![](/i/fill.gif) |
![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
You can find the image I spoke of included with the Pov-Ray documents under
display_gamma (Section 3.1.2.3.2 Setting your Display Gamma).
- Grim
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
I found the Gamma image in the help text and will try that out (again).
I have used a white background or black background for simplicity to post
the pictures. Typically my favourite is the reflective coppery one from the
"WOODBOX.POV" in the samples, but it takes longer to render during
develoment.
Perhaps what I'm not so sure about is the comment to:
>> Move the light back by at least 350 units in ...
as I'm unclear as to why that looks a bit better. Given my measurements of a
table sized 20x30, then 350 units would be about 30 feet away - I guess that
makes sense, but I had some lights about 200 units. Is there a technical
reason for the lights to be further way?
Any additional comments or suggestions welcome.
--
Stephen
and
Kathie
(S+K)
"GrimDude" <a36### [at] bellsouth net> wrote in message
news:41c44f76$1@news.povray.org...
> What you have here is great, yet overwhelmed with light (I believe).
>
> Somewhere on the Pov server is a gif image used to adjust your monitor for
> the proper gamma value. You can ask about it in povray.general, or if I
come
> across it later I'll reply again here. Whichever value it comes up with is
> the value you want to use as the assumed gamma.
>
> White backgrounds are good, too. You don't have to use white, though. I
> prefer to think of the background as the environment in which I would
> normally find the object I am rendering. I use white only after I have
> control of the lighting. However, I do use a sky_sphere rather than a
plane.
>
> sky_sphere { pigment { gradient y poly_wave 1.5 color_map {
> [0 color rgb .99]
> [.35 color rgb 1.33] } } }
>
> This isn't written in stone, though, so even if it gives you good results
> continue to experiment. I'm one guy that always breaks the rules in search
> of my personal 'artistic impression.'
>
> For wod, though, I think I'd be going in a different direction. I'd
probably
> use a slightly blue, or even yellow light source. But again, this is not
as
> important as control of the light is. In your image I see the light_source
> as overwhelming the object. Move the light back by at least 350 units in
> every direction. That's a good place to start, anyway. If you need a light
> in one area in order to bring out some details, don't be afraid to add a
> diminished light value that is also shadowless (so it doesn't leave clues
as
> to its presence. Something like:
>
> light_source { <#,#,#> rgb .3 shadowless }
>
> Good luck.
>
> - Grim
>
>
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Units are basically arbitrary. What I meant was to move the lights further
away. Try something like"
light_source { <-350,350,-475> color rgb 1 }
and if you insist upon using three lights, just sapce them apart on the
x-axis, or something similar. You won't see the swamping of light upon a
object and overall intensity should appear diminished. The reason for using
greater separation distances is to aid in the control of lighting. Lighting
will appear to be much more realistic without having to rely upon parallel
rays, for instance.
Experiment! You can't break anything that I know of! I can't wait to see
your results!
- Grim
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Thanks for the feedback.
But the reason I was using three lights was to get **enough** illumination.
Just one light doesn't seem to cut it, so I'm wondering if there's something
basically wrong in my approach or system.
So the question would be - in any typically arbitrary scene such as modeling
a TV or a plant or a table or desk, can you illuminate the scene with one
light source which would approximately simulate one 100W lightbulb in the
room, at 6-18 feet away?
My answer would be - nope- need multiple lights. What's that tell you?
I'll try the gamma correction right now.
Stephen
--
"GrimDude" <a36### [at] bellsouth net> wrote in message
news:41c4ee3f$1@news.povray.org...
> Units are basically arbitrary. What I meant was to move the lights further
> away. Try something like"
>
> light_source { <-350,350,-475> color rgb 1 }
>
> and if you insist upon using three lights, just sapce them apart on the
> x-axis, or something similar. You won't see the swamping of light upon a
> object and overall intensity should appear diminished. The reason for
using
> greater separation distances is to aid in the control of lighting.
Lighting
> will appear to be much more realistic without having to rely upon parallel
> rays, for instance.
>
> Experiment! You can't break anything that I know of! I can't wait to see
> your results!
>
> - Grim
>
>
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Stephen wrote:
> Thanks for the feedback.
>
> But the reason I was using three lights was to get **enough** illumination.
> Just one light doesn't seem to cut it, so I'm wondering if there's something
> basically wrong in my approach or system.
>
> So the question would be - in any typically arbitrary scene such as modeling
> a TV or a plant or a table or desk, can you illuminate the scene with one
> light source which would approximately simulate one 100W lightbulb in the
> room, at 6-18 feet away?
