Forum: Carrara


Subject: How did he model this?

Unc opened this issue on Jan 09, 2002 ยท 5 posts


Unc posted Wed, 09 January 2002 at 5:26 PM

Hi, Im a new convert to Carrara Studio after using Truespace for the past 2 years. After seeing Rustboy I bought my copy over the Christmas hols and have been playing with it none stop since then. However I still need to get the hang of modelling using the Vertex modeller. Anyway to cut to the chase I thought I would have a go at modelling a shape I had seen at this web site.... http://elements.infinitycubed.com/main.php (Its the one in the top left corner. No matter how I try I can't seem to do it. I think its my logic and not the technic that is wrong. Any ideas. Cheers, Unc


willf2 posted Wed, 09 January 2002 at 9:24 PM

IF you are good in solving math problems then this would be a great exersise to do with the forula function. I've seen a few of these "gordon" type surface models around & they are amazing. Sorry I can't come up with the URL let alone the math for the problem (I got into art 'cause I hated math!).


MarkBremmer posted Thu, 10 January 2002 at 10:32 AM

Hi Unc, Yep, it's a formula. (although I think it could be done in the spline modeler as well) I have a formula for it from a PovRay file that I've been slowly translating into Carrara. The two programs use different syntax, and frankly, I'm a little unsure as to script the variables in the last portion of the formula. Here is the original source file (I'm assuming the dx, dy caclulation should actually be applied as a deformer as shown in the prelim calc area. Double values such as vv and uu should be translated as v and u respectively-I think! ;) pi translates as PI): //various constants uumin = 0 uumax = 4pi vvmin = 0 vvmax = 2pi a=1 b=0.3 c=0.5 d=0.3 //preliminary calculations r=a+bcos(1.5uu) xx=rcos(uu) yy=rsin(uu) zz=csin(1.5uu) dx=-1.5bsin(1.5uu)cos(uu)-(a+bcos(1.5uu))sin(uu) dy=-1.5bsin(1.5uu)sin(uu)+(a+bcos(1.5uu))cos(uu) dz=1.5ccos(1.5uu) //Derivatives qn=vnormalize() //Vector operatons qvn=vnormalize() ww=vcross(qn,qvn) //points and normals x1=xx+d(qvn.xcos(vv)+ww.xsin(vv)) //Calculate the y1=yy+d*(qvn.ycos(vv)+ww.ysin(vv)) //points. ww.x is the z1=zz+dww.zsin(vv) //x value of ww vector nx1=qvn.xcos(vv)+ww.xsin(vv) //Normals needed to ny1=qvn.ycos(vv)+ww.ysin(vv) //make smooth triangles nz1=ww.z*sin(vv)






AzChip posted Thu, 10 January 2002 at 4:14 PM

Hi, Unc -- Yep, I think that Mark is probably on the right track, although I did have a go at it in the free-form (or spline) modeler (I'm using RayDream Studio -- very similar archetecture, but some different terminology). I was able to get some pretty interesting twisted tube things, but only had about fifteen minutes to play around with it, so nothing exactly like what appeared in this guy's site.... With enough time, though, I'd try starting with a circle in the spline modeler, use a torus extrusion preset, and then start dragging verteces around until you get something approaching what you're looking for. Pretty simple concept, but actually getting it to do what you want is another matter entirely. Hope this is helpful. - Dex


Unc posted Thu, 10 January 2002 at 6:48 PM

Thanks Guys I really appreciated your answers. I gotta say I've only had this package for a week now, but I find more and more features everyday. Im going to have to find time though and study the formula feature though. I used python scripting a bit in truespace, but this seem slightly different. I'll let you know how I get on. Cheers, Unc