My first test, using a modified version of Gilles Tran's cloud macro, 
then tga2df3.exe, and a Poser head. Note that I first closed /all/ the 
holes in the mesh2 object before using it. This is absolutely essential.

99 slices.

Not bad at all.

Thomas

Post a reply to this message

Attachments:
Download 'gt_df3_test.png' (71 KB)

Preview of image 'gt_df3_test.png'

Now, the previous image was based on a df3 generation without focal blur 
(Aperture parameter in Gilles Tran's macro). With a bit of focal bur 
(Aperture 1) you get a more fuzzy image like this.

More transformations can be obtained by the use of warp {turbulence} 
within the density block of the media but be aware that the basic shape 
of the object is soon lost, and that may not be the intention.

Thomas

Post a reply to this message

Attachments:
Download 'gt_df3_test.png' (61 KB)

Preview of image 'gt_df3_test.png'

On 09/23/2013 10:53 AM, Thomas de Groot wrote:
> Now, the previous image was based on a df3 generation without focal blur
> (Aperture parameter in Gilles Tran's macro). With a bit of focal bur
> (Aperture 1) you get a more fuzzy image like this.
>
> More transformations can be obtained by the use of warp {turbulence}
> within the density block of the media but be aware that the basic shape
> of the object is soon lost, and that may not be the intention.
>
> Thomas
And you completely took away our saying we are green with envy...

These look great and are much more detailed with respect to the head 
shape than what I did with the ghost image. As to your last statement 
about warp { turbulence }, remember Jamie's trick to scale turbulence 
with respect to an object/pattern. It is a trick I used in the ghost 
image. Namely:

// Credit: Jaime Vives Piqueres for warp turbulence scaling. Tierra zip. 
www.ignorancia.org.

#macro WarpShapesA_2 ()
     scale 1.0/0.9000
     warp {
         turbulence 0.1500
         octaves    6
         lambda     2.0000
         omega      0.6500
     }
     scale 0.9
#end

So to have a lot of scaled down, but heavier, turbulence, you first 
scale the "thing" up, apply the turbulence warp, then scale everything 
back to the original size.

Bill P.

Post a reply to this message

Thomas de Groot <tho### [at] degrootorg> wrote:
> Now, the previous image was based on a df3 generation without focal blur
> (Aperture parameter in Gilles Tran's macro). With a bit of focal bur
> (Aperture 1) you get a more fuzzy image like this.
>
> More transformations can be obtained by the use of warp {turbulence}
> within the density block of the media but be aware that the basic shape
> of the object is soon lost, and that may not be the intention.

Cool!

I'm concerned that the focal blur only blurs the volume along 2 axes.

Also, adding turbulence can result in very slow renders. Sometimes it's possible
to hide a granite pattern in a volume-based pigment_pattern instead... this is
usually faster than using turbulence. Or, you can try to get away with fewer
turbulence octaves... but of course I'm always trying to maximize my PPS :)

Sam

Post a reply to this message

On 23-9-2013 17:11, William F Pokorny wrote:
> And you completely took away our saying we are green with envy...

LOL

>
> These look great and are much more detailed with respect to the head
> shape than what I did with the ghost image. As to your last statement
> about warp { turbulence }, remember Jamie's trick to scale turbulence
> with respect to an object/pattern. It is a trick I used in the ghost
> image. Namely:

Yes, I knew that. It is a trick I use quite often indeed. Didn't 
remember that it came from Jaime ;-)

Obviously there are more things that can be achieved, so I go on 
exploring...

Thomas

Post a reply to this message

On 23-9-2013 18:52, Samuel Benge wrote:
> Cool!

<grin>

>
> I'm concerned that the focal blur only blurs the volume along 2 axes.

Really? I don't think so. When you follow the tracing of the slices, you 
can see that blurring also occurs along the z-axis. This is taken into 
account in the final df3.

>
> Also, adding turbulence can result in very slow renders. Sometimes it's possible
> to hide a granite pattern in a volume-based pigment_pattern instead... this is
> usually faster than using turbulence. Or, you can try to get away with fewer
> turbulence octaves... but of course I'm always trying to maximize my PPS :)

Yes, I noticed the slow down. One of the things I want to try is indeed 
what you suggest.

Thomas

Post a reply to this message

Exploring ghosts.

This could be some kind of ectoplasma. Using the First def3 density 
file, with an added spherical density within the same media.

Thomas

Post a reply to this message

Attachments:
Download 'gt_df3_test.png' (92 KB)

Preview of image 'gt_df3_test.png'

...or like this.

Thomas

Post a reply to this message

Attachments:
Download 'gt_df3_test.png' (88 KB)

Preview of image 'gt_df3_test.png'

>Thomas de Groot  on date 24/09/2013 15.59 wrote:
> ...or like this.
>
> Thomas
This is creepy!
Paolo

Post a reply to this message

Thomas de Groot <tho### [at] degrootorg> wrote:
> ...or like this.

Hmm, that /is/ kinda creepy.

Reminds me of a dream I had once... I dreamed that when we die, out bodies began
to enlarge slowly, and pretty soon we were big, translucent human shapes -- much
like clouds -- floating in the skies.

I guess that it would be true in one sense, because the electromagnetic energy
that our bodies produce (including the EM energy coming from our brains) expands
outward spherically (more or less). Which star systems are 36 light years away
from us, anyway? My ever-expanding "sphere" has probably reached them by now ;)
(Assuming, of course, that the universe is continuous and my EM field hasn't
encountered problems along the way, not to mention that fact that Sol and its
subjects don't stay put.)

Oh yeah, real ghosts would look different than they appear in the movies, BTW.
Real ghosts would probably show veins, organs, bones, etc. But nobody really
thinks about that, so it's all good :)

Sam

Post a reply to this message