Text archives Help
- From: David E DeMarle <
>
- To: Carson Brownlee <
>
- Cc:
- Subject: [Manta] Re: Re: [vtk-developers] Mandelbulb challenge
- Date: Thu, 21 Jan 2010 11:22:09 -0500
On Thu, Jan 21, 2010 at 11:19 AM, Carson Brownlee
<
>
wrote:
>
We implemented ray traced julian sets last year in CUDA based off of
>
http://www.devmaster.net/forums/showthread.php?t=4448
>
and it was extremely fast. There's no memory used to store geometry and
>
no need for acceleration structures. That exact intersection routine
>
could be used in Manta.
very cool
>
>
If you want something where you can zoom into infinite detail then
>
rendering geometry from vtkMantaPolyDataMapper definitely seems out of
>
the question. Does a VTK Mapper that skips the polygonization step fit
>
into the traditional VTK pipeline?
>
I agree that vtkMantaPolyDataMapper is the wrong approach and yes
there are many vtk mapper's that do not polygonalize. The volume
rendering pipeline has many examples.
>
Carson
>
>
On Thu, 2010-01-21 at 09:46 -0500, David E DeMarle wrote:
>
> I suspect one might _render_ the implicit function directly with Manta
>
> (http://mantawiki.sci.utah.edu/manta/index.php/Main_Page) assuming
>
> there is an intersection/sampling function in Manta suitable for that
>
> specific class of function.
>
>
>
> If there is such a function, one could then choose either interactive
>
> or really good looking (ie multiple samples per pixel, whitted
>
> effects, global illumination etc), ray tracing engine parameters.
>
>
>
> LANL has developed an interface between VTK and Manta, so you might
>
> specify the function and manipulate it in VTK, then render the
>
> geometry in Manta. You can find the interface in the
>
> ParaView/Plugins/Manta source directory, even though it works fine
>
> when compiled directly against VTK.
>
>
>
> For efficiency you would probably want a new VTK/Manta mapper that
>
> skips the polygonalization step, and sends the function parameters
>
> directly to manta.
>
>
>
> have fun!
>
>
>
> David E DeMarle
>
> Kitware, Inc.
>
> R&D Engineer
>
> 28 Corporate Drive
>
> Clifton Park, NY 12065-8662
>
> Phone: 518-371-3971 x109
>
>
>
>
>
>
>
> On Thu, Jan 21, 2010 at 8:45 AM, Bill Lorensen
>
> <
>
>
> wrote:
>
> > Will,
>
> >
>
> > Some early Siggraph work may be enlightening:
>
> > Generation and display of geometric fractals in 3-D
>
> > http://portal.acm.org/citation.cfm?id=801263
>
> > and
>
> > Ray tracing deterministic 3-D fractal
>
> > http://portal.acm.org/citation.cfm?id=74333.74363&type=series
>
> >
>
> > I recall that they do on-the-fly rendering rather then generate a volume.
>
> >
>
> > Bill
>
> >
>
> > On Thu, Jan 21, 2010 at 6:43 AM, Will Schroeder
>
> > <
>
>
> > wrote:
>
> >> Yesterday Jason Shepherd at Sandia posted a link to stunning
>
> >> visualizations
>
> >> of an attempt to visualize the 3D Mandelbrot set
>
> >> http://www.skytopia.com/project/fractal/mandelbulb.html#renders
>
> >>
>
> >> Does anybody in the community want to take a stab at adding this
>
> >> capability
>
> >> to VTK? Somewhere out there are hidden geniuses just waiting for a
>
> >> challenge
>
> >> like this, I can feel it :-)
>
> >>
>
> >> My initial, uninformed thoughts on this would be to approach it one of
>
> >> two
>
> >> ways:
>
> >> 1. Create an implicit function and then use vtkSampleFunction to sample
>
> >> it.
>
> >> This would provide very quick and dirty initial results and enable us
>
> >> to get
>
> >> out heads around the mathematics.
>
> >> 2. Create a custom class to produce a 3D image dataset, and then render
>
> >> it
>
> >> using one of the many rendering techniques in VTK. This would provide
>
> >> greater flexibility in generating and defining data, including producing
>
> >> coloration, etc.
>
> >>
>
> >> Of course if you're really good you'll do this using the GPU :-) Maybe
>
> >> support streaming to a region, etc.
>
> >>
>
> >> I'd love to have this because, well it's really cool :-), but also as a
>
> >> test
>
> >> case/data source for large data, streaming, HPC, volume rendering and
>
> >> so on.
>
> >> If there is interest we'll create some wiki pages and kick the project
>
> >> off.
>
> >>
>
> >> Cheers,
>
> >>
>
> >> --
>
> >> William J. Schroeder, PhD
>
> >> Kitware, Inc.
>
> >> 28 Corporate Drive
>
> >> Clifton Park, NY 12065
>
> >>
>
> >> http://www.kitware.com
>
> >> (518) 881-4902
>
> >>
>
> >> _______________________________________________
>
> >> Powered by www.kitware.com
>
> >>
>
> >> Visit other Kitware open-source projects at
>
> >> http://www.kitware.com/opensource/opensource.html
>
> >>
>
> >> Follow this link to subscribe/unsubscribe:
>
> >> http://www.vtk.org/mailman/listinfo/vtk-developers
>
> >>
>
> >>
>
> >>
>
> > _______________________________________________
>
> > Powered by www.kitware.com
>
> >
>
> > Visit other Kitware open-source projects at
>
> > http://www.kitware.com/opensource/opensource.html
>
> >
>
> > Follow this link to subscribe/unsubscribe:
>
> > http://www.vtk.org/mailman/listinfo/vtk-developers
>
> >
>
> >
>
>
>
Archive powered by MHonArc 2.6.16.