 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I've found tons of great heightmaps for planets, and moons at
http://bang.lanl.gov/solarsys/eng/homepage.htm
Since I've been playing around with planets lately, I'd love to use some
of these, as heightfields, or bumpmaps for spheres.
Problem: It uses a weird colour scale, that does not convert to
greyscale properly.
Any advice on using these?
(BTW: Look under Mars or Venus for good high res examples)
Simon
http://home.istar.ca/~sdevet
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Simon de Vet wrote:
> I've found tons of great heightmaps for planets, and moons at
> http://bang.lanl.gov/solarsys/eng/homepage.htm
>
> Since I've been playing around with planets lately, I'd love to use some
> of these, as heightfields, or bumpmaps for spheres.
>
> Problem: It uses a weird colour scale, that does not convert to
> greyscale properly.
>
> Any advice on using these?
>
> (BTW: Look under Mars or Venus for good high res examples)
>
> Simon
> http://home.istar.ca/~sdevet
Isn't there some palette included on this great page of yours? If not, you
could ask the creator for it. He should thank you for the advice.
JK
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
It looks to me like these are not pure height fields, but rather, height
field renderings. If you look closely, you can see shadows around the
features, probably created by embossing the original height field data to
give this 3-D effect. If you try using these as height fields, you'll
probably be disappointed with the results.
Simon de Vet wrote in message <35E### [at] istar ca>...
>I've found tons of great heightmaps for planets, and moons at
>http://bang.lanl.gov/solarsys/eng/homepage.htm
>
>Since I've been playing around with planets lately, I'd love to use some
>of these, as heightfields, or bumpmaps for spheres.
>
>Problem: It uses a weird colour scale, that does not convert to
>greyscale properly.
>
>Any advice on using these?
>
>(BTW: Look under Mars or Venus for good high res examples)
>
>Simon
>http://home.istar.ca/~sdevet
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
David Cook wrote:
> It looks to me like these are not pure height fields, but rather, height
> field renderings. If you look closely, you can see shadows around the
> features, probably created by embossing the original height field data to
> give this 3-D effect. If you try using these as height fields, you'll
> probably be disappointed with the results.
Still, they're a lot better than what I have right now.
For the record, what I have right now is absolutely nothing.
See what I mean? I'm not interested in making a topographical globe of Venus
to see door to door, I just want to play around with some heightfields based
on actual places, and these looked the most interesting....
Simon
http://home.istar.ca/~sdevet
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I should have said that you'd probably be better off using these as texture
maps than as height fields. The only drawback with this method (if this is
suitable for your application) is that since the renderings are embossings
of the flattened planet surface, when you wrap it around a sphere the
surface features shadows are inconsistent with the way the lighting should
follow the curvature of the planet.
However, if by "door to door" you mean that you're developing a VR model,
then texture mapping won't give you the elevation data you're looking for.
Have you tried contacting Calvin J. Hamilton, the site author, to ask him
where the original topo data that he used to make his images can be
acquired?
BTW, here's another site that has some good planetary maps:
http://maps.jpl.nasa.gov/
Hope this helps,
-- Dave
Simon de Vet wrote in message <35E41F6C.80D24035@istar.ca>...
>
>
>David Cook wrote:
>
>> It looks to me like these are not pure height fields, but rather, height
>> field renderings. If you look closely, you can see shadows around the
>> features, probably created by embossing the original height field data to
>> give this 3-D effect. If you try using these as height fields, you'll
>> probably be disappointed with the results.
>
>Still, they're a lot better than what I have right now.
>
>For the record, what I have right now is absolutely nothing.
>
>See what I mean? I'm not interested in making a topographical globe of
Venus
>to see door to door, I just want to play around with some heightfields
based
>on actual places, and these looked the most interesting....
>
>Simon
>http://home.istar.ca/~sdevet
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
David Cook wrote:
> I should have said that you'd probably be better off using these as texture
> maps than as height fields. The only drawback with this method (if this is
> suitable for your application) is that since the renderings are embossings
> of the flattened planet surface, when you wrap it around a sphere the
> surface features shadows are inconsistent with the way the lighting should
> follow the curvature of the planet.
I know.. I already have some image maps of planets, and I've already used them
in scenes.... And they have nicer colours than these :)
> However, if by "door to door" you mean that you're developing a VR model,
> then texture mapping won't give you the elevation data you're looking for.
> Have you tried contacting Calvin J. Hamilton, the site author, to ask him
> where the original topo data that he used to make his images can be
> acquired?
