Burningwing

18th October 2003 14:49 UTC

Help 3D ball
im quiet new 2 superscopes but i know the basics of 2D and try 2 learn 3D
now my question is how do i make a 3D ball i did get a circle that is moving but what 2 put in z1???
INT
n=200;r=0.5; mx=0;my=0;mx=0;dst=2;rx=0;ry=0;rz=0;rdx=1;rdy=1;rdz=1;p=3.14159265;p2=2*p;p3=180/p

onbeat
rdx=rand(3)+1;rdy=rand(3)+1;rdz=rand(4)+1;

perframe
fr=fr+0.01;mz=1+sin(fr); rx=rx+rdx;ry=ry+rdy;rz=rz+rdz; xs=sin(rx/p3);ys=sin(ry/p3);zs=sin(rz/p3);xc=cos(rx/p3);yc=cos(ry/p3);zc=cos(rz/p3);

per point

d=0.7;
x1=sin(r)*d;
y1=cos(r)*d;
z1=0;
y2=y1*xc-z1*xs; z2=y1*xs+z1*xc;
x2=z2*ys+x1*yc; z3=z2*yc-x1*ys;
x3=x2*zc-y2*zs; y3=y2*zc+x2*zs;
x4=mx+x3; y4=my+y3; z4=mz+z3;
x=x4/(1+z4/dst);y=y4/(1+z4/dst);



OBS i didet write the code only changed it a bit!!

PLS HELP ME

Burningwing

18th October 2003 16:26 UTC

iv found 1 way
i=i*p;
x1=sin(i)*sin(i*30)*d;
y1=cos(i)*d;
z1=sin(i)*cos(i*30)*d;
but does it exist any other way 2 make a ball that dont look like that?

Deamon

18th October 2003 17:56 UTC

I always use the way you posted the second time.

UIUC85

18th October 2003 21:21 UTC

well maybe if you wanted to do a geodesic sphere... but the code for that would be a pain... well in comparison to the normal sphere we use. Do you understand why that second bit of code works as a sphere?

Nic01

18th October 2003 23:54 UTC

Is that as in you want to know or checking if he knows?

I don't know what a geodesic sphere is [mind explaining?], but here's another way :
(parallels/*******ns, requires 2 scopes)
(stripped from UCD's Groovy Saturn, Whacko II)
(Try to understand how it works, then customize it to your liking)

Parallels :

n=261;nc=25;nm=.2;sf=1.2

red=cr;green=cr;blue=cr;
cr=1-cv;cg=1-cv;cb=1-cv;
rad=cos(mp*1.57);
py=cos(rp/nc*6.28)*rad;pz=sin(rp/nc*6.28)*rad;px=sin(mp*1.57);
x1=px*cz+py*sz;y1=py*cz-px*sz;
y2=y1*cx+pz*sx;z2=pz*cx-y1*sx;
x3=x1*cy+z2*sy;z3=z2*cy-x1*sy+2.5;
x=if(above(z3,0.1),x3/z3,x);y=if(above(z3,0.1),y2/z3*1.33,y);
rp=if(equal(cv,0),(rp+1)%nc,rp);
mp=if(equal(cv,0),mp,mp+nm);
cv=if(equal(rp,0),1-cv,0);
x=x*sf;y=y*sf;

*******ns :

n=261;nc=50;nm=.2;sf=1.2

rad=1;
px=cos(rp/nc*6.28)*rad;py=sin(rp/nc*6.28)*rad;pz=0;
y0=py*cos(mp*3.14);z0=py*sin(mp*3.14);
x1=px*cz+y0*sz;y1=y0*cz-px*sz;
y2=y1*cx+z0*sx;z2=z0*cx-y1*sx;
x3=x1*cy+z2*sy;z3=z2*cy-x1*sy+2.5;
x=if(above(z3,0.1),x3/z3,x);y=if(above(z3,0.1),y2/z3*1.33,y);
rp=if(equal(cv,0),(rp+1)%nc,rp);
mp=if(equal(cv,0),mp,mp+nm);
cv=if(equal(rp,0),1-cv,0);
x=x*sf;y=y*sf;

--

Optimize by yourself, plug in the missing parts, it's all in the life of a coder.

