POV-Ray : Newsgroups : povray.general : best way to plot 3D density field Server Time
7 Aug 2024 19:23:50 EDT (-0400)
  best way to plot 3D density field (Message 1 to 9 of 9)  
From: Dr  Paul Fons
Subject: best way to plot 3D density field
Date: 5 Aug 2001 19:00:46
Message: <pfons-0608010800520001@192.168.73.3>
Hi I am a newcomer to povray (just downloaded it and read some of the
documentation).  I was curious as to the best way to plot a three
dimensional set of experimental data (on a mesh).  Is it possible to plot
the data as a fog-like substance or an isosurface?  Are there any examples
out there I could bootstrap from.  If it matters, the data is actually a
wavefunction determined from tem structure factor data.  The idea is to
make beautiful graphics from the data and compare it previously existing
data (and show the improvements).  Thanks for any help you can offer. I
can be found at pau### [at] aistgojp.  Thanks in advance.

Dr. Paul Fons
National Institute for Advanced Industrial Science & Technology
Tsukuba, Japan


Post a reply to this message

From: Bill DeWitt
Subject: Re: best way to plot 3D density field
Date: 5 Aug 2001 19:28:49
Message: <3b6dd6b1$1@news.povray.org>
"Dr. Paul Fons" <pfo### [at] maccom> wrote :
>
> Hi I am a newcomer to povray (just downloaded it and read some of the
> documentation).

    You are about to have either some fun or some frustration.

> I was curious as to the best way to plot a three
> dimensional set of experimental data (on a mesh).

    Boy am I ever the wrong person to answer this question... so I won't.

> Is it possible to plot
> the data as a fog-like substance or an isosurface?

    Yes. This much I know for sure. Someone will be along in a hour or
twelve who can tell you how.

> Are there any examples
> out there I could bootstrap from.

    Probably. Again, keep checking back.

> If it matters, the data is actually a
> wavefunction determined from tem structure factor data.  The idea is to
> make beautiful graphics from the data and compare it previously existing
> data (and show the improvements).

    Yow.

> Thanks for any help you can offer. I
> can be found at pau### [at] aistgojp.  Thanks in advance.

    Hope this helps... I know I hate it when I have to wonder if anyone will
ever answer a question.


Post a reply to this message

From: Andrea Ryan
Subject: Re: best way to plot 3D density field
Date: 5 Aug 2001 21:58:21
Message: <3B6DF7DC.2CC33041@global2000.net>
Maybe a heightfield would work.  There is an example in the tutorial
section of the documentation.  If you can format your data as an image
of varying color, you could use it for a heightfield.  Heightfields are
just special meshes that take the vertical positions of their triangles
from the image file.  The documentation has some details.

If you data is in a text file, you could try using the file functions
like #fopen and #read to access it and use it as the heights of
cylinders or boxes arranged in an array.  I have seen something like
this on http://www.webelements.com/ .  They have images that use
cylinders and boxes of different heights to show how properties of
chemical elements differ.

Brendan Ryan

"Dr. Paul Fons" wrote:
> 
> Hi I am a newcomer to povray (just downloaded it and read some of the
> documentation).  I was curious as to the best way to plot a three
> dimensional set of experimental data (on a mesh).  Is it possible to plot
> the data as a fog-like substance or an isosurface?  Are there any examples
> out there I could bootstrap from.  If it matters, the data is actually a
> wavefunction determined from tem structure factor data.  The idea is to
> make beautiful graphics from the data and compare it previously existing
> data (and show the improvements).  Thanks for any help you can offer. I
> can be found at pau### [at] aistgojp.  Thanks in advance.
> 
> Dr. Paul Fons
> National Institute for Advanced Industrial Science & Technology
> Tsukuba, Japan


Post a reply to this message

From: Dr  paul Fons
Subject: Re: best way to plot 3D density field
Date: 6 Aug 2001 02:24:12
Message: <pfons-0608011524110001@etld1220.a02.aist.go.jp>
In article <3B6DF7DC.2CC33041@global2000.net>, Andrea Ryan
<ary### [at] global2000net> wrote:

> Maybe a heightfield would work.  There is an example in the tutorial
> section of the documentation.  If you can format your data as an image
> of varying color, you could use it for a heightfield.  Heightfields are
> just special meshes that take the vertical positions of their triangles
> from the image file.  The documentation has some details.
> 
> If you data is in a text file, you could try using the file functions
> like #fopen and #read to access it and use it as the heights of
> cylinders or boxes arranged in an array.  I have seen something like
> this on http://www.webelements.com/ .  They have images that use
> cylinders and boxes of different heights to show how properties of
> chemical elements differ.
> 
> Brendan Ryan
> 

 Pardon my ignorance, but a heightfield would seem to imply a
two-dimensional surface with an array of height data e.g. fz(x,y).  What I
have is actually a four dimensional set of data Fintensity(x,y,z). 
(actually the data is complex so the data is five dimensional)  For the
moment I just want to plot the intensity (perhaps the phase as color
later) for a mesh of points in three-space.  I have just played around
with a glow demo in MacMegaPOV and it seems like a possible solution -
next would be to set up a camera and stroll around the object to generate
an animation (e.g. quicktime or such).  Another possible solution would
seem to be to generate an isosurface from the numerical data. This way the
symmetry of the data would be clearer - perhaps a combination of two such
techniques would be best.  Someone must have done something like this
before and have an opinion -- e.g. plotting temperature data in three
space or some such.  Any opinions are more than welcome.  I definately do
*not* know what I am doing with POV so any opinions are most welcome.
         Paul Fons
         pfo### [at] maccom
National Institute for Advanced Industrial Science & Technology


Post a reply to this message

From: Adrien Beau
Subject: Re: best way to plot 3D density field
Date: 6 Aug 2001 04:43:46
Message: <3B6E58C6.DBD1696F@sycomore.fr>
I am not into that kind of renders, but here are some ideas that
seem feasible. It's up to you to conduct further researches.

