I've been doing more work on better blobs, here's the results. These 
blobs are as easy to control, since they use a bounded function, unlike 
the exponential blobs where adding one blob affects the whole object.

They are quite a bit slower as an isosurface with user-defined 
functions, I plan to make a pattern patch though. Being compiled alone 
will make it much faster, and it will make some optimizations possible. 
I have no idea how to add it to the blob object, or even if it is 
practical, but an object with the isosurface solving method could still 
make more optimizations than a simple pattern.

The smooth ones are the isosurface versions. I tried to duplicate the 
"open" one with blobs, but it would require a fifth blob component at 
the center, so I went back and tried to duplicate the real blob with the 
new blob function instead. As you can see, there are definite 
differences, you can't just throw it in as a replacement, but it is 
close.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message

Attachments:
Download 'smoothblob1.png' (16 KB) Download 'smoothblob2.png' (17 KB) Download 'blob.png' (15 KB)

Preview of image 'smoothblob1.png'

Preview of image 'smoothblob2.png'

Preview of image 'blob.png'

On Fri, 18 Oct 2002 12:59:50 -0400, Christopher James Huff <chr### [at] maccom>
wrote:
> I've been doing more work on better blobs, here's the results.

For sure: they are no blobs like in iso csg library ? That's different
solution ?

ABX

Post a reply to this message

In article <ibi0ruoakuga1i9ret2alv8bhp92u1voon@4ax.com>,
 ABX <abx### [at] abxartpl> wrote:

> For sure: they are no blobs like in iso csg library ? That's different
> solution ?

These are isosurfaces using a new field function for components that 
smooths out the second derivative discontinuities. I'm working on a new 
version of the blob pattern that uses this function, which will make 
using isosurface "blobs" with this function easier and faster, and will 
probably try a new blob primitive as well.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message

Christopher James Huff wrote:
> 
> These are isosurfaces using a new field function for components that
> smooths out the second derivative discontinuities. I'm working on a new
> version of the blob pattern that uses this function, which will make
> using isosurface "blobs" with this function easier and faster, and will
> probably try a new blob primitive as well.

The nice thing about isosurface blobs is that you can use arbitrary shapes
as components.  This would probably be difficult with a pattern.

Christoph

-- 
POV-Ray tutorials, IsoWood include,                 
TransSkin and more: http://www.tu-bs.de/~y0013390/  
Last updated 13 Aug. 2002 _____./\/^>_*_<^\/\.______

Post a reply to this message

In article <3DB### [at] gmxde>,
 Christoph Hormann <chr### [at] gmxde> wrote:

> The nice thing about isosurface blobs is that you can use arbitrary shapes
> as components.  This would probably be difficult with a pattern.

I don't quite understand what you mean. The pattern will support 
whatever shapes I add to it, which will be a lot more than the 
isosurface primitive supports. It could even include a function 
primitive. But the main point of doing it this way is that it will be 
much faster then user defined functions...of course they will always be 
more flexible.

I've got spherical components working, BTW. However, I'm finding some 
very ugly stuff in the function and pattern code...don't look for it too 
soon.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message

Christopher James Huff wrote:
> 
> > The nice thing about isosurface blobs is that you can use arbitrary shapes
> > as components.  This would probably be difficult with a pattern.
> 
> I don't quite understand what you mean. The pattern will support
> whatever shapes I add to it, which will be a lot more than the
> isosurface primitive supports. It could even include a function
> primitive. But the main point of doing it this way is that it will be
> much faster then user defined functions...of course they will always be
> more flexible.

If you can use an arbitrary function/shape as a blob component it would
not be bounded and therefore much less efficient.  The lack of bounding is
the great disadvantage of isosurface blobs.

Christoph

-- 
POV-Ray tutorials, IsoWood include,                 
TransSkin and more: http://www.tu-bs.de/~y0013390/  
Last updated 13 Aug. 2002 _____./\/^>_*_<^\/\.______

Post a reply to this message

In article <3DB05F16.95DFDADC@gmx.de>,
 Christoph Hormann <chr### [at] gmxde> wrote:

> If you can use an arbitrary function/shape as a blob component it would
> not be bounded and therefore much less efficient.  The lack of bounding is
> the great disadvantage of isosurface blobs.

They wouldn't be the only kind of component. Functions would not be 
automatically bounded, but it wouldn't make it impossible to bound the 
other components, and they could also be manually bounded.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message