If you need instant gratification, then local scenery is for you. Even x-plane only releases a new world every major release. So let's not be too hasty throwing around which way is 'better'. When complete, I hope to engage more with the community for input (what could be better), testing( scenery designers), and development( shader writers, postGRE folks) I am again, rewriting large portions from scratch, to guarantee a manifold mesh throughout. The ws2.0 is not ( hence - the broken urban shader ). A very large issue came up which made most of the work I did on terragear for ws2.0 useless. Issues were found and reported.Ģ years ago, I started looking at LOD. A team visually inspected favorite areas for about 2 months, and ws 2.0 was released. So, after 3 years, the toolchain actually generated all 2 million tiles on its own - no manual intervention. Then, once complete, how do you visually inspect 2 million tiles for correctness? The actual numbers ( for ws1.0) were not that big, but there were many that needed manual tweaking to generate them. So, when constructing 2 million+ tiles for FG, you would have maybe 0.5% fail - that's 10 thousand tiles that would not build. numerical roundoff generated errors - no matter how hard you try to fix incorrect data - you end up breaking something else. triangulation - terragear had setup resource limits to detect infinite loops during triangulation. Some issues were worked around, but many issues remained. clipping ( 5 years ago we used GPC ) : Ralf Gehrlich found many issues with this, but it was one of the more popular clipping libraries. They helped me find issues in the toolchain, and even provided patches for fixes as well.Īfter a few months, my goals shifted, as it became apparent that the tools themselves were just not robust enough to work for the entire planet. soitanen and wlbragg come to mind as people who started such projects while I was working on the tools. I support those generating such scenery ( maybe not as much as I should, but not for lack of enthusiasm, mostly a lack of time ). In fact, the toolchain for ws2.0 was developed by generating Madeira. When I first began working on it, my goal was to make it easier to use for people making such custom scenery. It can all be visited and inspected in a matter of days by a single person. Local scenery is always good for providing small areas for good 'wow' effect. I've been working on terragear now for 5 years.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |