 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
These are the clouds you've heard me talk and talk about from time to
time... They could be better, but that's about as good as I can get them...
unless someone invents a special scattering type better suited for
simulating clouds. This uses type 5, because being able to control both
exctinction and eccentricity is key to getting them right. This rendered in
7 minutes, 11 seconds, with me doing other stuff on the PC (400MHz Pentium
II). Enjoy.
Post a reply to this message
Attachments:
Download 'klauds.jpg' (11 KB)
Preview of image 'klauds.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
These are the real clouds I was working on... I posted the one-layer method
before. That layer was only the top layer high above. This includes the
lower layer now. The render time shot up slightly, because the bottom layer
is thicker and requires more intervals (more, as a matter of fact, than I
used in this render). It took 7h, 36m 48s to render at this size. Is that
OK? Or is it too slow? You can always remove the top layer and fake it on
the sky_sphere... I think. (But you'd rather know it's really there,
wouldn't you? :)
Post a reply to this message
Attachments:
Download 'klauds2.jpg' (30 KB)
Preview of image 'klauds2.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
This is looking *very* nice, much better than the images you sent me a few
days ago :) Perhaps you might consider adding a slightly deeper blue to the
top of the skysphere (if that's what you're using for the blue background),
so that the blue doesn't look completely uniform at the top of the image,
since there is already a lighter horizon area.
They look great though, and very realistic IMO, I think that it's worth the
render time when it comes out looking like that!
--
Lance.
http://come.to/the.zone
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Tony[B]" wrote:
>
> These are the real clouds I was working on... I posted the one-layer method
> before. That layer was only the top layer high above. This includes the
> lower layer now. The render time shot up slightly, because the bottom layer
> is thicker and requires more intervals (more, as a matter of fact, than I
> used in this render). It took 7h, 36m 48s to render at this size. Is that
> OK? Or is it too slow? You can always remove the top layer and fake it on
> the sky_sphere... I think. (But you'd rather know it's really there,
> wouldn't you? :)
>
Shot up slightly??? From 7 minutes, 11 seconds to 7h, 36m 48s is
_slightly_?
Or am I mistaken?
Sorry to say, but I wouldn't call these clouds realistic. (Unfortunately
I can't put my finger on problem. shape? or colouring? : greenish-blue,
from scattering probably, I don't think that it could be seen in
reality.)
But for artistical pictures, which do not aim for photorealism, these
clouds could be very useful.
Light source seems to be quite close to clouds, as shadows seem to be
too big for given clouds. Or are there clouds, which are not visible?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Vahur Krouverk wrote:
>Sorry to say, but I wouldn't call these clouds realistic. (Unfortunately
>I can't put my finger on problem. shape? or colouring? : greenish-blue,
>from scattering probably, I don't think that it could be seen in
>reality.)
>
Had the same problem, so while walking to the groceries I looked up a lot.
Tony, it is as if your picture is blurred. Although clouds should look
'fluffy' their boundaries are quite sharp. Maybe a steeper gradient near
the edges of the pattern will make them 'sharper'.
>But for artistical pictures, which do not aim for photorealism, these
>clouds could be very useful.
Agree.
Ingo
--
Photography: http://members.home.nl/ingoogni/
Pov-Ray : http://members.home.nl/seed7/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Wow Tony, them's nice and fluffy.
--
Cheers
Steve email mailto:ste### [at] zeropps uklinuxnet
%HAV-A-NICEDAY Error not enough coffee 0 pps.
web http://www.zeropps.uklinux.net/
or http://start.at/zero-pps
5:19pm up 9 days, 4:40, 3 users, load average: 1.07, 1.06, 1.02
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> This is looking *very* nice, much better than the images you sent me a few
> days ago :)
Excellent. =)
>Perhaps you might consider adding a slightly deeper blue to the
> top of the skysphere (if that's what you're using for the blue
background),
> so that the blue doesn't look completely uniform at the top of the image,
> since there is already a lighter horizon area.
I'll give it a try...
> They look great though, and very realistic IMO, I think that it's worth
the
> render time when it comes out looking like that!
Super! :) I'm very glad you think that.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Shot up slightly??? From 7 minutes, 11 seconds to 7h, 36m 48s is
> _slightly_? Or am I mistaken?
Good point. OK: The render time shot up dramatically, when we compare a
300x200 render with 15 intervals to a 720x480 render with 15+40 intervals
(soon to be 15+45 intervals).
> Sorry to say, but I wouldn't call these clouds realistic. (Unfortunately
> I can't put my finger on problem. shape? or colouring? : greenish-blue,
> from scattering probably, I don't think that it could be seen in
> reality.)
Hmm... I could adjust the color curve in Photoshop after rendering...
> But for artistical pictures, which do not aim for photorealism, these
> clouds could be very useful.
OK, at least I've acheived something.
> Light source seems to be quite close to clouds, as shadows seem to be
> too big for given clouds. Or are there clouds, which are not visible?
The clouds are contained in 2 individual boxes. The first one goes
rom -<1000000,0,1000000> to <1000000,3000,1000000>, and is 4500 units above
the ground plane. The second one goes from -<1000000,0,1000000> to
<1000000,2000,1000000>, and is 15000 units above the ground plane. The
light_source is at y*10000000, and the camera is at <10000,400,-30000>. I
moved it up that high in order to see if the shadows were showing up,
something which when closer to the plane is hard to confirm. What do you
suggest?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Had the same problem, so while walking to the groceries I looked up a lot.
> Tony, it is as if your picture is blurred. Although clouds should look
> 'fluffy' their boundaries are quite sharp. Maybe a steeper gradient near
> the edges of the pattern will make them 'sharper'.
I can do that. I'll adjust this and re-render... but mind you: there are
"generic" realistic clouds, and there are "real" realistic clouds. The first
ones are what we would identify in a render as "realistic", and the second
ones are the ones we might happen to see in real life, and complain about
how "fake" they look. :)
> >But for artistical pictures, which do not aim for photorealism, these
> >clouds could be very useful.
>
> Agreed.
Cool. :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Wow Tony, them's nice and fluffy.
Yup. Ah just aim tuh please y'all! :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Tony[B]" wrote:
>
> OK, at least I've acheived something.
>
That's for sure! I've yet to achieve such results...
> > Light source seems to be quite close to clouds, as shadows seem to be
> > too big for given clouds. Or are there clouds, which are not visible?
>
> The clouds are contained in 2 individual boxes. The first one goes
> rom -<1000000,0,1000000> to <1000000,3000,1000000>, and is 4500 units above
> the ground plane. The second one goes from -<1000000,0,1000000> to
> <1000000,2000,1000000>, and is 15000 units above the ground plane. The
> light_source is at y*10000000, and the camera is at <10000,400,-30000>. I
> moved it up that high in order to see if the shadows were showing up,
> something which when closer to the plane is hard to confirm. What do you
> suggest?
Actually these shadows seem to be too dark, but this is probably result
from one light source. Adding other light source or using radiosity
(hmm, maybe not, considering current rendering time) could improve this.
And now when I think about it, it seems to me, like this scene setup
(camera in such high position) makes it somewhat unrealistic. If camera
is set in "normal" height, then this scene could look more realistic, I
guess. Or perhaps downscaling of clouds will help?
For displaying shadows one can use heightfield and place camera on top
of hill, I guess. Too bad, that rendering time is so big, as it does not
allow to experiment very much...
HTH,
Vahur
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> That's for sure! I've yet to achieve such results...
<big grin> <proud look at my clouds>
> Actually these shadows seem to be too dark, but this is probably result
> from one light source.
I have 2 light_sources, actually, coming from the same spot, one has an
intensity of 1.3, and affects the media (media_attenuation on,
media_interaction on), the other has an intensity of 0.45 and does not
affect the media (media_attenuation off, media_interaction off).
> And now when I think about it, it seems to me, like this scene setup
> (camera in such high position) makes it somewhat unrealistic.
As I said before, I only moved it up that high to see if there were any
shadows or not. This is not mean to be a scene, or to develop into anything.
This was only done to make clouds. I will now copy-paste them into other
scenes, and adapt them as needed.
> Too bad that rendering time is so big, as it does not
> allow to experiment very much...
