POV-Ray : Newsgroups : povray.general : Someday I'll understand media - right after I get women figured out and work out the tensions in the Middle East. Server Time
26 Nov 2024 18:41:07 EST (-0500)
  Someday I'll understand media - right after I get women figured out and work out the tensions in the Middle East. (Message 1 to 10 of 10)  
From: Some Yahoo
Subject: Someday I'll understand media - right after I get women figured out and work out the tensions in the Middle East.
Date: 31 Jan 2010 14:06:55
Message: <4b65d4cf$1@news.povray.org>
Look, I'm not looking for super accuracy here, just a little media to scatter my light
sources.

right now I' looking down a long corridor so I made this.

#declare media_box=box{
	<-10,-10,-50><10,30,200>
	hollow
	pigment{color rgbf 1}
	interior{
		media {
		  intervals 9
		  scattering { 1, rgb 0.03 }
		  samples 1, 10
		  confidence 1/10
		  variance 1/1000
		  ratio 0.9
		}
	}
}

however, when I use it my render takes many hours to complete.
is there something that can produce a simple, easy haze to scatter my light and take,
oh say 10 minutes?


Post a reply to this message

From: Warp
Subject: Re: Someday I'll understand media - right after I get women figured out and work out the tensions in the Middle East.
Date: 31 Jan 2010 14:36:32
Message: <4b65dbc0@news.povray.org>
Some_Yahoo <no### [at] waycom> wrote:
> however, when I use it my render takes many hours to complete.

  Scattering media can be very slow if not optimally used.

> is there something that can produce a simple, easy haze to scatter my light and
take, oh say 10 minutes?

  The image here didn't take hours to render (although it didn't render in
a few seconds either):

http://wiki.povray.org/content/Documentation:Tutorial_Section_3.7#A_more_advanced_example_of_scattering_media

-- 
                                                          - Warp


Post a reply to this message

From: Some Yahoo
Subject: Re: Someday I'll understand media - right after I get women figured outand work out the tensions in the Middle East.
Date: 31 Jan 2010 14:36:58
Message: <4b65dbda@news.povray.org>
The entire file - is here.
http://pastebin.com/ff4fdf48

Some_Yahoo wrote:
> Look, I'm not looking for super accuracy here, just a little media to 
> scatter my light sources.
> 
> right now I' looking down a long corridor so I made this.
> 
> #declare media_box=box{
>     <-10,-10,-50><10,30,200>
>     hollow
>     pigment{color rgbf 1}
>     interior{
>         media {
>           intervals 9
>           scattering { 1, rgb 0.03 }
>           samples 1, 10
>           confidence 1/10
>           variance 1/1000
>           ratio 0.9
>         }
>     }
> }
> 
> however, when I use it my render takes many hours to complete.
> is there something that can produce a simple, easy haze to scatter my 
> light and take, oh say 10 minutes?


Post a reply to this message

From: Alain
Subject: Re: Someday I'll understand media - right after I get women figured outand work out the tensions in the Middle East.
Date: 31 Jan 2010 16:49:20
Message: <4b65fae0$1@news.povray.org>

> Look, I'm not looking for super accuracy here, just a little media to
> scatter my light sources.
>
> right now I' looking down a long corridor so I made this.
>
> #declare media_box=box{
> <-10,-10,-50><10,30,200>
> hollow
> pigment{color rgbf 1}
> interior{
> media {
> intervals 9
> scattering { 1, rgb 0.03 }
> samples 1, 10
> confidence 1/10
> variance 1/1000
> ratio 0.9
> }
> }
> }
>
> however, when I use it my render takes many hours to complete.
> is there something that can produce a simple, easy haze to scatter my
> light and take, oh say 10 minutes?

The default sampling method is method 3. The documentations are about 
the old method 1. The settings given are not good for method 3.

Remove intervals 9. It realy bogs you down. Default to intervals 1 and 
MUST be left at that value.

Use samples 9 instead. The second samples value is totaly ignored and 
best left out.
Also, remove ratio 0.9. It's no longer needed.

