Hi there,

Since the IRTC is back again, it seemed a good idea to jump in and make 
  an image. Staying on-topic of rebirth, plants and seeds are a suitable 
topic to make a not to complicated scene. So, there i go, making a 
dandelion seed (just flying away from the flower). The problem now is, 
how to make the seed head? At the moment it is a scaled sphere, but it 
should be more like this image (yes, i know, i am shamelessly 
reproducing the image :-)):

http://stevehickey.files.wordpress.com/2008/05/dandelion_seed.jpg

As you can see, there are dimples in the surface. My idea would be to 
make a surface with blobs. But how to place the blobs? The dimples 
should be evenly distributed over the surface (like a golfball). The 
only thing that comes to mind is to place the dimples randomly, and then 
iterativly let them "find" the place where they are the least close to 
their neighbours. Needles to say, this is complicated and a project on 
its own.

So, any pointers?

Thanks in advance,

JWV

Post a reply to this message

Attachments:
Download 'irtc.jpg' (35 KB)

Preview of image 'irtc.jpg'

JWV <Willempie_39___@___hotmail.com> wrote:
> Hi there,
>
> Since the IRTC is back again, it seemed a good idea to jump in and make
>   an image. Staying on-topic of rebirth, plants and seeds are a suitable
> topic to make a not to complicated scene. So, there i go, making a
> dandelion seed (just flying away from the flower). The problem now is,
> how to make the seed head? At the moment it is a scaled sphere, but it
> should be more like this image (yes, i know, i am shamelessly
> reproducing the image :-)):
>
> http://stevehickey.files.wordpress.com/2008/05/dandelion_seed.jpg
>
> As you can see, there are dimples in the surface. My idea would be to
> make a surface with blobs. But how to place the blobs? The dimples
> should be evenly distributed over the surface (like a golfball). The
> only thing that comes to mind is to place the dimples randomly, and then
> iterativly let them "find" the place where they are the least close to
> their neighbours. Needles to say, this is complicated and a project on
> its own.
>
> So, any pointers?
>
> Thanks in advance,
>
> JWV

are you talking about the seed or the flower head?  If you used a blob for the
flower head, the placement may not be too complex.  You could start with an
ordered placement to maintain the correct separation and perturb them randomly,
or, like you mentioned, use a random placement that checks to make sure no two
are too close.  The ordered start method is easiest and can produce acceptable
results with some tinkering.

I would probably do it as cut-aways - have the head be a larger sphere, and
place the group of smaller ones at a fixed location (of distance radius from
the center of the large sphere), then use a series of ordered rotations to
place them on the surface.

Alternatively, you could use a mesh (which is probably what I would use).  That
may also make making the petals easier.


-Reactor

Post a reply to this message

JWV wrote:

> So, any pointers?

Maybe this can help you get started, even though your
flower head isn't quite spherical:

http://www.ogre.nu/sphere.htm

Post a reply to this message

Actually, the dimples might also be arranged similar
to a fibonacci spiral like seeds of a sunflower:

http://en.wikipedia.org/wiki/Fibonacci_number
http://taystar.com/morespirals2.html

Post a reply to this message

JWV <Willempie_39___@___hotmail.com> wrote:

> As you can see, there are dimples in the surface. My idea would be to
> make a surface with blobs. But how to place the blobs? The dimples
> should be evenly distributed over the surface (like a golfball). The
> only thing that comes to mind is to place the dimples randomly, and then
> iterativly let them "find" the place where they are the least close to
> their neighbours. Needles to say, this is complicated and a project on
> its own.
>
> So, any pointers?
>
> Thanks in advance,
>
> JWV

Here's another way to think about this problem: Imagine the surface as a lattice
of sphere sweeps instead of a solid surface with dimples. That's what it looks
like to me, especially the lower left area of the photo where the surface
appears to be woven out of fibers. The dimple method would not re-create this
effect.

That looks like a fun project, and you're off to a great start!

Regards,
Dave Blandston

Post a reply to this message

> As you can see, there are dimples in the surface. My idea would be to
> make a surface with blobs. But how to place the blobs? The dimples
> should be evenly distributed over the surface (like a golfball). The
> only thing that comes to mind is to place the dimples randomly, and then
> iterativly let them "find" the place where they are the least close to
> their neighbours. Needles to say, this is complicated and a project on
> its own.
>
> So, any pointers?
>
> Thanks in advance,
>
> JWV

Here's another way to look at the problem: Instead of a solid surface with
dimples, imagine the surface as a lattice of sphere sweeps. It looks to me like
the surface in the photo actually consists of woven fibers, which the dimples
you're considering would not replicate. The diameter of a sphere sweep is
variable, which would allow you to replicate the slightly blobby appearance of
the fibers.

That looks like a fun project, and you're off to a great start!

Regards,
Dave Blandston

Post a reply to this message

Sorry for the double-posting...

Post a reply to this message

Christian Froeschlin <chr### [at] chrfrde> wrote:
> Actually, the dimples might also be arranged similar
> to a fibonacci spiral like seeds of a sunflower:
>
> http://en.wikipedia.org/wiki/Fibonacci_number
> http://taystar.com/morespirals2.html

"Might"? Judging from the latter page, they actually *are*. You did notice that
the article explicitly mentions dandelions, didn't you? :P

Post a reply to this message

clipka wrote:

>> http://taystar.com/morespirals2.html
> 
> "Might"? Judging from the latter page, they actually *are*. You did notice that
> the article explicitly mentions dandelions, didn't you? :P

I did notice, but I also heard rumours that the internet *might*
contain a few pages with partially incorrect content ;)

Post a reply to this message