David Megginson writes: > Curtis L. Olson writes: > > > I think Norman was talking about some sort of automated magic carpet > > mode.
Exactly > That would be slow, use minimal visibility and turn off any calls that actually render anything, ssgCullAndDraw, pre(post)RenderSky ect, and it should be 'quite fast'. > but it would work at first. The problem is that > we'd have to rerun the whole thing whenever the base scenery changed. Hmm.. Reruning the static scenery tools when the static scenery changes almost seems like a logical thing todo .... > It will be a lot more flexible and simpler to allow placing objects on > the ground automatically, once we work out some of the problems. David - You still haven't answered my question about your problem "Is the points containing tile loaded when you are trying to get it's elevation" ?? If it isn't then you just need to load it before making your query to the elevation routine and then it should just work using a similar method to what Jim uses for Tower placement If there is still a 'chicken / egg' problem then I guess you will need yet another queue similar to the 'defered model queue' but I suspect that just adding the dynamic objects into the DM queue might suffice hard to tell with out actually seeing the code you are trying to use But IMO this is static scenery and should be processed offline when at all feasable as we already have a 'stutter' issue and FGFS itself < with proper tuning as above > is IMO the logical 'filter' Norman _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
