Accumulator / Solar panel ratio

[Ojelle]

I wouldnt use that one at all. Those 4 accumulators are 1 tile wider wisth (I never now how to type witch -.-) means you cant walk between the grid. In my game I have this (first one of the post) coppied about 80-100 times. Walking between is a min requirement then

[Killkrog]

I dont need to walk between them
All built by robots

[Tnarg]

I think I read some where that in 13 accumulators hold twice as much power making the new ratio 50 solar to 21 Acc?

Edit: I was wrong ignore me.

[Qon]

I think I read some where that in 13 accumulators hold twice as much power making the new ratio 50 solar to 21 Acc?

Not in changelog. viewtopic.php?f=3&t=27101

[Ojelle]

I dont need to walk between them
All built by robots

Neither do I need to for building them. But when I need to go somewhere else, sometimes I'm to lazy to get a train and then I walk below them. Got a grid of those (20*10 or so) and its quite a long walk around that. I prefer a few paths trough then

[Jupiter]

Why not simplificate the whole solar panel and accu system to 1:1 :p

That would take the fun away of all the calculating ^^
And the search for a practial -perfectratio- set-up

We think very much alike!!

[Jupiter]

Guys, I'm not sure about the numbers on the wiki. It says that they are at least up to date up to 0.11.8. We're in 0.13.2 right now..

And the percentages of Day, Dusk, Night and Dawn don't match up with the stated number of ticks.
Does anyone know what the correct values are?

Maybe the holy 0.84 ratio isn't correct anymore

[Qon]

Guys, I'm not sure about the numbers on the wiki. It says that they are at least up to date up to 0.11.8. We're in 0.13.2 right now..

Maybe the holy 0.84 ratio isn't correct anymore

But they haven't changed anything about solar panels and accumulators, right? The wiki hasn't been updated because the game has not been updated in that area. q:

But the 0.84 ratio is pretty worthless outside some very specific applications anyways and should only be used for general guidance. It's almost never something you would need to follow strictly.

[Jupiter]

Guys, I'm not sure about the numbers on the wiki. It says that they are at least up to date up to 0.11.8. We're in 0.13.2 right now..

Maybe the holy 0.84 ratio isn't correct anymore

But they haven't changed anything about solar panels and accumulators, right? The wiki hasn't been updated because the game has not been updated in that area. q:

But the 0.84 ratio is pretty worthless outside some very specific applications anyways and should only be used for general guidance. It's almost never something you would need to follow strictly.

There is a mod that shows you the time of the day and the current brightness called EvoGui. If you watch and write down at what times the brightness starts to change, or stops changing (day->dusk, dusk-> night etc) you get the following:

02:00 night-dawn: the brightness starts increasing from 14%
06:00 dawn-day: brightness stops increasing at 100%
18:00 day-dusk: brightness starts decreasing
22:00 dusk-night: brightness is now 14%

I also timed the whole thing with a stopwatch on my phone (twice) and it lasted for 7 mins and 1.64 sec and for 7 mins and 6.13 sec.
This of course includes the intermittent auto saves and the occasional lagspike (I'm doing this on my lappy).

The total duration of a day seems to agree with what the wiki says (which says 25000 ticks = 6 minutes and 56.67 secs) but the duration of the individual phases do not (except for the day phase).
Right now it seems that dusk, night and dawn all have the same duration.

So yes, they did change this.

Edit: yes, I know for sure that the mod shows the correct time and brightness level because the changes in the shown brightness level coincide with the visual changes in brightness.

[Qon]

02:00 night-dawn: the brightness starts increasing from 14%
06:00 dawn-day: brightness stops increasing at 100%
18:00 day-dusk: brightness starts decreasing
22:00 dusk-night: brightness is now 14%
Edit: yes, I know for sure that the mod shows the correct time and brightness level because the changes in the shown brightness level coincide with the visual changes in brightness.

The solar panel efficiency does not jump to 14%. Just look at your power graph. The brightness levels does not perfectly coincide with solar panel efficiency.
You could also easily measure the panel effectiveness during a cycle by placing a single panel on it's own network with a radar and let it run for a while. Look at the 10m or 1h average and see the power curve for yourself, with it's average output. Measuring brighness is wrong if it's unrelated to what you acually want measured.

