Not in the league of Mr Lipka's train engine, but an image I spent some time on
and a result with which I am happy. About a dozen CSG objects used in four
'object-functions' along with some standard POVRay functions stuffed in a single
isosurface (excepting the plane object).
Bill

Post a reply to this message

Attachments:
Download 'objectpatchex12.jpg' (51 KB)

Preview of image 'objectpatchex12.jpg'

William Pokorny schrieb:
> Not in the league of Mr Lipka's train engine, but an image I spent some time on
> and a result with which I am happy. About a dozen CSG objects used in four
> 'object-functions' along with some standard POVRay functions stuffed in a single
> isosurface (excepting the plane object).

Even Mr Lipka is amazed about this "isosurfacizing" thing... I've toyed 
multiple times with the idea of ditching my whole 
beveled-thing-framework stuff and trying whether the same can be had 
with the isosurfaces approach. The only thing that's holding me back is 
that I guess I'd lose steam, figuratively speaking, if I tried to switch 
engines while rolling on the track.

Post a reply to this message

Do you have any idea how to approach the complex modeling required for your 
kind of beveled surfaces using isosurfaces? Does anybody in this newsgroup? 
(Non rhetorical question - if anybody does, please tell.)

I for my part do not know how to model this properly by using just a single 
function. I doubt you would gain anything by using isosurfaces - better keep 
on using CSG. Your model starts to look really good, though I did prefer the 
black colour you used in your next to last post... even for a preview I find 
the idea of a white steam locomotive somewhat strange. ;-)

Post a reply to this message

message de news: 4abf677b$1@news.povray.org...
>I find  the idea of a white steam locomotive somewhat strange. ;-)
Even for a U.N. train? :-)

Marc

Post a reply to this message

TC schrieb:
> Do you have any idea how to approach the complex modeling required for your 
> kind of beveled surfaces using isosurfaces? Does anybody in this newsgroup? 
> (Non rhetorical question - if anybody does, please tell.)

Isosurfaces generally require a "blurred" representation of an object, 
with the object boundary not being defined as a 0->1 transition, but a 
smooth gradient.

With such a blurred representation, the object can simply be extended by 
a certain amount (by changing the threshold), which will automatically 
smoothen the object.

It might also allow to interpolate, blurring the object even more, which 
would automatically round off even without changing the threshold; this 
will also round off not only edges, but also crevices.

> I for my part do not know how to model this properly by using just a single 
> function. I doubt you would gain anything by using isosurfaces - better keep 
> on using CSG.

I will - that's a given. As I said, I don't want to risk "losing steam".

 > Your model starts to look really good, though I did prefer the
> black colour you used in your next to last post... even for a preview I find 
> the idea of a white steam locomotive somewhat strange. ;-) 

Don't worry - the "false color" texturing is just for test renders to 
verify the geometry :-)

Post a reply to this message

Hi TC,
:-) I too would stick with CSG at this point, not knowing the cost for the CSG
object as isosurface approach. 100K primitives - wow.

The complexity / difficulty however would not be in the modeling - that would be
much simpler than what Mr Lipka has already done. If the train is a complete
'object' in the POVRay sense, you stick the train into the ObjectAsIso()
function and you have a very complex, single isosurface (1).

The hard part is getting the beveling into the code treating the train
as an isosurface. And, the hard part there is not coming up with ways to bevel -
existing interpolation techniques would work - it is doing that beveling in an
amount of compute resource any of us would tolerate.

I consider my hacked C++ compiled ObjectAsIso() implementation 'just barely'
tolerable for compute resource on today's most current multi-core processors. I
personally do not see a way to a beveling implementation that is not quite a bit
more expensive, but, who knows.

Bill
---------

(1) - Stretching a bit currently because all the csg behind the isosurface
'treatment' of the train is seen as one object by POVRay's existing search
mechanisms. The render time of this 'object-as-iso' approach goes up
dramatically as the CSG count in the object goes up. This single train object
would need to be broken into smaller object-as-isosurface objects to be a
practical render --> but, this piece-wise, build it up approach is how we do big
CSG objects today. Also practically we are unlikely to want the same beveling
magnitude across every shape of the whole train. The beveling would
likely be tuned for each object.

"TC" <do-not-reply@i-do get-enough-spam-already-2498.com> wrote:
> Do you have any idea how to approach the complex modeling required for your
> kind of beveled surfaces using isosurfaces? Does anybody in this newsgroup?

Post a reply to this message

"TC" wrote:
> Do you have any idea how to approach the complex modeling required for your
> kind of beveled surfaces using isosurfaces? Does anybody in this newsgroup?
> (Non rhetorical question - if anybody does, please tell.)
>
If you can assume the bevel has tighter curvature than surfaces of the objects
being joined, it should be just a matter of restricting the second derivative
(curvature) of your approximation function in that region to fall within the
desired beveling range.  In other words, if your object surfaces have a minimum
radius of curvature of, let's say, one unit, and you want a bevel radius of
about 0.1 units, limit your isosurface approximation to have a second derivative
resulting in a minimum curvature radius of 0.1.

I'd have to work out the math to verify this, but intuitively this should be a
matter of limiting the isosurface's second partial derivatives to all be less
than the inverse of the desired bevel curvature, 10 in this example.  I have a
hunch that this requirement will carry through fairly cleanly to an algorithm
used to determine an approximating function, but probably don't have time to
pursue it further at present.

Unfortunately, I have to put this problem on my stack of things I'd rather do
than what those paying my tuition want to see (which also has to do with
numerical differentiation and integration).

But I hope this helps a bit.

David Wagner

Post a reply to this message

High!

m_a_r_c wrote:

> message de news: 4abf677b$1@news.povray.org...
>> I find  the idea of a white steam locomotive somewhat strange. ;-)
> Even for a U.N. train? :-)
> 
> Marc 

I heard they currently are building Afghanistan's very first 
long-distance railway line... I would love to see such clean white 
humanitarian trains on the tracks!

See you on www.khyberspace.de!

Yadgar

Now playing: Dirty Dog (ZZ Top) - would make good railroad music!

Post a reply to this message

news:4ac0ecb5$1@news.povray.org...
>
> I heard they currently are building Afghanistan's very first long-distance 
> railway line... I would love to see such clean white humanitarian trains 
> on the tracks!

Probably a likely target for blowing up... :-(

Thomas

Post a reply to this message