Just made some tests with an Erosion routine I programmed some time ago.  The
result has some similarity with ridged multifractal but it is an editing
function modifying any existing heightfield.  

And it is quite fast (about 800 milliseconds for a single run on a 256x256
point heightfield on my K6/500)

The first one is the starting heightfield, the second one after about 15 runs
with smoothing between, the third after 15 additional runs.  

Please tell me what you think of it, I wonder if anybody has done something
similar before. 

I am already planning some addition to stop erosion process, when a certain
slope is reached, which would be more realistic.

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Attachments:
Download 'erode1.jpg' (21 KB) Download 'erode2.jpg' (20 KB) Download 'erode3.jpg' (20 KB)

Preview of image 'erode1.jpg'

Preview of image 'erode2.jpg'

Preview of image 'erode3.jpg'

Hmm, they look more like they're in reverse order. Tall, pointy peaks should get
rounded off.

--
David Fontaine     <dav### [at] faricynet>     ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/

Post a reply to this message

David Fontaine wrote:
> 
> Hmm, they look more like they're in reverse order. Tall, pointy peaks should get
> rounded off.
> 

no, they are in right order :-)

Have you ever looked at a mountain region from above: most mountains have quite
plane sides and very sharp ridges.

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

In article <396A1D01.C701A4CD@schunter.etc.tu-bs.de>, 
chr### [at] gmxde wrote:

> Have you ever looked at a mountain region from above: most mountains 
> have quite plane sides and very sharp ridges.

Yes, but they eventually get smoothed over by time and weather. Sharp 
points and steep sides shear off and the debris accumulates at the 
bottom, and the mountain becomes more hill-like and rounded.
Your filter looks more like a "time machine", reversing the normal 
erosion process.

-- 
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

i think it depends on whether it is wind erosion or water erosion.  wind
rounds, water peaks.

if you could define areas most and least likely to erode, that would be
cool.  for example, areas with a lot of ground cover don't erode as fast
as open areas.  dirt erodes faster than rock, etc.

Post a reply to this message

Chris Huff wrote:
> 
[...]
> 
> Yes, but they eventually get smoothed over by time and weather. Sharp
> points and steep sides shear off and the debris accumulates at the
> bottom, and the mountain becomes more hill-like and rounded.
> Your filter looks more like a "time machine", reversing the normal
> erosion process.
> 

It depends on the material, of course hillsides do not get infinitely steep, but
it can be quite extreme like in the attached photo.

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Attachments:
Download 'mount1.jpg' (8 KB)

Preview of image 'mount1.jpg'

"Christoph Hormann" <Chr### [at] schunteretctu-bsde> wrote :
>
> Have you ever looked at a mountain region from above: most mountains have
quite
> plane sides and very sharp ridges.

    Some mountains look sharp because they have been recently pushed up.
Even old mountains look that way if they are still being pushed up. But old
mountains in areas with no recent plate activity are rounded by erosion.
Your photo is clearly one of an upthrust formation of some sort. In a
billion years it will be a rounded boulder.

--
I get paid to do -anything- while connected to the internet.
Please use my referral id to find out more
http://www.getpaid4.com/?billdewitt

Post a reply to this message

ryan constantine wrote:
> 
> i think it depends on whether it is wind erosion or water erosion.  wind
> rounds, water peaks.
> 

IMO this type of erosion is mainly caused by the changing temperatures
destroying the structure of the material at the surface.  Of course water and
wind are major factors for transportation of material.

> if you could define areas most and least likely to erode, that would be
> cool.  for example, areas with a lot of ground cover don't erode as fast
> as open areas.  dirt erodes faster than rock, etc.

That's a good idea, it is only a problem how to define those areas.  

BTW, my code partly already does this, because it erodes more on steeper areas,
where there is normally fewer vegetation.  

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Can a routine be made to do wind erosion?  Desert landscapes could be made with
these mushroom shaped rocks that get eroded more at the bottom.  However, I'm not
sure if that can be done with height fields.
Brendan

Christoph Hormann wrote:

> Just made some tests with an Erosion routine I programmed some time ago.  The
> result has some similarity with ridged multifractal but it is an editing
> function modifying any existing heightfield.
>
> And it is quite fast (about 800 milliseconds for a single run on a 256x256
> point heightfield on my K6/500)
>
> The first one is the starting heightfield, the second one after about 15 runs
> with smoothing between, the third after 15 additional runs.
>
> Please tell me what you think of it, I wonder if anybody has done something
> similar before.
>
> I am already planning some addition to stop erosion process, when a certain
> slope is reached, which would be more realistic.
>
> Christoph
>
> --
> Christoph Hormann <chr### [at] gmxde>
> Homepage: http://www.schunter.etc.tu-bs.de/~chris/
>
>   ------------------------------------------------------------------------
>  [Image]  [Image]  [Image]

Post a reply to this message

Bill DeWitt  wrote:
> 
[...]
> 
>     Some mountains look sharp because they have been recently pushed up.
> Even old mountains look that way if they are still being pushed up. But old
> mountains in areas with no recent plate activity are rounded by erosion.
> Your photo is clearly one of an upthrust formation of some sort. In a
> billion years it will be a rounded boulder.
> 

Agreed, but on the other hand, the sharp structures often correspond with the
river network, which suggests, that they are also partly formed by erosion.  

