Board index FlightGear Development Scenery

osm2city.py development

Questions and discussion about enhancing and populating the FlightGear world.

Re: osm2city.py development

Postby slawekmikula » Sun Dec 15, 2019 5:35 pm

I've generated tiles #3269379 and #3285760 (centre of Cracow, Poland). It was generating two tiles on i7/16GB for ca. 8 hours.

My settings:

Code: Select all
OVERLAP_CHECK_CONVEX_HULL=True
OVERLAP_CHECK_CH_BUFFER_STATIC=15

TEXTURES_REGIONS_EXPLICIT = ["de", "gb", "us"]

BOUNDARY_WEST = 19.75
BOUNDARY_EAST = 20.0
BOUNDARY_NORTH = 50.125
BOUNDARY_SOUTH = 50.0

NO_ELEV = False
FG_ELEV = '/mnt/media/loty/flightgear/source/slawekmikula/fgbin/bin/fgelev'
FG_ELEV_CACHE = False
PROBE_FOR_WATER=True
FLAG_STG_BUILDING_LIST = True
FLAG_STG_BUILDING_LIST_2019_1 = False


Effect (new Compositor, no ALS, git, intel card):

Image
Image
Image

Overall urban density as in real life. Building structure OK. There is a lot of Z-fighting with the roads. What's your opinion and what should be corrected in the settings. There is also quite a lot of orange road segments (debug info?).
slawekmikula
 
Posts: 83
Joined: Sun Feb 19, 2017 9:31 am

Re: osm2city.py development

Postby slawekmikula » Sun Dec 15, 2019 10:24 pm

Additional info (but I think understandable). I'm using Compositor branch. Using ALS renderer the result is different. The same is on intel as on NVidia card (840M). First I thought that it was blame to intel card, but on nv the result is the same. I think this is because of the shader is not ported to the Compositor framework. Nonetheless i'm wondering why some of the buildings does indeed have textures ?

Image
slawekmikula
 
Posts: 83
Joined: Sun Feb 19, 2017 9:31 am

Re: osm2city.py development

Postby powoflight » Mon Dec 16, 2019 1:53 pm

I found 2 tiles (maybe more) where all OSM Buildings in a tile disappear once again.

I am testing this with Merspieler's the "osm2city worldbuild" V1 tile e010n40.

One is by LOWL austria Graz: tile 3203641.
See https://youtu.be/w4Hted4oQDs

The second one is in Rom close to LIRF: tile 3154169
Fly with ufo from LIRF East along the Tiber close to the Colloseum the tile (look to north) disapears.

There is nothing extraordinary with --log-level=Bulk

I have tested it with :

ubuntu 18.04
FG 18.3 + Change Materials to Use Buildings by Shaders (Experimental)
FG 19.1.2 stable + Change Materials to Use Buildings by Shaders (Experimental)
FG 19.2 Materials allready exists

win 7
19.1 + Change Materials to Use Buildings by Shaders (Experimental)

Everywhere the same result.

powoflight write
viewtopic.php?f=5&t=22809&p=333359#p332318

vanosten write
Please note that I experience that a whole tile with scenery can disappear suddenly.
viewtopic.php?f=5&t=22809&start=1020#p341645
So have Change Materials to Use Buildings by Shaders

Merspieler write
viewtopic.php?f=5&t=22809&start=1020#p341647

I am not shure this is the right place to communicate but how can i debug to find the problem? Maybe its a core problem? I dont know,
powoflight
 
Posts: 154
Joined: Fri Mar 25, 2016 10:04 am
Location: LOWW
Version: 2020.1
OS: win7 Ubuntu 18.04

Re: osm2city.py development

Postby merspieler » Mon Dec 16, 2019 2:26 pm

powoflight could you please let us know your system details (ie. RAM and GPU...)

I'm experiencing the same in the area I'm currently working on (around YBCS... start on rwy15 and go stright...)

Some debugging would be highly appreceated. tho I don't know how to do so.
If everything is going against you, keep in mind that airplanes take off against the wind, not with it.
merspieler
 
Posts: 718
Joined: Thu Oct 26, 2017 10:43 am
Location: Wish to be in YBCS
Callsign: JST935. ORI1711
IRC name: merspieler
Version: next
OS: Debian Bullseye

Re: osm2city.py development

Postby slawekmikula » Mon Dec 16, 2019 3:51 pm

powoflight wrote in Mon Dec 16, 2019 1:53 pm:I found 2 tiles (maybe more) where all OSM Buildings in a tile disappear once again.


