Hi all,

a long time has passed since I wrote again in here How is everybody?

All right, I have a question which is vital in using GS for the little experiment I am intending to do. I was playing around with some low polys I had constructed with 3ds max; textured them; normal mapped them; specular map added too and all that stuff...

Some objects:

(Everything is low Poly and the Room itself has about 1300 Tris)


(A Book - the Pages would be "concave" I pressume)

... then I have created an indoor scene ("The Inn") as just one level. My intention is to produce just one but very complex level - I don't fear low FPS as long it's playable a bit. Just want to push the graphical possibilities into the sky.

All right, I startet putting one Object into a standard level scene in WED (a hollowed block with one light). It was this Object:


(This is the most complex model in the scene - it is a mid poly Model with about 1000 Tris. I consider Models above 1k Tris as "mid Poly" - this consideration may vary from artist to artist)

As I tried to preview the level I got the message it would be a concave object and cannot be implemented within the level (and it was'nt there as expectet). I am not sure if I understood the concave issue correctly, but as the manual and the corresponging First Game Tutorial states: Concave objects are assumed to be indentations into an object - no matter if it's just a slight lay in into a box or more complex indentations like in this chair shown above.

Or in other words: More complex objects which are not constructed out of almost perfectly allined boxes are impossible to render with the GS Engine. Is that true? If so... how do I make this Chair being rendered within the GS Engine? Well, yeah - I know - it's not possible to render concaves, but is there a workaround. If not: Everything is lost and I unfortunately have to give up the experiment with GS.

To explain it: I do not want to rip off this pretty simple model to getting somehow the structure which would be rendered in GS in any way, because modeling this Object is the standard way to do low Poly Art for games. Well... that's not really the reason, it's more like: One Object - one UV-Unwraped Texture... it's the common and simple workflow anyone layout their Game models. Restricted to plain Boxmodels would just push up the complexity of modelling Objects for GS and would have an enormous negative impact into the common low P Workflow I followed for many years too (and would decrease the otherwise possible complexity of any thinkable model I have in mind and constructed for other engines).

I can't believe that... I have to miss something; I have to misunderstood something about those model construction...

... please tell me if I am wrong about all that ('cause I am new to GS and can't tell if I have missed something).

Thanks in advanced



Tristan