Mueen Nawaz wrote:

> 	Yes, the idea of starting from the bottom up used to appeal to me
> greatly earlier (I'm a math/theoretical physics/engineering person).

Euclidean proofs and all that :)


> However, the attitude of just getting the job done is occasionally
> dominant now.
> 
> 	I should have said in my initial post that the question was not
> intended for objects like meshes and patches - I think they were more or
> less meant to be done by other utilities. My worry is that if I rely too
> heavily on those utilities, I'll end up using them to create items I
> could easily make using CSG in POV-Ray.

CSG with the "pure" primitives is a very engaging intellectual puzzle. 
If I worry about anything, it is more likely that I get sidetracked with 
CSG improvisations when there is so much else to learn.  But really 
neither of us is getting anything done while we are worrying so much.

I tend to view CSG as being to CG as like classic burin engraving is to 
intaglio.  It can produce results of unparalleled beauty, that stem 
directly from the precision, discipline, needed to work with it, and a 
severe vision that can exploit certain stylizations it imposes.

> 
> 	I like programming, and I like math. I'm also much below average when
> it comes to drawing (by hand). Raytracing was my first experience where
> I could make images I liked (even using the computer). I like the whole
> "setup" of raytracing. I like to be able to build each object at a time,
> placed at the origin, at some unit scale, and then do the necessary
> transformations when I'm putting it altogether.

Then I guess you have found a home. :)


> 
> 	I haven't worked too hard with other software. They're a bit less
> intuitive to me, perhaps because of my general inability to draw. JPatch
> is easy to use, though. It's just that I keep struggling with issues
> such as when I export to POV-Ray, the item I created is not centered,
> not to the right scale, etc. Fixing that is a pain. I'm sure
> JPatch/other software could remedy this problem easily, but I don't know
> how.

I am sure you don't need me to tell you this, but that is supposed to be 
the appeal of integrated solutions. That you can avoid those problems. 
When you use an eclectic, best of breed approach, involving a mix of 
programs, you do have to assemble a consistent workflow and maintain 
certain disciplines of nomenclature, file storage, and standards for 
passing deliverables.  Else you *will* go crazy.


> 
> 	I looked at Blender and it seemed like quite a bit to bite on. I don't
> really have much time, so if I were to learn to use a utility, it'd have
> to be something I can just fiddle with 1-2 hours a week, and remember
> the salient details that I learned after a whole week has gone by. I'm
> not sure that's doable with Blender if you're a beginner.
> 
> 	Haven't tried Wings3D.

It's all about locating points correctly in space, ie, the vertices that 
will describe he surface you want.  It doesn't matter what tool you use 
if you can't solve the puzzle of where the points need to be.  By the 
same argumen,t you might as well use the modeller you like.  I use 
Wings.  Silo also has its fans among some very talented people here.  It 
would be my next choice, probably.


> 
> 	I use Linux, and that in itself is a constraint.

I know a few Linux users here are using Wings.

> 
> 	Is there any site where all the patterns are explained in detail (i.e.
> shown). I'm thinking of something along the lines of:
> 
> http://www-public.tu-bs.de:8080/~y0013390/pov/cr_metric.html

I think someone actually tried to make an menu insert file like that.

> 
> 	Where the pattern is "shown" in detail for various parameters. That
> would be immensely helpful. A lot of the patterns in the documentation
> are described mathematically, but I think images will make it easier to
> see the effect of the math and various parameters.

There really are too many parameters I think.  But there was an effort 
going on to explore some of this kind of atlasing of textures in the pov 
wiki.  That  effort takes the approach of an atlas of possibilities 
categorized under real world effects like metals, bark, leather, etc. 
It is a thorny problem, however one approaches it, because the basic 
issue is one of "-ishness"  There is no "bark" texture, there are just 
effects that can produce a "barkish" look in under certain situations.

Post a reply to this message