Well, by saying "see them door to door" (and that was a typo... I meant 'sell'
*sigh*) I was just baing a bit sarcastic. All I meant was that I didn't need
anything super scientificly accurate here, just something to play around with
that was not just generated at random....
> BTW, here's another site that has some good planetary maps:
> http://maps.jpl.nasa.gov/
I know. Love ths site! I made some custom planets using their maps, and a lot of
tweaking.
They're at:
http://home.istar.ca/~sdevet/Planet1.jpg
http://home.istar.ca/~sdevet/Planet2.jpg
http://home.istar.ca/~sdevet/Planet3.jpg
If anyone's interested.. just simple, silly little things.... but I like it.
I wanted the heightfields so I could render them with a waterlevel set, to
create realistic continents/oceans for custom planets, instead of relying on
flukes in the images to do the same..... that's all... :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi Simon and David,
I noticed your thread on planetary renderings, and I'm
interested in this, too. I have a friends who's writing a
game and he asked me to render some planets for him.
I was looking through _The Science of Fractal Images_ by
Pietgen and Saupe, and In the color plates I saw a cosine-offset
surface generation method used by R.F. Voss. I looked at that
for a while and realized I could code something like that in
Perl and make a pgm output file and use that as a bump_map
on a sphere.
However, when I try to render it, I don't see any bumps
on the sphere. Do you see anything erroneous about this
bump map statement?
#declare PLANET=sphere { <0, 0, 0>, 1
normal {
bump_map {
pgm "surfaces/jbrp.1.pgm"
bump_size 10.0
map_type 2
}
}
pigment {
Gray
//image_map {
// pgm "surfaces/jbrp.0.pgm"
// map_type 1
// interpolate 2
//}
}
scale 6
translate <0, 4, 0>
}
If that is correct, could you suggest anyother reason
I might not be able to display bumps on the sphere?
Attached is my surface map. Would you mind taking a look
at it?
Much appreciated
--jed
Simon de Vet wrote:
>
> I've found tons of great heightmaps for planets, and moons at
> http://bang.lanl.gov/solarsys/eng/homepage.htm
--
# Jed Reynolds, mailto:jed### [at] surfsoftcom http://www.surfsoft.com/~jed
# "...I've seen the old geeks talk to the young geeks, and they speak
# the same language. There's no age gap or anything." --Steve Wozniak
Post a reply to this message
Attachments:
Download 'us-ascii' (177 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
From what I see you are making you bump map to
big for the sphere size your using. Try a range between
0 - 1 to start with. Try the following and see if it makes
a difference:
normal {
bump_map {
pgm "surfaces/jbrp.1.pgm"
bump_size 0.5
map_type 1 }
// scale <0,0,0,> // try adding scaling here
}
Also I believe the map_type 2 is cylindrical and not spherical.
Check the docs. The image you provided looks like it should
do something even if it's wrong. When I get off line I'll
take a closer look at it and will let you know if I find anything
that wll work for you.
Please in the future post images like this in one of the
binaries groups. There are several people on this NG that
pay for their internet connection on a time basis and can't
afford to wait for your message to load. One of the Pov Team
members, who frequents this group, points this out regularly.
Ken Tyler
Jed Reynolds wrote:
> Hi Simon and David,
>
> I noticed your thread on planetary renderings, and I'm
> interested in this, too. I have a friends who's writing a
> game and he asked me to render some planets for him.
>
> I was looking through _The Science of Fractal Images_ by
> Pietgen and Saupe, and In the color plates I saw a cosine-offset
> surface generation method used by R.F. Voss. I looked at that
> for a while and realized I could code something like that in
> Perl and make a pgm output file and use that as a bump_map
> on a sphere.
>
> However, when I try to render it, I don't see any bumps
> on the sphere. Do you see anything erroneous about this
> bump map statement?
>
> #declare PLANET=sphere { <0, 0, 0>, 1
>
> normal {
> bump_map {
> pgm "surfaces/jbrp.1.pgm"
> bump_size 10.0
> map_type 2
> }
> }
>
> pigment {
> Gray
> //image_map {
> // pgm "surfaces/jbrp.0.pgm"
> // map_type 1
> // interpolate 2
> //}
> }
> scale 6
> translate <0, 4, 0>
> }
>
> If that is correct, could you suggest anyother reason
> I might not be able to display bumps on the sphere?
>
> Attached is my surface map. Would you mind taking a look
> at it?
>
> Much appreciated
>
> --jed
Post a reply to this message
|
|
| |
| |
|
|
From: Dan Connelly
Subject: Re: bump_maps [Re: Planetary Heightfields: Now what?]