UnConeD

19th October 2003 05:08 UTC

That's old code though ;). Lacks some common optimizations and correct aspect ratio correction (assume 4/3 ratio).

A geodesic sphere is where you start off with one of the platonic solids, and subdivide the faces into triangles. You then project the vertices back onto a unit sphere.
When you start with an icosahedron, you can nearly-even distribution of triangles, because the amount of triangles touching at a vertex is at least 5, at most 6.

For a tetrahedron for example, you'd have vertices where only 3 triangles touch, which is a lot different from 6.

http://www.gardendome.com/spheres.JPG

On the left, an icosahedral geosphere on the right an octahedral geosphere.

Would be lots of handwork to do this in AVS because there's no easy way to automatically generate one (lack of arrays and looping).

dirkdeftly

19th October 2003 06:06 UTC

hello. i am atero. i have been on these forums for many moons. it is a pleasure to meet you. through much spiritual meditation and chanting of the verses of the great buddha, i have become a mean bastard, and over time you will grow to loathe me, like the british hate the united states.

on that note, stop typing like a four year old you fucking retard.


welcome to the avs forums.

uNDefineD

19th October 2003 08:39 UTC

Originally posted by Atero
on that note, stop typing like a four year old you fucking retard.

UIUC85

19th October 2003 09:08 UTC

Yeah unconed that's why I mentioned the arrays in the wishlist forums lol. Not a very plausible idea, but it'd be kinda neat just for techie purposes. Nice models too :eek:
Nic01, yeah I was asking if he understood why it works.

dirkdeftly

19th October 2003 17:55 UTC

well ndd, i could ask you why do my posts bother you, couldn't i...but that would be mean....

uNDefineD

20th October 2003 00:14 UTC

Because I'm fucking sick of you hanging shit on newbies just because they don't type in a way you'd prefer. Does that answer your question? Now for the third time: answer mine.

sidd

20th October 2003 10:48 UTC

x1=sin(i)*sin(i*30)*d;
y1=cos(i)*d;
z1=sin(i)*cos(i*30)*d;

noone is going to care explain why it works?

well:

if you have x1=sin(i*30); z1=cos(i*30) then your scope is making a circle shaped thing on the x/z plane. only the circle is drawn more than once (looksee the *30 instead of *2pi).

then you make y1=cos(i), so that it stretches the scope along the y axis, dispersing it such that most of the circles are at near -1 and 1, and it thins out around 0 on the y axis.

then you times what you have in x1 and z1 by sin(i), that way the circle get smaller and smaller as they approach the start and end of i. (which we have just assigned to -1 and 1 on the y axis.) but they are the original size in the exact middle of i (which is 0 on the y axis)

so now what you have is a spiral shaped thing that is stretched out to form a spiraly circle.. YAAAY!

then you times all the axis by your constant d to resize evenly.

You should also times i by 2pi before all this so that you know exactly what you are coding (eg, using i*30 would then make the circle have 30 spirals exactly.)

dirkdeftly

20th October 2003 15:38 UTC

ndd, you just don't get the point do you...you can't bitch about me bitching about n00bs and their typing habits and say i'm wrong for doing it, cos you're doing the exact same thing.

Burningwing

20th October 2003 15:39 UTC

hmmm did u all forget what my question was stop geting angry in here

UnConeD

20th October 2003 17:42 UTC

Atero/ndd: Mirror mirror on the wall, who's the naggiest bitch of them all?

UIUC85

20th October 2003 19:42 UTC

You guys done posting flames and fairy tale rhymes yet?

dirkdeftly

21st October 2003 06:19 UTC

i was a while ago :rolleyes: anyway i think sidd answered your question pretty clearly....