I've struggled for a long time to try and achieve an erosion effect in POV that
is the least bit convincing.  I'm still not there, but perhaps I've made some
headway.  I've been using proximity functions to simulation a little erosion.
Maybe you know some better ways?

Post a reply to this message

Attachments:
Download 'erosion01.jpg' (89 KB)

Preview of image 'erosion01.jpg'

"Kirk Andrews" <kir### [at] tektonartcom> wrote:
> I've struggled for a long time to try and achieve an erosion effect in POV that
> is the least bit convincing.  I'm still not there, but perhaps I've made some
> headway.  I've been using proximity functions to simulation a little erosion.
> Maybe you know some better ways?

I stumbled across this a while ago:

http://geomorph.sourceforge.net/crack_network/en/The%20Crack%20Network%20Tool.html

I can't comment as to the technique's competence, as I know next to nothing
about the subject.

-Mike

Post a reply to this message

Kirk Andrews wrote:
> I've struggled for a long time to try and achieve an erosion effect in POV that
> is the least bit convincing.  I'm still not there, but perhaps I've made some
> headway.  I've been using proximity functions to simulation a little erosion.
> Maybe you know some better ways?

What exactly are you after? Deep ravines and washed-out plains? Or just 
simple textural wear and tear on ridges and outside rock edges? If it's 
the latter, I have a proximity macro I use for that sort of thing, one 
that's not bundled with the SSS macros. If it's the former, well, there 
are programs out there to achieve it:

World Machine:
http://www.world-machine.com/

^great erosion, but the demo version, while lasting forever, only lets 
you export small height maps. I tried tiling them in POV, as WM supports 
tiled landscapes, but as it turns out WM's tiling and erosion features 
don't not work well together. Sure, the maps are tiled, but there will 
always be a visible ridge at the seams :(

Wilbur:
http://www.ridgenet.net/~jslayton/wilbur.html

^I don't use this program very much. It's a little buggy, and the 
erosion isn't too good :(

You've probably already checked out these programs, but I listed them 
just in case. I really wish some talented programmer would make a 
*freeware* terrain program that has an awesome erosion algorithm, *and* 
exports in the 16-bit grayscale format.

Sam

Post a reply to this message

stbenge wrote:
> I really wish some talented programmer would make a 
> *freeware* terrain program that has an awesome erosion algorithm, *and* 
> exports in the 16-bit grayscale format.

Google pulls up some interesting papers:
http://cgg-journal.com/2002-2/03/index.htm
http://oddlabs.com/download/terrain_generation.pdf

Alas, you'll need to find a hacker with more free time than me. Or 
possibly offer me significant bribes to do this instead of the work I 
should be doing. :-P


-- 
William Tracy
afi### [at] gmailcom -- wtr### [at] calpolyedu

I disapprove of what you say, but I will defend to the death your right 
to say it.
     -- Evelyn Beatrice Hall, frequently mis-attributed to Voltaire

Post a reply to this message

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

stbenge wrote:
> You've probably already checked out these programs, but I listed them
> just in case. I really wish some talented programmer would make a
> *freeware* terrain program that has an awesome erosion algorithm, *and*
> exports in the 16-bit grayscale format.
> 
	What about hflab? I'm not sure about the 16 bits output, but it is
GPL and has erosion abilities. Plus, the source is available if
someone cares to improve it:

http://www.bealecorner.org/best/gforge/

		Jerome
- --
+------------------------- Jerome M. BERGER ---------------------+
|    mailto:jeb### [at] freefr      | ICQ:    238062172            |
|    http://jeberger.free.fr/     | Jabber: jeb### [at] jabberfr   |
+---------------------------------+------------------------------+
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkk6LdYACgkQd0kWM4JG3k9OHACfdWw+Bp3E0Asf66GQx+DPkVFq
bgMAni4ZAyTqJWgXAb/XdDI7q3EaSdpS
=rieK
-----END PGP SIGNATURE-----

