POV-Ray : Newsgroups : povray.binaries.images : Cloudscape Server Time: 29 Jan 2020 15:40:09 GMT
  Cloudscape (Message 1 to 10 of 42)  
Goto Latest 10 Messages Next 10 Messages >>>
From: Bruno Cabasson
Subject: Cloudscape
Date: 6 Dec 2019 10:00:01
Message: <web.5dea256ec4bd29762931b8760@news.povray.org>
Hi there !

Long time so see... I can't say I'm back, but after a while in hibernation
concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
+ac0.99 +r4".

Regards


Post a reply to this message


Attachments:
Download 'cloudscape1_2.png' (793 KB)

Preview of image 'cloudscape1_2.png'
cloudscape1_2.png


 

From: Thomas de Groot
Subject: Re: Cloudscape
Date: 6 Dec 2019 12:10:50
Message: <5dea454a$1@news.povray.org>
Op 06/12/2019 om 10:54 schreef Bruno Cabasson:
> Hi there !
> 
> Long time so see... I can't say I'm back, but after a while in hibernation
> concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
> Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
> +ac0.99 +r4".
> 
> Regards
> 

Good to see you here again indeed! Those clouds of yours were always my 
fovourites.

-- 
Thomas


Post a reply to this message

From: Bruno Cabasson
Subject: Re: Cloudscape
Date: 6 Dec 2019 12:50:01
Message: <web.5dea4e0a5bb70f082931b8760@news.povray.org>
Thomas de Groot <tho### [at] degrootorg> wrote:
> Op 06/12/2019 om 10:54 schreef Bruno Cabasson:
> > Hi there !
> >
> > Long time so see... I can't say I'm back, but after a while in hibernation
> > concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
> > Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
> > +ac0.99 +r4".
> >
> > Regards
> >
>
> Good to see you here again indeed! Those clouds of yours were always my
> fovourites.
>
> --
> Thomas

Thanks Thomas! For my clouds, I was lacking specific built-in functions that can
express density for the medias involed within a spherical shell.

-) One for the atmosphere, an exponentially decreasing function like
A*exp(-kh/T), where k is the order of the funtion, h is the height wrt ground, T
is shell thickness, currently expressed as a function :

// Atmosphere density function A*exp(-kh/T), h being height wrt ground (whish it
were native ...). _R is the planet diameter at ground level.
#declare shell_exp = function (x, y, z, _h, _t, _a, _k, _R)
{
    _a*exp(-_k*(sqrt(x*x+(y+_R)*(y+_R)+z*z) - (_h + _R))/_t)
}

-) Another would be a series of classical patterns (wrinkles works quite fine),
but contained in a spherial shell. I hope it can be expressed mathematically.

-) And yet another would be a spherical shell shaped fog. I use scattering media
to imitate, but the behaviour is not really correct.

Another thing I noticed :  the is a noticeable différence in the render if the
camera is at the edge of the media container but inside, or if it is just
outside. It has also implications with radiosity, so far I could see.

Here is the exact same scene but with sun elevation is 10° instead of 0.5. All
the rest is the same values, except the sun power and the parmeters for the fog
media. Colors are obtained automatically thanks to media, mainly the atmosphere
with rayleigh scattering.

Regards


Post a reply to this message


Attachments:
Download 'cloudscape1_3.png' (765 KB)

Preview of image 'cloudscape1_3.png'
cloudscape1_3.png


 

From: pascal
Subject: Re: Cloudscape
Date: 6 Dec 2019 12:50:18
Message: <5dea4e8a$1@news.povray.org>
Le 06/12/2019 à 10:54, Bruno Cabasson a écrit :
> Hi there !
> 
> Long time so see... I can't say I'm back, but after a while in hibernation
> concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
> Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
> +ac0.99 +r4".
> 
> Regards
> 
Nice one !
Do you use a special macro for the clouds ?

P.


Post a reply to this message

From: jr
Subject: Re: Cloudscape
Date: 6 Dec 2019 16:10:01
Message: <web.5dea7c6f5bb70f08feeb22ff0@news.povray.org>
hi,

"Bruno Cabasson" <bru### [at] cabassoncom> wrote:
> Thomas de Groot <tho### [at] degrootorg> wrote:
> > Op 06/12/2019 om 10:54 schreef Bruno Cabasson:
> > > Hi there !
> > > Long time so see... I can't say I'm back, but after a while in hibernation
> > > concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
> > > Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
> > > +ac0.99 +r4".
> > > ...
> Here is the exact same scene but with sun elevation is 10° instead of 0.5.
> ...

lovely light(ing).  impressed.


regards,jr.


Post a reply to this message

From: Bruno Cabasson
Subject: Re: Cloudscape
Date: 6 Dec 2019 16:30:01
Message: <web.5dea810b5bb70f08dc39a140@news.povray.org>
pascal <gos### [at] gmailcom> wrote:
> Le 06/12/2019 à 10:54, Bruno Cabasson a écrit :
> > Hi there !
> >
> > Long time so see... I can't say I'm back, but after a while in hibernation
> > concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
> > Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
> > +ac0.99 +r4".
> >
> > Regards
> >
> Nice one !
> Do you use a special macro for the clouds ?
>
> P.

No. Clouds are from a density function, based on standard patterns (wrinkles
here), within a spherical shell container (difference of spheres). You tweak
turbulence, thresholds and media parameters in order to get desisred result. A
bit tricky because of vast range of combinations and render time, especially
when using radiosity.

However, I use a marcro that builds a layer, with numerous parameters, one of
them is the density function. As of now, I can have 4 layers :

1) high altitude clouds (typ. low density),
2) mid altitude clouds ('normal' density),
3) ground to layer 2, for beam effects (typ. low density)
4) simulated fog layer, just above the ground.

plus the atmosphere itself, of course.

B.


Post a reply to this message

From: Thomas de Groot
Subject: Re: Cloudscape
Date: 7 Dec 2019 07:50:14
Message: <5deb59b6$1@news.povray.org>
Op 06/12/2019 om 13:48 schreef Bruno Cabasson:
> Thomas de Groot <tho### [at] degrootorg> wrote:
>> Op 06/12/2019 om 10:54 schreef Bruno Cabasson:
>>> Hi there !
>>>
>>> Long time so see... I can't say I'm back, but after a while in hibernation
>>> concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
>>> Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
>>> +ac0.99 +r4".
>>>
>>> Regards
>>>
>>
>> Good to see you here again indeed! Those clouds of yours were always my
>> fovourites.
>>
>> --
>> Thomas
> 
> Thanks Thomas! For my clouds, I was lacking specific built-in functions that can
> express density for the medias involed within a spherical shell.
> 
> -) One for the atmosphere, an exponentially decreasing function like
> A*exp(-kh/T), where k is the order of the funtion, h is the height wrt ground, T
> is shell thickness, currently expressed as a function :
> 
> // Atmosphere density function A*exp(-kh/T), h being height wrt ground (whish it
> were native ...). _R is the planet diameter at ground level.
> #declare shell_exp = function (x, y, z, _h, _t, _a, _k, _R)
> {
>      _a*exp(-_k*(sqrt(x*x+(y+_R)*(y+_R)+z*z) - (_h + _R))/_t)
> }
> 
> -) Another would be a series of classical patterns (wrinkles works quite fine),
> but contained in a spherial shell. I hope it can be expressed mathematically.
> 
> -) And yet another would be a spherical shell shaped fog. I use scattering media
> to imitate, but the behaviour is not really correct.
> 
> Another thing I noticed :  the is a noticeable différence in the render if the
> camera is at the edge of the media container but inside, or if it is just
> outside. It has also implications with radiosity, so far I could see.
> 
> Here is the exact same scene but with sun elevation is 10° instead of 0.5. All
> the rest is the same values, except the sun power and the parmeters for the fog
> media. Colors are obtained automatically thanks to media, mainly the atmosphere
> with rayleigh scattering.
> 
> Regards
> 

Lately, I played (again) with the macro SkySim developed by Scott Boham 
in 2013:
http://news.povray.org/povray.binaries.scene-files/thread/%3C51d13061@news.povray.org%3E/

In particular, it nicely shows the sky colour changes as the Sun rises 
(or sets).

-- 
Thomas


Post a reply to this message

From: Bald Eagle
Subject: Re: Cloudscape
Date: 7 Dec 2019 13:25:01
Message: <web.5deba7465bb70f084eec112d0@news.povray.org>
"Bruno Cabasson" <bru### [at] cabassoncom> wrote:
> ... I was lacking specific built-in functions that can
> express density for the medias involed within a spherical shell.

> ... (wish it were native ...).

> -) Another would be a series of classical patterns (wrinkles works quite fine),
> but contained in a spherial shell. I hope it can be expressed mathematically.

I think all of that is certainly doable.

If you take whatever pattern you want to use, you can wrap it around into a
sphere using polar coordinates, as Mike Williams shows here:

#declare F=function {f_mesh1
  (x, y, z, 0.15, 0.15, 1, 0.02, 1) - 0.03}

isosurface {
  function { F(f_ph(x,y,z), f_r(x,y,z)-1, f_th(x,y,z)) }
  ...

http://www.econym.demon.co.uk/isotut/substitute.htm


and as I'm sure you already know, all manner of new functions can obviously be
integrated into a scene, if you try hard enough, for long enough.

http://news.povray.org/povray.binaries.images/thread/%3Cweb.5d3c5fe27be432e04eec112d0%40news.povray.org%3E/?ttop=428926



Perhaps you could start by making a proof-of-concept include file with the
density functions, and then maybe that could be translated to source by Jerome
Grimbert, Bill Pokorny, or someone else who understands the necessary changes
and additions to be done in the various places.

If there's a simple density pattern that you'd like to try in source, maybe see
about just replacing the definition for one of the native POV-Ray patterns that
is redundant (there are 2 or 3) and use that as your cloud pattern.  Then just
make a POV-Ray build with those edits and see how it works.

If I understand things correctly, the new source is hellish to follow, due to
the the threading, but the older versions are simpler and more straightforward.

Bill


Post a reply to this message

From: Bruno Cabasson
Subject: Re: Cloudscape
Date: 7 Dec 2019 16:00:01
Message: <web.5debcaa65bb70f08dc39a140@news.povray.org>
"Bald Eagle" <cre### [at] netscapenet> wrote:
> "Bruno Cabasson" <bru### [at] cabassoncom> wrote:
> > ... I was lacking specific built-in functions that can
> > express density for the medias involed within a spherical shell.
>
> > ... (wish it were native ...).
>
> > -) Another would be a series of classical patterns (wrinkles works quite fine),
> > but contained in a spherial shell. I hope it can be expressed mathematically.
>
> I think all of that is certainly doable.
>
> If you take whatever pattern you want to use, you can wrap it around into a
> sphere using polar coordinates, as Mike Williams shows here:
>
> #declare F=function {f_mesh1
>   (x, y, z, 0.15, 0.15, 1, 0.02, 1) - 0.03}
>
> isosurface {
>   function { F(f_ph(x,y,z), f_r(x,y,z)-1, f_th(x,y,z)) }
>   ...
>
> http://www.econym.demon.co.uk/isotut/substitute.htm
>
>
> and as I'm sure you already know, all manner of new functions can obviously be
> integrated into a scene, if you try hard enough, for long enough.
>
>
http://news.povray.org/povray.binaries.images/thread/%3Cweb.5d3c5fe27be432e04eec112d0%40news.povray.org%3E/?ttop=4289
26
>
>
>
> Perhaps you could start by making a proof-of-concept include file with the
> density functions, and then maybe that could be translated to source by Jerome
> Grimbert, Bill Pokorny, or someone else who understands the necessary changes
> and additions to be done in the various places.
>
> If there's a simple density pattern that you'd like to try in source, maybe see
> about just replacing the definition for one of the native POV-Ray patterns that
> is redundant (there are 2 or 3) and use that as your cloud pattern.  Then just
> make a POV-Ray build with those edits and see how it works.
>
> If I understand things correctly, the new source is hellish to follow, due to
> the the threading, but the older versions are simpler and more straightforward.
>
> Bill

Since all that time, I never dared to really dive into POV-Ray source code and
make a build ... Maybe I'll give it a try someday.

Mike William's tutorial is nothing but marvelous. However, I don't thing a
coordinate change will make much difference and I guess this will lead to a
scaling problem. I try to model real world and my Earth is a sphere of 6378 km
in radius. Size of clouds is some 100's meters ...


As my scene file is quite short, I produce the code here. Perhaps one can
play with the (quite numerous) tweaking variables, in the "USER DECLARATIONS"
section, in particular the SUN_ELEVATIONS array (change first value), and the
USE_LAYER_xxx variables.


// Persistence of Vision Ray Tracer Scene Description File
// File: cloudscape.pov
// Vers: 3.8
// Desc: Cloudscape experiment
// Date: 2019
// Auth: Bruno Cabasson
// +bs8 +a0.05 +am3 +ac0.99 +r4


#version 3.8

#include "colors.inc"

global_settings
{
  assumed_gamma 1
  max_trace_level 50
}

#default {finish{ambient 0}}

// ======================== ENGINE DECLARATIONS ===============================
// General constants
#declare m = 1;
#declare km = 1000*m;
#declare cm = 0.01*m;

// Earth
#declare Re = 6378*km;

// Cameras
#declare VIEW_CAM = 0;
#declare FISHEYE_CAM = 1;
#declare ALTITUDE_CAM = 2;

// Atmosphere density function A*exp(-kh/T), h being height wrt ground (whish it
were native ...).
#declare shell_exp = function (x, y, z, _h, _t, _a, _k, _R)
{
    _a*exp(-_k*(sqrt(x*x+(y+_R)*(y+_R)+z*z) - (_h + _R))/_t)
}
// ====================== END OF ENGINE DECLARATIONS ==========================


// ======================== USER DECLARATIONS ===============================
#declare QUICK = yes;

// Camera
#declare CAM_HEADING = 0;
#declare CAM_ELEVATION = 10;
#declare CAM_ANGLE = 80;
#declare CAM_H = 2*m;
#declare CAM = VIEW_CAM;

// Radiosity
#declare USE_RAD = yes;
#declare RAD_BRIGHTNESS = 1.5;
#declare RAD_QUICK_COUNT = 50;
#declare RAD_COUNT = 400;
#declare RAD_USE_NOCACHE = no;
#declare RAD_START = 0.08;
#declare RAD_PASSES = 4;
#declare RAD_RECURSION_LIMIT = 2;

// Sun
#declare SUN_ELEVATIONS = array [7] {1.5, 3, 12, 20, 5, 0.5, -0.5};
#declare SUN_HEADING = 0;
#declare SUN_POWER = 0.5;

// Atmosphere
#declare Ta = 50*km;
#declare Ha = -1*cm;
#declare LAMBDA_VIOLET = 380;
#declare LAMBDA_BLUE = 440;
#declare LAMBDA_GREEN = 510;
#declare LAMBDA_RED = 650;
#declare BASE_SKY_COLOR = rgb vnormalize(<pow(LAMBDA_VIOLET/LAMBDA_RED, 4),
pow(LAMBDA_VIOLET/LAMBDA_GREEN, 4),  pow(LAMBDA_VIOLET/LAMBDA_BLUE, 4)>);
#declare DENSITY_POWER_FACTOR = 1;
#declare ATMO_INTERVALS = 1;
#declare ATMO_SAMPLES = 20;
#declare ATMO_QUICK_INTERVALS = 1;
#declare ATMO_QUICK_SAMPLES = 5;
#declare TURBIDITY = 1;

// Layers
#declare USE_LAYER_1 = no;
#declare USE_LAYER_2 = yes;
#declare USE_LAYER_3 = no           ;
#declare USE_LAYER_FOG = yes;

// Layer 1: high low-density clouds
#declare L1_H = 8000*m;
#declare L1_T = 800*m;
#declare L1_SCATTERING_TYPE = 3;
#declare L1_SCATTERING = 3;
#declare L1_SCATTERING_COLOR = White;
#declare L1_ABSORPTION = 0.1;
#declare L1_ABSORPTION_COLOR = White;
#declare L1_METHOD = 3;
#declare L1_INTERVALS = 3;
#declare L1_SAMPLES = 5;
#declare L1_QUICK_INTERVALS = 1;
#declare L1_QUICK_SAMPLES = 5;
#declare L1_DENSITY = density
{
    granite
    scale <40, 1, 10>
    warp {turbulence 0.1 octaves 10 lambda 2 omega 0.65}
    scale L1_T
    rotate 30*y
    translate 5500*m*x
    color_map {[0.12 rgb 1][0.17 rgb 0]}
}

// Layer 2: main clouds : normal height quite opaque cloud cover
#declare L2_H = 800*m;
#declare L2_T = 400*m;
#declare L2_SCATTERING_TYPE = 2;
#declare L2_SCATTERING = 0.5;
#declare L2_SCATTERING_COLOR = White;
#declare L2_ABSORPTION = 0.1;
#declare L2_ABSORPTION_COLOR = White;
#declare L2_METHOD = 3;
#declare L2_INTERVALS = 3;
#declare L2_SAMPLES = 30;
#declare L2_QUICK_INTERVALS = 1;
#declare L2_QUICK_SAMPLES = 50;
#declare L2_DENSITY = density
{
    wrinkles translate <3, 6, 9>
    scale <10, 1, 10>
    warp {turbulence 0.2 octaves 10 lambda 2.1 omega 0.65}
    scale L2_T
    translate 6500*m*z + 8500*m*x
    color_map {[0.15 rgb 0.08/L2_T][0.6 rgb 0.3/L2_T][0.7 rgb 1]}
}

// Layer 3: Ground-to-Layer2 haze for scattering beams
#declare L3_H = 1*cm;
#declare L3_T = L2_H - L3_H - 1*m;
#declare L3_SCATTERING_TYPE = 2;
#declare L3_SCATTERING = 1;
#declare L3_SCATTERING_COLOR = White;
#declare L3_ABSORPTION = 0.1;
#declare L3_ABSORPTION_COLOR = White;
#declare L3_METHOD = 3;
#declare L3_INTERVALS = 1;
#declare L3_SAMPLES = 20;
#declare L3_QUICK_INTERVALS = 1;
#declare L3_QUICK_SAMPLES = 10;
#declare L3_DENSITY = density
{
    function {shell_exp(x, y, z, L3_H, L3_T, 0.1, 1.5, Re)}
}

// Fog layer
#declare FOG_H =  1*cm;
#declare FOG_T = 50*m;
#declare FOG_SCATTERING_TYPE = 1;
#declare FOG_SCATTERING = 1;
#declare FOG_SCATTERING_COLOR = White;
#declare FOG_ABSORPTION = 0.2;
#declare FOG_ABSORPTION_COLOR = White;
#declare FOG_METHOD = 3;
#declare FOG_INTERVALS = 1;
#declare FOG_SAMPLES = 40;
#declare FOG_QUICK_INTERVALS = 1;
#declare FOG_QUICK_SAMPLES = 20;
#declare FOG_DENSITY = density
{
    function{shell_exp(x, y, z, FOG_H, FOG_T, 0.15, 1.5, Re)}
}

// ====================== END OF DECLARATIONS ==========================


// ====================== ENGINE ==========================

// Radiosity
#if (USE_RAD)
global_settings
{
  radiosity
  {
    media on
    brightness RAD_BRIGHTNESS
    #if (QUICK)
        count RAD_QUICK_COUNT
        recursion_limit 1
    #else
        count RAD_COUNT
        recursion_limit RAD_RECURSION_LIMIT
    #end
    #if (RAD_USE_NOCACHE)
        no_cache
    #else
        pretrace_start RAD_START
        pretrace_end RAD_START/pow(2, RAD_PASSES - 1)
        nearest_count 20
        error_bound 1
        low_error_factor 0.8
        minimum_reuse 0.25
        adc_bailout 0.01/2
    #end
  }
}
#end

#if (clock_on)
    #declare SUN_ELEVATION = SUN_ELEVATIONS[frame_number-1];
#else
    #declare SUN_ELEVATION = SUN_ELEVATIONS[0];
#end


#declare ViewCam = camera
{
  location  0
  angle CAM_ANGLE
  right     x*image_width/image_height
  look_at   z
  rotate -CAM_ELEVATION*x
  rotate CAM_HEADING*y
  translate CAM_H*y
}

#declare FisheyeCam = camera
{
  fisheye
  location  0
  angle 180
  right     x*image_width/image_height
  look_at   z
  rotate -90*x
  rotate CAM_HEADING*y
  translate CAM_H*y
}

#declare AltitudeCam = camera
{
  location  0
  angle CAM_ANGLE
  right     x*image_width/image_height
  look_at   -y
  translate 100*Ta*y
}

#declare Cams = array[3] {ViewCam, FisheyeCam, AltitudeCam}
camera {Cams[CAM]}


// Ground
#declare Earth = sphere
{
    0, Re
    pigment {White}
    finish {diffuse 1}
    translate -Re*y
}

//Sun
#declare SUN_BASE_POWER = 1;
#declare SUN_COLOR = White;
#declare Sun = light_source {
  <0, 0, 0>            // light's position (translated below)
  color rgb SUN_BASE_POWER*SUN_POWER*SUN_COLOR  // light's color
  //area_light 3800*x, 3800*y, 6, 6  adaptive 1 circular orient
  looks_like {sphere{0, 2000*km pigment{color rgb 1} finish {diffuse 0 emission
10}}}
  translate 150000*km*z
  media_attenuation on
  parallel point_at -z
  rotate -SUN_ELEVATION*x
  rotate SUN_HEADING*y
}

// Atmosphere
#declare DENSITY_BASE_POWER = 4;
#declare AtmoScatterColor = BASE_SKY_COLOR;

#if (QUICK)
    #declare AtmoIntervals = ATMO_QUICK_INTERVALS;
    #declare AtmoSamples = ATMO_QUICK_SAMPLES;
#else
    #declare AtmoIntervals = ATMO_INTERVALS;
    #declare AtmoSamples = ATMO_SAMPLES;
#end

#declare M_Rayleigh = media
{
    method 3 intervals AtmoIntervals samples AtmoSamples jitter 0 variance 1/256
    scattering {4, color AtmoScatterColor/Ta extinction 1}
    density
    {
        function{shell_exp(x, y, z, Ha, Ta, TURBIDITY,
DENSITY_POWER_FACTOR*DENSITY_BASE_POWER, Re)}
    }
}


#declare Atmo = difference
{
    difference
    {
        sphere {0, Re + Ha + Ta translate -Re*y}
        sphere {0, Re + 0.01*m  translate -Re*y}
    }

    hollow
    pigment {rgbt 1}
    interior
    {
        media {M_Rayleigh}
    }
}

// Layers
#macro MakeLayer (_h, _thick, _type, _scatter, _scatter_color,  _absorb,
_absorb_color, _density, _method, _intervals, _samples)
    #local _ret = difference
    {
        sphere {0, Re + _h + _thick translate -Re*y}
        sphere {0, Re + _h translate -Re*y}
        hollow
        pigment {rgbt 1}
        interior
        {
            media
            {
                method _method
                intervals _intervals
                samples max(_samples/5, 5), _samples
                jitter 0
                variance 1/1024
                confidence 0.99
                #if (_scatter != 0) scattering {_type,
_scatter*_scatter_color/_thick extinction 0.25 } #end
                #if (_absorb != 0) absorption _absorb*_absorb_color/_thick #end
                density {_density}
            }
        }
    }
    _ret
#end

#if (QUICK)
    #declare L1Intervals = L1_QUICK_INTERVALS;
    #declare L1Samples = L1_QUICK_SAMPLES;
    #declare L2Intervals = L2_QUICK_INTERVALS;
    #declare L2Samples = L2_QUICK_SAMPLES;
    #declare L3Intervals = L3_QUICK_INTERVALS;
    #declare L3Samples = L3_QUICK_SAMPLES;
    #declare FOGIntervals = FOG_QUICK_INTERVALS;
    #declare FOGSamples = FOG_QUICK_SAMPLES;
#else
    #declare L1Intervals = L1_INTERVALS;
    #declare L1Samples = L1_SAMPLES;
    #declare L2Intervals = L2_INTERVALS;
    #declare L2Samples = L2_SAMPLES;
    #declare L3Intervals = L3_INTERVALS;
    #declare L3Samples = L3_SAMPLES;
    #declare FOGIntervals = FOG_INTERVALS;
    #declare FOGSamples = FOG_SAMPLES;
#end