[Jupiter]

02:00 night-dawn: the brightness starts increasing from 14%
06:00 dawn-day: brightness stops increasing at 100%
18:00 day-dusk: brightness starts decreasing
22:00 dusk-night: brightness is now 14%
Edit: yes, I know for sure that the mod shows the correct time and brightness level because the changes in the shown brightness level coincide with the visual changes in brightness.

The solar panel efficiency does not jump to 14%. Just look at your power graph. The brightness levels does not perfectly coincide with solar panel efficiency.
You could also easily measure the panel effectiveness during a cycle by placing a single panel on it's own network with a radar and let it run for a while. Look at the 10m or 1h average and see the power curve for yourself, with it's average output. Measuring brighness is wrong if it's unrelated to what you acually want measured.

You completely misread all of what I said.
I didn't 'measure' the brightness. I used a mod called EvoGui that simply shows the current brightness alongside the time of day. And I am also not saying that the brightness suddenly jumps to 14%. What I'm saying is that at 18:00 it STARTS decreasing (linearly but that's not the point) and at 22:00 it is at 14% and stops decreasing.

But the point I wanted to bring across is that the numbers concerning the length of each cycle within a day as shown on the wiki are not correct (any more).
Firstly because several of the numbers shown conflict with each other in the following sense:
The table on this page (wiki.factorio.com/index.php?title=Game-day) says that dusk would last for 5000 ticks. It also says that it starts after 25% of the day has passed and it ends at 42%. This means that dusk comprises 42-25 = 17% of a whole day (a complete day being 25000 ticks). 17% of 25000 ticks = 4250 ticks. This is different from the stated 5000 ticks. So either one (or both) have to be incorrect.
The same applies to the night phase and the dawn phase (but not for the day phase, 12500 ticks really is 50% of 25000 ticks)

Secondly, my measurements of when the different phases start and end are not consistent with the wiki.
What I measured was is that the dusk, night and dawn phases all are of the same length. They all last for 1/6 of a complete day (with the day phase making up for the rest of the complete day). This contradicts the number of ticks and number of seconds each one would last according to the wiki.

I hope I made myself clear this time.
I also hope that anyone with the ability to edit the wiki finds this post and changes it.

[Qon]

And you still missed my point. Have solar panel output actually changed? That is the only thing that matters here. All other details are secondary. No one should change the wiki, because there's no info suggesting the data is false. I measured the solar panel output a while back, and it is still 42kW average with no mention in any changelong since about solar panel alterations. Now the game nychthemeron average could be the same even if they had changed the game day and night lengths but that could still be detected from the power output graphs.

If solar panels wer eoutputting 100% when the brightness was 0% would you still use the brightness levels to measure solar panel output? Or would you directly measure solar panel output to learn the solar panel output?

Also, inb4 you remember that you have a mod enabled altering the timescale or solar panels lol.

[Jupiter]

I don't have any mods that mess with day length.

And yes, average solar panel output really is different. This is because the dusk and dawn phases are shorter.

I also did the math to recalculate the new ratio using the info in the first post and it is now:
(copied from first post and edited)

n / m = (E_acc / P') * (o / c) = [(t1 + t3) * [t2 + t3 * (t1 + t3) / T] / T] * (o/c)

o = 60 kW
c = 5000 kJ
t1 = 0.5 (as before)
t2 = 1/6 (=0.16666667)
t3 = 1/6 (=0.16666667) (which is counted twice, visible in the graph in the first post)
T = 1.0

Times are in fraction of day, so they must be multiplied by (25 000 / 60). With these numbers, the numerical application gives us :

n / m = [(0.5 + 1/6) * [1/6 + 1/6*((0.5+1/6) / 1.0] / 1.0] * (25 000 / 60) * (60 / 5000) = 0.92592592592592592592592592592593

... No one should change the wiki, because there's no info suggesting the data is false....

I just pointed out an inconsistency in the wiki.... so yes the data on the page on Game Day really is false. Then there are my measurements...

[DaveMcW]

LuaSurface.darkness (which the mod uses) is capped at 0.85, it is completely useless for calculating solar panels.

Anyone who thinks 0.84 is wrong should test it with 50 solar panels, 42 accumulators, and 7 radars before trying to do math.

