As requested, here's a tutorial that covers using media like atmosphere.
It's a good read and I think most people will find it informative. 
Unfortunately is seemed to prove that in certain areas media falls
short.  It's got a few hours worth of rendering on it.

http://members.xoom.com/POVRAY3/media2.html

Or you can get to it through the frames here

http://members.xoom.com/POVRAY3.

I added some animations not too long ago.  The one with the moving water
was done in POV.

-Mike

Post a reply to this message

Mike <Ama### [at] aolcom> wrote:
: Unfortunately is seemed to prove that in certain areas media falls
: short.

  Atmosphere seems to be much better than media.
  I think it would be a very good idea to support atmosphere again, not as
a substitute of the atmospheric media, but as an alternative.
  Does anyone else agree with this? Perhaps povteam will listen to the
crowd...

-- 
main(i){char*_="BdsyFBThhHFBThhHFRz]NFTITQF|DJIFHQhhF";while(i=
*_++)for(;i>1;printf("%s",i-70?i&1?"[]":" ":(i=0,"\n")),i/=2);} /*- Warp. -*/

Post a reply to this message

Nieminen Mika wrote:

> Mike <Ama### [at] aolcom> wrote:
> : Unfortunately is seemed to prove that in certain areas media falls
> : short.
>
>   Atmosphere seems to be much better than media.
>   I think it would be a very good idea to support atmosphere again, not as
> a substitute of the atmospheric media, but as an alternative.
>   Does anyone else agree with this? Perhaps povteam will listen to the
> crowd...

I never got around to using atmospheric media and know little of it's
capabilities. Perhaps a better request would be to incorporate
different sampling methods into the media feature itself. As has been
mentioned Chris Young choose the monte carlo version of sampling
but there are other methods available. I know too little to comment
more, but it seems possible the feature can be added on to, to increase
it's effectiveness.

--
Ken Tyler

tyl### [at] pacbellnet

Post a reply to this message

Great tut, Mike. Bad media. Say it with me everyone, "Bad media, bad!"
I didn't find mention of sky_sphere there so I will mention it here. It
seems to me that an infinite background is impossible to work with when
using the uncontained form of media; meaning 'sky_sphere', 'plane' (as
ground or floor only), or 'background' are not compatible to the
atmospheric media idea.
If I use a regular sphere or box as a massive sky container replacement
the media is far more useable it seems.

Mike, you might have the most "damaging evidence" tutorial against media
out there but it's a good one none-the-less!


Mike wrote:
> 
> As requested, here's a tutorial that covers using media like atmosphere.
> It's a good read and I think most people will find it informative.
> Unfortunately is seemed to prove that in certain areas media falls
> short.  It's got a few hours worth of rendering on it.
> 
> http://members.xoom.com/POVRAY3/media2.html
> 
> Or you can get to it through the frames here
> 
> http://members.xoom.com/POVRAY3.
> 
> I added some animations not too long ago.  The one with the moving water
> was done in POV.
> 
> -Mike

-- 
 omniVERSE: beyond the universe
  http://members.aol.com/inversez/POVring.htm
=Bob

Post a reply to this message

Ken wrote:
> 
> I never got around to using atmospheric media and know little of it's
> capabilities. Perhaps a better request would be to incorporate
> different sampling methods into the media feature itself. As has been
> mentioned Chris Young choose the monte carlo version of sampling
> but there are other methods available. I know too little to comment
> more, but it seems possible the feature can be added on to, to increase
> it's effectiveness.

I agree.  The biggest problem is the monte-carlo sampling.  I just sent
a message to Chris about it (before reading this).  I modified my version
of POV to use uniform samples up to the minimum samples (and monte-carlo
after that).  When I set min and max samples the same (so you never go
above the min samples), media is just as smooth as a halo (I haven't
tried atmosphere).  The render time for the same number of samples is
almost the same, too (media is still a touch slower, but that may be
my compiler).

Have you heard of any other sampling techniques?

