Re: [Flightgear-devel] Multiplayer Location Transformation toLat/Lon/Alt
Hi, Is there a plan to switch to OSG? Just wondering. I didn't know. I do not know of concrete plans, but here and then there are roumors. I for my own would apprecheate that, since it looks very promising. ... also a few weeks ago a small cvs checkin with some 'flightgear' path in it slipped into osg's cvs. So there must be people on it behind the scenes. :) I think the current math utilities are self-documented adequately. One of my obstacles has been learning the definition of the multiplayer interface, so I know where to start. I think I am beginning to understand now. It appears the position is cartesian (ecef), but is the difference between the player location and the center of _his_ current tile. The position is in the fourth row of the 4x4, in the first three columns. The upper left 3x3 represents the orientation. That depends a bit. The scenery center is in current cvs now locateable at any position. This fixes problems with roundoff and jumpy 3d objects near the eyepoint. In fact it is allways near the current eyepoint. There has been much talk here about reworking the multiplayer model. I'm guessing that the assumption that the other players are all on the same tile will not be required in future versions. Are the current thoughts toward sending absolute position across the wire? Would that be geodetic or cartesian? (just gathering current thoughts, not definitive answers for where this is heading) It looks like they will be cartesian. And IMO this is the preferred way, even if people find lat/lon more intuitive, cartesian coordinates are much more handy for nearly every kind of computation required to be done in this area. Greetings Mathias -- Mathias Fröhlich, email: [EMAIL PROTECTED] ___ Flightgear-devel mailing list Flightgear-devel@flightgear.org http://mail.flightgear.org/mailman/listinfo/flightgear-devel 2f585eeea02e2c79d7b1d8c4963bae2d
RE: [Flightgear-devel] Multiplayer Location Transformation toLat/Lon/Alt
Well, if my hackery here must server as a reference for that, I think we should improove the framework around that stuff and we should document that better. By improoving that framework, I can well imagine to abstract from the vector classes provided from plib, so that for that time we really switch to Open SceneGraph, we can just change that abstraction to be compatible with the osg::* classes instead of changes several hundred places where they occure. Added to my allways growing todo list. :) Apart from that, you may look at http://www.flightgear.org/Docs/Scenery/CoordinateSystem/Coordinate System.html Greetings Mathias -- Mathias Fröhlich, email: [EMAIL PROTECTED] Thanks, Mathias. Is there a plan to switch to OSG? Just wondering. I didn't know. I think the current math utilities are self-documented adequately. One of my obstacles has been learning the definition of the multiplayer interface, so I know where to start. I think I am beginning to understand now. It appears the position is cartesian (ecef), but is the difference between the player location and the center of _his_ current tile. The position is in the fourth row of the 4x4, in the first three columns. The upper left 3x3 represents the orientation. There has been much talk here about reworking the multiplayer model. I'm guessing that the assumption that the other players are all on the same tile will not be required in future versions. Are the current thoughts toward sending absolute position across the wire? Would that be geodetic or cartesian? (just gathering current thoughts, not definitive answers for where this is heading) Highest Regards, Jim A ___ Flightgear-devel mailing list Flightgear-devel@flightgear.org http://mail.flightgear.org/mailman/listinfo/flightgear-devel 2f585eeea02e2c79d7b1d8c4963bae2d
