I used my blob macros another way (than in my
"Two blobs" posting) to create this image.

This time there is only one blob and many 
small spheres.


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Sorry. I forgot to enclose the image.

Here it is.


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Attachments:
Download 'blobtest20.jpg' (44 KB)

Preview of image 'blobtest20.jpg'

A dumbbell with no-slip grip might have been a better choice of words  :-)

Seeing objects follow the contours of a 'blob' is no less than amazing to me.
Have you done more exotic shapes?

Bob

Post a reply to this message

Bob Hughes wrote:
> 
> A dumbbell with no-slip grip might have been a better choice of words  :-)

I have never had problems with dumbbells 
slipping out of my hands !  =)


> Seeing objects follow the contours of a 'blob' is no less than amazing to me.

I'm only using some macros (in plain POV)
to make a 3D-map of the accumulated "field 
strengths" of the different components in 
the blob. 

Other macros then use this 3D-map to 
select points in space with the "right" 
"field strength".

These points can again be used for new 
components in a blob or for placing other
objects.


> Have you done more exotic shapes?

No not yet. 

I don't have much experience with blob
modelling. And I have to many other 
projects to take care of, so I haven't
got enough time to experiment much 
with these macros.
(But I have ideas of some shapes :-)

At this moment my macros can handle 
any blobs made of sphere components. 
They can have different radius and 
strength, but they cannot be scaled.

(Maybe some day I'll improve them so 
that they can  take scaled spheres and 
cylinders too.)


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

That looks just plain kinky :)

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

In article <3967E4EA.5D8BED36@online.no>, Tor Olav Kristensen 
<tor### [at] onlineno> wrote:

> I'm only using some macros (in plain POV)
> to make a 3D-map of the accumulated "field 
> strengths" of the different components in 
> the blob. 
> 
> Other macros then use this 3D-map to 
> select points in space with the "right" 
> "field strength".
> 
> These points can again be used for new 
> components in a blob or for placing other
> objects.

Seems like using trace() in MegaPOV would be a much more efficient 
solution...it would also work with cylinder components, and scaled 
components.
Or if it is really the field strength you want, you could use 
eval_pattern() with the blob pattern...though this would also require 
MegaPOV.
Or you could use vtransform() to allow transformations to be applied to 
the components in your macros. Again, this would require MegaPOV.

-- 
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/

Post a reply to this message

Chris Huff wrote:
> 
> In article <3967E4EA.5D8BED36@online.no>, Tor Olav Kristensen
> <tor### [at] onlineno> wrote:
> 
> > I'm only using some macros (in plain POV)
> > to make a 3D-map of the accumulated "field
> > strengths" of the different components in
> > the blob.
> >
> > Other macros then use this 3D-map to
> > select points in space with the "right"
> > "field strength".
> >
> > These points can again be used for new
> > components in a blob or for placing other
> > objects.
> 
> Seems like using trace() in MegaPOV would be a much more efficient
> solution...it would also work with cylinder components, and scaled
> components.

As I understand from the documentation, 
trace() can not provide the initial 
points to "shoot" vectors from.

Nor can it provide the directions in 
which to point the vectors so that 
they point towards the blob surface.
(Or is there available a "field 
direction" function in MegaPOV ?)


For complex blob shapes it will not be 
enough to "shoot" several vectors from 
a circle centre towards the circle 
surface (nor from the surface towards 
the centre).

But my macros can make use of trace() 
to find the exact positions of the 
blob surface once it knows all the 
points to "shoot" from and in which 
direction to "shoot" the vectors.


> Or if it is really the field strength you want, you could use
> eval_pattern() with the blob pattern...though this would also require
> MegaPOV.

I agree. This sounds like a faster way 
to determine the "file strength" at a
point.


> Or you could use vtransform() to allow transformations to be applied to
> the components in your macros. Again, this would require MegaPOV.

I haven't read anything about 
vtransform() yet. 

Is it a MegaPOV feature ?


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Margus Ramst wrote:
> 
> That looks just plain kinky :)

And what associations, exactly, do 
you get when looking at this image ?

=)


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

In article <396A7DDA.970C6F5A@online.no>, Tor Olav Kristensen 
<tor### [at] onlineno> wrote:

> As I understand from the documentation, trace() can not provide the 
> initial points to "shoot" vectors from.

It doesn't...how would it know where you want to look?


> Nor can it provide the directions in which to point the vectors so 
> that they point towards the blob surface. (Or is there available a 
> "field direction" function in MegaPOV ?)

There is not, though it wouldn't be too difficult to figure out with a 
macro and eval_pattern().


> For complex blob shapes it will not be enough to "shoot" several 
> vectors from a circle centre towards the circle surface (nor from the 
> surface towards the centre).
> 
> But my macros can make use of trace() to find the exact positions of 
> the blob surface once it knows all the points to "shoot" from and in 
> which direction to "shoot" the vectors.

One possible way of doing it would be to trace() in a 3D grid, 
continuing on after each intersection on the same line. This wouldn't 
provide a very even coverage, though...actually, it would probably give 
something like the image you posted.
You might also trace in random directions from the center of each 
component. This wouldn't work for every blob, but it should produce some 
interesting results.


> I agree. This sounds like a faster way to determine the "file 
> strength" at a point.

You mean "field strength"? :-)
It could also take a blob object which has already been defined, if you 
are using the latest version of my MegaPOVPlus blob pattern patch(which 
is available at my web site and probably will be in the next MegaPOV).


> I haven't read anything about vtransform() yet. 
> Is it a MegaPOV feature ?

Yes. It just takes a transformation and a vector, and returns the 
transformed vector. Sort of like vrotate(), but can also be used with 
scale, translate, matrix, and complex combinations of these(and also 
declared transforms).

-- 
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/

Post a reply to this message