True, but I'm patient, and have other things to do besides sittig here and
waiting for it to render. :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Well, crew, here's the story. I tried and tried (and tried and...)
to make the edges sharper without increasing render-time, but I'm afraid
it's unavoidable. Unless you add some noise or something, you're all just
going to have to live with the fact that the sharper they are, the more
visible the "planes" are. For example, the ones in the background (and the
ones in my post from the 26th), are rendered with about 12 intervals,
because the edges are really smooth, but the ones in front (the nice, big
fluffy ones) had to be rendered with 50 intervals in this post, and I'm
afraid that that might not be enough! (Zoom in, you'll see what I mean). So,
anyway, here they are. I consider them done. I won't do any more work on
them. Let some other crazy POVer with a faster processor rack his brains
perfecting them. I'm gonna get back to nice, quick-rendering grass...
Especially since I want to exploit my extra 128MB of fresh-installed RAM. :)
Oh, by the way, these took 1 hour, 40 minutes, 17 seconds to render
on my 400MHz Pentium II with 256MB of RAM (joy!) at a resolution of 300x200.
Post a reply to this message
Attachments:
Download 'klauds3.jpg' (12 KB)
Preview of image 'klauds3.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Some days ago, I was watching the sky, and thinking of the huge amount
of cloud images posted here, I said to myself : "If I took a photograph
of it, and posted it on pbi, pretending it's made with POV-Ray, would
they say that it looks realistic ?"... and the more I think of it,
the more it seems that the answer will be "no" is most cases.
The sky looks so often unrealistic ! So, I think there's a huge
difference between what people "idealizes" to be a realistic sky,
and the sheer reality of skieS...
So, to some exent, your clouds (as many others I've seen here
in the past months) *might* actually be realistic ! Just one of the
millions of millions possible skies...
A good support of this thinking is Dave Merchant's "delight" image,
in the "sea" IRTC.
That said, this "quest" for good-looking POV-Ray skies is exciting,
as it shows how you can have a thousand approaches to the same problem,
and how we evolved since the days of the old "sky" texture (you know,
the ugly one that can be seen on "skyvase").
Fabien.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3a4e326b@news.povray.org>, "Tony[B]" <ben### [at] catholicorg>
wrote:
> For example, the ones in the background (and the ones in my post from
> the 26th), are rendered with about 12 intervals, because the edges
> are really smooth, but the ones in front (the nice, big fluffy ones)
> had to be rendered with 50 intervals in this post, and I'm afraid
> that that might not be enough! (Zoom in, you'll see what I mean).
I suspect the problem is that you are using too many intervals, and that
method 3 can only do it's anti-aliasing within an interval. Since you
have so many intervals, many of which are entirely in flat areas which
only need a few samples, you are pushing up the minimum total number of
samples.
Try simply increasing the number of samples and adjusting the
aa_threshold and aa_level values instead, so it can skimp out on the
flat areas...you shouldn't even need to have more than 1 interval in
MegaPOV.
--
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/
<><
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
These look great Tony, I hope you've made the code
available somewhere. I'm sure that soon someone will
start to build on the progress that you have made here.
Thanks.
--
Cheers
Steve email mailto:ste### [at] zeroppsuklinuxnet
%HAV-A-NICEDAY Error not enough coffee 0 pps.
web http://www.zeropps.uklinux.net/
or http://start.at/zero-pps
10:51pm up 12 days, 10:13, 3 users, load average: 1.14, 1.24, 1.17
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I'm using method 2. Method 3 changes the color of the clouds in nassty
ways... Besides, you can use whatever method you want, and jitter and
whatnot. These are mine, and this is how they stay.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I could post the code to pbsf or ptsf... which would you prefer? Do you want
comments?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Looking very good, how about including a simple HF for the groud, might help
with the quest for realism :)
--
Rick
POV-Ray News & Resources - http://povray.co.uk
Kitty5 WebDesign - http://kitty5.com
Hi-Impact web site design & database driven e-commerce
TEL : +44 (01625) 266358 - FAX : +44 (01625) 611913 - ICQ : 15776037
PGP Public Key
http://pgpkeys.mit.edu:11371/pks/lookup?op=get&search=0x231E1CEA
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Patience... I'm working on a real scene that'll have them... :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Tony[B] wrote:
>
> I'm using method 2. Method 3 changes the color of the clouds in nassty
> ways...
Unless you use mpov... ;)
Jerome
--
******************************* Jérôme M. BERGER
* Doctor Jekyll had something * mailto:ber### [at] inamecom
* to Hyde... * http://www.enst.fr/~jberger
*******************************
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Unless you use mpov... ;)
I know! That's why I'm pushing for your fixes to be in MegaPOV! I don't want
this to happen any more. :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
