I am trying to simulate the extinction of light inside a MIST volume, for
different sky conditions.
The simulations seems to work well using "sun" as the light source, but I
have problems when using an overcast sky or a sunny sky with no sun (-s
option).
The light sources for the MIST volume are "sun" and "sky" which is defined
using glow.
Did MIST work for glow type light sources? If not, what do you recommend to
use to simulate this case, with diffuse light coming from the sky?
Thank you in advance,
Guedi
···
---
Arch. Guedi Capeluto, D.Sc.
Faculty of Architecture and Town Planning
Technion - Israel Institute of Technology
Technion City, Haifa 32000. Israel.
Tel: +972.4.8294013 Fax: +972.4.8294617
email: [email protected]
hi,
glow is considered as a light source by the direct component only within the
given distance (4th number of glow definition) and I guess "mist" doesn't
consider glow with 0 (or -1) because it is not a direct light source.
a quick and probably useless (**) trick is to change the sky description
from
skyfunc glow skyglow
0
0
3 1 1 1
to
skyfunc light skyglow
0
0
3 1 1 1
this way the 1st bounce is now direct but -ds doesn't work on the sky and
there is just a big point source above so ... (**)
another trick (this time should be ok) is to create a big polygonal volume
of brightfunc-light-sources with the sky distribution... in this case you
could be able to use -ds.
I am trying to simulate the extinction of light inside a MIST volume, for
different sky conditions.
The simulations seems to work well using "sun" as the light source, but I
have problems when using an overcast sky or a sunny sky with no sun (-s
option).
The light sources for the MIST volume are "sun" and "sky" which is defined
using glow.
Did MIST work for glow type light sources? If not, what do you recommend to
use to simulate this case, with diffuse light coming from the sky?
Thank you in advance,
Guedi
---
Arch. Guedi Capeluto, D.Sc.
Faculty of Architecture and Town Planning
Technion - Israel Institute of Technology
Technion City, Haifa 32000. Israel.
Tel: +972.4.8294013 Fax: +972.4.8294617
email: [email protected]
___________________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses.
ERRATA CORRIGE
apparently -ds doesn't do the trick of subdividing the source for the mist...
if you are simulating light from windows this shouldn't be an issue
but if you have an open volume of atmosphere you may see bright areas of
mist in corrispondence of the sources centres (depending on your mist settings).
I guess the reason is that it is more efficient to only test the centre
as usually sources are small compared to the scene dimensions... in your
case it is the opposite...
any ideas how to solve this (a part using tons of small sources)?
giulio
-- Messaggio originale --
From: Giulio Antonutto <[email protected]>
To: 'Radiance general discussion' <[email protected]>
Subject: RE: [Radiance-general] Using MIST for overcast sky
Date: Fri, 16 Apr 2004 14:56:43 +0100
Reply-To: Radiance general discussion <[email protected]>
hi,
glow is considered as a light source by the direct component only within
the
given distance (4th number of glow definition) and I guess "mist" doesn't
consider glow with 0 (or -1) because it is not a direct light source.
a quick and probably useless (**) trick is to change the sky description
from
skyfunc glow skyglow
0
0
3 1 1 1
to
skyfunc light skyglow
0
0
3 1 1 1
this way the 1st bounce is now direct but -ds doesn't work on the sky and
there is just a big point source above so ... (**)
another trick (this time should be ok) is to create a big polygonal volume
of brightfunc-light-sources with the sky distribution... in this case you
could be able to use -ds.
I am trying to simulate the extinction of light inside a MIST volume, for
different sky conditions.
The simulations seems to work well using "sun" as the light source, but
I
···
-----Original Message-----
have problems when using an overcast sky or a sunny sky with no sun (-s
option).
The light sources for the MIST volume are "sun" and "sky" which is defined
using glow.
Did MIST work for glow type light sources? If not, what do you recommend
to
use to simulate this case, with diffuse light coming from the sky?
Thank you in advance,
Guedi
---
Arch. Guedi Capeluto, D.Sc.
Faculty of Architecture and Town Planning
Technion - Israel Institute of Technology
Technion City, Haifa 32000. Israel.
Tel: +972.4.8294013 Fax: +972.4.8294617
email: [email protected]
___________________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses.
_______________________________________________
Radiance-general mailing list [email protected] http://www.radiance-online.org/mailman/listinfo/radiance-general
Unfortunately, mist is never going to work with distant sources, because if your raise any extinction to an infinite power, no matter how close to one it starts out, you get a zero result. (From RwR Chap. 14, eq. 14.7, we see that light is absorbed as the power of distance.) I just had a similar discussion with another user (Pablo Escanilla <[email protected]>), who was simulating fog on a roadway.