At a very simply level, you could define a lot of spheres from
your data. You could make their radius, or their color or texture,
be the F(x,y,z) you are trying to plot. Spheres are the simplest
objects POV handles. Note that you can deform them (using scale,
rotate, translate), giving them an elongated form which could be
suited to represent some derivative information (flow direction,
etc.).

You could also perhaps replace them by blobs, where you can
actually code the intensity of your data as the strength of
the blob. But blobs merge together, and that might not be the
kind of display you want.

Finally, if you want more of an ethereal substance you can
travel around and into, I think media is what you need.
Especially, POV has a density_file type of pattern. It uses
a file that contains a 3D bitmap pattern. At each voxel you
can specify an density value. You could then create a box,
fill it with the media, and prepare for time-consuming renders.

I do not know megaPOV and such patches, so they might have
some even nicer feature.

So, you might want to read the help file (and search these
newsgroups) about

	sphere blob media density_file

Hope this helps.

-- 
Adrien Beau - adr### [at] freefr - http://adrien.beau.free.fr
 Mes propos n'engagent que moi et en aucun cas mes employeurs


Post a reply to this message

From: Michael Andrews
Subject: Re: best way to plot 3D density field
Date: 6 Aug 2001 07:08:07
Message: <3B6E7BF4.DF0BC1E@reading.ac.uk>
Hi Paul,

The best place to start would be to look at the df3 demo files. With the
Windows MegaPoV version they come under the demos\df3 directory. The
'make_df3.pov' creates a .df3 file from a series of slices through a
texture, 'use_df3.pov' takes the created .df3 and uses it as a media
density pattern.

The standard POVRay help file describes the .df3 file format in the
'Density_File' section under 'Patterns'; it is a very simple format so
you should be able to convert your TEM data.

Hope this helps a little,
	Mike Andrews.

"Dr. Paul Fons" wrote:
> 
> Hi I am a newcomer to povray (just downloaded it and read some of the
> documentation).  I was curious as to the best way to plot a three
> dimensional set of experimental data (on a mesh).  Is it possible to plot
> the data as a fog-like substance or an isosurface?  Are there any examples
> out there I could bootstrap from.  If it matters, the data is actually a
> wavefunction determined from tem structure factor data.  The idea is to
> make beautiful graphics from the data and compare it previously existing
> data (and show the improvements).  Thanks for any help you can offer. I
> can be found at pau### [at] aistgojp.  Thanks in advance.
> 
> Dr. Paul Fons
> National Institute for Advanced Industrial Science & Technology
> Tsukuba, Japan


Post a reply to this message

From: Jamie Davison
Subject: Re: best way to plot 3D density field
Date: 6 Aug 2001 10:23:19
Message: <MPG.15d8b9476fbc091c9899b5@news.povray.org>
On Mon, 06 Aug 2001 08:00:24 +0900, Dr. Paul Fons wrote...
> Hi I am a newcomer to povray (just downloaded it and read some of the
> documentation).  I was curious as to the best way to plot a three
> dimensional set of experimental data (on a mesh).  Is it possible to plot
> the data as a fog-like substance or an isosurface?  Are there any examples
> out there I could bootstrap from.  If it matters, the data is actually a
> wavefunction determined from tem structure factor data.  The idea is to
> make beautiful graphics from the data and compare it previously existing
> data (and show the improvements).  Thanks for any help you can offer. I
> can be found at pau### [at] aistgojp.  Thanks in advance.

You might want to look at the df3 section of the documentation.

More specifically, the Density_file section in the media area of the 
documents.

And then, I assume, ask here again if that's not what you're after :)

Bye for now,
     Jamie.


Post a reply to this message

From: Mark James Lewin
Subject: Re: best way to plot 3D density field
Date: 6 Aug 2001 18:00:57
Message: <3B6F1148.9733FE8B@yahoo.com.au>
If you would like to "plot" a fog, and have the data as ascii, you could try
df3Maker at http://www.geocities.com/m_j_lewin/index.html. It can generate df3
files that others here have mentioned, but you do not have to generate any
extra textures etc. The downside is that you have to use Windows. The site has
some info on using df3Maker, and some examples as well.

MJL


Post a reply to this message

From: Tom Melly
Subject: Re: best way to plot 3D density field
Date: 8 Aug 2001 08:56:16
Message: <3b7136f0@news.povray.org>
"Dr. Paul Fons" <pfo### [at] maccom> wrote in message
news:pfo### [at] 19216873.3...

<snip>

Any chance of you posting some of the data so we can play with it?


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.