Are you pulling data from EDSC now? Or should I send my new distances/coordinates to you? All of my data is submitted to EDSC now. Preferably that's the only place I have to submit it now
Discussion What is the most efficient way to crowdsource the 3D system coordinates
- Thread starter wolverine2710
- Start date
Are you pulling data from EDSC now? Or should I send my new distances/coordinates to you? All of my data is submitted to EDSC now. Preferably that's the only place I have to submit it now
Found another system where the nav panel and galaxy map show different values. LHS 266. The nearby Yin system is off by 0.01. Ross 89 is off by 0.01. Will have to stop testing for today, but tomorrow I will check if those values change when I get to the stations in that system (one of the 150000 book reading cruises).
Are you pulling data from EDSC now? Or should I send my new distances/coordinates to you? All of my data is submitted to EDSC now. Preferably that's the only place I have to submit it now
Keep sending them to TGC. I'll sync with that once I've written the tools to do so in the next few days.
- - - - - Additional Content Posted / Auto Merge - - - - -
Just used it on my internal stuff and seems OK. Thanks for your hard work
EDIT: And alt codes don,t seem to work in the galmap (i.e. ALT + 0225)
Yep, I've ticketed that. Probably be good if you ticket it too as it's likely to get ignored unless a lot of people mention it.
Last edited:
I used a homegrown A* algorithm in my little excel version that I use for myself, which works nicely. A key step, though, was to limit the number of systems in the search to a reasonable bubble around the start and destination systems before starting the algorithm. Otherwise time issues start mounting up, as excel and VBA are not the fastest for this sort of thing.
I have a version using cpan.perl.org's AI:
athfinding::AStar and whenever it asks for neighbours I simply supply those within the defined jump range. Then obviously distance to them from current position is the 'cost' and their distance to the destination is the 'heuristic'. For now I have a postgresql DB with the data in locally that I just do queries on against the x,y,z co-ordinate columns (which IIRC I didn't even put indexes on yet) for that 'neighbours' search.I have a version using cpan.perl.org's AI:
Interesting approach
I did some racing in B3 on the Buckeyball run in the old B3 pill, and if you are racing and need to refuel somewhere along the line, then defining the heuristics starts becoming a bit trickier. This is because in the race, the time is the main factor, not necessarily the LY distance, which is the obvious heuristic. So then one starts trying to equate jump times, time to a base from the star to refuel, etc, etc to LY per jump to try and get the right algorithm.
It needs to figure out for example, do I jump A,B,C, Refuel at C, then jump to D, and finish at E. Or is it better to jump A, B, F, refuel at F, and finish back at E. The second option has less jumps and possibly less distance, but if the closest base at F is much further than at C, then first option might be better, time wise.
I never got round to building that in properly, buts its a nice challenge.
Last edited:
Indeed, my current implementation is simply "cover the distance in least jumps". To do a proper fuel/time thing you'd need the full characteristics of your FSD and Fuel Scoop, and some estimate of the time spent per star when fuelling or not. I think we're veering off-topic here nowInteresting approach
I did some racing in B3 on the Buckeyball run in the old B3 pill, and if you are racing and need to refuel somewhere along the line, then defining the heuristics starts becoming a bit trickier. This is because in the race, the time is the main factor, not necessarily the LY distance, which is the obvious heuristic. So then one starts trying to equate jump times, time to a base from the star to refuel, etc, etc to LY per jump to try and get the right algorithm.
It needs to figure out for example, do I jump A,B,C, Refuel at C, then jump to D, and finish at E. Or is it better to jump A, B, F, refuel at F, and finish back at E. The second option has less jumps and possibly less distance, but if the closest base at F is much further than at C, then first option might be better, time wise.
I never got round to building that in properly, buts its a nice challenge.
