Board index FlightGear Development Scenery

osm2city.py development

Questions and discussion about enhancing and populating the FlightGear world.

Re: osm2city.py development

Postby danw » Fri Dec 28, 2018 6:23 pm

The Kuakini helipad appears to be messed up by mapping of a pedestrian way as a street: https://www.openstreetmap.org/way/421680496. The helipad is sitting on top of the pedestrian way.
danw
 
Posts: 17
Joined: Wed Aug 08, 2018 2:47 am

Re: osm2city.py development

Postby vanosten » Wed Jan 02, 2019 5:49 pm

I need some help. I cannot find out where some of the APRON surfaces (dark grey) are coming from, which I need to get rid of some of the roads. I "guess" I am taking all I can from apt-dat, but I might be wrong. Someone know what ID the grey areas have in apt-dat - or where they else might come from?

Picture from PHNL:
Image
Maintaining osm2city
vanosten
 
Posts: 385
Joined: Sat Sep 25, 2010 5:38 pm
Location: Denmark - but I am Swiss
Callsign: HB-VANO
Version: latest
OS: Win 10 and Ubuntu

Re: osm2city.py development

Postby legoboyvdlp » Wed Jan 02, 2019 10:36 pm

They are in the scenery files - modifying the apt.dat will not remove the apron, you need to regenerate the scenery using terragear.
User avatar
legoboyvdlp
 
Posts: 6418
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: osm2city.py development

Postby xDraconian » Thu Jan 03, 2019 12:01 am

vanosten wrote in Wed Jan 02, 2019 5:49 pm:I cannot find out where some of the APRON surfaces (dark grey) are coming from


They are present in apt.dat for FlightGear.
Within WED, however, they are present within +21-158.dsf

The apron represented in the picture has these coordinates[partial list] within the FlightGear apt.dat.
Code: Select all
110 1 0.25 10.0000 Apron
111  21.33805731 -157.92361879
111  21.33843694 -157.92628005
111  21.33501980 -157.92695138
111  21.33461695 -157.92722396
111  21.33454924 -157.92665336
111  21.33442060 -157.92666790
xDraconian
 
Posts: 159
Joined: Sun Jan 21, 2018 5:53 am
Version: Git
OS: Linux Mint

Re: osm2city.py development

Postby vanosten » Thu Jan 03, 2019 7:21 pm

Ok, thanks. I read already 110 and related nodes, so I must have a reference hanging around somewhere else for the roads, even though I remove/split them with overlap check.
Maintaining osm2city
vanosten
 
Posts: 385
Joined: Sat Sep 25, 2010 5:38 pm
Location: Denmark - but I am Swiss
Callsign: HB-VANO
Version: latest
OS: Win 10 and Ubuntu

Re: osm2city.py development

Postby danw » Fri Jan 04, 2019 7:57 pm

Overall, I'm pretty happy with the results of the scenery that I generated for Charleston SC, except the waterfront along the peninsula - buildings and roads ends up in the water. Using the PROBE_FOR_WATER parameter set to true fixes this problem, but then that portion of the scenery looks incorrect. Would it be feasible to shift the position of the nodes,ways,etc in the database prior to performing scenery generation in order to compensate for the misalignment between terragear and openstreetmap scenery?
danw
 
Posts: 17
Joined: Wed Aug 08, 2018 2:47 am

Re: osm2city.py development

Postby vanosten » Sun Jan 06, 2019 9:48 am

I am not sure whether I would do it in the database or just on the fly (I guess I would do the later). Given that the misalignment is no consistent in any direction, then you would have to pick a vector, which improves it for some stuff and does make it worse for other stuff. Also you would have to define in which area to stretch or squeeze stuff, so you do not get overlap/gaps with non-moved parts. So essentially you would have to build a tool, which graphically lets you distort stuff.
The easier way to fix this is to create a custom scenery, where you use OSM data and/or change the boundaries of scenery data and then use Terragear based on this data. That has been described elsewhere - and is not easy.

Maybe I missed something.
Maintaining osm2city
vanosten
 
Posts: 385
Joined: Sat Sep 25, 2010 5:38 pm
Location: Denmark - but I am Swiss
Callsign: HB-VANO
Version: latest
OS: Win 10 and Ubuntu

Re: osm2city.py development

Postby legoboyvdlp » Sun Jan 06, 2019 9:55 am

While not easy it is indeed... considerably easier than manually adjusting every position in the database. The major problem is lack of shapefile landuse data in the USA, especially Martin's custom scenery shapefiles.
User avatar
legoboyvdlp
 
Posts: 6418
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: osm2city.py development

Postby danw » Sun Jan 06, 2019 4:59 pm

In the Charleston case, I was thinking that I could create a bounding region around the peninsula and only shift nodes in that region. I would probably run the region north of the airport up past the Summerville, Hanahan regions. I know this would probably create some road discontinuities at the edges of the shifted region, but it would be in an area of little interest, so probably it would be okay.

