Long Range Navigation

The new route planner is great, it really makes the job of navigating much easier. However, it's still rather inconvenient to plot routes over long distances.

What would make this MUCH MUCH easier, would be the ability to select a star and have the galactic map show a line drawn from that star to the current ship position.

Nothing else needed beyond that, no changes to the route planner, just draw a simple line from a specified star to where I am.
 
The new route planner is great, it really makes the job of navigating much easier. However, it's still rather inconvenient to plot routes over long distances.

What would make this MUCH MUCH easier, would be the ability to select a star and have the galactic map show a line drawn from that star to the current ship position.

Nothing else needed beyond that, no changes to the route planner, just draw a simple line from a specified star to where I am.

You can pretty much do that yourself. Not exactly, but works for me: 5mins 57seconds into this video http://youtu.be/EeL9fJSmYMQ
 
That's similar to what I do now - except that the distances I'm navigating are not 160 LY, but 7,000 LY.

I need to find point after point after point on a route towards my destination. I'm practised at it, I'm even pretty good at it. But just drawing a line would make the task so much more accurate.
 

Deleted member 38366

D
-- Deleted --
 
Last edited by a moderator:
What I dont understand is why you have to jump star to star anyways? why cant i just keep re-injecting fuel into my HD until i need to fuel scoop or until I arrive. *Cough* Future Update *Cough*
 
I always wondered by the Galaxy Map takes such a long time to calculate the route (?)

Since it's a one-way route pathfinding, it shouldn't take a modern CPU longer than a few seconds even for quite long Ranges to a known destination (calculations should be prioritized towards it anyway, instead of building that route cluster in a spherical manner)...

I believe it's because it's calculating the routes ahead of a destination being selected. As part of the spiderweb generation.

I'm pretty sure we agree with each other that doing it that way is not what we believe is the optimal solution. I'd far rather see that spiderweb being drawn in a process that's independant of route planning, and have a separate task of being able to select a destination and calculate a route to it.

I'm fine with a distance limitation for gameplay reasons, or for technical ones. Like I say, what we have is much better than what we had. But for long distance journeys (multi-thousand LY) it can be hard to judge how much to ascend or descend relative to the galactic plane each leg. At 40-60LY per leg (I don't generally wait for the full range to be available), we're talking over 100 legs for this trip..... each way
 
I always wondered why the Galaxy Map takes such a long time to calculate the route (?)

Since it's a one-way route pathfinding, it shouldn't take a modern CPU longer than a few seconds even for quite long Ranges to a known destination (calculations should be prioritized towards it anyway, instead of building that route cluster in a spherical manner)...

If this is such a big issue, precalculated route tables should be stored on the Client, then you could basically use them instantly with only their optimization (for your precise range) running in the meantime - and being cached locally as well within a certain radius.
At least that's how I would do it...

It's because it's not actually calculating the route you want. It currently calculates all routes using the spiderweb system. Your route only comes available to plot once it's one jump from a star that's currently part of the web.

Hopefully this will be expanded on to allow us to select any star and plot a route, cancelling the spider web and instead logically selecting only stars that are closer to the destination.

It's possible, no doubt. Just a matter of it being deemed high enough of a priority.
 
am i right in thinking my CPU does not max out when it's plotting all the jumps? if so, i've got 8 cores of goodness, and i don't mind them taking a hammering if it means i can get that 100Ly route plotted in seconds instead of minutes!

thanks for the link to the co-pilot. i might give that a go and see how it compares. looks like it's a trade off between start time and in-flight time (game based one ebing nice and quick as you're target comes pre-selected).
 
When planning longer trips (which I do regularly, as I have multiple Hubs ~200 LY apart), I just enter the destination and look at my current position marker from there, then move the selector towards it until I can plot a route. For the more regular destinations, I know the general direction, so I just pick a system that roughly points there.

It's funny - the VR forums are constantly filled with threads about how the galaxy map doesn't work in the rift, whereas I think it's one of the greatest things to use in VR. Never used the map on a screen, but being in the map definitely enhances my sense of orientation.
 
The same way they interact now??? All those billions of star systems, its not like they are booming with players at every one anyways. Besides, those who would want to interact can hang around the feds systems.
 
When planning longer trips (which I do regularly, as I have multiple Hubs ~200 LY apart), I just enter the destination and look at my current position marker from there, then move the selector towards it until I can plot a route. For the more regular destinations, I know the general direction, so I just pick a system that roughly points there.

It's funny - the VR forums are constantly filled with threads about how the galaxy map doesn't work in the rift, whereas I think it's one of the greatest things to use in VR. Never used the map on a screen, but being in the map definitely enhances my sense of orientation.

It's the same process for multi-thousand lightyear trips too. But it's a lot harder to judge the vertical progress you should make each time. That line drawn from eventual destination to current location would make that a pleasure rather than a chore.

Certainly the tools we have now permit the job of navigation to be done. And the 3rd party sites like co-pilot work extremely well for known systems and shorter jump ranges (any time you have a jump radius greater than the left hand panel will show, navigation star to star becomes a real chore).
 
Check out Elite Copilot as well - very handy for plotting long journeys.
It is pretty useless for long distance travel, since it can plot route only through already explored systems. So if you are in some backwater system, you are out of luck. And if you are 500 Ly from civilization, you are out of luck too.

It is good tool, but very limited in use. Frontier should just allow people to tap into the star map completely.
 
Last edited:
am i right in thinking my CPU does not max out when it's plotting all the jumps? if so, i've got 8 cores of goodness, and i don't mind them taking a hammering if it means i can get that 100Ly route plotted in seconds instead of minutes!

Unfortunately having 8 cores does nothing to help unless the processing is divided into multiple threads. I rather suspect the spiderweb process that the nav planner depends on is running in a single thread, and thus using only one of the cores.

I have this dream that in the future I'll be able to set a destination 25k light years away, and have the system navigate me towards it, calculating the next segment of the route each it needs to as it goes.
 
I suspect and hope that the current "route planner" will be expanded on and improved. All I can say for the current system is that it's better than nothing.
 
Unfortunately having 8 cores does nothing to help unless the processing is divided into multiple threads. I rather suspect the spiderweb process that the nav planner depends on is running in a single thread, and thus using only one of the cores.

I have this dream that in the future I'll be able to set a destination 25k light years away, and have the system navigate me towards it, calculating the next segment of the route each it needs to as it goes.
i get the feeling the spider web ball would be a great candidate for parallel processing. so i'm thinking it could be split over all CPU cores, or maybe even go the CUDA route and use the hundreds of GPU cores many desktop gaming PC's have these days!

even if the current code is not good for breaking down like that i bet there are other algorithms out there which are perfect for the job. it can be a real joy finding a different way of doing things that makes your code run 10 times faster.

still remember when i found out about binary search. take a big ordered list of things and try to find the one you want. the first idea a young programmer will have is to loop through each one in turn and check if it's the right one. so on average if you have 1 million items to search through it will take 500K checks to find the one you want.

compare that to a binary search which will only take 20 searches to find the one you want out of 1 million!

the route planner is a great first step, and i'm so glad they got something implemented before launch. but it does feel like there is a lot of optimisation to be done
 
The new route planner is great, it really makes the job of navigating much easier. However, it's still rather inconvenient to plot routes over long distances.

What would make this MUCH MUCH easier, would be the ability to select a star and have the galactic map show a line drawn from that star to the current ship position.

Nothing else needed beyond that, no changes to the route planner, just draw a simple line from a specified star to where I am.

Agreed +1
 
Back
Top Bottom