-Nathan

Post a reply to this message

Nathan Kopp wrote:

> Ken wrote:
> >
> > I never got around to using atmospheric media and know little of it's
> > capabilities. Perhaps a better request would be to incorporate
> > different sampling methods into the media feature itself. As has been
> > mentioned Chris Young choose the monte carlo version of sampling
> > but there are other methods available. I know too little to comment
> > more, but it seems possible the feature can be added on to, to increase
> > it's effectiveness.
>
> I agree.  The biggest problem is the monte-carlo sampling.  I just sent
> a message to Chris about it (before reading this).  I modified my version
> of POV to use uniform samples up to the minimum samples (and monte-carlo
> after that).  When I set min and max samples the same (so you never go
> above the min samples), media is just as smooth as a halo (I haven't
> tried atmosphere).  The render time for the same number of samples is
> almost the same, too (media is still a touch slower, but that may be
> my compiler).
>
> Have you heard of any other sampling techniques?
>
> -Nathan

 No more than a few vague mentions from Chris in his infrequent
public announcements. I think I might have seen some info at
the RTNews site but could not say in which issue.

http://www.acm.org/tog/resources/RTNews/html/

--
Ken Tyler

tyl### [at] pacbellnet

Post a reply to this message

Bob Hughes wrote:

> Great tut, Mike. Bad media. Say it with me everyone, "Bad media, bad!"
> I didn't find mention of sky_sphere there so I will mention it here. It
> seems to me that an infinite background is impossible to work with when
> using the uncontained form of media; meaning 'sky_sphere', 'plane' (as
> ground or floor only), or 'background' are not compatible to the
> atmospheric media idea.
> If I use a regular sphere or box as a massive sky container replacement
> the media is far more useable it seems.
>
> Mike, you might have the most "damaging evidence" tutorial against media
> out there but it's a good one none-the-less!
>

Not altogether true. I'm going to be posting a small image in the
images group today that uses the background color feature. The
media container will again be a single hollow plane with the camera
inside of the planes normal. It is possible to have the virtual backgroud
objects work with media as you will see later on.

--
Ken Tyler

tyl### [at] pacbellnet

Post a reply to this message

Ken wrote:

> Nathan Kopp wrote:
>
> > Ken wrote:
> > >
> > > I never got around to using atmospheric media and know little of it's
> > > capabilities. Perhaps a better request would be to incorporate
> > > different sampling methods into the media feature itself. As has been
> > > mentioned Chris Young choose the monte carlo version of sampling
> > > but there are other methods available. I know too little to comment
> > > more, but it seems possible the feature can be added on to, to increase
> > > it's effectiveness.
> >
> > I agree.  The biggest problem is the monte-carlo sampling.  I just sent
> > a message to Chris about it (before reading this).  I modified my version
> > of POV to use uniform samples up to the minimum samples (and monte-carlo
> > after that).  When I set min and max samples the same (so you never go
> > above the min samples), media is just as smooth as a halo (I haven't
> > tried atmosphere).  The render time for the same number of samples is
> > almost the same, too (media is still a touch slower, but that may be
> > my compiler).
> >
> > Have you heard of any other sampling techniques?
> >
> > -Nathan
>
>  No more than a few vague mentions from Chris in his infrequent
> public announcements. I think I might have seen some info at
> the RTNews site but could not say in which issue.
>
> http://www.acm.org/tog/resources/RTNews/html/
>
> --
> Ken Tyler
>
> tyl### [at] pacbellnet

Here is a link to what looks like a usenet discussion about the
relevant problems encountered using monte carlo sampling methods.
While not directly tied into media and scattering the discussion focuses
on the illuminary properties and raytracing problems and benefits
associated with it's use.

http://www.acm.org/tog/resources/RTNews/html/rtnv10n2.html#art6

I found it enlightening :)

--
Ken Tyler

tyl### [at] pacbellnet

Post a reply to this message

THANK YOU!

Mike wrote:

> As requested, here's a tutorial that covers using media like atmosphere.
> It's a good read and I think most people will find it informative.
> Unfortunately is seemed to prove that in certain areas media falls
> short.  It's got a few hours worth of rendering on it.
>
> http://members.xoom.com/POVRAY3/media2.html
>
> Or you can get to it through the frames here
>
> http://members.xoom.com/POVRAY3.
>
> I added some animations not too long ago.  The one with the moving water
> was done in POV.
>
> -Mike

Post a reply to this message

Thanks for the URL... actually, I'm changing my earlier suggestion... I think
there should be an option to go back to the old sampling method (uniform
samples with adaptive anti-aliasing).  I'm working on code right now.

-Nathan

Ken wrote:
> 
> Here is a link to what looks like a usenet discussion about the
> relevant problems encountered using monte carlo sampling methods.
> While not directly tied into media and scattering the discussion focuses
> on the illuminary properties and raytracing problems and benefits
> associated with it's use.
> 
> http://www.acm.org/tog/resources/RTNews/html/rtnv10n2.html#art6
> 
> I found it enlightening :)
>

Post a reply to this message