I'm sure this is a naive question, but why would updating the positions in the database difficult? Wouldn't I just be iterating over all of the nodes and doing a vector addition and updating the positions in a bounded region? Based on what I have now, I don't think it's too difficult to determine the vector (everything is consistently shifted to the southeast), and I can always iterate it a few times until I get the effect that I'm looking for.
danw
 
Posts: 17
Joined: Wed Aug 08, 2018 2:47 am

Re: osm2city.py development

Postby vanosten » Sun Jan 06, 2019 7:11 pm

xDraconian wrote in Thu Jan 03, 2019 12:01 am:
vanosten wrote in Wed Jan 02, 2019 5:49 pm:I cannot find out where some of the APRON surfaces (dark grey) are coming from


They are present in apt.dat for FlightGear.
Within WED, however, they are present within +21-158.dsf

The apron represented in the picture has these coordinates[partial list] within the FlightGear apt.dat.
Code: Select all
110 1 0.25 10.0000 Apron
111  21.33805731 -157.92361879
111  21.33843694 -157.92628005
111  21.33501980 -157.92695138
111  21.33461695 -157.92722396
111  21.33454924 -157.92665336
111  21.33442060 -157.92666790


I must be missing the obvious, but I really do not get it. When I look in the apt.dat.gz file and search fro PHNL, I only get lines starting with 1, 100, 21, 10, 14, 15, , 18, 19, 5x. So I do not even get the APRONs. Are they defined somehow later in the file - or should I instead of looking at $FGDATA/Airports/apt.data.gz into another file?

(It actually bothers me to see the 10 code, which is not defined in the 1000 file format, but is in the 715 version.)
Maintaining osm2city
vanosten
 
Posts: 385
Joined: Sat Sep 25, 2010 5:38 pm
Location: Denmark - but I am Swiss
Callsign: HB-VANO
Version: latest
OS: Win 10 and Ubuntu

Re: osm2city.py development

Postby xDraconian » Mon Jan 07, 2019 5:51 pm

vanosten wrote in Sun Jan 06, 2019 7:11 pm:When I look in the apt.dat.gz file and search fro PHNL, I only get lines starting with 1, 100, 21, 10, 14, 15, , 18, 19, 5x. So I do not even get the APRONs.

Appears the apt changes were not merged then.
You can find the airport definitions I used for Oahu (a.k.a Honolulu) within fgmeta/fg-from-scratch (test.apt.dat).
https://sourceforge.net/p/flightgear/fgmeta/ci/next/tree/fg-from-scratch/CustomSceneryProjects/Test/
xDraconian
 
Posts: 159
Joined: Sun Jan 21, 2018 5:53 am
Version: Git
OS: Linux Mint

Re: osm2city.py development

Postby vanosten » Mon Jan 07, 2019 8:23 pm

xDraconian wrote in Mon Jan 07, 2019 5:51 pm:Appears the apt changes were not merged then.
You can find the airport definitions I used for Oahu (a.k.a Honolulu) within fgmeta/fg-from-scratch (test.apt.dat).
https://sourceforge.net/p/flightgear/fgmeta/ci/next/tree/fg-from-scratch/CustomSceneryProjects/Test/


Thank you very much for sharing. The I understand better. And I was able to test my code with your test.apt.dat :-)
Maintaining osm2city
vanosten
 
Posts: 385
Joined: Sat Sep 25, 2010 5:38 pm
Location: Denmark - but I am Swiss
Callsign: HB-VANO
Version: latest
OS: Win 10 and Ubuntu

Re: osm2city.py development

Postby xDraconian » Wed Jan 09, 2019 5:17 am

vanosten wrote in Mon Jan 07, 2019 8:23 pm:And I was able to test my code with your test.apt.dat :-)

Glad to learn that you got it working.
xDraconian
 
Posts: 159
Joined: Sun Jan 21, 2018 5:53 am
Version: Git
OS: Linux Mint

Re: osm2city.py development

Postby vanosten » Fri Jan 18, 2019 9:29 pm

In case you are running the unrelease NEXT version of FG, you can try out shader based buildings as part of osm2city sceneries. I have published a new scenery for 4 tiles around Edinburgh (EGPH): https://my.pcloud.com/publink/show?code ... GusHtzozAV.
You need to have a recent version of NEXT and configure according to https://osm2city.readthedocs.io/en/late ... perimental.

Please note that I experience that a whole tile with scenery can disappear suddenly.

Image
Image
Maintaining osm2city
vanosten
 
Posts: 385
Joined: Sat Sep 25, 2010 5:38 pm
Location: Denmark - but I am Swiss
Callsign: HB-VANO
Version: latest
OS: Win 10 and Ubuntu

Re: osm2city.py development

Postby merspieler » Fri Jan 18, 2019 11:01 pm

I've experiensed disappearing tiles before... when i've build a dense area of munich... sometimes some tiles disappear...


I think, the reason why it happens on your linux on ram full but not on win is, that on linux the "GPU"s ram is as well in the ram while on your win your GPU runs out of space... just one thing i could think about...
Love at first flight
Checkout Autopush. An improvment to the pushback to make your life easier.
merspieler
 
Posts: 218
Joined: Thu Oct 26, 2017 10:43 am
Location: GCLP
Callsign: LH454, HG3840, LX64
IRC name: merspieler
Version: 2018.3.1
OS: Linux

PreviousNext

Return to Scenery

Who is online

Users browsing this forum: Exabot [Bot] and 3 guests