 |
|
|
|
|
|
| |
| |
|
|
From: Roberto Amorim
Subject: Re: More Faked Volumetrics Update (31k total, 2 jpegs)
Date: 8 Apr 2004 15:51:48
Message: <4075ad54@news.povray.org>
|
|
|
| |
| |
|
|
> I attached two more images. The first one has the surface normal in
> place and uses aa. The second uses no surface normal, but still has aa.
> The render times are diplayed in the lower right corner of each image.
Number two looks better (IMHO) and renders faster. Looks like a winner to
me.
Nice technique, BTW.
Regards!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Uzytkownik "Samuel Benge" <stb### [at] hotmail com> napisal w wiadomosci
news:407### [at] hotmailcom...
> Hello everyone! Here's an idea I've been thinking of for quite a while.
> I'm pretty sure somebody has done something like this before... it might
> have been a grass texture, though.
>
> What you see is actually a flat plane layered with over eighty
> semitransparent textures. Each additional texture is scaled slightly
> larger than the last, and uses a slightly altered color_map (actually a
> pigment_map, but it doesn't really need to be one anymore). The texture
> is then translated to the camera's X & Z coordinates to achieve a more
> proper illusion.
>
> I can't really see a practical application for this trick. Even though
> this image rendered in a little over five minutes, it can get very slow
> when antialiasing is applied :( Also, POV-Ray tends to throw an error
> if the number of texture layers goes over 100.
It may come handy at rendering backdrops to aerial scenery (e.g. a flying
aeroplane).
>
> I still have yet to try it with radiosity. There's no reason why it
> shouldn't work.
>
> The code has been placed (a little too hastily) to
> povray.binaries.scene-images.
>
> Questions and comments are always welcome-
>
> -Samuel Benge
>
Respectfully
Matt
----------------------------------------------------------------------------
----
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Samuel Benge wrote:
> Hello everyone! Here's an idea I've been thinking of for quite a while.
> I'm pretty sure somebody has done something like this before... it might
> have been a grass texture, though.
<snip:technical />
BRILLIANT idea! It may cause some challenges for the renderer if objects
are within those planes. Have you found any trouble with, say, a
sphere in the middle of the clouds?
--
Respectfully,
Dan P
http://<broken link>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Samuel Benge nous apporta ses lumieres ainsi en ce 2004/04/08 15:33... :
> Samuel Benge wrote:
>
>> I can't really see a practical application for this trick. Even
>> though this image rendered in a little over five minutes, it can get
>> very slow when antialiasing is applied :(
>
>
>
> Update:
> I just tried it again with antialising, and the render time didn't
> really go up that much. Evidently, the last time I tried aa the
> texture had a lot of noise in it. Also, removing the surface normal
> considerably decreases render time.
>
> I attached two more images. The first one has the surface normal in
> place and uses aa. The second uses no surface normal, but still has
> aa. The render times are diplayed in the lower right corner of each
> image.
>
> -Samuel Benge
>
>
>
>
> ------------------------------------------------------------------------
>
>
> ------------------------------------------------------------------------
>
Second one looks beter for a flyby over some clouds, the first should be
beter for a snowy field. For me anyway.
Alain
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <4075d862$1@news.povray.org>,
Dan P <dan### [at] yahoocom> wrote:
> BRILLIANT idea! It may cause some challenges for the renderer if objects
> are within those planes. Have you found any trouble with, say, a
> sphere in the middle of the clouds?
Well, since it's on a 2D surface, it'd look a bit odd...
BTW, I vaguely recall some very good sky_sphere pigments done using this
technique. One advantage there is that you don't have any intersection
or finish/normal calculations, so it renders faster. The layered pigment
feature of sky_spheres seems to be overlooked far too much.
--
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:
> In article <4075d862$1@news.povray.org>,
> Dan P <dan### [at] yahoocom> wrote:
>
>
>>BRILLIANT idea! It may cause some challenges for the renderer if objects
>> are within those planes. Have you found any trouble with, say, a
>>sphere in the middle of the clouds?
>
>
> Well, since it's on a 2D surface, it'd look a bit odd...
>
> BTW, I vaguely recall some very good sky_sphere pigments done using this
> technique. One advantage there is that you don't have any intersection
> or finish/normal calculations, so it renders faster. The layered pigment
> feature of sky_spheres seems to be overlooked far too much.
I misunderstood what he did; I thought he used many planes with seperate
transparent image_maps. I should have looked at the source; my bad.
--
Respectfully,
Dan P
http://<broken link>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christopher James Huff wrote:
> BTW, I vaguely recall some very good sky_sphere pigments done using this
> technique. One advantage there is that you don't have any intersection
> or finish/normal calculations, so it renders faster. The layered pigment
> feature of sky_spheres seems to be overlooked far too much.
Looks like something I'll have to try. Thanks.
-Samuel Benge
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Samuel Benge" <stb### [at] hotmailcom> wrote in message
news:407### [at] hotmailcom...
> I can't really see a practical application for this trick. Even though
> this image rendered in a little over five minutes, it can get very slow
> when antialiasing is applied :( Also, POV-Ray tends to throw an error
> if the number of texture layers goes over 100.
>
> -Samuel Benge
>
*drool* Practical applications are overrated and can alwasy come later :)
this alone is cool.
Post a reply to this message
|
|
| |
| |
|
|
From: Tim Nikias v2 0
Subject: Re: More Faked Volumetrics (35k jpeg)
Date: 9 Apr 2004 12:06:35
Message: <4076ca0b@news.povray.org>
|
|
|
| |
| |
|
|
Note that this only works when looking at the origin. Try looking at a spot
300 units to the right/left. The scaling scews the volumetric effect
outwards. That aside, neat thing.
--
"Tim Nikias v2.0"
Homepage: <http://www.nolights.de>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Dan P wrote:
> BRILLIANT idea! It may cause some challenges for the renderer if
> objects
> are within those planes. Have you found any trouble with, say, a
> sphere in the middle of the clouds?
Why not a plane? (Or both, scnr)
Dave
--
Dave "tPassive" Vogt| ruby -e "n=gets.to_i;a=(2..n).to_a;a.each{
Linux user #225040| |x|puts x;(x*2).step(n,x){|j|a.delete j}}"
____________________|_____________________________ |
PGP Key: http://frozenbrain.com/public_key.asc \____________/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
