After spending a while trying to figure out how to generate realistic,
fast, foolproof clouds in POV, I found my ideal solution.

The water (plane with turbulated crackle normal) and boat (model I'm
building IRL) are just junk objects to fill in the scene a bit.

I am very impressed with the results! Hopefully, this will allow me to
get an IRTC entry in on time (haven't started yet, less than a week to
go!)


Simon

Post a reply to this message

Attachments:
Download 'seaskytest1jpg.jpg' (66 KB)

Preview of image 'seaskytest1jpg.jpg'

Those clouds are the best I've seen POV produce so far. Extraordinarily
realistic. AWSOME ROLEX! even.


--
Paul Vanukoff



"Simon de Vet" <sde### [at] istarca> wrote in message
news:39A717D0.ADBEA992@istar.ca...
> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
>
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
>
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
>
>
> Simon
>


----------------------------------------------------------------------------
----

Post a reply to this message

Wow!  These are wonderful clouds!  Can you make different kinds of clouds?
Brendan

Simon de Vet wrote:

> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
>
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
>
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
>
> Simon
>
>   ------------------------------------------------------------------------
>  [Image]

Post a reply to this message

Whoa!  Simon, take a bow!

Post a reply to this message

"Simon de Vet" <sde### [at] istarca> wrote

    Well, I say that just about does it.... as soon as we all figure out how
you did that we can start looking for some other ultimate test of
raytracing...

Post a reply to this message

Ok, how'd you do it?  Wow.

Simon de Vet wrote:

> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
>
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
>
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
>
> Simon
>
>   ------------------------------------------------------------------------
>  [Image]

Post a reply to this message

Simon de Vet wrote in message <39A717D0.ADBEA992@istar.ca>...
>After spending a while trying to figure out how to generate realistic,
>fast, foolproof clouds in POV, I found my ideal solution.



These clouds suck. I mean, seriously, why post this junk.

I have absolutely no interest whatsoever as to your code.

Honestly, sometimes I wonder why people post this cr*p.

If you do not 'read' this as sarcastic you would be mistaken.

Peter Warren
Proud Povray user since 1996

ps. how long?

Post a reply to this message

Peter Warren wrote:

> ps. how long?

How long to figure it out? About 2 days.How long to render? About 2
minutes, with radiosity (1 min without)
How long till I explain the method? Tomorrow, I hope. I want to read more
ego-boosting comments first :)


Simon

Post a reply to this message

Ahh crap that's it I'm going home!:-)

Simon de Vet wrote:

> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
>
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
>
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
>
> Simon
>
>   ------------------------------------------------------------------------
>  [Image]

--
Come visit my web site:-) : http://www.geocities.com/~thomaslake/

Post a reply to this message

"Bill DeWitt" <the### [at] earthlinknet> wrote in message
news:39a733fe$1@news.povray.org...
>     Well, I say that just about does it.... as soon as we all figure out
how
> you did that we can start looking for some other ultimate test of
> raytracing...

How about lightning?

Post a reply to this message

Like these: http://www.stormguy.com/light1.htm

Post a reply to this message

Fair enough, but all I have are questions. :)

Grim

Post a reply to this message

Hey,

<ego_boosting_comment> Cool image! </ego_boosting_comment>

Now some observations :) Firstly, thought it is hard to tell with the water,
there doesn't seem to be any shadows cast, either on the water surface or on
other clouds. Also, judging from the ship the lighting is quite low, and
almost directly to the left of the ship - yet teh lighting on the clouds
seems to be quite high and from a source somewhere behind the camera. So I
am guessing that the clouds aren't a real-time lit object. I don't want to
hur any feeling here, but ... imagemap?  Just a guess, can't wait to find
out how you actually did it :)

Equiprawn

Post a reply to this message

WOW those clouds are the best!! You can be extremely proud of them. If
it's really procedural and renders so fast, I *very much* look forward
to see the code!! I can't imagine how you came so close to reality with
a relative simple routine (must be simple for a fast render). Can one
fly through them?  I suppose no..

But don't forget to post tomorrow.  8-))  The water is also great btw,
and the horisont.


