Route Plotting in the Core: an explanation
- Thread starter Esvandiary
- Start date
I don't know what happened in the meantime since this thread was started, because this is my first trip to the core of the galaxy, but doing short jumps (an economic route) causes a nasty freeze of the game client.
That calculation formula doesn't seem to apply at all in my case or for other people that I've talked to. With fast jumps (range ~33ly), I can't plot any course of 2 jumps or more without the game lagging (it's better than a complete freeze, at least I can exit to main menu and start over). And I don't know where the problem started exactly. I was jumping 500ly per play session before reaching 1000ly from the core and then I ran into this behaviour, it wasn't a gradual degradation, from plotting 500ly, it went down to having to plot only a single jump. Now, if it would depend that much on the client machine, then I'd expect to see some CPU starvation. Yet in my case, the CPU tops at 20% usage with none of the cores being really strained. And the game is almost unresponsive. There are multiple issues that impact this behaviour and it would be nice to see some feedback from the developers.
That calculation formula doesn't seem to apply at all in my case or for other people that I've talked to. With fast jumps (range ~33ly), I can't plot any course of 2 jumps or more without the game lagging (it's better than a complete freeze, at least I can exit to main menu and start over). And I don't know where the problem started exactly. I was jumping 500ly per play session before reaching 1000ly from the core and then I ran into this behaviour, it wasn't a gradual degradation, from plotting 500ly, it went down to having to plot only a single jump. Now, if it would depend that much on the client machine, then I'd expect to see some CPU starvation. Yet in my case, the CPU tops at 20% usage with none of the cores being really strained. And the game is almost unresponsive. There are multiple issues that impact this behaviour and it would be nice to see some feedback from the developers.
You shouldn't take the number that comes out from the formula as godlike, it's an indication. Try to play with it. There's a magic number that suits your jumprange if you coreward or in the other direction. My jumprange is 31.07 and my magic number going coreward was 982 ly range, getting away from the core i had to raise it a little bit.
For sure it's not a matter of distance. If you don't strike the curret number it can crash calculating 2 jumps like 30.
.
I'm still using part of the formula to set plot far away from the core, i notice that uses all full potential of my jumprange. If i go random the last 5, 6 jumps i use less jumprange, like 26. So i was loosing about 35 ly every 1000.
.
BTW when elite get struck calculating plots in the core it's not your pc end fault. I'm not a programmer but i guess my processor could calculate all the alternatives in a split second if the coding was good.
.
As a final note i hope they'll fix this, becouse when the route plotter get stuck, it gets really well and closing the gal map is useless, you have to go log out and back in, wich is not exactly fun.
You can still plot multiple jumps across hundred of light years. In my case that range is limitted to one jump and ~33ly. 950ly >>> 33ly. No? Basically what I'm trying to say is that formula doesn't really explain the problems that these mechanic has at the moment.
Yeah, this was one thing they made worse in 1.3 - previously you could just cancel out of the galaxy map (slooooowwwlllyyy) and go back in to do a fresh attempt. Now it keeps lagging even when you exit the galaxy map.
Trying to plot any multiple-jump route will cause the problem, since it'll try to calculate the most optimal set of jumps whether it's two jumps or twenty. If you plot a route of e.g. 65Ly whilst in the core, that would probably work reasonably well, for instance - since that'd be two max-range jumps and it wouldn't have very many options to try.
And as for it being able to calculate it within a split second, I doubt it - the density of stars in the core means that unless the jumps are close to the max range, it has millions upon millions of options to try. That takes time.
Last edited:
+1 Thanks for this. I am a long way out at the moment but I am sure this will be a big help.
Last edited:
I just logged into the 1.4 stress test build for the first time and found myself in an Anaconda 2000K above the core. So I tried to plot a route straight down and... the route plotter still sucks.
Plotting the first 1000K wasn't great, but it only reduced my frame rate to 15 FPS - the star density isn't so great up there and it was still possible to try different length routes until one plotted. And then I tried to plot down to Sag A*. The frame rate was reduced to 1 FPS and the game was virtually unusable.
