1. Can I do a sweep on a 3D object in Moray 3.2? Ex: rotational sweep on
a cube around the axis which is the longest line segment that can fit
inside it cube.

2. The bezier patch has control points outside the mesh, so how is it
not qualified as a NURBS? Just because it's a surface?

3. Some modelers let you make polylines/splines and 2D shapes that you
can use as paths and cross-sections for elaborate sweeps and extrusions.
Ex: sweeping a circle perpendicular to a circular path creates a torus.
Programs also offer morphing to make a pipe whose cross-section turns
from circular to square. Moray doesn't offer too many 2D shapes, except
infinite planes and disks. Can I do sweeps along a path in Moray?

4. If I want to do an origami-like model, how do I specify properties of
paper? i.e. Paper folds perpendicular to straight creases. Bezier
patches behave more like rubber. What should I do?

Post a reply to this message

moray can do some of what you want, but it's difficult in the current
version.  you may want to try spatch instead and then export to moray. 
in spatch you can do sweeps of splines (including circles to make your
torus) and while it wouldn't be called morphing, you could make a pipe
go from circle to square as long as you made both with the same number
of points.  hopefully somewhere down the road moray will include a patch
modeller like spatch since in my opinion it doesn't look like it would
be too hard to make.  hamapatch is also out there but has varying
stability from release to release (i.e. not all the bugs are out).  no
pov modeller can do solid (3d) sweeps.  i also hope that down the road
moray will include a mesh modeller as well.

Stan wrote:
> 
> 1. Can I do a sweep on a 3D object in Moray 3.2? Ex: rotational sweep on
> a cube around the axis which is the longest line segment that can fit
> inside it cube.
> 
> 2. The bezier patch has control points outside the mesh, so how is it
> not qualified as a NURBS? Just because it's a surface?
> 
> 3. Some modelers let you make polylines/splines and 2D shapes that you
> can use as paths and cross-sections for elaborate sweeps and extrusions.
> Ex: sweeping a circle perpendicular to a circular path creates a torus.
> Programs also offer morphing to make a pipe whose cross-section turns
> from circular to square. Moray doesn't offer too many 2D shapes, except
> infinite planes and disks. Can I do sweeps along a path in Moray?
> 
> 4. If I want to do an origami-like model, how do I specify properties of
> paper? i.e. Paper folds perpendicular to straight creases. Bezier
> patches behave more like rubber. What should I do?

Post a reply to this message

> 2. The bezier patch has control points outside the mesh, so how is it
> not qualified as a NURBS? Just because it's a surface?

Correct me if I am wrong( and I very well could be.) I think that a bezier
patch is a uniform rational surface in that the contol points all have the
same weight. In a NURBS surface the control points can have different
weights and one point may affect the shape of the curve more than others. Of
coarse, I am not the expert. I'll bet Warp would know.

Post a reply to this message

Halbert <hha### [at] capitalnet> wrote:
: Correct me if I am wrong( and I very well could be.) I think that a bezier
: patch is a uniform rational surface in that the contol points all have the
: same weight. In a NURBS surface the control points can have different
: weights and one point may affect the shape of the curve more than others. Of
: coarse, I am not the expert. I'll bet Warp would know.

  Why me?

  And what you said sounds quite right to me (but I'm not a NURBS expert).

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

>   Why me?

Because you seem to have a great deal more math background than I and NURBS
is a very mathy subject!

Post a reply to this message

Halbert <hha### [at] capitalnet> wrote in message
news:399877d1$1@news.povray.org...
> > 2. The bezier patch has control points outside the mesh, so how is it
> > not qualified as a NURBS? Just because it's a surface?
>
> Correct me if I am wrong( and I very well could be.) I think that a bezier
> patch is a uniform rational surface in that the contol points all have the
> same weight. In a NURBS surface the control points can have different
> weights and one point may affect the shape of the curve more than others.
Of
> coarse, I am not the expert. I'll bet Warp would know.
>

A bezier patch is very similar to a NURBS surface. In fact, if you have a
NURBS
surface with the minimum number of points in it, it is exactly the same
thing as
a bezier surface.

The problem with beziers is that they only naturally form a single small
piece
of surface with only a limited number of points in it - NURBS surfaces are
what
are called "piecewise", which means that they are actually a bunch of small
surfaces that are glued together to make one large one.

NURBS is basically a mechanism for stringing together a whole bunch of
beziers such that all the beziers are guaranteed to be smooth to one another
and
are easier to manipulate.

If you just use individual beziers, it becomes very difficult to get a high
degree of
smoothness between them. You can get 2 adjacent curves to share tangents ok
(which is how Adobe Illustrator works), but it is difficult to get them to
share curvatures also.

The thing that makes NURBS good isn't just that they have control points
outside of
the mesh - it's that you can have any number of control points that you want
to make
a large surface that is one very smooth piece.

It is possible for a bezier to have different weights on every control
point - that's what
"rational" means.

The confusing part about NURBS is that there are 2 totally different
mechanisms that
change the way that a control point can affect the curve - one is the weight
of the
control point, and one is the knot sequence for the curve. Beziers can have
weights,
but they don't have knots.


      - Michael

Post a reply to this message