Date: 26 Sep 1998 22:05:57
Message: <360D8F65.EAFCBFDC@flash.net>
|
|
|
| |
| |
|
|
I had no problem seeing features.... but I suggest
the following changes :
1. The map field is defined from y=0 to y=1. You want
y=-1 to y=+1. Therefore you need to apply a translation
and a scale.
2. specify a finish.
3. Your filename is different than the one you posted (jbrp.0.pgm).
This may be an issue.
4. "interpolate 2" should almost always be used with bump maps
to prevent pixellization artifacts.
Consider the following scene :
sphere { <0, 0, 0>, 1
normal {
bump_map {
pgm "c:\tmp\jbrp.0.pgm" // adjust this line
bump_size 10.0
map_type 2
interpolate 2 // this is important
}
translate -0.5 * y
scale 2 * y
}
pigment {
color rgb 0.5
}
finish {
ambient 0.2 diffuse 0.8 specular 0.25 roughness 0.02
}
}
camera {
location -3 * z
look_at 0
}
light_source {
<50, 50, -100>
color rgb 1
}
Jed Reynolds wrote:
>
> However, when I try to render it, I don't see any bumps
> on the sphere. Do you see anything erroneous about this
> bump map statement?
>
> #declare PLANET=sphere { <0, 0, 0>, 1
>
> normal {
> bump_map {
> pgm "surfaces/jbrp.1.pgm"
> bump_size 10.0
> map_type 2
> }
> }
>
> pigment {
> Gray
> //image_map {
> // pgm "surfaces/jbrp.0.pgm"
> // map_type 1
> // interpolate 2
> //}
> }
> scale 6
> translate <0, 4, 0>
> }
--
http://www.flash.net/~djconnel/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I took the liberty of playing around with it for a while
and I think you will like what I have come up with.
Anyway take a look and see what you think.
I used the bump map image as an image map to add
highlighting. The filtered white layer on top was needed
becasue the pgm file had too many dark areas to get
much surface definition (try removing it and you'll see
what I mean). The filtering lets the underlying surface
show through and really adds to the appearance of depth.
I still don't know if this is what you wanted the final image
to look like but it's definately interesting looking.
camera{location z*-3 look_at 0}
light_source{<0,30,-3>rgb 1}
sphere{<0,0,0>,1
normal{
bump_map{pgm "surfaces/jbrp.1.pgm"
bump_size 50 map_type 1 interpolate 2}
translate -.5 scale 1}
pigment{
image_map{pgm "surfaces/jbrp.1.pgm"
map_type 1 interpolate 2}
translate -.5 scale 1}
pigment{rgb 1 filter .8}
finish{ambient .3 diffuse 1}
}
Ken Tyler
Jed Reynolds wrote:
> Hi Simon and David,
>
> I noticed your thread on planetary renderings, and I'm
> interested in this, too. I have a friends who's writing a
> game and he asked me to render some planets for him.
>
> I was looking through _The Science of Fractal Images_ by
> Pietgen and Saupe, and In the color plates I saw a cosine-offset
> surface generation method used by R.F. Voss. I looked at that
> for a while and realized I could code something like that in
> Perl and make a pgm output file and use that as a bump_map
> on a sphere.
>
> However, when I try to render it, I don't see any bumps
> on the sphere. Do you see anything erroneous about this
> bump map statement?
>
> #declare PLANET=sphere { <0, 0, 0>, 1
>
> normal {
> bump_map {
> pgm "surfaces/jbrp.1.pgm"
> bump_size 10.0
> map_type 2
> }
> }
>
> pigment {
> Gray
> //image_map {
> // pgm "surfaces/jbrp.0.pgm"
> // map_type 1
> // interpolate 2
> //}
> }
> scale 6
> translate <0, 4, 0>
> }
>
> If that is correct, could you suggest anyother reason
> I might not be able to display bumps on the sphere?
>
> Attached is my surface map. Would you mind taking a look
> at it?
>
> Much appreciated
>
> --jed
>
> Simon de Vet wrote:
> >
> > I've found tons of great heightmaps for planets, and moons at
> > http://bang.lanl.gov/solarsys/eng/homepage.htm
>
> --
> # Jed Reynolds, mailto:jed### [at] surfsoftcom http://www.surfsoft.com/~jed
> # "...I've seen the old geeks talk to the young geeks, and they speak
> # the same language. There's no age gap or anything." --Steve Wozniak
>
> --------------------------------------------------------------------------------
>
> Name: jbrp.0.pgm
> jbrp.0.pgm Type: PGM Image (image/x-portable-graymap)
> Encoding: 7bit
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