Simon de Vet wrote:
> 
> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
> 
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
> 
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
> 
> Simon

Post a reply to this message

I sure hope you aren't jerking us around, because I'm impressed and that
never happens!

-Mike

Post a reply to this message

Simon de Vet wrote:
> 
> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
> 
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
> 
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
> 

I'm really a bit disturbed by this.  To sum it up:

Extremely realistic, foolproof clouds taking only 1 or 2 minutes to render and
that were developed during 2 days.  Sorry, but i think there has to be something
wrong :-)

1 / 2 minutes for w/o radiosity would be quite impossible for 3d clouds. 
Equiprawn has also made some observations in that direction.  There are strong
hints towards image_map (i really hope not !), even though it could be some
procedural texture.

We are all waiting for the solution :-)

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

you know, i think you have cracked it! - top notch clouds!!

Rick

Post a reply to this message

Waiting for the code for them there clouds:

Maybe it's some sort of clue as to how you made these, but
I see a faint terquoise band about two inches down from
the top of the image, it fades away when I apply some gamma
correction (so only visible if you darken it perhaps).

OK lets see a night time scene or a twilight scene with 
these, sunsets might be interesting. 

-- 
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

 12:22pm  up 4 days, 16:40,  2 users,  load average: 1.94, 1.42, 1.15

Post a reply to this message

Image map, not procedural.

Simple, but not necessarily easy to implement convincingly. It is my
observation that merging "reality" with CG effectively, highlights not
so much the technical expertise of the the user, but their artistic
skill and understanding of the medium.

I think this effort is pretty good so far and I look forward to seeing
the final piece.

Abe


Simon de Vet wrote:
> 
> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
> 
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
> 
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
> 
> Simon
> 
>   ------------------------------------------------------------------------
>  [Image]

Post a reply to this message

"Christoph Hormann" <Chr### [at] schunteretctu-bsde> wrote :
>
>  There are strong
> hints towards image_map

    Say it ain't so!

Post a reply to this message

This looks suspiciously like the original source was a photo...!

"Simon de Vet" <sde### [at] istarca> wrote in message 
news:39A717D0.ADBEA992@istar.ca...
> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
> 
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
> 
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
> 
> 
> Simon
>

Post a reply to this message

"Simon de Vet" <sde### [at] istarca> wrote in message
news:39A717D0.ADBEA992@istar.ca...
> After spending a while trying to figure out how to generate realistic,
> fast, foolproof clouds in POV, I found my ideal solution.
>
> The water (plane with turbulated crackle normal) and boat (model I'm
> building IRL) are just junk objects to fill in the scene a bit.
>
> I am very impressed with the results! Hopefully, this will allow me to
> get an IRTC entry in on time (haven't started yet, less than a week to
> go!)
>
>
> Simon
>

 [image]

      Hi Simon.
                    You have created a bit of a storm, I fear! :o)  Fair
play to you. I think that your clouds are extremely good.  Equiprawn has
commented that they can't see any shadows cast onto other clouds, but I can
see some, as they would/should be.

  Just a thought,  I believe that waves are also very hard to do. Could your
clouds somehow  be used for this? I've a feeling that it is a possibility,
ie; white water, waves crashing, etc. I think that this would also look
real.

         ~Steve~

Post a reply to this message

Equiprawn wrote:

> Hey,
>
> <ego_boosting_comment> Cool image! </ego_boosting_comment>
>
> Now some observations :) Firstly, thought it is hard to tell with the water,
> there doesn't seem to be any shadows cast, either on the water surface or on
> other clouds. Also, judging from the ship the lighting is quite low, and
> almost directly to the left of the ship - yet teh lighting on the clouds
> seems to be quite high and from a source somewhere behind the camera. So I
> am guessing that the clouds aren't a real-time lit object. I don't want to
> hur any feeling here, but ... imagemap?

Ding ding! We have a winner!

I guess this shows that, on occasion, imagemaps aren't the terrible solution
they're made out to be.

I am planning, however, to create some POV code to generate semi-realistic
clouds procedurally. In my experimentation, I have made some progress, and I
think it should be do-able.

In the mean time, with time running out before the IRTC entry is due, I found a
very reasolable stop-gap solution, as all the positive comments attest to.


Simon

Post a reply to this message

Ahhhhh, you got us all excited for nothing!

Here I thought someone found the grail...

-Mike

Post a reply to this message

Btw, just getting your goat, assuming you have one.  ;)

-Mike

Post a reply to this message

Simon de Vet wrote in message <39A74B70.9EF7F5E0@istar.ca>...
>
>
>Peter Warren wrote:
>
>> ps. how long?
>
>How long to figure it out? About 2 days.How long to render? About 2
>minutes, with radiosity (1 min without)
>How long till I explain the method? Tomorrow, I hope. I want to read more
>ego-boosting comments first :)


2 minutes?
Consider you ego boosted.

Peter Warren
Proud Povray user since 1996

Post a reply to this message

Image_map and IRTC? Bite your tongue!

Grim

Post a reply to this message

Simon de Vet wrote:
> 
> Ding ding! We have a winner!
> 
> I guess this shows that, on occasion, imagemaps aren't the terrible solution
> they're made out to be.
> 

Thanks for the day of hopeful doubt :-)

This one really isn't that bad, even though i'm not so sure, whether light
position of scene and image_map fit together.  NTL, it's well done, the fact
that nobody was able to tell exactly that it is image_map speaks for it

> I am planning, however, to create some POV code to generate semi-realistic
> clouds procedurally. In my experimentation, I have made some progress, and I
> think it should be do-able.
> 

But not with that rendering time :-)

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Wow!

Post a reply to this message

I think the person who said "AWESOME ROLEX" was the most accurate this
time. Nicely done, tho. You fooled (almost) everybody!

Simon de Vet wrote:
> 
> Equiprawn wrote:
> 
> > Hey,
> >
> > <ego_boosting_comment> Cool image! </ego_boosting_comment>
> >
> > Now some observations :) Firstly, thought it is hard to tell with the water,
> > there doesn't seem to be any shadows cast, either on the water surface or on
> > other clouds. Also, judging from the ship the lighting is quite low, and
> > almost directly to the left of the ship - yet teh lighting on the clouds
> > seems to be quite high and from a source somewhere behind the camera. So I
> > am guessing that the clouds aren't a real-time lit object. I don't want to
> > hur any feeling here, but ... imagemap?
> 
> Ding ding! We have a winner!
> 
> I guess this shows that, on occasion, imagemaps aren't the terrible solution
> they're made out to be.
> 
> I am planning, however, to create some POV code to generate semi-realistic
> clouds procedurally. In my experimentation, I have made some progress, and I
> think it should be do-able.
> 
> In the mean time, with time running out before the IRTC entry is due, I found a
> very reasolable stop-gap solution, as all the positive comments attest to.
> 
> Simon

Post a reply to this message

Christoph Hormann wrote:

> Simon de Vet wrote:
>
> > I am planning, however, to create some POV code to generate semi-realistic
> > clouds procedurally. In my experimentation, I have made some progress, and I
> > think it should be do-able.
> >
>
> But not with that rendering time :-)

I'm not so sure about this.

I've looked at other people's cloud techniques, and they all use 3-d to generate
3-d looking clouds. Some use media, or isosurfaces, or stacked planes, but all use
actuall objects.

When I look at the clouds around me, most are way too far away to look 3-d. 3-d
objects are usually shaded based on the angle of the incomming light (diffuse
shading) and on self shadowing. Clouds, on the other hand, have a bit of this, but
the vast majority of the shading comes from the extinction of light as it passes
through the water vapour.

I've been looking at clouds not as 3-d objects, but as 2-d objects, painted onto a
giant sphere. If treated this way, I think that most of the effects can be
generated with complex textures on a skysphere, which are very quick to render.

http://freespace.virgin.net/hugo.elias/models/m_main.htm has a good (award
winning) example of a pigment based cloud render.

In addition, pigments can generate clouds other than puffy cumulus, like cirrus,
stratus, and realistic haze.


I'm attaching a test image I made before I gave up on this technique, due to time
limitations. It's a very early image, with no pseudo-3-d effects, but it shows how
pigments can be used to create the shape of distant thunderclouds. It's all one
pigment, generated with the 'pigment control trick' and MegaPov's spline.


