Plotting your own jumps

[Peter Powers]

What myth? The route planner is slow because it calculates every possible route (see below). It probably uses A* or a derivative as the other route planners do, but that's not the problem. The others only calculate the route that really interests you, and that's what is important.

I don't want this spaghetti, I only want 1 route to my target.

hum? what you have there on your image shows all *jumps* from that system, not "every possible route", thats completely unrelated.

 

[saviornt]

Actually, the system is calculating all jumps from all systems, that's why it is so abnormally slow. That's why I'm thinking it's using the Dijkstra's algorithm. With A*, you could use waypoints, and it would calculate the shortest path possible to hit all the waypoints.

edit: of course we won't know absolutely what algorithm that they are using unless they actually tell us.

 

[Andargor]

hum? what you have there on your image shows all *jumps* from that system, not "every possible route", thats completely unrelated.

You are arguing semantics. The route planner calculates the route for all possible destinations for the first jump, then for the second jump using the algorithm, and so on until it reaches 100 LY. The point is that it's useless, the only important thing is the one route to where you want to go.

If you enjoy it like it is, more power to you.

 

[steevaavoo]

You are arguing semantics. The route planner calculates the route for all possible destinations for the first jump, then for the second jump using the algorithm, and so on until it reaches 100 LY. The point is that it's useless, the only important thing is the one route to where you want to go.

If you enjoy it like it is, more power to you.

Speaking of semantics and the actual point - I would contend the assertion that it's useless. I personally find it pretty useful. Though I certainly agree that it would benefit from calculating the route one is asking for on the fly based on selected destination.

Then it would be very useful. Fortunately nothing like this is set in stone and over time will be improved.

@Andargor - Love your signature text by the way. Have you come up with an Elite-related version of the Litany Against Fear?

 

[Datalink]

My computer's ability to use the map is somewhat limited, I will pick a destination, then start writing down the route from that destination as soon as it's path is resolved, if I'm lucky, I'm within 30 LY of my start point before my computer overheats and shuts down (it's a 2-in-1 Tablet with a 4 core Atom CPU, the fact it runs ED at all is miraculous). Then in another 10 minutes of having the computer off... usually not my choice, I will finish the route, then after another 10 minutes, I'll start flying toward my destination with my route on a weaker computer's text file.

My format tends to look a bit like this:
...
34.76LY LP 915-27
41.49LY AO
48.25LY LHS 2891
...
I won't bother updating distances listed but the distance helps me plan my routes better, as I will know when to take breaks to drop off scan data.
If I'm doing pure exploration (I am an explorer) it will also tell me when to dock to prevent excess backtracking... I have lost system data to players interdicting me because they couldn't be bothered to scan me for cargo modules. I've since replaced the hauler with a more combat oriented ship, which should help. I'll know for sure when I reach my next destination, and pay an alliance fine for a paint swap incident kicking me back into the docking bay.

There's been moments I've considered going outside given we're in winter at my area and just playing ED in the cold... the advantage would be the computer would benefit from the improved cooling, the downside is I would be in the same cooling environment and wouldn't benefit as much.

I wonder if the display of the available routes might contribute to the lag. I do know the jump range calculations appear to be based partially on distance, and partially on prior routes not intersecting a star. I wouldn't know any more than anyone else on the topic unless I could somehow see the formula and functional logic of the sorting routine. I'm kinda with Saviornt, in that it feels like a Dijkstra style sort.

I'm sure in future updates the mapping system will be improved, for now... it seems to be a better off switch for my computer than the power button.

 

[Peter Powers]

Actually, the system is calculating all jumps from all systems, that's why it is so abnormally slow. That's why I'm thinking it's using the Dijkstra's algorithm. With A*, you could use waypoints, and it would calculate the shortest path possible to hit all the waypoints.

edit: of course we won't know absolutely what algorithm that they are using unless they actually tell us.

the simple fact that you can chose a heuristic (most economic / fastest route), kinda gives away that its not dijkstra's.

 

[Peter Powers]

You are arguing semantics.

no i don't, there is quite a difference.

The route planner calculates the route for all possible destinations for the first jump, then for the second jump using the algorithm, and so on until it reaches 100 LY.

again: no thats not what it is doing, what you showed in the picture is finding reachable systems by distance - basically data that you require for the route finding, and yes - thats showing more data than the actual route-finding requires, as at that time you haven't selected a destination yet.
Now you could argue if a starmap should display a starmap, or if it should be able to show possible jumps,
but thats not route planning - its not what the websites do, and its not compareable. And no, its not the same thing as "plotting all possible routes".

the actual calculation on a route to take is what happens when you click the "plot route" button.

Here, have a video that shows what i mean (once processing is done..):
[video=youtube;BquJ5DM-T_o]https://www.youtube.com/watch?v=BquJ5DM-T_o[/video]

Now the funny part is: the display of possible jumps gets slower the higher your jump range (because more distances need to be calculated), while the actual route finding should get faster (because less systems are on the relevant tree).

 

[Mikey2305]

@datalink... You can play ED with an atom processor. That is absolutely brilliant. Perhaps you should upgrade to D class modules. And not get too near the sun... Should be ok today


Have been using the cmdr prog this morning and am very impressed. I think it will live alongside my ED window from now on. Hopefully the ingame one will be refined and improved over time..

 

[Datalink]

@Mikey2305, believe it or not dropping out of hyperspace is a point I don't have as much problem with, it either slideshows through hyperspace or stops... then drops me at the star, I scoop fuel and I go on to my station, my scanning or my next waypoint, depending on my goals at any given time.

However, map view will actually seem to freeze, if I let it resolve out to the full 100LY, there's a good chance I'll crash. Then again, a lot of stuff crashes me... or my computer overheats... my ship... usually only hits 110% heat if I'm really trying to scoop, when I can, I do fit bigger fuel scoops (my current ship has a class 3 scoop normally) this means I'm usually on my way with just a short skim of the corona...

I do appreciate the program crash recovery, it's actually saved me from a white dwarf once or twice, but I really would prefer not to have crashes or thermal shutdowns in the first place.

I really need to find a way to get a better computer.

 

[Tubit]

allow us to set manual waypoints.

I seem to remember that I've set manual routes before by clicking the "plot course" for each individual jump. I don't think it deletes the previous route you've set. It just adds to the existing one.