On Wednesday 04 April 2007 16:28, Curtis Olson wrote: > On 4/3/07, Anders Gidenstam wrote: > > Hi! > > > > There seems to be an elevation problem in Scandinavia in the 0.9.10 world > > scenery: between N60 and N61 the terrain elevation is 0. Everything is > > flat. I flew due west from the Baltic sea to the Atlantic at 300ft MSL > > along the southern edge of the area without hitting terrain.. The cliffs > > were spectacular ;) > > > > Here is an Atlas picture from another user: > > > > http://pics.ww.com/d/48167-1/fgfs-altproblem.png > > > > I don't know if/how far the problem area extends further to the east. > > > > There is also a lake at N60.19 E15.10 (mpmap coordinates) that has > > its surface at -580ft in FlightGear. There could be more such strangeness > > but this is what I've found so far. > > > > It would be nice if someone who has the raw data handy could check > > whether it too has problems in this area or if this is something that > > went wrong in the world scenery generation. > > There are a couple issues that could be happening here. The SRTM data ends > at +60 latitude. So from that point on up to the north pole, we have to > use a different data set which off the top of my head might be the GLOBE 30 > arcsec data set? SRTM also pots out a global 30 arcsec data set which > might have the regions outside of +/- 60 latitude covered from other data > sets ... can't remember now off the top of my head. > > There is a point (maybe +62 latitude?) where FlightGear switches tile > widths. This seemed like a good idea at first (and it is for a couple > reasons), but the tile boundaries at this division are not handled > correctly in the code so you can see some missing triangles. > > Generally, if altitudes are just plain wrong, we would need to go back and > check the source terrain data ... although there may not be a quick/easy > solution. > > Curt.
It may be that this is actually due to the srtm data itself. I downloaded the V2 3 arcsec srtm data for the British Isles region to have a look at it in Grass and the N60 'tiles' appear to be both incomplete and incorrect. They are incomplete in that they do not cover a full 1x1 deg region - the width (E-W) is ok but they only seem to cover about 2/5 of the N-S (lat) axis. The data for this strip appears to be incorrect in that areas of sea are shown as land, albeit with lots of small lakes. I've attached a small screen grab of a section of a grass display. LeeE
<<attachment: Screenshot.jpg>>
------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