Simon

Post a reply to this message

Attachments:
Download 'cloudtest1.jpg' (43 KB)

Preview of image 'cloudtest1.jpg'

A total cheat, mister!

Sure it looks good, but imagemaps are not very flexible, cause they are
not very real... Okay I use them too, but I have found that the greatest
realism is achieved when objects are modelled like in real life.

Well, your clouds are in the horisont, so unless you plan an animation
of them, or a sundown, it doesn't matter if they're real or not. And as
long as you put enough real things into the picture, people will look at
mostly them, and not the clouds... In which case you can receieve praise
with a clean conscience.


- Hugo


Simon wrote:

>> I guess this shows that, on occasion, imagemaps aren't the terrible
>> solution they're made out to be.

Post a reply to this message

Simon de Vet wrote:
> 
> Christoph Hormann wrote:
> 
> >
> > But not with that rendering time :-)
> 
> I'm not so sure about this.
> 
> I've looked at other people's cloud techniques, and they all use 3-d to generate
> 3-d looking clouds. Some use media, or isosurfaces, or stacked planes, but all use
> actuall objects.
> 
> When I look at the clouds around me, most are way too far away to look 3-d. 3-d
> objects are usually shaded based on the angle of the incomming light (diffuse
> shading) and on self shadowing. Clouds, on the other hand, have a bit of this, but
> the vast majority of the shading comes from the extinction of light as it passes
> through the water vapour.
> 

Your description of the clouds lighting seems perfectly right, but i don't agree
about the distance.  Clouds like the ones in your scene are about 3-6 km high. 
In a large scale movement animation or a fast motion film you could perfectly
see the 3d structure of those clouds.  

> I've been looking at clouds not as 3-d objects, but as 2-d objects, painted onto a
> giant sphere. If treated this way, I think that most of the effects can be
> generated with complex textures on a skysphere, which are very quick to render.
> 
> http://freespace.virgin.net/hugo.elias/models/m_main.htm has a good (award
> winning) example of a pigment based cloud render.
> 

I have seen that one before and i like it too, but NTL, the lighting used there
is not very convincing for roundish clouds, because it only makes clouds lighter
according to their general distance to the sun, while each cloud should be
somewhat lighter on it's sun side and dark on the opposite side.  

> In addition, pigments can generate clouds other than puffy cumulus, like cirrus,
> stratus, and realistic haze.

I agree, especially for high cirrus clouds the plain texture approach seems most
efficient.  

> 
> I'm attaching a test image I made before I gave up on this technique, due to time
> limitations. It's a very early image, with no pseudo-3-d effects, but it shows how
> pigments can be used to create the shape of distant thunderclouds. It's all one
> pigment, generated with the 'pigment control trick' and MegaPov's spline.
> 

That already looks quite promising for some kind of near horizon cumulus
clouds.  It only will be quite difficult to achieve realism on both near the sun
and on the opposite side.  

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Simon de Vet wrote:
--snip--
> 
> I've looked at other people's cloud techniques, and they all use 3-d to generate
> 3-d looking clouds. Some use media, or isosurfaces, or stacked planes, but all use
> actuall objects.
>
--snip--

If I may beat my own drum for a sec, I posted a non-media 2D cloud
example not too far back. The method provided for some 3D lighting
effect. For the results, the rendering times were pretty fast. I'll take
the liberty of sticking in another example my pigment clouds below.
(This was from an effort to copy the technique from the TGLTLSBFSSP page
which you mentioned).

I have always thought that simple bump mapping should do the trick of
providing the correct shadows to pigment. So far though, I have had very
little success with it in POV. Not sure why ...

Abe

Post a reply to this message

Attachments:
Download 'clouds_simple.jpg' (33 KB)

Preview of image 'clouds_simple.jpg'

Regarding the clouds at Hugo Elias' page, he _does_ apply cloud shading 
via bump mapping in the snippet of pseudo-code at the end. You can also
see it in the example image.

Abe

Post a reply to this message

"Paul Vanukoff" <van### [at] primenetcom> wrote in message
news:39a71fe1@news.povray.org...
>
> AWSOME ROLEX! even.

