I don't necessarily think that Source is the wrong engine. There are a number of paths I'm ready to follow to optimize things and I'm not too worried about it's capacity to handle the setting. And I've got 2 years to figure it out.
A couple of the things I'm currently doing or planning on are... - The player has been scaled down, so the potential map size is now on the order of a few kilometers across. - There are no world brushes required, it's all displacement surface terrain (which draws extremely efficiently). - I've started creating a new system that is similar to the current detail object system, but much more detailed in it's efforts to maintain good frame rate, etc. - Detail props are created by a new pass in the vbsp build phase which accesses the database to get actual tree/forest/vegetation composition to determine what to place where (not based on material/blend levels) - At that time, various visualization optimizations can be pre-calculated. - Rather than use a fixed maximum draw distance, the distance will change dynamically based on the users location and what they are looking at. - When in a dense forest stand, the maximum draw distance for trees can be extremely short - you can't see very far. - When in a sparse area with fewer trees, the draw distance goes up (dynamically). - Trees don't need to get drawn with much more than a locked sprite when at a distance. - ONLY when users decide to fly up high will the system be stressed, but even then... - Trees viewed from above can quickly be replaced with bump mapped material views on the displacements using a new shader which varies what is drawn based on user distance. Think blend textures where the blend value changes based on how far away you are from it. Up close, you get ground dirt texture, at a long distance, you get green tree tops for example. - Users won't typically go flying around like magic, but will instead use more GIS-like overview map interface to pick points to move to, get data on, etc. - Having the VGUI system available, a variety of pretty detailed user interactions and GIS visualizations of the data behind the surrounding terrain is possible. But anyway, about that CZero content... Quoting John Sheu <[EMAIL PROTECTED]>: > It's also about using the right tools for the right job. For the medical > simulation, the Source engine is suitable. For the forestry visualization, > it is not. > > On Tuesday 30 August 2005 12:12 pm, Tim Holt wrote: > > Smell away, the point is to create a visualization system, not a > multiplayer > > frag fest :^) Also there's no LOD nor any number of other optimizations > going > > on (yet). > > > > Also, smell the fact that it is a paying job to develop it over a 2 year > period > > (just started 2 months ago). I'm trying to wave a sort of "over here!" > flag > > for some of you developers who want to go beyond just the amateur mod > scene. > > It's not just about entertainment. > > -- > I > think poem > that never as a > I shall see a lovely as binary tree > > _______________________________________________ > To unsubscribe, edit your list preferences, or view the list archives, please > visit: > http://list.valvesoftware.com/mailman/listinfo/hlcoders > > _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlcoders
