Route Plotting in the Core: an explanation

Hate to say it, but that introduces a lot more complexity than just not doing the fuel check. If you're doing something like that, why not just add an option such that filtering the galmap also filters which stars are available for routing? There, no more unscoopables on your routes. :p

I agree that a better way of doing "fully boosted" routes would be helpful, though.
As a software developer I'm firmly of the opinion that it's my job to martyr myself, and take on certain pains in order to let the designers and other stakeholders do what they want, ideally in a way that enhances those designs and business requirements. Making routes completely avoid unscoopables would not do that. ;)
 
As a software developer I'm firmly of the opinion that it's my job to martyr myself, and take on certain pains in order to let the designers and other stakeholders do what they want, ideally in a way that enhances those designs and business requirements. Making routes completely avoid unscoopables would not do that. ;)

As a software developer I think This is a bit of exaggeration. Clients often don't even know themselves what they want until they see the software in action. Usually it's compromise of constraints and what can be done with reasonable allocation of resources in your team. Completely martyring yourself to the whims of others will only end in a negative result for both.
 
As a software developer I think This is a bit of exaggeration. Clients often don't even know themselves what they want until they see the software in action. Usually it's compromise of constraints and what can be done with reasonable allocation of resources in your team. Completely martyring yourself to the whims of others will only end in a negative result for both.
Thus the qualifying word, "certain" modifying "pains". ;) Most of the engineers that I've managed take a minimum effort route towards completion of requirements...and it's almost never sufficient.
 
why not just add an option such that filtering the galmap also filters which stars are available for routing? There, no more unscoopables on your routes. :p


You could use it for more than just scoopable filtering. You could use it in Neutron Fields to filter for "non-sequence" + "A stars". Then plot a route through the fields while hitting mostly high value BH, NS, and WD, and be able to scoop often enough to not run out of fuel. All without having to replot.

Likewise, you could use to filter for systems that are more likely to have terraformables. Actually the applications for this kind of filtering would be huge even inside the bubble. Say, if you are pirate and want to find a path only through lawless systems, etc.
 
Last edited:
Defintely explained that for me. I was looking at entering sag a* but found this topic instead. Still a great help and gave me a bit more understanding of the route planner.
 
I'd actually like to know how many jumps there are in an eco-kylie in the thick of the core?
Hah, interesting question... I'm still only a little way past WP16, I'll give it a go this evening :p
Not the middle of the core, but still potentially an informative test!

You could use it for more than just scoopable filtering. You could use it in Neutron Fields to filter for "non-sequence" + "A stars". Then plot a route through the fields while hitting mostly high value BH, NS, and WD, and be able to scoop often enough to not run out of fuel. All without having to replot.
Indeed, more power in choosing which systems we go via would be very helpful indeed.
 
Last edited:
I was wondering if the formula works the same for shorter distances. Let's say I'm 500 LY from the core and want to plot a route that's 500 or 550 at maximum. If I replaced the 1000 with 500 would it still work? Or will the results be inaccurate.
 
Last edited:
I was wondering if the formula works the same for shorter distances. Let's say I'm 500 LY from the core and want to plot a route that's 500 or 550 at maximum. If I replaced the 1000 with 500 would it still work? Or will the results be inaccurate.

No, the formula won't work as it is because the D * 2 factor is based on the 1000 LY distance and is independent of the number of jumps. If you want a route of around 500 LY, then you should halve that factor (i.e just use D) and then it should work.

In general, if you wish to plot other distances, first use the formula for 1000 LY. Then calculate the distance per jump (just divide the plot distance that it gives you by the number of jumps) and use any multiple of that.


 
No, the formula won't work as it is because the D * 2 factor is based on the 1000 LY distance and is independent of the number of jumps. If you want a route of around 500 LY, then you should halve that factor (i.e just use D) and then it should work.

In general, if you wish to plot other distances, first use the formula for 1000 LY. Then calculate the distance per jump (just divide the plot distance that it gives you by the number of jumps) and use any multiple of that.



Yeah I figured it was something like that. I want my program to have distance selection and thought that the D*2 was based on the 1000ly. Thank you so much for the advice.
 
Last edited:
ASC;3638053[FONT=arial said:
In general, if you wish to plot other distances, first use the formula for 1000 LY. Then calculate the distance per jump (just divide the plot distance that it gives you by the number of jumps) and use any multiple of that.[/FONT]


...which is basically what my spreadsheet does :)
 
I have a Jump Range of 32.12Ly
I am precisely 0Ly from Sag A* (I am IN Sag A*)

Can someone talk me through the math please? I got a magic number of 956.25Ly but have crashed my game several times attempting to plot unfortunately.

The sentence:
"If you have a "max optimal range" of 980Ly, you need to plot much lower than that on the edges of the core if you want to get a good plot."
Is an incredibly useless statement. Do I set just one jump at a time? Bit daft as it negates the need for any of the math.
Define MUCH lower???

The explanation pictures are very good. I can viusualise what's happening much better, but I could use help with the actual calculations.
 