You can also leave confidence and variance out and use the default values.
This gives something like:
interior{ media { scattering { 1, rgb 0.03 } samples 20}

This will be much faster and gives the same, or beter, result.

As said, scatering is the slowest kind of media, but you can make it 
even slower with bogus parameters.


Alain


Post a reply to this message

From: Reactor
Subject: Re: Someday I'll understand media - right after I get women figured out and= work out the tensions in the Middle East.
Date: 31 Jan 2010 16:50:01
Message: <web.4b65f9e8ce1251af01717f0@news.povray.org>
Some_Yahoo <no### [at] waycom> wrote:
> Look, I'm not looking for super accuracy here, just a little media to scatter my
light sources.
>
> right now I' looking down a long corridor so I made this.
>
> #declare media_box=box{
>  <-10,-10,-50><10,30,200>
>  hollow
>  pigment{color rgbf 1}
>  interior{
>   media {
>     intervals 9
>     scattering { 1, rgb 0.03 }
>     samples 1, 10
>     confidence 1/10
>     variance 1/1000
>     ratio 0.9
>   }
>  }
> }
>
> however, when I use it my render takes many hours to complete.
> is there something that can produce a simple, easy haze to scatter my light and
take, oh say 10 minutes?


Yes, your media settings are less than optimal.  I don't know if this is an
older scene or not, but the default method for media is now 3.  As such, the
intervals should be left at one, and the samples only needs one value.  You can
read more about this in the documentation.

I changed your settings to as below and it renders in about 3 minutes @ 800 x
600 (no aa) on a Q9550 using the beta.  If your are using 3.6, it'll be longer
but fairly close to the ten minute mark if your machine is in the 2.8GHz range.

//-------------
    interior{
        media {
          intervals 1
          scattering { 1, rgb 0.01 }
          samples 10
          confidence 0.9
          variance 1/1000
          ratio 0.9
        }
    }
//-------------

The important difference is the intervals - the number of samples is basically
multiplied by the number of intervals, so your render will take longer.
In any case, if I may offer some constructive criticism on the scene in general,
it will be much more realistic to use attenuation for the lights, both media and
distance.  This may require additional tuning of light brightness, but this can
be done without the media.  You also may want to consider using radiosity.

Currently, the media in the scene is overwhelmingly bright.  If this is not on
purpose, then you can thin the media or boost the absorption (either by the
extinction keyword or adding an absorption component).

   HTH
-Reactor


Post a reply to this message

From: Some Yahoo
Subject: Re: Someday I'll understand media - right after I get women figured out and=work out the tensions in the Middle East.
Date: 1 Feb 2010 01:05:49
Message: <4b666f3d@news.povray.org>
OK, as to where the settings came from - everyone do this:

Insert - Special Effects - Atmospheric Media

That was my starting point.

...and yes, I am adding a density so it will thin down - can't wait to post the
results...


---
Oh, and I was talking to the PASTEBIN guy and here is his reply...

"If you can write a GeSHi syntax highlighting definition for it, I'd be happy to add
it."

is there an off-chance anyone has done this already and/or knows what it is?


Post a reply to this message

From: waggy
Subject: Re: Someday I'll understand media - right after I get women figuredout and=work out the tensions in the Middle East.
Date: 1 Feb 2010 09:00:47
Message: <4b66de8f$1@news.povray.org>
Some_Yahoo wrote:
> Oh, and I was talking to the PASTEBIN guy and here is his reply...
> 
> "If you can write a GeSHi syntax highlighting definition for it, I'd be 
> happy to add it."
> 
> is there an off-chance anyone has done this already and/or knows what it 
> is?

I did a partial GeSHi description of POV-Ray's SDL some time ago for my 
wiki, but now it looks like it's included with newer versions of the 
wiki software.

http://www.dokuwiki.org/syntax#syntax_highlighting

It'll take a little digging through the source to find it, though.

