Discussion What is the most efficient way to crowdsource the 3D system coordinates

RedWizzard

RedWizzard
Harbinger said:
I then test all 126 unique permutations (no repetition) of 4 star groups which return x,y,z coordinates which will fit in a 1/32 grid. I then run a count of the number of matching coordinates on each axis. The coordinate with the highest count is the one selected.

Typically there are 120+ matching positions for each axis (many times the full 126). I'd like to think a 95%+ agreement on position on each axis is pretty good result.
Click to expand...

Are you counting on each axis independently? If so, then I think you could end up choosing coordinates that were never generated by any set of references. E.g if your generated coordinates were this:
X Y Z
1 1 1
1 1 1
1 1 1
1 3 2
2 2 2
2 2 3
2 2 3
3 2 3
3 3 3
Would your algorithm pick [1, 2, 3] for the final coordinates? Or would it pick [1, 1, 1]? If it's [1, 2, 3] that would seem counter-intuitive to me, though I doubt it would happen with real data.

I'm doing something similar. I also test all the combinations of the reference stars, though I only select groups of 3 stars. Each 3 star combination gives me 2 candidate coordinates. I then pick the single set of coordinates from all the candidates that gives the lowest error when comparing the supplied distances and the calculated distances.
 
Last edited:

TunaMage

TunaMage
Biteketkergetek said:
coordinate consumers
Click to expand...
Lol - I like that phrase :)

As and when the whole galaxy opens up, I think it will be impossible for anyone or any group to try and get comprehensive data for any seriously large amount of stars.

I think what is likely to happen is there will be 'bubbles' of space (probably similar size to our current 'pill') mapped out by players or groups of players, around points of interest in the galaxy. That, and maybe long narrow corridors, as I envision the FGE guys will be doing.
 

ShadowGar

Banned
ShadowGar
TunaMage said:
Lol - I like that phrase :)

As and when the whole galaxy opens up, I think it will be impossible for anyone or any group to try and get comprehensive data for any seriously large amount of stars.

I think what is likely to happen is there will be 'bubbles' of space (probably similar size to our current 'pill') mapped out by players or groups of players, around points of interest in the galaxy. That, and maybe long narrow corridors, as I envision the FGE guys will be doing.
Click to expand...

I think once FD focuses on an API we need to all push to get certain things added to the API to allow us to map everything. Once we get that API, we could once again "scrape" all of that data and automate the entry process into whatever programs at the time need the data.
 

graham.reeds

G
The trade sites need to ask players if they would give them the distance from their location to another with the option of "No, just take me to the damn profit!"
 

Biteketkergetek

Biteketkergetek
TunaMage said:
Lol - I like that phrase :)

I think what is likely to happen is there will be 'bubbles' of space (probably similar size to our current 'pill') mapped out by players or groups of players, around points of interest in the galaxy. That, and maybe long narrow corridors, as I envision the FGE guys will be doing.
Click to expand...

Well, thats what we are really from TD's point of view at least :p But people could also improve it without worrying of breaking TD itself. The SQL schema is generic enough, the only problem I see for the future is that it does allow two stations to have the same name yet. Surely that will be fixed, if this will be more common (looking at the SQL I saw one such instance for the bubble).

Interestingly, the idea of bubbles and corridors are exactly what I used for unit tests for my routing algorithm. I wanted to see if it works on thousands of stars, but in Beta 1 there were only some 25 or so systems.
 

TunaMage

TunaMage
Biteketkergetek said:
Interestingly, the idea of bubbles and corridors are exactly what I used for unit tests for my routing algorithm. I wanted to see if it works on thousands of stars, but in Beta 1 there were only some 25 or so systems.
Click to expand...

Yep I do the same for mine :D
What I do basically, is before the route finding algorithm even kicks off, I create a ovoid of stars around the starting and ending stars to contain the amount of stars the route finding algoritm checks. The size of the ovid 'tunnel' is selectable by the user so if a route isn't found, they can increase the search area.
 

wolverine2710

Tutorial & Guide Writer
wolverine2710
RedWizard. Did you perhaps have time to enhance your tool furhter? Would be happy to test it. And any change you and Harbinger had time to talk to each other?
 