[Jupiter]

LuaSurface.darkness (which the mod uses) is capped at 0.85, it is completely useless for calculating solar panels.

Anyone who thinks 0.84 is wrong should test it with 50 solar panels, 42 accumulators, and 7 radars before trying to do math.

I was not testing solar panel capacity.
I was just measuring the brightness on different times of the day.

And yes, the LuaSurface.darkness value is capped at 0.85 but that is just because the actual brightness never goes lower than 15. (well actually 14.999...something... which gets rounded down by EvoGui to 14%). That is just how the game works, not a limitation of the API.

But listen, I've presented my findings. I've drawn conclusions and I've spelled out all logic and interpretation of my findings which have nothing to do with solar panels, just time and brightness levels. I don't know why you keep coming back to solar panels. Their average/effective output are only influenced indirectly.

So I'm going to let this rest, or maybe create another thread to get other people in on this.

[Qon]

I was not testing solar panel capacity.
I was just measuring the brightness on different times of the day.

But listen, I've presented my findings. I've drawn conclusions and I've spelled out all logic and interpretation of my findings which have nothing to do with solar panels, just time and brightness levels. I don't know why you keep coming back to solar panels. Their average/effective output are only influenced indirectly.

Liar.

And yes, average solar panel output really is different. This is because the dusk and dawn phases are shorter.

I also did the math to recalculate the new ratio using the info in the first post and it is now:
n / m = [(0.5 + 1/6) * [1/6 + 1/6*((0.5+1/6) / 1.0] / 1.0] * (25 000 / 60) * (60 / 5000) = 0.92592592592592592592592592592593

[Jupiter]

I was not testing solar panel capacity.
I was just measuring the brightness on different times of the day.

But listen, I've presented my findings. I've drawn conclusions and I've spelled out all logic and interpretation of my findings which have nothing to do with solar panels, just time and brightness levels. I don't know why you keep coming back to solar panels. Their average/effective output are only influenced indirectly.

Liar.

And yes, average solar panel output really is different. This is because the dusk and dawn phases are shorter.

I also did the math to recalculate the new ratio using the info in the first post and it is now:
n / m = [(0.5 + 1/6) * [1/6 + 1/6*((0.5+1/6) / 1.0] / 1.0] * (25 000 / 60) * (60 / 5000) = 0.92592592592592592592592592592593

Almost all of what you said so far in response to my posts show that you either did not read or misread my posts, including this case.

[Qon]

Almost all of what you said so far in response to my posts show that you either did not read or misread my posts, including this case.

Nice save. Everyone will now believe you instead of me. Clever.

I think people who read your posts can draw their own conclusions and don't really need me or you to tell them what interpretation is the correct one. Anyone can read and re-read your posts as much as they want until you edit or delete them. And if you do then they will be able to see that too and it just might look a little bit suspicious. I'm done, bye.

[Jupiter]

I fully concede on all points.

This is the screenshot that convinced me:


More details at viewtopic.php?f=5&t=28049&p=177823#p177823

[ziron999]

original ratio guess = 0.84

float numOfSolarPanels = 1;
const float powerOfSolarPanels = 60; //vanilla value
float numOfAccumlators = 100; //calculate per 100 accumlators
const float powerOfAccumlators = 5000; //vanilla value
const float ratioGuess = 0.85;

if ((numOfSolarPanels * powerOfSolarPanels) / (numOfAccumlators * powerOfAccumlators) == ratioGuess)
{
optimal
}
else if ((numOfSolarPanels * powerOfSolarPanels) / (numOfAccumlators * powerOfAccumlators) > ratioGuess)
{
numOfSolarPanels += 0.1f;
numOfAccumlators -= 0.1f;
}
else if ((numOfSolarPanels * powerOfSolarPanels) / (numOfAccumlators * powerOfAccumlators) < ratioGuess)
{
numOfSolarPanels -= 0.1f;
numOfAccumlators += 0.1f;
}
else
{
wtf?!?!?!?!?! something didn't calculate right...
}

original solution:
70 solar to every 100 accumulators
0.85 solution:
70.9(71) to every 100 accumulators is 0.8508 pretty close...honestly it's so minor that 7 to every 10 should be fine even if 0.85 is right.

this is quick programming there is problems...if it can't find the the exact value it will keep looping between the else if's but oh well it works for my purpose.