Real atmospheres are quite a bit more complex than setting a global constant value for extinction and albedo. Real atmospheres change with altitude, often dramatically, so it doesn't become pitch dark as soon as the fog rolls in. Simulating large scale atmospherics in Radiance means creating enormous "slabs" of mist with different extinctions at different altitudes. It can be done, but it's a bit of a pain and, as far as I know, no one has yet made the attempt.
As to sampling the sky dome, this is never going to work. You are much better off setting a sensible value for -av, which gets figured into the mist calculation as an appropriate multiplier on albedo. This was my advice to Pablo.
-Greg
From: "Guedi Capeluto" <[email protected]>
Date: April 16, 2004 2:06:43 AM PDT
Hi,
I am trying to simulate the extinction of light inside a MIST volume, for different sky conditions.
The simulations seems to work well using "sun" as the light source, but I have problems when using an overcast sky or a sunny sky with no sun (-s option).
The light sources for the MIST volume are "sun" and "sky" which is defined using glow.
Did MIST work for glow type light sources? If not, what do you recommend to use to simulate this case, with diffuse light coming from the sky?
My intention was not simulating light extinction through the atmosphere but
once light reaches a mist object which is on the ground, (I exactly know the
extinction coefficient within it), analyze the light (direct from sun and
diffuse from sky dome) extinction inside this small volume.
Since I need to know the light levels (illuminance) inside the volume at
different depths, it seems to me that the -av approach will not work.
Unfortunately, mist is never going to work with distant sources,
because if your raise any extinction to an infinite power, no matter
how close to one it starts out, you get a zero result. (From RwR Chap.
14, eq. 14.7, we see that light is absorbed as the power of distance.)
I just had a similar discussion with another user (Pablo Escanilla
<[email protected]>), who was simulating fog on a roadway.
Real atmospheres are quite a bit more complex than setting a global
constant value for extinction and albedo. Real atmospheres change with
altitude, often dramatically, so it doesn't become pitch dark as soon
as the fog rolls in. Simulating large scale atmospherics in Radiance
means creating enormous "slabs" of mist with different extinctions at
different altitudes. It can be done, but it's a bit of a pain and, as
far as I know, no one has yet made the attempt.
As to sampling the sky dome, this is never going to work. You are much
better off setting a sensible value for -av, which gets figured into
the mist calculation as an appropriate multiplier on albedo. This was
my advice to Pablo.
-Greg
From: "Guedi Capeluto" <[email protected]>
Date: April 16, 2004 2:06:43 AM PDT
Hi,
I am trying to simulate the extinction of light inside a MIST volume,
for different sky conditions.
The simulations seems to work well using "sun" as the light source, but
I have problems when using an overcast sky or a sunny sky with no sun
(-s option).
The light sources for the MIST volume are "sun" and "sky" which is
defined using glow.
Did MIST work for glow type light sources? If not, what do you
recommend to use to simulate this case, with diffuse light coming from
the sky?
Well, then you can try Pillo's suggestion of simulating the sky as many small sources, but I don't hold out much hope that you'll get an accurate answer. What local phenomenon are you trying to simulate, exactly? A puff of smoke?
-Greg
···
Begin forwarded message:
From: "Guedi Capeluto" <[email protected]>
Date: April 18, 2004 9:06:50 AM PDT
Thank you very much for your answer.
My intention was not simulating light extinction through the atmosphere but
once light reaches a mist object which is on the ground, (I exactly know the
extinction coefficient within it), analyze the light (direct from sun and
diffuse from sky dome) extinction inside this small volume.
Since I need to know the light levels (illuminance) inside the volume at
different depths, it seems to me that the -av approach will not work.
Our intention is to simulate light extinction inside a tree (yes yes...a
tree) or a group of trees with the purpose of studying light penetration
according to different morphologies. The mist approach looked to me much
simpler than modeling the whole tree (or a lot of trees), considering than
even in this case the model will be just an approximation.
Anyway I will try constructing a dome from polygons
brightfunc-light-sources, as Giulio suggested.
Guedi
···
-----Original Message-----
From: Greg Ward [mailto:[email protected]]
Sent: Monday, April 19, 2004 1:30 AM
To: Guedi Capeluto; Radiance general discussion
Cc: [email protected]
Subject: Re: Using MIST for overcast sky
Hi Guedi,
Well, then you can try Pillo's suggestion of simulating the sky as many
small sources, but I don't hold out much hope that you'll get an
accurate answer. What local phenomenon are you trying to simulate,
exactly? A puff of smoke?
-Greg
Begin forwarded message:
From: "Guedi Capeluto" <[email protected]>
Date: April 18, 2004 9:06:50 AM PDT
Thank you very much for your answer.
My intention was not simulating light extinction through the
atmosphere but
once light reaches a mist object which is on the ground, (I exactly
know the
extinction coefficient within it), analyze the light (direct from sun
and
diffuse from sky dome) extinction inside this small volume.
Since I need to know the light levels (illuminance) inside the volume
at
different depths, it seems to me that the -av approach will not work.