With patience you can try slowly moving the map and replotting. If you get the right distance it does plot instantly but if you get it wrong it seems just as bad as 1.3.
Plotting the first 1000K wasn't great, but it only reduced my frame rate to 15 FPS - the star density isn't so great up there and it was still possible to try different length routes until one plotted. And then I tried to plot down to Sag A*. The frame rate was reduced to 1 FPS and the game was virtually unusable.
With patience you can try slowly moving the map and replotting. If you get the right distance it does plot instantly but if you get it wrong it seems just as bad as 1.3.
Great to see this level of detail, thanks.
For me, the route plotter also fails badly in very sparse areas - I ended up getting stuck for days in the "basement" as I couldn't plot a route back up to the bubble. I could see nearby stars (~50Ly) that I'd recently visited, but the plotter couldn't find any route back. I moved a few systems off in another direction, and suddenly the route opened up again (nothing to do with fuel/jump range).
For me, the route plotter also fails badly in very sparse areas - I ended up getting stuck for days in the "basement" as I couldn't plot a route back up to the bubble. I could see nearby stars (~50Ly) that I'd recently visited, but the plotter couldn't find any route back. I moved a few systems off in another direction, and suddenly the route opened up again (nothing to do with fuel/jump range).
The uploader doesn't want to accept my XLS File...
I'm about to try my run to Sag A*, I've written a little excell worksheet to do the numbers for me. if the formula works, I'll write a small program (Windows/Wine Only I'm afraid, I only know BlitzPlus!) that doesn't take up as much real estate as Excel.
https://dl.dropboxusercontent.com/u/79150668/Sweet Jump Calculator.xls
I'm about to try my run to Sag A*, I've written a little excell worksheet to do the numbers for me. if the formula works, I'll write a small program (Windows/Wine Only I'm afraid, I only know BlitzPlus!) that doesn't take up as much real estate as Excel.
https://dl.dropboxusercontent.com/u/79150668/Sweet Jump Calculator.xls
Amazing! Tried this after several unsuccessful plotting attempts when departing SagA. Formula resulted in 978.75 - et voilà - plot comleted almost instantly. Thanks for figuring this out and sharing it! I'll just assume there was no black magic involved.
OP used some balck magic indeed for the formula.
This was mine too. The OP is my new favorite person. I love exploring around the core, but getting there is always rough. Now its instant. Cant give enough reps to you!!!
Yup.
I just raised a bug report on this, with screenshot evidence...
Thread: Galaxy Map "Route Planner" does not use full ship jump range
I actually got even better evidence from a jump plotted shortly after posting that... A route that went...
A->B->C->D->E
...in "Fastest" plot mode.
At "A", I was able to manually select system "C", and jump directly to it. My plotted route stayed intact in the galaxy map, as "C->D->E".
I was then able to manually select system "E", and jump directly to that.
A 4-jump route... able to be completed in 2 jumps by bypassing the route plotter jumps and selecting perfectly valid jumps by hand.
Reckon it's a 1.4 issue.
Very helpful thank you!
I'm on my first long run, heading to the core and who knows where to next. I'm currently 12Kly away from SgrA*, and so far can't complain about the route planner. I usually try to go as close to 1000ly plots as I can and it only takes a few seconds to calculate.
However, I'm starting to wonder if the problem with plotting does happen, would it help if I filtered out some of the less interesting star types on the GalMap?
This thread has brought to my attention that I still don't know if the route planner will use the filtered star types or not. Will have to check when I get home.
But still, 12Kly out from GC, roughly 300Ly below the plane and plotting like a boss. I hope it doesn't get too bad as I move closer...
However, I'm starting to wonder if the problem with plotting does happen, would it help if I filtered out some of the less interesting star types on the GalMap?
This thread has brought to my attention that I still don't know if the route planner will use the filtered star types or not. Will have to check when I get home.
But still, 12Kly out from GC, roughly 300Ly below the plane and plotting like a boss. I hope it doesn't get too bad as I move closer...
Even if you're not close to the core you can test out this assumption. If you filter out the stars and the route plotter selects stars to jump to that were filtered out, then, no, it won't...