>
> My answer would be - nope- need multiple lights. What's that tell you?
>
>
> I'll try the gamma correction right now.
From an advanced newbie...
Realistically simulating lamplight (or any other kind of real-world
light source) in POV-Ray is a good deal more complicated than simply
defining a light source at a certain distance, with a certain
value/intensity, and hoping for the best.
First of all, a default light source has no falloff. The light rays will
be equally bright whether they're 100 units away or 10,000 units. The
only difference is how many rays hit the object (since a default light
source emits light in all directions, placing the light source very far
away spreads the rays out more, thereby making an object *appear*
darker). In reality, it's just receiving fewer beams.
However, putting your light source very close isn't always the answer
either, since you'll have high concentrations of light beams in a small
area, causing washed-out spots.
GrimDude is simply saying that you'll get smoother coverage with your
lights - the rays will be more spread-out, reducing blown highlights -
by moving your light sources farther away. I would have said exactly the
same thing myself if I had noticed the problem (it really didn't strike
me). When I'm not trying to achieve photorealism, I quite often put my
lights 500+ units away, even if my object is only a few units each
dimension. If I had a very large object, eg. 30x30x30, I would probably
put my lights even farther, such as 800. That's just how I tend to
operate, however. If you really want a photorealistic effect, the rules
all change.
There are ways to make the light sources behave realistically. POV-Ray
supports light falloff and fading, which, when set properly (an area I
can't advise on, unfortunately) can produce quite realistic results. If
you want to look these up, look for the 'falloff' and 'fade_distance'
keywords in the Help file.
Incidentally, Jaime Vives Piqueres has done some research into the
physical properties of light and has created some macros for realistic
light simulation. I am still learning how to use them myself, but you
might find them interesting. http://ignorancia.org/lightsys.php
Hope this helps somehow. As long as you keep in mind that I'm not an
expert and thus take all this with a grain of salt, I think you'll be
okay. :)
~Mike
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Mike Thorn wrote:
| First of all, a default light source has no falloff. The light rays
will
| be equally bright whether they're 100 units away or 10,000 units. The
| only difference is how many rays hit the object (since a default light
| source emits light in all directions, placing the light source very
far
| away spreads the rays out more, thereby making an object *appear*
| darker). In reality, it's just receiving fewer beams.
|
There appears to be a small misunderstanding on how light sources
work here. For regular light sources (ie not area lights), there is
exactly one ray of light for each point of your object no matter what
the distance to the light source. If you don't use anything special
like falloff or semi-transparent objects, that light ray will have
the exact intensity specified in the light source definition. As a
result, the object will have about the same brightness no matter how
far the light source is situated. However the brightness is affected
by the angle between the light ray and the surface: it is maximal if
the ray hits the surface straight on and diminishes to zero when the
ray is nearly parallel to the surface. This is why placing the light
source farther away makes for more uniform illumination: the farther
away, the closer to parallel the rays.
Incidentally, your description ("rays spread out more") is more or
less what happens in real life and the reason why real light sources
get dimmer with distance.
| There are ways to make the light sources behave realistically. POV-Ray
| supports light falloff and fading, which, when set properly (an area I
| can't advise on, unfortunately) can produce quite realistic
results. If
| you want to look these up, look for the 'falloff' and 'fade_distance'
| keywords in the Help file.
|
To get realistic results, you should use "fade_power 2" and set the
fade_distance to the size of your light bulb, then give a *very* high
value for light intensity (you do know that intensity is not
constrained to the 0-1 range, don't you? In fact you can even create
negative light sources wich allows for some interresting special
effects...). 'falloff' has nothing to do with it and has meaning only
with spotlights and cylindrical lights: it defines how much light you
get as you get further from the center of the beam.
BTW, if you intend to use cylindrical lights, read the doc
carefully, they probably don't do what you think. To get the right
effect, you'll have to look at the 'parallel' keyword :p
Jerome
- --
******************************
* Jerome M. Berger *
* mailto:jbe### [at] ifrance com *
* http://jeberger.free.fr/ *
******************************
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
iD8DBQFBxSXpqIYJdJhyixIRAtD6AJ9/Pr7e2Fq16s0OIHA9exblcJ9NHgCeKBO0
AYq6+ef9Ev9ahbfnsBNmEBQ=
=DYOL
-----END PGP SIGNATURE-----
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Hi and thanks for the info.
In rating myself, I know I'm a LONG way from some of the experts here, given
what I have seen (lots of positive challenges ahead), but my strong math and
programming skills suggest that I'm a reasonably capable newbie.