Biteketkergetek

Biteketkergetek
TunaMage said:
Yep I do the same for mine :D
What I do basically, is before the route finding algorithm even kicks off, I create a ovoid of stars around the starting and ending stars to contain the amount of stars the route finding algoritm checks. The size of the ovid 'tunnel' is selectable by the user so if a route isn't found, they can increase the search area.
Click to expand...

Is it online somewhere? Mine is in my signature, with link to sources at the corner of the page.
 

RedWizzard

RedWizzard
wolverine2710 said:
RedWizard. Did you perhaps have time to enhance your tool furhter? Would be happy to test it. And any change you and Harbinger had time to talk to each other?
Click to expand...

I've just uploaded an update (direct link). Mainly updated the entry.html page:
  1. Reworked changing the reference systems: you can edit any of them now.
  2. Checks for systems that already have data (you get a warning but can still add the system if you want to). At the moment this only checks for systems that exist in the systems.json file, not any you've added.
  3. Commander name can optionally be entered.
I've updated the systems.json file with the systems I've confirmed (will add the one's you've done tomorrow). When entering a new system you can add distances to those derived systems but they aren't used for trilateration . It can be good data for confirmation though. Unrecognised systems (with distances) also get stored in the json output but are otherwise ignored.

I haven't caught up with Harbinger yet. I'll send him a PM tomorrow (I'm about to sleep). I suspect we're going to get most of the beta2 systems done ourselves and at this point doing it manually is not too much work. I don't know if FD are planning to expand the area in beta3 or not, we've still got plenty of time to get everything hooked up to a webservice before then anyway.
 

wolverine2710

Tutorial & Guide Writer
wolverine2710
RedWizzard said:
I suspect we're going to get most of the beta2 systems done ourselves and at this point doing it manually is not too much work. I don't know if FD are planning to expand the area in beta3 or not, we've still got plenty of time to get everything hooked up to a webservice before then anyway.
Click to expand...

If I've counted correctly we have 423 sytems atm. Means 570-423=147 left. We started with the 307 of Michael Brookes. 423-307=119. Hence left is about the same amount of what has been done so far.

Not that funny but do able. I COULD update my OP to ask ppl for help and explain that its not automized entirely and point them to the released tools. But then again during the first crowd source effort in the Slopeys BPC thread we had basically not a single tool (to verify our findings) and still we ended up with enough data.

Gonna have a look at your tools and test it. Perhaps its possible to have the systems.json in a shared google doc or so and have that updated by me or others and downloaded by volunteers so to minimize double entries. Have to think about this. On the one hand I like to have the data asap. On the other hand its good to have a good infrastructure so things are automized - just in case SB3 will have new stars. At this point the tools are being perfected so perhaps some data has to be entered again.....

As I'm just the guy who started this thread and the real HARD work is done by you and the other ACES. What do the rest of you think?
 
Last edited:

TunaMage

TunaMage
Biteketkergetek said:
Is it online somewhere? Mine is in my signature, with link to sources at the corner of the page.
Click to expand...

Nope, not yet, sorry. Its a macro driven spreadsheet with VBA doing the back end calcs. VBA is probably the worst thing to use for this due to speed issues, but it's the only programming I've taught myself so far :eek:. What it does do, though is make one focus on keeping search ranges as tight as possible. :)
 

Harbinger

Volunteer Moderator
Harbinger
RedWizzard said:
Are you counting on each axis independently? If so, then I think you could end up choosing coordinates that were never generated by any set of references. E.g if your generated coordinates were this:
X Y Z
1 1 1
1 1 1
1 1 1
1 3 2
2 2 2
2 2 3
2 2 3
3 2 3
3 3 3
Would your algorithm pick [1, 2, 3] for the final coordinates? Or would it pick [1, 1, 1]? If it's [1, 2, 3] that would seem counter-intuitive to me, though I doubt it would happen with real data.

