 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hey, this seems a very promising approach!
Let me guess where the idea comes from:
A thread in cgrr started last Sunday, right?
Is it much change to the plain pov code?
-Hans-
Thomas Willhalm wrote:
>
> Hello,
>
> this is a snapshot of my current work on fur. I'm sorry for the file
> size, but jpg didn't like this image. I have now implemented a new
> lighting model, which was developped by Kajiya and Kay. I have also
> added the possibility to provide a pigment statement to scattering
> media.
>
> In future, I plan to support other geometries like planes, tori, discs
> and perhaps even blobs. Furthermore, I want to add curliness.
>
> The rendertime has been 30 minutes on a Pentium II 400 running linux.
> Memory consumption (as reported by POV-Ray) has been 85 KB.
>
> Thomas
>
> [Image]
>
> --
> http://www.fmi.uni-konstanz.de/~willhalm
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hans-Detlev Fink <hdf### [at] pecos nospamde> writes:
> Hey, this seems a very promising approach!
> Let me guess where the idea comes from:
> A thread in cgrr started last Sunday, right?
You're wrong. I stopped following the discussions in cgrr, because
there was too much traffic. Furthermore, I'm working on fur for
more than several weeks now.
> Is it much change to the plain pov code?
I try to keep the changes small. I have redesigned my approach several
times. Now, the main part is in two new files named fur.h and fur.c.
Of course, there have to be some changes in parstxt.c and media.c. I
also have to change frame.h to include a pointer to the pigment and
the fur struct in the media struct.
Thomas
--
http://www.fmi.uni-konstanz.de/~willhalm
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas Willhalm wrote:
>
> Hans-Detlev Fink <hdf### [at] pecosnospamde> writes:
>
> You're wrong. I stopped following the discussions in cgrr, because
> there was too much traffic. Furthermore, I'm working on fur for
> more than several weeks now.
Oh sorry, I talked about c.g.r.renderman which is far from
overcrowded ;-). It was there that I first stepped upon the
idea of 'hair by shader'.
>
> > Is it much change to the plain pov code?
>
> I try to keep the changes small. I have redesigned my approach several
> times. Now, the main part is in two new files named fur.h and fur.c.
> Of course, there have to be some changes in parstxt.c and media.c. I
> also have to change frame.h to include a pointer to the pigment and
> the fur struct in the media struct.
Sounds very interesting. Hopefully you will give us an announcement
(and of course the code) when you feel the code is ready for the public.
Have you any early idea how your approach rates in terms of render
time as compared with traditional (object based) procedures?
-Hans-
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hans-Detlev Fink <hdf### [at] pecosnospamde> writes:
> Thomas Willhalm wrote:
> >
> > Hans-Detlev Fink <hdf### [at] pecosnospamde> writes:
> >
> > > Is it much change to the plain pov code?
> >
> > I try to keep the changes small. I have redesigned my approach several
> > times. Now, the main part is in two new files named fur.h and fur.c.
> > Of course, there have to be some changes in parstxt.c and media.c. I
> > also have to change frame.h to include a pointer to the pigment and
> > the fur struct in the media struct.
>
> Sounds very interesting. Hopefully you will give us an announcement
> (and of course the code) when you feel the code is ready for the public.
No, I will keep them secret :-) (Just kidding)
However, don't expect them in near future. Up and then I will
show a picture here to demonstrate the progress.
> Have you any early idea how your approach rates in terms of render
> time as compared with traditional (object based) procedures?
I think that it's slower, but according to the papers that I read
about hypertexture the resulting images are better at distant views.
Furthermore, the memory consumption is quite low, because everything
is computed "on the fly".
Thomas
--
http://www.fmi.uni-konstanz.de/~willhalm
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Whee... I can see some very interesting things coming up.. (ever noticed that
the beneath of some leaves are "furred") This is good, perhaps vapour still, but
not so dense that we choke...
Thomas Willhalm wrote:
>
> Hello,
>
> this is a snapshot of my current work on fur. I'm sorry for the file
> size, but jpg didn't like this image. I have now implemented a new
> lighting model, which was developped by Kajiya and Kay. I have also
> added the possibility to provide a pigment statement to scattering
> media.
>
> In future, I plan to support other geometries like planes, tori, discs
> and perhaps even blobs. Furthermore, I want to add curliness.
>
> The rendertime has been 30 minutes on a Pentium II 400 running linux.
> Memory consumption (as reported by POV-Ray) has been 85 KB.
>
> Thomas
>
> [Image]
>
> --
> http://www.fmi.uni-konstanz.de/~willhalm
--
//Spider -- [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
And the meek'll inherit what they damn well please
Get ahead, go figure, go ahead and pull the trigger
Everything under the gun
--"Sisters Of Mercy" -- "Under The Gun"
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas Willhalm wrote in message ...
>
>Hello,
>
>this is a snapshot of my current work on fur. I'm sorry for the file
>size, but jpg didn't like this image. I have now implemented a new
>lighting model, which was developped by Kajiya and Kay. I have also
>added the possibility to provide a pigment statement to scattering
>media.
>
>In future, I plan to support other geometries like planes, tori, discs
>and perhaps even blobs. Furthermore, I want to add curliness.
>
>The rendertime has been 30 minutes on a Pentium II 400 running linux.
>Memory consumption (as reported by POV-Ray) has been 85 KB.
>
>Thomas
>
>
Please don't take my criticism to personal or anything like that. This is by
far better than any attempt I have had (which totals 0) But anyway, to me it
looks like a heightfield of random noise wrapped around a sphere and
blurred. I know this isn't what you did actually, I just think it looks like
that. Id say it looks quite like the bristles on a toothbrush as well.
Curl would definately add to the fur realism. The edges look pretty good,
especially at the bottom and top-right. I think it is the dark grey that
makes things look awkward.
But still, nice work. I hope you post pics of furthur developments.
R.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