~David


Post a reply to this message

From: waggy
Subject: Re: Someday I'll understand media - right after I get women figuredoutand=work out the tensions in the Middle East.
Date: 1 Feb 2010 09:16:13
Message: <4b66e22d@news.povray.org>
I wrote:
> I did a partial GeSHi description of POV-Ray's SDL some time ago for my 
> wiki, but now it looks like it's included with newer versions of the 
> wiki software.
> 
> http://www.dokuwiki.org/syntax#syntax_highlighting
> 
> It'll take a little digging through the source to find it, though.
> 
> ~David
Never mind.  Though I did find the file in the code for DokuWiki, it 
appears POV-Ray is supported by GeSHi itself now.  Perhaps I misunderstood.

http://qbnz.com/highlighter/

Getting more coffee now,

~David


Post a reply to this message

From: Alain
Subject: Re: Someday I'll understand media - right after I get women figuredout and=work out the tensions in the Middle East.
Date: 1 Feb 2010 22:13:32
Message: <4b67985c$1@news.povray.org>

> OK, as to where the settings came from - everyone do this:
>
> Insert - Special Effects - Atmospheric Media
>
> That was my starting point.
>

That bit of code absolutely NEED to be updated asap.

In fact, several of those inserts need some update.



Alain


Post a reply to this message

From: Alain
Subject: Re: Someday I'll understand media - right after I get women figuredoutand work out the tensions in the Middle East.
Date: 1 Feb 2010 22:22:19
Message: <4b679a6b$1@news.povray.org>


>> Look, I'm not looking for super accuracy here, just a little media to
>> scatter my light sources.
>>
>> right now I' looking down a long corridor so I made this.
>>
>> #declare media_box=box{
>> <-10,-10,-50><10,30,200>
>> hollow
>> pigment{color rgbf 1}
>> interior{
>> media {
>> intervals 9
>> scattering { 1, rgb 0.03 }
>> samples 1, 10
>> confidence 1/10
>> variance 1/1000
>> ratio 0.9
>> }
>> }
>> }
>>
>> however, when I use it my render takes many hours to complete.
>> is there something that can produce a simple, easy haze to scatter my
>> light and take, oh say 10 minutes?
>
> The default sampling method is method 3. The documentations are about
> the old method 1. The settings given are not good for method 3.
>
> Remove intervals 9. It realy bogs you down. Default to intervals 1 and
> MUST be left at that value.
>
> Use samples 9 instead. The second samples value is totaly ignored and
> best left out.
> Also, remove ratio 0.9. It's no longer needed.
>
> You can also leave confidence and variance out and use the default values.
> This gives something like:
> interior{ media { scattering { 1, rgb 0.03 } samples 20}
>
> This will be much faster and gives the same, or beter, result.
>
> As said, scatering is the slowest kind of media, but you can make it
> even slower with bogus parameters.
>
>
> Alain

Change  Special effects|Athmospheric media as follow:

// atmospheric media can be generated by adding a media statement
// to the scene not attached to any specific object

media {    // atmospheric media sample
   scattering { 1, rgb 0.03 }
   samples 10
}

Then change Material|Media to:

// general media statement
// put in interior or in scene file for atmospheric media
media {
   // (---general values---)
   samples 10             // minimum and maximum number of samples taken 
per interval [1,1]
   // (---media types---)
   absorption rgb<0,1,0>  // absorbing media, block light of specified color
   //emission rgb<0,1,0>  // emitting media, emit light of specified color
   /*
   scattering {           // scattering media
     1,                   // scattering type 1=isotropic; 2=Mie haze; 
3=Mie murky
                          // 4=Rayleigh; 5=Henyey-Greenstein
     rgb<0,1,0>           // color
     //eccentricity 0.25  // eccentricity for type 5 [0.0]
     //extinction 1.0     // for balancing amount of absorption [1.0]
   }*/
   // (---method---)
   //method 1             // old Povray 3.1 method
   //method 2             // new method with even distribution
   method 3               // adaptive sampling. New default
   aa_threshold 0.1     // accuracy threshold for method 3 [0.1]
   aa_level 4           // maximum recursion depth for method 3 [4]
   jitter 0.5           // randomness for method 2&3
   // (---density---)
   /*
   density {
     spherical            // any pattern
     color_map {...}      // color_map
     //density_map {...}  // or alternatively density_map
   }*/

   // translate etc.
}



Alain


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.