.I think we're veering off-topic here now
We are indeed
- possibly another thread on route finding methodology is appropriate, if such a thread doesn't already exist...Indeed, my current implementation is simply "cover the distance in least jumps". To do a proper fuel/time thing you'd need the full characteristics of your FSD and Fuel Scoop, and some estimate of the time spent per star when fuelling or not. I think we're veering off-topic here now
Can also do a poor mans approach and just square all the weights. This should give a route with more but shorter jumps and a more fuel efficient route in ED.
Maybe someone can resurrect the thread where we where given the fuel cost formula.
@TornSoul: I noticed that if I resubmit a system to TGC I get status entries for all the cr increments for the distances but I don't get a system status entry for the cr increment for the system itself. Is that expected or is it a bug?
If you resubmit *exactly* the same information twice then:
- Distances will have cr incremented - they where afteral confirmed again
- System cr will not be incremented, as TGC detects that there is no new information to do trilateration with, and thus don't. Therefor system cr is not incremented (only happens on successful trilateration)
So yes that's expected behavior.
- Distances will have cr incremented - they where afteral confirmed again
- System cr will not be incremented, as TGC detects that there is no new information to do trilateration with, and thus don't. Therefor system cr is not incremented (only happens on successful trilateration)
So yes that's expected behavior.
Last edited:
Done. /10char
A small update, I've fully implemented database search on my app now (route finder only for now, more inc). If you need to check if a system exists in the database you can do there rather fast (is synched via AJAX so no need to submit), my database is updated daily from EDSC. (reminder on the link: http://www.elitedangerouscentral.com/)
It needs to figure out for example, do I jump A,B,C, Refuel at C, then jump to D, and finish at E. Or is it better to jump A, B, F, refuel at F, and finish back at E. The second option has less jumps and possibly less distance, but if the closest base at F is much further than at C, then first option might be better, time wise.
I never got round to building that in properly, buts its a nice challenge.
This is what my current algorithm behind the map (apart from the usual octree based A*) does: keep track of how much fuel is available for the live A* nodes, along with the maximum fuel possible at the same node. If fuel runs out, replace it with the maximum possible fuel value. This method produces valid routes when other tools possibly can't: it generates the necessary detours to take fuel. In some sense my routes were worse, though: they were generally longer, but only when detours were actually necessary.
Fuel use can be calculated according to the FB Tuz's spreadsheet. Basically 3 values are necessary for precise fuel calculation: all you need is the jump range, FSD max jump fuel and fuel power.
Ok, that makes sense.
Is "Destination" a real star in the game?
I noticed two stars near Dahan in the beta's star list (B2?): Destination and Training
I got the impression these were on the list from the FD dump and represented stars only accessible in the tutorials.
Has anyone confirmed these are in the game?
[edit]
That being said, "Training" seems to have been culled, which is good:
Last edited:
They're not visitable but they do show up in trade data and can be searched in the galaxy map. I've ticketed both these issues. As well as those two there are more: Test, Test2, Test3, Singlelighttest (IIRC).
What trade data is that? ingame or out of game tools?
wolverine2710
Tutorial & Guide Writer
Was AWOL for a week (holdiday, no internet). Have to do a LOT of reading up. Just a few questions - haven't (yet) read the last week posts ;-(
What is the status of TGC by TornSoul?
Is is already possible to add other data then distances. Ie adding Systemnames to TGC and other data?
Is currently TOR by RedWizzard leading or is it TGC by TornSoul?
RedWizzard: I think you mentioned last week you wanted to add TGC support to your tool. Has this been implemented?
JesusFreke: I think you mentioned last week you wanted to add TGC support to your tool. Has this been implemented?
What is the status of TGC by TornSoul?
Is is already possible to add other data then distances. Ie adding Systemnames to TGC and other data?
Is currently TOR by RedWizzard leading or is it TGC by TornSoul?
RedWizzard: I think you mentioned last week you wanted to add TGC support to your tool. Has this been implemented?
JesusFreke: I think you mentioned last week you wanted to add TGC support to your tool. Has this been implemented?
Last edited: