Another experiment, after several tweaks... now without leaves...

Of course, commentaries are most welcomed.

Fernando Gonzalez del Cueto
Mexico.

Post a reply to this message

Attachments:
Download 'DryTree_01.jpg' (105 KB)

Preview of image 'DryTree_01.jpg'

The scatter of small branches and twigs in the extremeties is most
convincing.  That the thickness of branches at the first node come from
the supporting trunk is less convincing.  Have you posted the code?



> Another experiment, after several tweaks... now without leaves...
>
> Of course, commentaries are most welcomed.
>
> Fernando Gonzalez del Cueto
> Mexico.
>
>  [Image]

Post a reply to this message

"Jim Charter" <jrc### [at] aolcom> wrote in message
news:3C1CC73B.5FB59F27@aol.com...
> The scatter of small branches and twigs in the extremeties is most
> convincing.  That the thickness of branches at the first node come from
> the supporting trunk is less convincing.  Have you posted the code?

Is it too thin? or too thick?

Mmmh... I coded the coordinate-generation program in VB and use I/O in POV
to read the file.
Would you still be interested in either code?

Thanks for the comments,

Fernando.


>
> > Another experiment, after several tweaks... now without leaves...
> >
> > Of course, commentaries are most welcomed.
> >
> > Fernando Gonzalez del Cueto
> > Mexico.

Post a reply to this message

@news.povray.org:

> Another experiment, after several tweaks... now without leaves...
> 
> Of course, commentaries are most welcomed.

Some of the tree branches look like they're growing back on themselves, 
which gives the impression that the tree is either sick or in dire need of 
pruning.  You might want to consider changing the code so the branches 
always point away from the center of the tree.

Post a reply to this message

> 
> "Jim Charter" <jrc### [at] aolcom> wrote in message
> news:3C1CC73B.5FB59F27@aol.com...
> > The scatter of small branches and twigs in the extremeties is most
> > convincing.

(Agreed)

> >  That the thickness of branches at the first node come from
> > the supporting trunk is less convincing.  Have you posted the code?
> 
> Is it too thin? or too thick?

The trunk should still be thicker IMHO.

-- 
David Fontaine  <dav### [at] faricynet>  ICQ 55354965
My raytracing gallery:  http://davidf.faricy.net/

Post a reply to this message

"Coridon Henshaw" <che### [at] sympaticoca> wrote in message
news:Xns### [at] 204213191226...

> @news.povray.org:
>
> > Another experiment, after several tweaks... now without leaves...
> >
> > Of course, commentaries are most welcomed.
>
> Some of the tree branches look like they're growing back on themselves,
> which gives the impression that the tree is either sick or in dire need of
> pruning.  You might want to consider changing the code so the branches
> always point away from the center of the tree.

Yes. The algorithm increases the randomness of the direction of the branches
proportionally to the "depth" of the branch. I also think it got a little
bit excesive.

Thanks for your comments.

Fernando.

Post a reply to this message

news:3c1d2e08@news.povray.org: 

> Yes. The algorithm increases the randomness of the direction of the
> branches proportionally to the "depth" of the branch. I also think it
> got a little bit excesive.

Something else I'll suggest is a higher minimum divergence angle for larger 
branches.  I noticed a few spots in both the dried and fall trees where 
major branches split at very shallow angles.  To me this makes the branches 
look more a bit too much like pipe manifolds rather than tree limbs.  :-)

Post a reply to this message

"Coridon Henshaw" <che### [at] sympaticoca> wrote in message
news:Xns### [at] 204213191226...

> news:3c1d2e08@news.povray.org:
>
> > Yes. The algorithm increases the randomness of the direction of the
> > branches proportionally to the "depth" of the branch. I also think it
> > got a little bit excesive.
>
> Something else I'll suggest is a higher minimum divergence angle for
larger
> branches.  I noticed a few spots in both the dried and fall trees where
> major branches split at very shallow angles.  To me this makes the
branches
> look more a bit too much like pipe manifolds rather than tree limbs.  :-)

Yes, you're right, that might improve the appearance of the bifurcations.
Now I'm experimenting with another thing :-) so I might not have the time to
implement this new ideas. However, this let me realize that modeling trees
is not so easy as I thought. I believed that some basic recursion code would
give a very realistic tree. I learned that's not the case!!!

Thanks for the comments,

Fernando.

Post a reply to this message

>
>
> Is it too thin? or too thick?

The supporting trunk seems too thin to me.

>
>
> Mmmh... I coded the coordinate-generation program in VB and use I/O in POV
> to read the file.
> Would you still be interested in either code?
>

Interested at the psuedo-code level actually.

Post a reply to this message

"Jim Charter" <jrc### [at] aolcom> wrote in message
news:3C1E7BAB.BDEE43ED@aol.com...
>
>
>
> >
> >
> > Is it too thin? or too thick?
>
> The supporting trunk seems too thin to me.
>
> >
> >
> > Mmmh... I coded the coordinate-generation program in VB and use I/O in
POV
> > to read the file.
> > Would you still be interested in either code?
> >
>
> Interested at the psuedo-code level actually.
>

Hi Jim. I've posted a newer experiment which I think it fixed some of the
problems you addressed on this thread.

The (latest) pseudo-code is basically a recursive algorithm which operates
like this:

A BRANCH object is defined more or less like a cylinder:
 end1 as a 3d vector
 end2 as a 3d vector
 thickness as a float

1. create an initial branch (the main trunk rooted in the ground)
2. decide how many levels of recursion you're going to accept.
3. decide how many subbranches are going to stem out of the initial branch
(in this example is fixed) and where are they going to stem (in this example
always at end2)
4. generate sub-sub-branches for each of the subbranches you've generated
until you've got until the deepest permitted level of recursion. The
thickness of each subbranch is diminished by a factor less than one. First I
experimented with a random value, but then I decided that a fixed value
simplified some things in POV and it didn't look so bad. Also the length of
the subbranch has to be shortened in some way: the latest example uses a
function which depends on the recursion depth. I think it looks OK.

Then in POV, I read the text file which contain all the coordinates and
thicknesses and place the cones in the those positions. In this example I
didn't add leaves, but in the newer example I added some elliptical discs at
the deepest branches.

I hope I got to explain myself more or less (my English is sometimes a bit
crappy :)

Thanks for your comments,

Fernando.

Post a reply to this message