Re . Triangle primitives / EARTH

New Message Reply Date view Thread view Subject view Author view

GRAUX Ludovic (LUDOVIC.GRAUX++at++siege.aerospatiale.fr)
Mon, 22 Jul 1996 15:13:36 +0200


      Hi,
     I also tried to modelize the earth, for a space simulation.
     I may not have understood your method. What are you doing with lines and
     points ? Are you trying to follow the coasts with the primitives ?

     Personnally, I have made a spheric node, with LODs and groups, to efficiently
     use the culling. My triangle mesh is uniform, without consideration of
     geography. I applied a fine texture of a geogreaphic planisphere.
     I was inspired both by the pfdNewSphereGSet routine of the libpfdu librairy
     of performer 2.0, and by the ARPA World exemple of the 'Friends' directory,
     to make my own uniform model. There are nice ideas in both those examples,
     but both do not adress good texcoordinates for spheric texture mapping. I
     made a texcoords system by converting the latitude and longitude angles of
     each vertex into texture coordinates. But I face a pb of circularity, that
     makes artefacts of texture mapping particularly in the poles.

     I wont try to explain in details my pb, but I would like to know where to
     find a good spheric texture coordinate generation algorithm (pfTexGen doesn't
     work for this).

     Another item about earth model : The nasty consequence of a regular triangle
     mesh is that the coasts get blurred when I get closer to the earth's surface,
     even with a well sharpened magfilter.
     I would be glad to find a model of the earth with a triangle mesh respecting
     the coasts, and geography in general. Does anyone know where to find this ?

     I hope to continue with you all this discussion about earth modelization.

     Michael Boccara
===================================List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
            Submissions: info-performer++at++sgi.com
        Admin. requests: info-performer-request++at++sgi.com


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:53:12 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.