"Xiaobin Wu" <xwu### [at] cisefuledu> wrote in news:firstname.lastname@example.org:
> It has been 10 days since we posted first version of pov-ray with
> subdivision. The number of responses and downloads has been
> certainly beyond our expectation. We are glad that if you found it
> useful. All the feedback has positively motivated us to make it a
> better program.
First of all kudos for your work on this POV-Ray patch.
I've some questions:
- When is subdivision performed? As a preprocessing step after parsing and
before rendering (actually creating a new, more detailed, mesh/mesh2
object and substituting the original object with the subdivided one) or
dynamically from the original mesh/mesh2 at rendering time, without
keeping in memory the subdivided one?
- I do not understand the new discard_boundary switch. It discards
boundary edges that do not meet which conditions?
May I suggest to look at Wings (www.wings3d.com: it has an export plugin
(http://www.midcoast.com.au/~rgcoy/software.html) that outputs models to
POV-Ray (in various ways) and is able to perform the subdivision. Even if
I think they adopt the Catmull-Clark with infinitely sharp creases scheme
it may be useful to create test models, "grab" some great free models,
look at the export plugin implementation (beware they use a new, strange
scripting language called Erlang).
Post a reply to this message