|
 |
----- Original Message -----
From: Andrew Clinton <cli### [at] ibm net>
Newsgroups: povray.general
Sent: Thursday, November 11, 1999 6:10 PM
Subject: Re: Putting light sources inside hollow objects
> Oops! I didn't realize this. Sorry about the mis-information above and
> thank you for pointing this out, Chris. It does seem unnatural for the
> hollow keyword to have no effect on the csg.
I think the point is that there is no obvious *use* for omitting hollow
objects from csg. If you don't want the object to cut anything out, don't
cut it... While one could write a macro in which a hollow-in-that-sense
object would act differently than any single thing now available, it would
be bad code in that semantic units would cross syntactic boundaries (like C
preprocessor macros without enough parentheses.)
My own $2E-2: "hollow" is very noninintuitive. When media is attached
to an object, I want to think of it as being the middle of the object, so
that an object with media inside is one which is *not* hollow. I guess I
would have preferred media-containment to be the default, with "empty" or
"no_media" for an object that excludes media. There are probably god
arguments for the way it was done, though; and I cannot say that the choice
made is in any way hard to work with.
-Robert Dawson
Post a reply to this message
|
|