I have a Jump Range of 32.12Ly
I am precisely 0Ly from Sag A* (I am IN Sag A*)

Can someone talk me through the math please? I got a magic number of 956.25Ly but have crashed my game several times attempting to plot unfortunately.

The sentence:
"If you have a "max optimal range" of 980Ly, you need to plot much lower than that on the edges of the core if you want to get a good plot."
Is an incredibly useless statement. Do I set just one jump at a time? Bit daft as it negates the need for any of the math.
Define MUCH lower???

The explanation pictures are very good. I can viusualise what's happening much better, but I could use help with the actual calculations.

Don't sweat, you're not the only one in this situation.

It just flat out doesn't work in some cases and there are multiple posts on different mediums saying the same thing, just as many as there are of those who say that this might work, last time I checked. I was there last a few months ago after reading this thread and some reddit posts. I got pretty decent specs too, i7, 970, plenty memory, SSD - so taxing algorithms wouldn't be that much of a problem, but any attempt got the game frozen with one of the CPU cores at 100% and the game being almost unresponsive. After multiple game processes killed and me continually decreasing the jump range, the best I could plot were a few jumps away 120ly-200ly (my jump range was 32.5ly). The process of calculating this was too slow by comparison to plotting one jump at the time and the "optimal" for the route planner changes the more you get out of Sag A. proximity.

The options are: don't get any deeper, if you posted on your way towards the centre of the galaxy or be patient in case you're already stuck deep in those last 2k ly. All in all, one by one jumps shouldn't slow you more than 3h in and another 3h out. It's not that big of a deal.

Another option would be to wait for the galaxy map's bookmark feature to be released. It would have saved me a lot of frustration back then.
 
Last edited:
Ok thanks.
Would still like someone to confirm my calculations though lol.

(also, as I said at the top. I am actually IN the Sag A* system ;-P )
 
I'm currently about 1300ly away from Sag A and i'm giving up. The thing just won't plot anymore. Could be my calculations are off, but i'm tired of moving through the stuttering map. I don't feel its worth plotting for only 2 jumps every time to get there. Want to get to Eol Prou too. I'll just start heading away from Sag A slowely till the plotting works again and then head to Eol Prou.

Its a big disappointment as there were absolutely zero problems all the way to 2000ly away from Sag A. i didn't even know there were problems with plotting in the core till today. Left from the bubble a month ago and my original intentions were not even reaching Sag A so its not a total loss(goal was 16000ly). Oh well, maybe in the future.

Probably going to take a month to reach Eol Prou and with my luck the station is already gone by the time i get there... Maybe i should just head back to the bubble and hand in my data to get elite status, though i'm afraid its not enough yet.
 
I'm currently about 1300ly away from Sag A and i'm giving up. The thing just won't plot anymore. Could be my calculations are off, but i'm tired of moving through the stuttering map. I don't feel its worth plotting for only 2 jumps every time to get there. Want to get to Eol Prou too. I'll just start heading away from Sag A slowely till the plotting works again and then head to Eol Prou.

Its a big disappointment as there were absolutely zero problems all the way to 2000ly away from Sag A. i didn't even know there were problems with plotting in the core till today. Left from the bubble a month ago and my original intentions were not even reaching Sag A so its not a total loss(goal was 16000ly). Oh well, maybe in the future.

Probably going to take a month to reach Eol Prou and with my luck the station is already gone by the time i get there... Maybe i should just head back to the bubble and hand in my data to get elite status, though i'm afraid its not enough yet.

Hi Cmdr, don't give up, try the Explorer Calculator app here. I've just travelled from north side of the core, through Sag A, and am now exiting (well, have left) the southern edge of the core. Using the app, I managed to get 987 - 993Ly routes plotted for my 37Ly Exploraconda, the Quo Vadis.

If you don't want to use the app, fire up your favourite spreadsheet tool and use the formula provided in the first post on this thread and let us know how you get on..

Both methods work, honest!!
 
Hi Cmdr, don't give up, try the Explorer Calculator app here. I've just travelled from north side of the core, through Sag A, and am now exiting (well, have left) the southern edge of the core. Using the app, I managed to get 987 - 993Ly routes plotted for my 37Ly Exploraconda, the Quo Vadis.

If you don't want to use the app, fire up your favourite spreadsheet tool and use the formula provided in the first post on this thread and let us know how you get on..

Both methods work, honest!!

Sorry, but no. Neither method work for me. I've tried for a half hour again with multiple different systems in the right range. Nothing. It get very nervous and frustrated from such a slow and stuttering image so i rage quit and next time, i'm going back.

Ca't even play right now, so frustrated am i.

Its a shame the devs haven't fixed the problem properly yet.
 
I have a feeling I'm doing this wrong. =/ According to the math, at a jump range of 39LY, at 11k from the core, my optimum plot is around 947ish. I still can't get it to plot quickly. I also have a question as to the plotting of shorter routes via this method? If I want to do a short hop to check out something neat looking, it seems to bog as badly as it does for a long route. I could use a hand to hold I think :(
 
Back
Top Bottom