The photo is not a very good example, very good are usually satellite pictures
where you can see the sharp edges in the light *and* the rivers.  

just found one (sorry for the many pictures), well in fact it's a radar image,
but it should work...

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Attachments:
Download 'mount2.jpg' (44 KB)

Preview of image 'mount2.jpg'

have you seen any of the tutorials and utilities made for positioning
items on a height field?  i think there is something out there that can
define areas of height fields to be worked on.  i think it is genesis
toolkit found at www.rhein-ruhr.de/~schrammel/index.htm.  i'm not sure
it's usefull, but there you go.

Christoph Hormann wrote:
> 
> ryan constantine wrote:
> >
> > i think it depends on whether it is wind erosion or water erosion.  wind
> > rounds, water peaks.
> >
> 
> IMO this type of erosion is mainly caused by the changing temperatures
> destroying the structure of the material at the surface.  Of course water and
> wind are major factors for transportation of material.
> 
> > if you could define areas most and least likely to erode, that would be
> > cool.  for example, areas with a lot of ground cover don't erode as fast
> > as open areas.  dirt erodes faster than rock, etc.
> 
> That's a good idea, it is only a problem how to define those areas.
> 
> BTW, my code partly already does this, because it erodes more on steeper areas,
> where there is normally fewer vegetation.
> 
> Christoph
> 
> --
> Christoph Hormann <chr### [at] gmxde>
> Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

> That's a good idea, it is only a problem how to define those areas.

Could you perhaps allow the user to provide an erosion_map for it?

Post a reply to this message

Christoph Hormann wrote:
> 
> Just made some tests with an Erosion routine I programmed some time ago.  The
> result has some similarity with ridged multifractal but it is an editing
> function modifying any existing heightfield.
> ...

Interesting.

Is it a POV-script doing this ?

And if so, then how is it done ?


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

Post a reply to this message

Maybe an animation would help illustrate your technique? (O.k., so I just
like animations a lot... :)

Regards,
Dave

"Christoph Hormann" <Chr### [at] schunteretctu-bsde> wrote in
message news:396A0D1D.2B382007@schunter.etc.tu-bs.de...
>
> Just made some tests with an Erosion routine I programmed some time ago.
The
> result has some similarity with ridged multifractal but it is an editing
> function modifying any existing heightfield.
>
> And it is quite fast (about 800 milliseconds for a single run on a 256x256
> point heightfield on my K6/500)
>
> The first one is the starting heightfield, the second one after about 15
runs
> with smoothing between, the third after 15 additional runs.
>
> Please tell me what you think of it, I wonder if anybody has done
something
> similar before.
>
> I am already planning some addition to stop erosion process, when a
certain
> slope is reached, which would be more realistic.
>
> Christoph
>
> --
> Christoph Hormann <chr### [at] gmxde>
> Homepage: http://www.schunter.etc.tu-bs.de/~chris/


----------------------------------------------------------------------------
----






----------------------------------------------------------------------------
----






----------------------------------------------------------------------------
----

Post a reply to this message

Christoph Hormann wrote in message
<396A32E4.CF4A5F18@schunter.etc.tu-bs.de>...
>That's a good idea, it is only a problem how to define those areas.


Allow a user to provide a second greyscale image that specifies the
"hardness" of parts of the map.

Mark

Post a reply to this message

Andrea Ryan wrote in message <396A333F.8A101D00@global2000.net>...
>Can a routine be made to do wind erosion?  Desert landscapes could be made
with
>these mushroom shaped rocks that get eroded more at the bottom.  However,
I'm not
>sure if that can be done with height fields.
>Brendan


It can't be done with a heightfield, but a .df3-based isosurface would work
quite well for this.

Mark

Post a reply to this message

Dave Blandston wrote:
> 
> Maybe an animation would help illustrate your technique? (O.k., so I just
> like animations a lot... :)
> 
> Regards,
> Dave
> 

Never did an animation before :-) so i will have to learn it first...
NTL, it is a good idea.

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Tor Olav Kristensen wrote:
> 
[...]
> 
> Interesting.
> 
> Is it a POV-script doing this ?
> 
> And if so, then how is it done ?
> 
> Tor Olav

No, it's a delphi program.

