Hi Stefan,

the spheres values/colors are set with a greyscale
(not really, it's a RGB but all channels have the same value ;-)
Give it a try with a RGB image, all channels different/desired
values. Should do the trick, too.

Yep, you're right, it's a modified old prj.
While sitting together with a friend, I've seen on his
table a speaker with a perforation of hundreds of holes
with different size. The size variation was something like a simple
radial greyscale gradient .
That's where the idea comes from: to lazy to scale hundreds of spheres by hand 
;-)
The REAL SOFT was only an animation test, how fast my
machine computes things like this, same as the nailboard-test.

Matthias

----- Original Message ----- 
From: "Beg-inner" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Wednesday, October 17, 2007 8:22 PM
Subject: Re: individual scaling with custom channels, how to ;-)


> Hi Matthias..
> 
> This is cool stuff, if I am not totally wrong I have seen you done this 
> before..(or maybe it was the same prj..)
> 
> Thx for sharing this, it gives a nice and really cool result...
> 
> btw..
> Just a thought..
> Could it be made working in also the 3rd dimension.. with spheres in also 
> the 3rd dimension.. and use another texture from the side ?
> So that both the Front Texture and the Side Texture contribute to the 
> result...(if you know what I mean..)
> Each sphere gets the value of Front multiplied with Side value..
> 
> Thx in advance..
> 
> Take Care
> Best Regards
> Stefan Gustafsson ( Beg-inner )
> A Proud Owner and User of Real3D and Realsoft3D..
> 
> > Use it like this:
> >
> > 1. Create a Object or level with objects
> > 2. Create 3 Tags as seen on the attached tags image (all as "Float" with 
> > value "1")
> >    (Note: The Tags will give a better and animatable control, if the 
> > script is assigned)
> > 3  Open the scripts tab and assign the script below, end of mail 
> > (copy'n'paste)
> >    Set "Command Language" to "JavaScript".
> >
> > // if you add the next lines at the end of the script below
> > // you should get an realtime feedback, if you move the Tags values slider
> >
> > print('Some Information:'+
> > '\nColorValue X= '+myCx+
> > '\nColorValue Y= '+myCy+
> > '\nColorValue Z= '+myCz+
> > '\n\nScaleValueX= '+mySx+
> > '\nScaleValue Y= '+mySy+
> > '\nScaleValue Z= '+mySz );
> >
> > // DO NOT USE IT ON MORE THAN ONE OBJECT
> > Why? Info Window output for 100s of objects? Hmmmmm, not good
> >
> >
> > 4. Use a macro or copy by hand as many objects as you need
> > 5. Create a simple vsl-material with a color gradient or use a texture
> > 6. Create a Parralel-Mapping(or other) with this material, which spans
> >    over all objects
> > 7. Select all objects and the material-mapping
> > 8. Use Map2Obj with the settings on the attached images and accept
> > 9. move the frame slider, the objects scale should switch
> >
> > Matthias
> >
> >
> > Some logical error in the files script,
> > here's a better one:
> >
> > //------script begins copy and paste begins
> > // One from the poor mans scripting corner
> >
> > //Get the color
> > myC = Self.GetColor();
> > //Get the Tags values
> > scX = Self.Get("scX");
> > scY = Self.Get("scY");
> > scZ = Self.Get("scZ");
> > //Split the color vektor
> > myCx = myC.x;
> > myCy = myC.y;
> > myCz = myC.z;
> > // Maybe we want to multiply or divide later
> > // we should make sure that we do not get zero values
> > // CHANGED NEXT LINES
> > // now if lower 0.001 previous if lower 0 does not make sense
> > if (myC.x < 0.001) {
> >    myCx = 0.001;
> > }
> > else if (myC.y < 0.001) {
> >    myCy = 0.001;
> > }
> > else if (myC.z < 0.001) {
> >    myCz = 0.001;
> > }
> >
> > mySx = myCx*scX;
> > mySy = myCy*scY;
> > mySz = myCz*scZ;
> >
> > Self.SetTotalScale(new r3Vect(mySx,mySy,mySz));
> >
> > //------script ends copy and paste ends
> 
> 
> --------------------------------------------------------------------------------
> 
> 
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.476 / Virus Database: 269.14.8/1064 - Release Date: 10/11/2007 
> 3:09 PM
> 

Reply via email to