Pablo will be always right on that so, if he says otherwise, listen to him.
AS i remember for animated stuff you should join the objects, else yes there will be issues around that, maybe there s something to do that i don t know, still you can keep textures separated, and anyway unless you plane to use blender libraries to use again the same skirt and hair on another model, there s no real reason to separate the objects. For the textures don't ask yourself any question, simply work at the highest resolution which isn't an issue for you and you ll downgrade the resolution when this becomes an issue. For the mesh this is another problem as changing the polycount will require re-mapping, re-weighting maybe re-animating, so basically try to not get crazy on the polycount, 15K poly is not that hight but it depends also on the rest of the scene, and how much such characters will be loaded. On Sat, 2009-10-10 at 02:03 +0200, res wrote: > On 09.10.2009 19:52, Jorrit Tyberghein wrote: > >> * First of all, what is the shader to be used? Is it OK to use the > >> ligthing_default for the body, and ligthing_default_binalpha for the > >> skirt + hairs? Another way to ask it: is it OK to provide diffuse + > >> specular + normal maps, or are there other maps that are more desired? > > > > I let Frank answer this one. > > These maps are fine. > And the _binalpha shaders were made exactly for use with binary alpha > (aka alpha test), so if you want these, that's the way to go. > > > > >> * Once rendered in CS, the colors of the mesh appear a little > >> gloomy/depressive. How can we improve that? Is it to be handled by the > >> shaders, by the lights, by the colors of the texture? > > > > Same > > Well, both. Giving the diffuse map more saturated colors will make it > look less gloomy. But so can do a different light setup (e.g. increased > intensity and/or a slight yellowish tint - like sunlight). > > >> * Is 14 000/15 000 triangles OK for a highly detailed model? Do you > >> prefer a model that can run on many hardware, or a model as nice as > >> possible (personally, I prefer the second choice)? > > > > That number is a bit on the high side but maybe it will work. I would > > first have to try it out to see that. But try to avoid getting higher then > > this. > > For comparison, frankie has ~2000 tris. > > > > >> * What are the best sizes for the textures: 1024x1024 for the main body? > >> 2048x2048? 512x512 for the normals and specular maps? > > > > Perhaps 1024x1024 for main body. > > Depends on the detail you want to have. 1024x1024 would probably already > give you decent detail if the model is well mapped. For comparison > again: the frankie model also has a 1024x1024 map - probably too much > resolution. (Of course, ideally you would give it just enough detail > that it looks good in the typical use case - but I don't think we know > that. Anyway, 1024 sounds very reasonable, 2048 is probably already very > high detail). > I would give all maps the same dimensions. > > -f.r. > > ------------------------------------------------------------------------------ > Come build with us! The BlackBerry(R) Developer Conference in SF, CA > is the only developer event you need to attend this year. Jumpstart your > developing skills, take BlackBerry mobile applications to market and stay > ahead of the curve. Join us from November 9 - 12, 2009. Register now! > http://p.sf.net/sfu/devconference > _______________________________________________ Crystal-main mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/crystal-main Unsubscribe: > mailto:[email protected]?subject=unsubscribe ------------------------------------------------------------------------------ Come build with us! The BlackBerry(R) Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9 - 12, 2009. Register now! http://p.sf.net/sfu/devconference _______________________________________________ Crystal-main mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/crystal-main Unsubscribe: mailto:[email protected]?subject=unsubscribe