I'm doing something similar. I also test all the combinations of the reference stars, though I only select groups of 3 stars. Each 3 star combination gives me 2 candidate coordinates. I then pick the single set of coordinates from all the candidates that gives the lowest error when comparing the supplied distances and the calculated distances.
Click to expand...

The former but as I said previously, typically there are at least 120/126 for every axis. It's impossible for the selected coordinates not to be the most frequent x,y,z grouping in this situation.
 

Harbinger

Volunteer Moderator
Harbinger
25 newly mapped systems:

  • LTT 18557 (-59.875, 19.15625, -25.28125).
  • Lawd 86 (-60.40625, 18.71875, -21.6875).
  • LHS 3749 (-69.5625, 12.46875, -22.4375).
  • LTT 18506 (-70.84375, 20.25, -18.09375).
  • G 262-26 (-66.09375, 15.8125, -11.625).
  • Hemsut (-53.875, 16.6875, -10.125).
  • LHS 494 (-50.5625, 11.46875, -6.1875).
  • Wolf 1084 (-51, 7.09375, -2.1875).
  • Ross 1069 (-50.53125, 10.625, 1.03125).
  • LFT 1487 (-55, 15.15625, 4.4375).
  • G 205-38 (-51.65625, 19.125, 11.40625).
  • BY Draconis (-48.3125, 21.1875, 7.9375).
  • LHS 3409 (-59.78125, 24.375, 8.53125).
  • LHS 446 (-44.3125, 31.625, 11.15625).
  • Laima (-46.03125, 36.8125, 14.125).
  • LP 180-17 (-56.84375, 37.71875, 12.71875).
  • ADS 10329 (-61.53125, 45.78125, 1.5).
  • Wredguia DE-N B48-1 (-65.28125, 55.375, -3.4375).
  • MS Draconis (-72.71875, 45.40625, -5.875).
  • Omega Draconis (-63.375, 39.90625, -10.125).
  • Dziban (-62.15625, 38.46875, -14.3125).
  • LHS 6354 (-65.28125, 27.34375, -16.125).
  • Wredguia DK-G C24-18 (-65.8125, 23.375, -8.53125).
  • Anemoi (-61.375, 22.25, 1.5625).
  • 36 Draconis (-65.6875, 35.375, -4.59375).

Verification

SQL for Smacker's fork of Trade Dangerous:
Code: 
INSERT INTO "System" VALUES(X,'LTT 18557',-59.875,19.15625,-25.28125,102,'2014-10-16 00:00:00');
INSERT INTO "System" VALUES(X,'Lawd 86',-60.40625,18.71875,-21.6875,102,'2014-10-16 00:00:00');
INSERT INTO "System" VALUES(X',LHS 3749',-69.5625,12.46875,-22.4375,102,'2014-10-16 00:00:00');
INSERT INTO "System" VALUES(X,'LTT 18506',-70.84375,20.25,-18.09375,102,'2014-10-17 12:36:53');
INSERT INTO "System" VALUES(X,'G 262-26',-66.09375,15.8125,-11.625,102,'2014-10-17 13:30:08');
INSERT INTO "System" VALUES(X,'Hemsut',-53.875,16.6875,-10.125,102,'2014-10-17 13:43:27');
INSERT INTO "System" VALUES(X,'LHS 494',-50.5625,11.46875,-6.1875,102,'2014-10-17 13:49:37');
INSERT INTO "System" VALUES(X,'Wolf 1084',-51,7.09375,-2.1875,102,'2014-10-17 13:55:32');
INSERT INTO "System" VALUES(X,'Ross 1069',-50.53125,10.625,1.03125,102,'2014-10-17 14:08:52');
INSERT INTO "System" VALUES(X,'LFT 1487',-55,15.15625,4.4375,102,'2014-10-17 14:15:08');
INSERT INTO "System" VALUES(X,'G 205-38',-51.65625,19.125,11.40625,102,'2014-10-17 14:22:46');
INSERT INTO "System" VALUES(X,'BY Draconis',-48.3125,21.1875,7.9375,102,'2014-10-17 14:28:58');
INSERT INTO "System" VALUES(X,'LHS 3409',-59.78125,24.375,8.53125,102,'2014-10-17 14:38:47');
INSERT INTO "System" VALUES(X,'LHS 446',-44.3125,31.625,11.15625,102,'2014-10-17 14:58:47');
INSERT INTO "System" VALUES(X,'Laima',-46.03125,36.8125,14.125,102,'2014-10-17 15:05:02');
INSERT INTO "System" VALUES(X,'LP 180-17',-56.84375,37.71875,12.71875,102,'2014-10-17 15:11:20');
INSERT INTO "System" VALUES(X,'ADS 10329',-61.53125,45.78125,1.5,102,'2014-10-17 15:23:41');
INSERT INTO "System" VALUES(X,'Wredguia DE-N B48-1',-65.28125,55.375,-3.4375,102,'2014-10-17 15:40:41');
INSERT INTO "System" VALUES(X,'MS Draconis',-72.71875,45.40625,-5.875,102,'2014-10-17 15:48:13');
INSERT INTO "System" VALUES(X,'Omega Draconis',-63.375,39.90625,-10.125,102,'2014-10-17 15:54:48');
INSERT INTO "System" VALUES(X,'Dziban',-62.15625,38.46875,-14.3125,102,'2014-10-17 16:01:59');
INSERT INTO "System" VALUES(X,'LHS 6354',-65.28125,27.34375,-16.125,102,'2014-10-17 16:34:53');
INSERT INTO "System" VALUES(X,'Wredguia DK-G C24-18',-65.8125,23.375,-8.53125,102,'2014-10-17 16:40:28');
INSERT INTO "System" VALUES(X,'Anemoi',-61.375,22.25,1.5625,102,'2014-10-17 16:46:13');
INSERT INTO "System" VALUES(X,'36 Draconis',-65.6875,35.375,-4.59375,102,'2014-10-17 17:04:25');
(The regular version of Trade Dangerous may need to remove "102," from each line, not sure if Smacker's changes to the database structure have been merged)
 

Harbinger

Volunteer Moderator
Harbinger
If I don't mess around I can do each system within around 5 minutes. Hell I can even type SAN GUARALARU without thinking about it much now. ;)
 

Smacker

Smacker
Harbinger said:
25 newly mapped systems:

  • LTT 18557 (-59.875, 19.15625, -25.28125).
  • Lawd 86 (-60.40625, 18.71875, -21.6875).
  • LHS 3749 (-69.5625, 12.46875, -22.4375).
  • LTT 18506 (-70.84375, 20.25, -18.09375).
  • G 262-26 (-66.09375, 15.8125, -11.625).
  • Hemsut (-53.875, 16.6875, -10.125).
  • LHS 494 (-50.5625, 11.46875, -6.1875).
  • Wolf 1084 (-51, 7.09375, -2.1875).
  • Ross 1069 (-50.53125, 10.625, 1.03125).
  • LFT 1487 (-55, 15.15625, 4.4375).
  • G 205-38 (-51.65625, 19.125, 11.40625).
  • BY Draconis (-48.3125, 21.1875, 7.9375).
  • LHS 3409 (-59.78125, 24.375, 8.53125).
  • LHS 446 (-44.3125, 31.625, 11.15625).
  • Laima (-46.03125, 36.8125, 14.125).
  • LP 180-17 (-56.84375, 37.71875, 12.71875).
  • ADS 10329 (-61.53125, 45.78125, 1.5).
  • Wredguia DE-N B48-1 (-65.28125, 55.375, -3.4375).
  • MS Draconis (-72.71875, 45.40625, -5.875).
  • Omega Draconis (-63.375, 39.90625, -10.125).
  • Dziban (-62.15625, 38.46875, -14.3125).
  • LHS 6354 (-65.28125, 27.34375, -16.125).
  • Wredguia DK-G C24-18 (-65.8125, 23.375, -8.53125).
  • Anemoi (-61.375, 22.25, 1.5625).
  • 36 Draconis (-65.6875, 35.375, -4.59375).

Verification

SQL for Smacker's fork of Trade Dangerous:
Code: 
INSERT INTO "System" VALUES(X,'LTT 18557',-59.875,19.15625,-25.28125,102,'2014-10-16 00:00:00');
INSERT INTO "System" VALUES(X,'Lawd 86',-60.40625,18.71875,-21.6875,102,'2014-10-16 00:00:00');
INSERT INTO "System" VALUES(X',LHS 3749',-69.5625,12.46875,-22.4375,102,'2014-10-16 00:00:00');
INSERT INTO "System" VALUES(X,'LTT 18506',-70.84375,20.25,-18.09375,102,'2014-10-17 12:36:53');
INSERT INTO "System" VALUES(X,'G 262-26',-66.09375,15.8125,-11.625,102,'2014-10-17 13:30:08');
INSERT INTO "System" VALUES(X,'Hemsut',-53.875,16.6875,-10.125,102,'2014-10-17 13:43:27');
INSERT INTO "System" VALUES(X,'LHS 494',-50.5625,11.46875,-6.1875,102,'2014-10-17 13:49:37');
INSERT INTO "System" VALUES(X,'Wolf 1084',-51,7.09375,-2.1875,102,'2014-10-17 13:55:32');
INSERT INTO "System" VALUES(X,'Ross 1069',-50.53125,10.625,1.03125,102,'2014-10-17 14:08:52');
INSERT INTO "System" VALUES(X,'LFT 1487',-55,15.15625,4.4375,102,'2014-10-17 14:15:08');
INSERT INTO "System" VALUES(X,'G 205-38',-51.65625,19.125,11.40625,102,'2014-10-17 14:22:46');
INSERT INTO "System" VALUES(X,'BY Draconis',-48.3125,21.1875,7.9375,102,'2014-10-17 14:28:58');
INSERT INTO "System" VALUES(X,'LHS 3409',-59.78125,24.375,8.53125,102,'2014-10-17 14:38:47');
INSERT INTO "System" VALUES(X,'LHS 446',-44.3125,31.625,11.15625,102,'2014-10-17 14:58:47');
INSERT INTO "System" VALUES(X,'Laima',-46.03125,36.8125,14.125,102,'2014-10-17 15:05:02');
INSERT INTO "System" VALUES(X,'LP 180-17',-56.84375,37.71875,12.71875,102,'2014-10-17 15:11:20');
INSERT INTO "System" VALUES(X,'ADS 10329',-61.53125,45.78125,1.5,102,'2014-10-17 15:23:41');
INSERT INTO "System" VALUES(X,'Wredguia DE-N B48-1',-65.28125,55.375,-3.4375,102,'2014-10-17 15:40:41');
INSERT INTO "System" VALUES(X,'MS Draconis',-72.71875,45.40625,-5.875,102,'2014-10-17 15:48:13');
INSERT INTO "System" VALUES(X,'Omega Draconis',-63.375,39.90625,-10.125,102,'2014-10-17 15:54:48');
INSERT INTO "System" VALUES(X,'Dziban',-62.15625,38.46875,-14.3125,102,'2014-10-17 16:01:59');
INSERT INTO "System" VALUES(X,'LHS 6354',-65.28125,27.34375,-16.125,102,'2014-10-17 16:34:53');
INSERT INTO "System" VALUES(X,'Wredguia DK-G C24-18',-65.8125,23.375,-8.53125,102,'2014-10-17 16:40:28');
INSERT INTO "System" VALUES(X,'Anemoi',-61.375,22.25,1.5625,102,'2014-10-17 16:46:13');
INSERT INTO "System" VALUES(X,'36 Draconis',-65.6875,35.375,-4.59375,102,'2014-10-17 17:04:25');
(The regular version of Trade Dangerous may need to remove "102," from each line, not sure if Smacker's changes to the database structure have been merged)
Click to expand...
Thanks! I have added these and they are sitting in the merge queue. Hopefully they will get added tonight.
 
You must log in or register to reply here.
Back
Top Bottom