Post a reply to this message

"Kirk Andrews" <kir### [at] tektonartcom> schreef in bericht 
news:web.493969b33e06781e198b63d90@news.povray.org...
> I've struggled for a long time to try and achieve an erosion effect in POV 
> that
> is the least bit convincing.  I'm still not there, but perhaps I've made 
> some
> headway.  I've been using proximity functions to simulation a little 
> erosion.
> Maybe you know some better ways?
>

Not easy overall. Until now, I have mainly used the f_ridge(), 
f_ridged_mf(), and f_hetero() functions as a basis for decent erosion 
features in isosurfaces. This is done also for instance by the program 
Leveller. However, that is not always convincing. Ideally, one would like to 
take a pre-defined landscape and apply erosion functions to it to wear it 
down and apply a sediment apron around standing ridges. That is basically 
what programs like GeoControl or World Machine are doing I believe, and very 
convincingly so.

Thomas

Post a reply to this message

"Thomas de Groot" <tDOTdegroot@interDOTnlANOTHERDOTnet> wrote:
> Not easy overall. Until now, I have mainly used the f_ridge(),
> f_ridged_mf(), and f_hetero() functions as a basis for decent erosion
> features in isosurfaces. This is done also for instance by the program
> Leveller. However, that is not always convincing. Ideally, one would like to
> take a pre-defined landscape and apply erosion functions to it to wear it
> down and apply a sediment apron around standing ridges. That is basically
> what programs like GeoControl or World Machine are doing I believe, and very
> convincingly so.
>
> Thomas

I wanted to do something like this for my Habitable Planet macro (see the Object
Collection). I also wanted to accurately add rivers and and other types of
falling whater based on this information. However it is something that I think
lies beyond my abilities.

-Mike

Post a reply to this message

"SharkD" <nomail@nomail> schreef in bericht 
news:web.493a9bd26f3c551ba19b81b0@news.povray.org...
>
> I wanted to do something like this for my Habitable Planet macro (see the 
> Object
> Collection). I also wanted to accurately add rivers and and other types of
> falling whater based on this information. However it is something that I 
> think
> lies beyond my abilities.
>

Well, you cannot access it now, but have you had a look at my Geomorph macro 
in the pov wiki? The name is unfortunately the same as the Linux Geomorph 
program. I think I was first.... ;-)

I used a large number of different functions to build isosurface landscape. 
I did it the hard way, now I would use much more the isoCSG set of macros, 
but I learned a lot about functions and the way to work with them.

Thomas

Post a reply to this message

"Thomas de Groot" <tDOTdegroot@interDOTnlANOTHERDOTnet> wrote:
> Well, you cannot access it now, but have you had a look at my Geomorph macro
> in the pov wiki? The name is unfortunately the same as the Linux Geomorph
> program. I think I was first.... ;-)

No, I must have missed that one. :)

-Mike

Post a reply to this message

"Thomas de Groot" <tDOTdegroot@interDOTnlANOTHERDOTnet> wrote:
> "Kirk Andrews" <kir### [at] tektonartcom> schreef in bericht
> news:web.493969b33e06781e198b63d90@news.povray.org...
> > I've struggled for a long time to try and achieve an erosion effect in POV
> > that
> > is the least bit convincing.  I'm still not there, but perhaps I've made
> > some
> > headway.  I've been using proximity functions to simulation a little
> > erosion.
> > Maybe you know some better ways?
> >
>
> Not easy overall. Until now, I have mainly used the f_ridge(),
> f_ridged_mf(), and f_hetero() functions as a basis for decent erosion
> features in isosurfaces. This is done also for instance by the program
> Leveller. However, that is not always convincing. Ideally, one would like to
> take a pre-defined landscape and apply erosion functions to it to wear it
> down and apply a sediment apron around standing ridges. That is basically
> what programs like GeoControl or World Machine are doing I believe, and very
> convincingly so.
>
> Thomas

I had never used f_ridged_mf().  I have almost always used some combination of
pigment patterns.  There certainly is some promise in it--thanks for
introducing it to me.

Just playing with it a bit:

Post a reply to this message

Attachments:
Download 'ridged02.jpg' (339 KB)

Preview of image 'ridged02.jpg'