#declare L1 = MakeLayer (L1_H, L1_T, L1_SCATTERING_TYPE, L1_SCATTERING,
L1_SCATTERING_COLOR, L1_ABSORPTION, L1_ABSORPTION_COLOR, L1_DENSITY, L1_METHOD,
L1Intervals, L1Samples)
#declare L2 = MakeLayer (L2_H, L2_T, L2_SCATTERING_TYPE, L2_SCATTERING,
L2_SCATTERING_COLOR, L2_ABSORPTION, L2_ABSORPTION_COLOR, L2_DENSITY, L2_METHOD,
L2Intervals, L2Samples)
#declare L3 = MakeLayer (L3_H, L3_T, L3_SCATTERING_TYPE, L3_SCATTERING,
L3_SCATTERING_COLOR, L3_ABSORPTION, L3_ABSORPTION_COLOR, L3_DENSITY, L3_METHOD,
L3Intervals, L3Samples)
#declare FOG = MakeLayer (FOG_H, FOG_T, FOG_SCATTERING_TYPE, FOG_SCATTERING,
FOG_SCATTERING_COLOR, FOG_ABSORPTION, FOG_ABSORPTION_COLOR, FOG_DENSITY,
FOG_METHOD, FOGIntervals, L3Samples)

// Scene setup
light_source {Sun}
object {Earth}
object {Atmo}
#if (USE_LAYER_1) object {L1} #end
#if (USE_LAYER_2) object {L2} #end
#if (USE_LAYER_3) object {L3} #end
#if (USE_LAYER_FOG) object {FOG} #end
// ====================== END OF ENGINE ==========================

// User objects
#declare Dome = sphere
{
    0, 2.5*m
    pigment {White}
}

#declare Slate = box
{
    -1.5*m, 1.5*m
    scale 0.05*y
    pigment {White}
}

object {Dome translate 17*m*z - 3*m*x}
object {Slate translate 2.75*m*y rotate 10*x translate 17*m*z - 3*m*x}


Post a reply to this message

From: Paolo Gibellini
Subject: Re: Cloudscape
Date: 9 Dec 2019 10:58:30
Message: <5dee28d6$1@news.povray.org>
Bruno Cabasson wrote on 06/12/2019 13:48:
> Thomas de Groot <tho### [at] degrootorg> wrote:
>> Op 06/12/2019 om 10:54 schreef Bruno Cabasson:
>>> Hi there !
>>>
>>> Long time so see... I can't say I'm back, but after a while in hibernation
>>> concerning POV-Ray, I recently revived my clouds. This render took 1h 45 mn on a
>>> Xeon quad E5-1620 V2 @ 3.7GHz, with latest version 3.8 and with "+a0.05  +am3
>>> +ac0.99 +r4".
>>>
>>> Regards
>>>
>>
>> Good to see you here again indeed! Those clouds of yours were always my
>> fovourites.
>>
>> --
>> Thomas
> 
> Thanks Thomas! For my clouds, I was lacking specific built-in functions that can
> express density for the medias involed within a spherical shell.
> 
> -) One for the atmosphere, an exponentially decreasing function like
> A*exp(-kh/T), where k is the order of the funtion, h is the height wrt ground, T
> is shell thickness, currently expressed as a function :
> 
> // Atmosphere density function A*exp(-kh/T), h being height wrt ground (whish it
> were native ...). _R is the planet diameter at ground level.
> #declare shell_exp = function (x, y, z, _h, _t, _a, _k, _R)
> {
>      _a*exp(-_k*(sqrt(x*x+(y+_R)*(y+_R)+z*z) - (_h + _R))/_t)
> }
> 
> -) Another would be a series of classical patterns (wrinkles works quite fine),
> but contained in a spherial shell. I hope it can be expressed mathematically.
> 
> -) And yet another would be a spherical shell shaped fog. I use scattering media
> to imitate, but the behaviour is not really correct.
> 
> Another thing I noticed :  the is a noticeable différence in the render if the
> camera is at the edge of the media container but inside, or if it is just
> outside. It has also implications with radiosity, so far I could see.
> 
> Here is the exact same scene but with sun elevation is 10° instead of 0.5. All
> the rest is the same values, except the sun power and the parmeters for the fog
> media. Colors are obtained automatically thanks to media, mainly the atmosphere
> with rayleigh scattering.
> 
> Regards
> 

This one is very realistic, perhaps it needs a few work near the horizon.

Paolo


Post a reply to this message

Goto Latest 10 Messages Next 10 Messages >>>

Copyright 2003-2008 Persistence of Vision Raytracer Pty. Ltd.