 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In POV 3.5 I would like to see the following implementations:
- True wavelength-dependent interference
- Diffraction
- Optical birefringence
- Polarization and polarizing filters, including ior-dependent polarized
reflection
- Non-uniform ior for object interiors
- Differentiation between absorption and interference filtration
<G> >;)
--
David Fontaine <dav### [at] faricy net> ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
David Fontaine wrote:
>
> In POV 3.5 I would like to see the following implementations:
> - True wavelength-dependent interference
> - Diffraction
> - Optical birefringence
> - Polarization and polarizing filters, including ior-dependent polarized
> reflection
> - Non-uniform ior for object interiors
> - Differentiation between absorption and interference filtration
...and a partridge in a pear tree.
Time for you to learn how to write your own patches, David.
--
Ken Tyler - 1400+ POV-Ray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 09 Aug 2000 14:37:37 -0500, David Fontaine <dav### [at] faricynet>
wrote:
>In POV 3.5 I would like to see the following implementations:
>- True wavelength-dependent interference
>- Polarization and polarizing filters, including ior-dependent polarized
>reflection
>- Differentiation between absorption and interference filtration
The main difficulty with the above features is conversion from rgb to
spectrum space. Simply said, it's impossible. It can be faked in a lot
of ways like rgb -> hsb conversion but they will not come close to the
real thing as spectra are usually very complex with emission and
absorption lines and regions and so on.
Now, if all colors were represented in spectrum space, things would be
much easier. I don't think users will love tweaking such colors,
though :)
>- Diffraction
>- Optical birefringence
Diffraction is tricky. I don't think it's impossible but it would
probably require insane render times (unless someone comes up with a
algorithm of following a wave front which does not require extensive
Monte Carlo sampling).
>- Non-uniform ior for object interiors
This has been discussed more than enough so I'll just skip it.
Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG e-mail : pet### [at] tagpovrayorg
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Peter Popov wrote:
> The main difficulty with the above features is conversion from rgb to
[snip]
I was of course only joking, I selectively listed the most
computation-intensive real-world properties and the ones that would require
massive restructuring of the entire POV code.
--
David Fontaine <dav### [at] faricynet> ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
David Fontaine <dav### [at] faricynet> wrote:
: In POV 3.5 I would like to see the following implementations:
: - True wavelength-dependent interference
: - Diffraction
: - Optical birefringence
: - Polarization and polarizing filters, including ior-dependent polarized
: reflection
: - Non-uniform ior for object interiors
: - Differentiation between absorption and interference filtration
It's easy to say, but how easy to implement? You are free to post your own
ideas about the algorithms that will implement those.
And by the way, about the non-uniform ior in objects interiors:
http://www.students.tut.fi/~warp/povVFAQ/languageVFAQ.html#variableior
--
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3991B301.80EBA665@faricy.net>, David Fontaine
<dav### [at] faricynet> wrote:
> In POV 3.5 I would like to see the following implementations:
> - True wavelength-dependent interference
> - Diffraction
> - Optical birefringence
> - Polarization and polarizing filters, including ior-dependent polarized
> reflection
> - Non-uniform ior for object interiors
> - Differentiation between absorption and interference filtration
You forgot these:
- True displacement mapping for all objects, using curved rays
- Relativistic effects, with adjustable light speed
- Differential ray tracing
- Custom BDRF functions
- 5x faster rendering and parsing
--
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Thu, 10 Aug 2000 08:17:43 -0500, Chris Huff <chr### [at] maccom>
wrote:
> - Relativistic effects, with adjustable light speed
This has already been implemented in another raytracer so it's not
that impossible.
Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG e-mail : pet### [at] tagpovrayorg
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <10n5ps4sjuvnbl6398bihmen322htdppue@4ax.com>, Peter Popov
<pet### [at] usanet> wrote:
> This has already been implemented in another raytracer so it's not
> that impossible.
So have some of the other things, but they would all require large
changes in the architecture and source code. Adding it to POV would
probably be impossible...it would probably be easier to write a new
renderer that supports those things.
--
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> - 5x faster rendering and parsing
there is a solution for this right? getting a computer that is 5X
faster :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <399311B7.AAB467A0@yahoo.com>, ryan constantine
<rco### [at] yahoocom> wrote:
> there is a solution for this right? getting a computer that is 5X
> faster :)
But that costs more than a POV upgrade. :-)
--
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Huff wrote:
> In article <10n5ps4sjuvnbl6398bihmen322htdppue@4ax.com>, Peter Popov
> <pet### [at] usanet> wrote:
>
> > This has already been implemented in another raytracer so it's not
> > that impossible.
>
> So have some of the other things, but they would all require large
> changes in the architecture and source code. Adding it to POV would
> probably be impossible...it would probably be easier to write a new
> renderer that supports those things.
There is a link to a POV patch (over v2.2) that does this in
http://povray.org/links/3D_Programs/POV-Ray_Unofficial_Patches_and_Front_End_Programs/POV-Ray_Patches_v2_2/
--
Bye
Pabs
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Huff <chr### [at] maccom> wrote...
> - True displacement mapping for all objects, using curved rays
Would we want to do it any other way?
> - Relativistic effects, with adjustable light speed
Ooooh... sounds like fun. :-) (Yes, I've seen the animations from the
renderer that already does this.)
> - Custom BDRF functions
Actually, this shouldn't be too far off as a MegaPov feature. (of course,
post-3.5)
> - 5x faster rendering and parsing
My favorite by far. Anybody want to volunteer to write this patch?
-Nathan
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
> David Fontaine <dav### [at] faricynet> wrote:
> : In POV 3.5 I would like to see the following implementations:
> : - True wavelength-dependent interference
> : - Diffraction
> : - Optical birefringence
> : - Polarization and polarizing filters, including ior-dependent polarized
> : reflection
> : - Non-uniform ior for object interiors
> : - Differentiation between absorption and interference filtration
>
> It's easy to say, but how easy to implement? You are free to post your own
> ideas about the algorithms that will implement those.
>
> And by the way, about the non-uniform ior in objects interiors:
>
> http://www.students.tut.fi/~warp/povVFAQ/languageVFAQ.html#variableior
I already know why it won't be implemented, you notice there was an evil
grinning smiley a few lines down... >;)
--
David Fontaine <dav### [at] faricynet> ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nathan Kopp wrote:
> > - 5x faster rendering and parsing
>
> My favorite by far. Anybody want to volunteer to write this patch?
Free and easy! Just change some jumpers on the motherboard... :)
--
David Fontaine <dav### [at] faricynet> ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
> And by the way, about the non-uniform ior in objects interiors:
>
> http://www.students.tut.fi/~warp/povVFAQ/languageVFAQ.html#variableior
BTW, this FAQ also talks about non-uniform transformations, saying they are only
possible in mesh modellers by moving the vertexes. Indded they are quite
possible in MegaPOV isosurfaces.
--
David Fontaine <dav### [at] faricynet> ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
David Fontaine <dav### [at] faricynet> wrote:
: BTW, this FAQ also talks about non-uniform transformations, saying they are only
: possible in mesh modellers by moving the vertexes. Indded they are quite
: possible in MegaPOV isosurfaces.
Firstly: You can't modify an arbitrary object with an isosurface.
Secondly: Yes, you can modify an isosurface function with another function.
But is this really a transformation? When you modify a function with another
function, you just create a third function. There are no specific
transformations involved.
Yes, you can, for example, convert a cylinder-shaped isosurface into a
cone-shaped isosurface. This is, however, done by actually creating another
function, not by transforming.
Of course it is possible to think that any transformations are possible
for an isosurface if you like.
--
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
David Fontaine <dav### [at] faricynet> wrote:
: Free and easy! Just change some jumpers on the motherboard... :)
I think that if you try to make your computer 5 times faster by overclocking
(and notice that it's not enough to se the MHz amount 5 times larger for
this), you will fry your CPU.
--
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Huff wrote:
>
> In article <399311B7.AAB467A0@yahoo.com>, ryan constantine
> <rco### [at] yahoocom> wrote:
>
> > there is a solution for this right? getting a computer that is 5X
> > faster :)
>
> But that costs more than a POV upgrade. :-)
For the user, yes. Doesn't cost the POV-Team a dime.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 11 Aug 2000 05:20:24 -0400, Warp wrote:
>David Fontaine <dav### [at] faricynet> wrote:
>: Free and easy! Just change some jumpers on the motherboard... :)
>
> I think that if you try to make your computer 5 times faster by overclocking
>(and notice that it's not enough to se the MHz amount 5 times larger for
>this), you will fry your CPU.
That depends on where you started. If I'd started with my K6/233 at 50 MHz,
then decided to overclock to five times that, it'd probably work (but it'd
run hot... my K6 is the last .35u K6 made, for just that reason.)
--
Ron Parker http://www2.fwi.com/~parkerr/traces.html
My opinions. Mine. Not anyone else's.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <39935BE7.AB4FF0BB@hotmail.com>, Pabs <pab### [at] hotmailcom>
wrote:
> There is a link to a POV patch (over v2.2) that does this in
> http://povray.org/links/3D_Programs/POV-Ray_Unofficial_Patches_and_Front_E
> nd_Programs/POV-Ray_Patches_v2_2/
Ok, I will have to expand the requirements: full simulation of frequency
shifting(Doppler effect), taking gravitational effects into account, etc.
--
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3993c4bd@news.povray.org>, Warp <war### [at] tagpovrayorg>
wrote:
> But is this really a transformation? When you modify a function with
> another function, you just create a third function. There are no
> specific transformations involved.
> Yes, you can, for example, convert a cylinder-shaped isosurface into a
> cone-shaped isosurface. This is, however, done by actually creating
> another function, not by transforming.
You can use a function as a pattern in a pigment, use a warp(a
non-uniform transform) on that pigment, and then use that pigment in the
function for the final isosurface.
--
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <399377e1@news.povray.org>, "Nathan Kopp" <Nat### [at] Koppcom>
wrote:
> Ooooh... sounds like fun. :-) (Yes, I've seen the animations from the
> renderer that already does this.)
Animations? All I saw were stills...but that was a while ago, maybe the
animations were added since then.
> > - Custom BDRF functions
>
> Actually, this shouldn't be too far off as a MegaPov feature. (of course,
> post-3.5)
How would it be implemented? The only thing I can think of is a set of
splines specifying diffuse, specular reflection, etc. for the different
angles and overriding the ordinary finish settings. Basically, a new
finish model.
Of course, it would help if I knew something about BDRF functions. :-)
My knowledge of them is currently similar to my knowledge of shaders: I
know what they do, but I don't know the specifics of what they are like.
--
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
>
> David Fontaine <dav### [at] faricynet> wrote:
> : BTW, this FAQ also talks about non-uniform transformations, saying they are only
> : possible in mesh modellers by moving the vertexes. Indded they are quite
> : possible in MegaPOV isosurfaces.
>
> Firstly: You can't modify an arbitrary object with an isosurface.
Wouldn't the proximity pattern allow such a thing?
OK, I know, it's not possible to give an example before 2050 because of
rendering times...
__ __ __ __ _
| | / \ / / |_ / |/
\/\/ \__/ /_ /_ |__ \_ |\
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <39945B44.5BBE192C@club-internet.fr>, Francois Dispot
<woz### [at] club-internetfr> wrote:
> Wouldn't the proximity pattern allow such a thing?
> OK, I know, it's not possible to give an example before 2050 because of
> rendering times...
The granularity would be a problem...but I am thinking of a way to add
per-object proximity functions, so some objects could have their own,
much computationally cheaper way of calculating proximity. The proximity
pattern might become useful in isosurfaces then...
My blob pattern isn't limited to blob components, it can also use boxes,
pigments, and other components are planned(torus, superellipsoid, etc.).
I have had good results with using it in isosurfaces.
--
Christopher James Huff - Personal e-mail: chr### [at] maccom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://homepage.mac.com/chrishuff/
TAG Web page: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
> Firstly: You can't modify an arbitrary object with an isosurface.
No. But with enough highly complicated pain-in-the-ass slow-calculating functions
almost any object can be translated into an isosurface...
> Secondly: Yes, you can modify an isosurface function with another function.
> But is this really a transformation? When you modify a function with another
> function, you just create a third function. There are no specific
> transformations involved.
> Yes, you can, for example, convert a cylinder-shaped isosurface into a
> cone-shaped isosurface. This is, however, done by actually creating another
> function, not by transforming.
> Of course it is possible to think that any transformations are possible
> for an isosurface if you like.
But isn't that the same thing? I mean, you could create a cylinder in POV and
transform it and really have POV be finding a new function, and get the same result.
--
David Fontaine <dav### [at] faricynet> ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
ryan constantine wrote:
> > - 5x faster rendering and parsing
>
> there is a solution for this right? getting a computer that is 5X
> faster :)
Ya know, if trends continue, we should have 16Ghz home PCs by 2006.
Although some say we are pushing the limits of what silicon-based chips
can take... (Fortunately we'll have copper-based and photonic chips soon
:) *crossing fingers*)
And while processors are growing at a rate of 2^x, storage capacity is
growing at a rate of 2^(2^x)!!!
--
David Fontaine <dav### [at] faricynet> ICQ 55354965
Please visit my website: http://www.faricy.net/~davidf/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
David Fontaine wrote:
>
> And while processors are growing at a rate of 2^x, storage capacity is
> growing at a rate of 2^(2^x)!!!
And the ability of Windoze to handle it is growing at the rate of
sqrt(x)...
But I'll be interested in that storage space when I get around to
making feature-length films. 90 minutes of 704x480 frames at 24fps
works out to over 130 GB.
Regards,
John
--
ICQ: 46085459
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
John VanSickle wrote:
>
> David Fontaine wrote:
> >
> > And while processors are growing at a rate of 2^x, storage capacity is
> > growing at a rate of 2^(2^x)!!!
>
> And the ability of Windoze to handle it is growing at the rate of
> sqrt(x)...
>
> But I'll be interested in that storage space when I get around to
> making feature-length films. 90 minutes of 704x480 frames at 24fps
> works out to over 130 GB.
Hmm, i guestimate 111 GB :) (900KB per frame .png average. Plus then
you need space to compress that on to. (And at least a couple of iterations
of test mpegs too...)
My new machine (coming next week, dual P3/933 pant pant) only has a 73GB SCSI.
I guess no feature films for me. :)
dik
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Dick Balaska wrote:
>
> John VanSickle wrote:
> >
> > David Fontaine wrote:
> > >
> > > And while processors are growing at a rate of 2^x, storage
> > > capacity is growing at a rate of 2^(2^x)!!!
> >
> > And the ability of Windoze to handle it is growing at the rate of
> > sqrt(x)...
> >
> > But I'll be interested in that storage space when I get around to
> > making feature-length films. 90 minutes of 704x480 frames at 24fps
> > works out to over 130 GB.
>
> Hmm, i guestimate 111 GB :) (900KB per frame .png average. Plus then
> you need space to compress that on to. (And at least a couple of
> iterations of test mpegs too...)
>
> My new machine (coming next week, dual P3/933 pant pant) only has a
> 73GB SCSI. I guess no feature films for me. :)
You can always zip down the png's. I was calculating based on
TGAs, which zip down to about 40% of original size.
Regards,
John
--
ICQ: 46085459
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3996F731.B984158D@erols.com> , John VanSickle
<van### [at] erolscom> wrote:
> You can always zip down the png's. I was calculating based on
> TGAs, which zip down to about 40% of original size.
The PNG image format already uses ZIP compression!
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
John VanSickle wrote:
> But I'll be interested in that storage space when I get around to
> making feature-length films. 90 minutes of 704x480 frames at 24fps
> works out to over 130 GB.
>
You have to store scenes as well. As someone from Pixar recently wrote
in c.g.r.renderman, average scene description size is 500 MB and big
scene is 2 GB (compressed!) per frame!. Add to this texture maps, which
could take even more space. This of course applies to RIB format and
perhaps POV-Ray will not require such amount of data.
(I think that you can't avoid texture images, if you want to render fast
and good scenes: ARM describes, how folks in Pixar at first wanted to
create all textures as procedural, but this proved to be too
inefficient).
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thorsten Froehlich wrote:
>
> In article <3996F731.B984158D@erols.com> , John VanSickle
> <van### [at] erolscom> wrote:
>
> > You can always zip down the png's. I was calculating based on
> > TGAs, which zip down to about 40% of original size.
>
> The PNG image format already uses ZIP compression!
Learn something new every day...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
John VanSickle <van### [at] erolscom> wrote:
: Learn something new every day...
How did you think png images are smaller than tga images?
--
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Warp" <war### [at] tagpovrayorg> wrote in message
news:3997f82f@news.povray.org...
| John VanSickle <van### [at] erolscom> wrote:
| : Learn something new every day...
|
| How did you think png images are smaller than tga images?
I know I've zip-filed them before and essentially got no further
compression.
Testing this out real quick using the maximum setting in my zipping program
I get the following.
Targa files (uncompressed) will go to about 0.4 times their original size,
while compressed TGA starts out at half that of uncompressed TGA then goes
to about the same size as the other. Jpg might actually increase in file
size. And indeed PNG remains the same (as seen through Windows Explorer
stats).
Bob
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
>
> John VanSickle <van### [at] erolscom> wrote:
> : Learn something new every day...
>
> How did you think png images are smaller than tga images?
I've never used png's, so I didn't know they were smaller.
Regards,
John
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
