Curtis L. Olson wrote:
Jon Stockill wrote:
Curtis L. Olson wrote:
SNIP
It still would be really nice though to be able to place objects at ground level at load time too.
Regards,
Curt.
When this happens, it would be nice to see some cycles dedicated to the problem of slope. A ranch house looks pretty silly hanging out on a 30 deg slope with its foundation exposed. On the other hand, I have seen many buildings in the real world that have ground entrances on floors that are 2, 3 or even 4 floors apart vertically. This would require special building models divided into sets as well as data assigned to each model regarding how it should be oriented on the slope. I'm not sure how this applies to the great guyed antennas up on the hill by KSFO, but they also exhibit a similar problem. It is also worth thinking about this if anyone ever decides to models a set of generic bridges that can be automatically placed.
<tangent>
Speaking of which, imagine the following. First, a DB with all the known antenna locations and heights in it. For giggles, let's generate a data set of bridges from our scenery data, and if it's available, dump some pre-existing data-set of water towers in there. Make some sweepingly broad generalizations about what these objects would look like based on their country of residence and assign those values as a base case. Now build a web interface that lets joe FGFS pilot and expert on his home town go browse and add an entry to any of those objects classifying it as looking like this or that type of bridge/tower/tank in that general color scheme. Joe pilot can also, on the honor system, specify a score of how sure he is about the data he is supplying (guess/pretty sure/it's in my back yard). Advertise it. Build a few dozen simple models of various generic types in said color schemes. Now take all that data and feed it into some process to place the right (ie highest scored) generic object at the right set of coords. A little bit of coding, a DB, a web site and some distributed observation and the FG world becomes a much better approximation of the real one. Of course, it would take a while to reap the benefits, but if it is set up with an eye for longevity it will just become more and more valuable as more people add data.
</tangent>
You'll find plenty of suitable data in VMAP0. I've tried adding power lines to scenery (well, the pylons, not the wires, although I suspect modeling those would keep the helicopter pilots on their toes) and it really does make a difference - makes navigation easier for a start - you can just follow the pylons all the way to the power station. Adding these things currently involves an awful lot of work though. FGSD provides one solution, but seems to like to segfault on my local scenery, so it's not really an option here.
I'm perfectly happy to generate this kind of data - I'd just like to know it's going to be some use before I start.
-- Jon Stockill [EMAIL PROTECTED]
_______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