BTW, that leads to an interesting idea for a new feature:

   height_field {
      pov {
         "hf.pov" 
         width=512
         height=512 
      }
   }

   or similar...

Recursive implementation could occupy povray forever :-)

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Mark Wagner wrote:
> 
[...]
> 
> Allow a user to provide a second greyscale image that specifies the
> "hardness" of parts of the map.
> 
> Mark

That's already planned (it's already possible to specify a B/W mask for total
exclusion of certain areas).  

The problem is how to generate that map, because in real world the amount of
erosion is influenced by a lot of factors like:

   -rock material
   -water amount
   -slope
   -hillside orientation
   -most often wind directions
   -existing vegetation
   -...

Some of these factors can be calculated from the heightfield, some have to be
specified by the user, but for example not every user drawn "geological map" is
possible, it is not reasonable if the material changes from granite to sand just
at the top of a mountain.

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

That's what I said.

Post a reply to this message

Ugh, forget I said anything...

--
David Fontaine     <dav### [at] faricynet>     ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/

Post a reply to this message

Just put some more samples on my web page, so if anybody is interested:

http://www.schunter.etc.tu-bs.de/~chris/erosion1.html

I also included an "inverse version" with rounded hills, it does not look as bad
as i thought :-)  Maybe I could try out some mixtures...


Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

That would be a relatively young mountain in your photo.

Christoph Hormann wrote:

> Chris Huff wrote:
> >
> [...]
> >
> > Yes, but they eventually get smoothed over by time and weather. Sharp
> > points and steep sides shear off and the debris accumulates at the
> > bottom, and the mountain becomes more hill-like and rounded.
> > Your filter looks more like a "time machine", reversing the normal
> > erosion process.
> >
>
> It depends on the material, of course hillsides do not get infinitely steep, but
> it can be quite extreme like in the attached photo.
>
> Christoph
>
> --
> Christoph Hormann <chr### [at] gmxde>
> Homepage: http://www.schunter.etc.tu-bs.de/~chris/
>
>   ------------------------------------------------------------------------
>  [Image]

--
Come visit my web site:-) : http://www.geocities.com/~thomaslake/

Post a reply to this message

Christoph Hormann wrote:

> It depends on the material, of course hillsides do not get infinitely steep, but
> it can be quite extreme like in the attached photo.

That is an example of water erosion, not wind. I have attached a photo from Glacier
Park to demonstrate...

--
David Fontaine     <dav### [at] faricynet>     ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/

Post a reply to this message

Attachments:
Download 'glacierpark.jpg' (43 KB)

Preview of image 'glacierpark.jpg'

"David Fontaine" <dav### [at] faricynet> wrote in message
news:398DDC5C.2F84F85E@faricy.net...
| Christoph Hormann wrote:
|
| > It depends on the material, of course hillsides do not get infinitely
steep, but
| > it can be quite extreme like in the attached photo.
|
| That is an example of water erosion, not wind. I have attached a photo
from Glacier
| Park to demonstrate...

I had said (and someone else was before me) about how the material being
carried away would make a lot of difference.  Good example of that.
Come on now, someone could render a copy of this photo can they not?  No
image mapping allowed though.

Bob

Post a reply to this message

Bob Hughes wrote:

> Come on now, someone could render a copy of this photo can they not?  No
> image mapping allowed though.

Someone would be quick to point out that the cloud in the center was
not realistic in appearance. Me for example :)

-- 
Ken Tyler - 1400+ POV-Ray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/

Post a reply to this message

Ken wrote:

> Someone would be quick to point out that the cloud in the center was
> not realistic in appearance. Me for example :)

Yes, you are correct. It only took 28 days, too (look at the post times). I am
immensely proud of my accomplishment. If you look closely, you will also
notice I modelled and very accurately textured some helicopter-window
plexiglass right in front of the camera. Come on, if Gilles can make Wet Bird,
who's to say I can't make this? :-) Of course, I owe a lot of thanks to Nathan
for all the great MegaPOV features.

--
David Fontaine     <dav### [at] faricynet>     ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/

Post a reply to this message

David Fontaine wrote:
> 
> That is an example of water erosion, not wind. I have attached a photo from Glacier
> Park to demonstrate...
> 

Your pict. shows a lot of different aspects of erosion: on the lower parts,
there is water erosion in "soft materials".  I don't see much wind erosion
anywhere.  

Most erosion effects on massive mountains IMO are "thermal erosion", meaning the
destruction of the rock structure due to temperature changes.  Of course, water
and wind influence these effects in certain ways.  

I think both photos show very much of this type of erosion, but it is not easy
to simulate, because it much depends on the internal structure of the rock.  

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message