I guess I should post another of my stern warnings against posting photos
and trying to pass them off as POV.

[ ;-) Good work]

--
Eric
--------------------
http://www.datasync.com/~ericfree
--------------------
"I don't like it, and I'm sorry I ever had anything to do with it."
- Erwin Schrodinger talking about Quantum Mechanics.

Post a reply to this message

Abe wrote:
> 
> Regarding the clouds at Hugo Elias' page, he _does_ apply cloud shading
> via bump mapping in the snippet of pseudo-code at the end. You can also
> see it in the example image.
> 
> Abe

Seems you are right, but the effect is not very strong.  I think simple bump
maps probably are not sufficient in many cases like this:

http://www.wolkenatlas.de/wolken/wo12625.htm

Of course these are somewhat extreme, but the general type of those clouds is
very common.  

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Christoph Hormann wrote:
> 
> Seems you are right, but the effect is not very strong.  I think simple bump
> maps probably are not sufficient in many cases like this:
> 
> http://www.wolkenatlas.de/wolken/wo12625.htm

An understatement.

> 
> Of course these are somewhat extreme, but the general type of those clouds is
> very common.

...and very desirable ;)

Abe

Post a reply to this message

Why are image maps so frowned upon in the POV community?  Almost all of
the big wig renderers use them solely.

GrimDude wrote:
> 
> Image_map and IRTC? Bite your tongue!
> 
> Grim

Post a reply to this message

Simon de Vet <sde### [at] istarca> wrote:
> I've looked at other people's cloud techniques, and they all use 3-d to
generate
> 3-d looking clouds. Some use media, or isosurfaces, or stacked planes, but
all use
> actuall objects.
[snip]
> I'm attaching a test image I made before I gave up on this technique, due
to time
> limitations. It's a very early image, with no pseudo-3-d effects, but it
shows how
> pigments can be used to create the shape of distant thunderclouds. It's
all one
> pigment, generated with the 'pigment control trick' and MegaPov's spline.

Yes, I think this method does hold a lot of promise.  A while ago I put
together a few animations of various types of clouds, which do include
sun-shading (see the Current Projects page at
http://www.geocities.com/ccolefax/projects.html).  The code is all quite
basic POV 3.1, and I've parametised most aspects of the clouds.  I think the
trick is to find pigments that look effective at high resolutions...

Post a reply to this message

The 3D structure of clouds can be seen if they are close and moving quickly in strong
winds.  I saw a thunderhead move over and away from us and it looked very 3D.  Some
dark clouds passed in front of it with visible motion.  The thunderhead was also
visibly swelling.
Brendan

Post a reply to this message

Nick Portelli wrote:

> Why are image maps so frowned upon in the POV community?  Almost all of
> the big wig renderers use them solely.

It's my personal belief that when I'm rendering, I'm not trying to win a
race, or stick to rules, but just make a nice looking picture. If an
imagemap can produce good results (as the comments have shown), and it
suits my situation, then I will use an imagemap. If I need more
flexibility, then I will use procedural textures.

There are no correct and incorrect tools, merely tools that are, or aren't
suited to the job. There is nothing more honourable in using a penknife
when a chainsaw is better suited to the job, and vice-versa. Both have
their uses, and there is no reaosn for conflict.


Simon

Post a reply to this message

"Nick Portelli" <por### [at] pilotmsuedu> wrote in message
news:39A9A67D.29090A1A@pilot.msu.edu...
> Why are image maps so frowned upon in the POV community?  Almost all of
> the big wig renderers use them solely.
>

It's not the image maps (a valid texturing process after all), but the
images (photos). Photographs should hold only a *very* minor role in any
rendering, particularly in the IRTC, where you want to set your work apart
from all the others. Image maps should be used *only* when nothing else will
do (as the Pov docs state). :)

There are a million valid uses for image maps, even in the IRTC. I would
not, however, use one as in this scene. I intend no offense upon Simon. I am
merely stating my opinion and advice.

Nothing says the big wig renderers are doing anything right. Those guys get
you hooked on image mapping, so they can sell you expensive texture
libraries. Thank the Pov gurus that you have procedural functionality!

Oops, inspiration has struck again. Ciao!

Grim

Post a reply to this message