up.

Once you get to counting SPS instead of SPM, you may as well start measuring in rockets per minute/second.

mrvn wrote: Mon Jul 15, 2019 10:31 amIt will never be "balanced better". No matter how you optimize it something that uses O(n) time will never compete with something that uses (O1) time. That's the whole definition of O() notation. Solar + accumulator will always be free power (UPS wise) and nothing you can do to nuclear power will ever compete with that. Just alone that inserters need to insert fuel and remove spend fuel cells means an O(n) cost. So even if you optimize nuclear power down to 0 CPU time otherwise the inserters will still kill you compared to solar + accumulators.

A bit offtopic but an algorithm that takes billion years for any amount of input is O(1) and an algorithm that uses .00000000000000000000000000000000001s*(n*n) is O(N^2). The big O notation defination has nothing to do with time. "Big O notation is a mathematical notation that describes the limiting behavior of a function when the argument tends towards a particular value or infinity". In computer science it is used to describe how an algorithm scales with inputsize. It does not describe how much real/computational time something takes. It is suprisingly common for an O(1) algorithm to be slower than a O(n) or O(n log n) algorithm for small input sizes.

Solar vs nuclear I would also consider the time it takes to build the panels/reactors vs time it saves for having a few % more ups. For how long do you have to play the save until you have saved the time it takes to lay all those panels. If it is more than the time you are planning to play that save then it is better to go for nuclear.

PunPun wrote: Sun Jul 21, 2019 4:04 pm

mrvn wrote: Mon Jul 15, 2019 10:31 amIt will never be "balanced better". No matter how you optimize it something that uses O(n) time will never compete with something that uses (O1) time. That's the whole definition of O() notation. Solar + accumulator will always be free power (UPS wise) and nothing you can do to nuclear power will ever compete with that. Just alone that inserters need to insert fuel and remove spend fuel cells means an O(n) cost. So even if you optimize nuclear power down to 0 CPU time otherwise the inserters will still kill you compared to solar + accumulators.

A bit offtopic but an algorithm that takes billion years for any amount of input is O(1) and an algorithm that uses .00000000000000000000000000000000001s*(n*n) is O(N^2). The big O notation defination has nothing to do with time. "Big O notation is a mathematical notation that describes the limiting behavior of a function when the argument tends towards a particular value or infinity". In computer science it is used to describe how an algorithm scales with inputsize. It does not describe how much real/computational time something takes. It is suprisingly common for an O(1) algorithm to be slower than a O(n) or O(n log n) algorithm for small input sizes.

Solar vs nuclear I would also consider the time it takes to build the panels/reactors vs time it saves for having a few % more ups. For how long do you have to play the save until you have saved the time it takes to lay all those panels. If it is more than the time you are planning to play that save then it is better to go for nuclear.

But we have a (practially) endless map and are continually expanding. So n does keep increasing. And with n > 100000000000000000 or proportionally a 1000000000 * 1000000000 tile base (which is within the possible map size but I guess exceeds possible memory) you .00000000000000000000000000000000001s*(n*n) function would become slower than a 1 function. Just takes time.

More practical even a single inserter for fuel for a single reactor is likely already more costly than solar.

mrvn wrote: Mon Jul 22, 2019 9:20 amBut we have a (practially) endless map and are continually expanding. So n does keep increasing. And with n > 100000000000000000 or proportionally a 1000000000 * 1000000000 tile base (which is within the possible map size but I guess exceeds possible memory) you .00000000000000000000000000000000001s*(n*n) function would become slower than a 1 function. Just takes time.

More practical even a single inserter for fuel for a single reactor is likely already more costly than solar.

Sigh. First off the map is nowhere near endless. The max size is infact 2000000x2000000 tiles. Second that was not the point. The big O notation does not indicate time at all. It describes how something scales with input. The first thing you learn in programming courses that talk about the big O notation is that it is not to be used to decide what algorithm to use. It can be used as a hint on what could be useful but the final decision needs to be made by actually benchmarking with a realistic use case for the project.

If you use three hours to manufacture and place down solar panels or 15 minutes to manufacture and place down a nuclear reactor that makes ups drop by 1% an extremely poorly designed reactor with lots of unneccessary heat/fuildpipes/tanks then the reactor is still more optimal choise until you play more than 275 hours after the reactor is built. If the reactor is properly made then the break even point should be thousands of hours.

This community really does have a problem of spending hours to save a few seconds.

Practically, how much RAM would one need to generate ALL map chunks ?