Some notes. I've also encountered building from the whole tile dissapear - mostly the most dense areas as the one i'm testing right now. osm2tiles generated in May this year does have this issue. Nonetheless current build DO NOT have this issue as far as I'm testing but for sure I would find it, if it appears. So please check with the most recent generation.
slawekmikula
 
Posts: 83
Joined: Sun Feb 19, 2017 9:31 am

Re: osm2city.py development

Postby merspieler » Mon Dec 16, 2019 4:05 pm

slawekmikula thanks for the hint...
If everything is going against you, keep in mind that airplanes take off against the wind, not with it.
merspieler
 
Posts: 718
Joined: Thu Oct 26, 2017 10:43 am
Location: Wish to be in YBCS
Callsign: JST935. ORI1711
IRC name: merspieler
Version: next
OS: Debian Bullseye

Re: osm2city.py development

Postby powoflight » Mon Dec 16, 2019 4:14 pm

my hwinfo

ubuntu 18.04.3
cpu i5 6600
mem 16GB
gpu geforce GTX 1060 6GB

latest FG2019.2 from 28.11.2019 (downloaded and compiled)
powoflight
 
Posts: 154
Joined: Fri Mar 25, 2016 10:04 am
Location: LOWW
Version: 2020.1
OS: win7 Ubuntu 18.04

Re: osm2city.py development

Postby powoflight » Mon Dec 16, 2019 4:21 pm

Some notes. I've also encountered building from the whole tile dissapear - mostly the most dense areas as the one i'm testing right now


I agree. Happens in Graz / Austria and in ROM / italy
But it is reproducible.

slawekmikula
which FG Version do use and when do you compiled it. My is : FG2019.2 from 28.11.2019 / VERSION=e25f4e39d2dc8876350c43bb7089bd6c57194e76
powoflight
 
Posts: 154
Joined: Fri Mar 25, 2016 10:04 am
Location: LOWW
Version: 2020.1
OS: win7 Ubuntu 18.04

Re: osm2city.py development

Postby powoflight » Mon Dec 16, 2019 4:35 pm

I was a little in stress.... misunderstood...
So i should regenerate the tiles with latest osm2city py.

Q to merspieler and slawekmikula
By the way . Which parameters do you use?
powoflight
 
Posts: 154
Joined: Fri Mar 25, 2016 10:04 am
Location: LOWW
Version: 2020.1
OS: win7 Ubuntu 18.04

Re: osm2city.py development

Postby merspieler » Mon Dec 16, 2019 4:40 pm

I think then we can rule out hw problems...

I'll update my osm2city and run a "small" test in the area around YBCS to see if it fixes the problem there.
If everything is going against you, keep in mind that airplanes take off against the wind, not with it.
merspieler
 
Posts: 718
Joined: Thu Oct 26, 2017 10:43 am
Location: Wish to be in YBCS
Callsign: JST935. ORI1711
IRC name: merspieler
Version: next
OS: Debian Bullseye

Re: osm2city.py development

Postby merspieler » Mon Dec 16, 2019 5:35 pm

I can confirm, that it doesn't happen (at least not for the tiles I've checked) with a current rebuild.

But I feel like there are missing buildings now. I'm currently rebuilding fg without compositor to check if it is does to that...
If everything is going against you, keep in mind that airplanes take off against the wind, not with it.
merspieler
 
Posts: 718
Joined: Thu Oct 26, 2017 10:43 am
Location: Wish to be in YBCS
Callsign: JST935. ORI1711
IRC name: merspieler
Version: next
OS: Debian Bullseye

Re: osm2city.py development

Postby wlbragg » Mon Dec 16, 2019 6:01 pm

I don't know if this is related, I have a few counties in Kansas built with a really old osm2city version, a version that didn't use the dbase method. A while back (month or two) it suddenly lost some buildings out of the blue. I assumed I just messed something up, but it didn't really make any sense because I really didn't do anything that would have accounted for the city portion of the installation to suddenly fail, especially partially. It was right around the time I rebuild FG, but not necessarily when I rebuilt FG with compositor.
It's possible this isn't necessarily an osm2city thing, but possibly a core FG issue. What is weird, is that it lost some buildings in a fairly tight group of buildings, it was a prison and it lost 90% of it but not all of it?
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
Intel i5 3570K AMDRX480
User avatar
wlbragg
 
Posts: 5725
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Linux/AMDRX480

Re: osm2city.py development

Postby merspieler » Mon Dec 16, 2019 6:09 pm

Dropping in the old scenery makes all buildings to show up (until they diasappear)...
Compositor cam to my mind as well... I just rebuilded fg without it but that gave me the same result..
If everything is going against you, keep in mind that airplanes take off against the wind, not with it.
merspieler
 
Posts: 718
Joined: Thu Oct 26, 2017 10:43 am
Location: Wish to be in YBCS
Callsign: JST935. ORI1711
IRC name: merspieler
Version: next
OS: Debian Bullseye

Re: osm2city.py development

Postby vanosten » Mon Dec 16, 2019 6:56 pm

I get confused with all the messages. Please use the newest build of osm2city to generate new sceneries, which are tested on a recent version of 2019.2 with ALS and without experimental compositor etc. If you then get consistent deviation, I can look into it.
Maintaining osm2city. Contributing with ground attack stuff to the OPRF FlightGear military-simulation community.
vanosten
 
Posts: 480
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 » Mon Dec 16, 2019 7:30 pm

Up to date osm2city and FG 2019.2.0: not all buildings are shown. Build with osm2city from 6 months ago on FG 2019.2.0 shows all buildings
If everything is going against you, keep in mind that airplanes take off against the wind, not with it.
merspieler
 
Posts: 718
Joined: Thu Oct 26, 2017 10:43 am
Location: Wish to be in YBCS
Callsign: JST935. ORI1711
IRC name: merspieler
Version: next
OS: Debian Bullseye

PreviousNext

Return to Scenery

Who is online

Users browsing this forum: No registered users and 3 guests