Ron Parker wrote:
> 
> Who is this "most people" you talk about?  Most modelers do that, but
> I would think most people make the box where they want it.  I know I
> do.

Only if you want it aligned to the major axes.
I usually consider it a good idea to create objects at the origin and then apply
the necessary transformations. I feel it gives more control over the final
placement of the object, and the look of the texture.

Margus

Post a reply to this message

On Mon, 13 Mar 2000 04:03:50 +0200, Margus Ramst wrote:
>Ron Parker wrote:
>> 
>> Who is this "most people" you talk about?  Most modelers do that, but
>> I would think most people make the box where they want it.  I know I
>> do.
>
>Only if you want it aligned to the major axes.
>I usually consider it a good idea to create objects at the origin and then apply
>the necessary transformations. I feel it gives more control over the final
>placement of the object, and the look of the texture.

That's true; I do create boxes that won't be axis-aligned at the origin, but
I still create them the correct size.

Post a reply to this message

Does anyone remember the shapes.inc file?
Make changes there and use predefined shapes from there.
Ken wrote:
> 
> Simon de Vet wrote:
> 
> > No matter how much I work on it, it's always a struggle. Having a default
> > I can scale is easier to understand. I can easily visualize a box scaled
> > to <1, 2, 5>, but a box defined by corners <-0.5, -1, -2.5>  <0.5, 1,
> > 2.5> is not something I can see in my mind.
> 
> Both methods require repetitious use before you become comfortable with
> them. When I was first learning Pov I bounced back and forth between
> various modellers I was evaluating and hand coding. Personally I found
> more power in the hand coding environment but had I found a modeller
> I was comfortable with I might have swung the other way. Regardless
> I still think that with practice you would learn to use the various
> control points of a box with the same ease that you do the scale feature
> in your modeller of choice.
> 
> Besides defining a unit box is no more difficult than
> 
> box{-.5,.5}
> 
> Defining a unit sphere
> 
> sphere{0,1}
> 
> Defining unit cylinders you have a choice of which axis to align it to
> 
> cylinder {x*-.5, x*.5, 1}
> cylinder {y*-.5, y*.5, 1}
> cylinder {z*-.5, z*.5, 1}
> 
> etc.
> 
> --
> Ken Tyler -  1300+ Povray, Graphics, 3D Rendering, and Raytracing Links:
> http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/

-- 
Mr. Art

"Often the appearance of reality is more important 
than the reality of the appearance."
Bill DeWitt 2000

Post a reply to this message

"mr.art" wrote:
> 
> Does anyone remember the shapes.inc file?
> Make changes there and use predefined shapes from there.

Actually for something like using a unit sized box I find it easier
to type

box{-.5,.5}

rather than

#include "shapes.inc"

object {cube}

-- 
Ken Tyler -  1300+ Povray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/

Post a reply to this message

Francois Labreque <fla### [at] attglobalnet> wrote in message
news:38CBAF9C.7C199D2B@attglobal.net...
>
> Something I'd like to be able to do is automatically place objects
> relative to others without having to do all the trig myself, but I guess

Now there's something I could have used at times - translate_relative as opposed to
the
existing translate_absolute.

Alf

http://www.peake42.freeserve.co.uk/
http://ourworld.compuserve.com/homepages/Alf_Peake/

Post a reply to this message

David Fontaine wrote:
> 
> Francois Labreque wrote:
> 
> > box {
> >     scale < 2, 2, 2 >
> >     translate < 5, 5, 5 >
> > }
> >
> > is actually more characters than
> >
> > box {
> >     < 5, 5, 5 >
> >     < 7, 7, 7 >
> > }
> 
> or you could do box { 5 7 } :-)

OK.  The choice of numbers wasn't probably the best to illustrate my
point, but I think it came across nevertheless.   Anyway, most of my
boxes look more like this:

box {
    < 2.875, 5.5, 3.6333333 >
    < 7.142857, 9.3, 4 >
}

Which is wayyyyyyyyy easier than 

box {
    scale < 4.267857, 3.8, 0.2666666 >
    translate < 7.642857, 9.8, 4.5 >
}

-- 
Francois Labreque | It is by caffeine alone I set my mind in motion, it
     flabreq      | is by the beans of Java that thoughts acquire speed,
        @         | the hands acquire shaking, the shaking becomes a 
  attglobal.net     warning, it is by caffeine alone I set my mind in
motion.
                               - Unknown

Post a reply to this message

"Francois Labreque" <fla### [at] attglobalnet> wrote :
>
> box {
>     < 2.875, 5.5, 3.6333333 >
>     < 7.142857, 9.3, 4 >
> }
>

    Modeler syndrome.

Post a reply to this message

Ron Parker <ron### [at] povrayorg> wrote in message
news:slr### [at] linuxparkerrfwicom...
> On Sun, 12 Mar 2000 07:56:08 -0500, Bill DeWitt wrote:
> >Since most people use a unit, at origin object, then scale
> >and translate, it could save tons of typing.
>
> Who is this "most people" you talk about?  Most modelers do that, but
> I would think most people make the box where they want it.  I know I
> do.

I'm suprised - I'm trying to write a short newbie tutorial on object
transformations, and one of my general rules is always create objects with a
known point at <0,0,0>. Don't you lose track of where you objects are going
to end up if you scale them? (or at least find it harder).

Post a reply to this message

Tom Melly wrote:

> I'm suprised - I'm trying to write a short newbie tutorial on object
> transformations, and one of my general rules is always create objects with a
> known point at <0,0,0>. Don't you lose track of where you objects are going
> to end up if you scale them? (or at least find it harder).

I completely agree. When I was a newbie, my permanent problem was that
I defined objects "directly" and always lost them after a few translates and
rotates.
I often had to "pile" transform statements to obtain a desired effect, and lost
a
lot of time looking for disappearing primitives (no VFAQ then !).
This problem was solved when I started creating them at <0,0,0>
and applied the transforms afterwards. I find it easier to apply textures too.
This can't be a general rule, though. Chain-like objects can be often created
more
easily using absolute coordinates (when using loops at least) since the end
points
of a primitive can be used as the starting points of the next.

G.

Post a reply to this message

On Mon, 13 Mar 2000 17:27:07 -0000, Tom Melly wrote:
>I'm suprised - I'm trying to write a short newbie tutorial on object
>transformations, and one of my general rules is always create objects with a
>known point at <0,0,0>. Don't you lose track of where you objects are going
>to end up if you scale them? (or at least find it harder).

If you create it where you want it, you don't need to scale it.  I don't see
the problem here.

Okay, to be honest, I still follow the rule of creating things at the origin.
But if I'm going to make a simple stool, I do this:

 union {
   box {<-6, 36, -6>,<6,37,6>}
   intersection {
     box {<-8,0,-8>,<8,36.5,8>}
     cylinder { <-5,0,-5>, <-4,36.5,-4>, .5 }
     cylinder { < 5,0,-5>, < 4,36.5,-4>, .5 }
     cylinder { <-5,0, 5>, <-4,36.5, 4>, .5 }
     cylinder { < 5,0, 5>, < 4,36.5, 4>, .5 }
   }
 }

That is, I create the parts of the stool where they should be, relative to the
fixed point on the stool that I have chosen to have at the origin (in this case,
a point on the floor directly under the center.)  To me, this looks a lot 
cleaner than trying to create and translate and rotate and scale all those 
components to get the desired result.  I will still translate and rotate the 
stool to put it where I want it (but I won't scale it, because it's already to 
scale.)

The exception, of course, is when I need an ellipsoid or a torus.  Those have to
be created at the origin.  (In fact, I wish there were an alternate syntax for
toruses that took a center and an axis so I could work with them without having
to wave my left hand around to figure out the rotations.)

-- 
These are my opinions.  I do NOT speak for the POV-Team.
The superpatch: http://www2.fwi.com/~parkerr/superpatch/
My other stuff: http://www2.fwi.com/~parkerr/traces.html

Post a reply to this message