Where I'm having some troubles is in fine-tuning things such as lighting.
Therefore I appreciate your comments that .... "you'd do this, this & that"
as a matter of *your* standards. In other words it works for you and it's
worth sharing the experience. YMMV, as they say.
I just re-lighted what was to be my last image of the night, and now the
last and second-last are attached for your consideration.
I just changed the Z component without regard to the direction of the
lights, so the cabinet is lit differently in each picture. I didn't mean for
this to become a comparison, but, the lights in picture 8 are further away
are at z*-300 and intensity of 1.0
light_source { < -30, 10, -300> color rgb <1.0,1.0,1.0> }
light_source { < 0, 20, -300> color rgb <1.0,1.0,1.0> }
while the lights in picture 7 are close to the chest at z*-30 and intensity
2.0:
light_source { < -30, 10, -30> color rgb <2.0,2.0,2.0> }
light_source { < 0, 20, -30> color rgb <2.0,2.0,2.0> }
Does anyone typically use an intensity (light colour) greater than 1.0? I
just tried this to get more light into the picture. Both scenes are lit with
the above lights and a sky sphere. This is getting better, although I do
like the highlight on the front left leg from the close-in light.
I have been tending to put lights reasonably close but quiet happy to move
them to 500-1000 units away based on a 30x30x30 object (or a 3x3x3 as you
suggested). I'm wondering if the wood-textures I'm using are part of my
issue. Usually I use {Green} when I'm prototyping or {Red & Blue & Yellow}
and these colours tend to show up well.
I don't intend to go so far as the photorealistic lighting i.e. falloff,
etc, unless I have a specific project in mind, but thanks for pointing me in
that direction for future reference.
I didn't mean to turn this conversation into such a long one, but I do
appreciate suggestions such as you have provided.
best regards
Stephen
--
Stephen
and
Kathie
(S+K)
"Mike Thorn" <mik### [at] realitycheckmultimedia com> wrote in message
news:41c50d30$1@news.povray.org...
> Stephen wrote:
> > Thanks for the feedback.
> >
> > But the reason I was using three lights was to get **enough**
illumination.
> > Just one light doesn't seem to cut it, so I'm wondering if there's
something
> > basically wrong in my approach or system.
> >
> > So the question would be - in any typically arbitrary scene such as
modeling
> > a TV or a plant or a table or desk, can you illuminate the scene with
one
> > light source which would approximately simulate one 100W lightbulb in
the
> > room, at 6-18 feet away?
> >
> > My answer would be - nope- need multiple lights. What's that tell you?
> >
> >
> > I'll try the gamma correction right now.
>
> From an advanced newbie...
>
> Realistically simulating lamplight (or any other kind of real-world
> light source) in POV-Ray is a good deal more complicated than simply
> defining a light source at a certain distance, with a certain
> value/intensity, and hoping for the best.
>
> First of all, a default light source has no falloff. The light rays will
> be equally bright whether they're 100 units away or 10,000 units. The
> only difference is how many rays hit the object (since a default light
> source emits light in all directions, placing the light source very far
> away spreads the rays out more, thereby making an object *appear*
> darker). In reality, it's just receiving fewer beams.
>
> However, putting your light source very close isn't always the answer
> either, since you'll have high concentrations of light beams in a small
> area, causing washed-out spots.
>
> GrimDude is simply saying that you'll get smoother coverage with your
> lights - the rays will be more spread-out, reducing blown highlights -
> by moving your light sources farther away. I would have said exactly the
> same thing myself if I had noticed the problem (it really didn't strike
> me). When I'm not trying to achieve photorealism, I quite often put my
> lights 500+ units away, even if my object is only a few units each
> dimension. If I had a very large object, eg. 30x30x30, I would probably
> put my lights even farther, such as 800. That's just how I tend to
> operate, however. If you really want a photorealistic effect, the rules
> all change.
>
> There are ways to make the light sources behave realistically. POV-Ray
> supports light falloff and fading, which, when set properly (an area I
> can't advise on, unfortunately) can produce quite realistic results. If
> you want to look these up, look for the 'falloff' and 'fade_distance'
> keywords in the Help file.
>
> Incidentally, Jaime Vives Piqueres has done some research into the
> physical properties of light and has created some macros for realistic
> light simulation. I am still learning how to use them myself, but you
> might find them interesting. http://ignorancia.org/lightsys.php
>
> Hope this helps somehow. As long as you keep in mind that I'm not an
> expert and thus take all this with a grain of salt, I think you'll be
> okay. :)
>
> ~Mike
Post a reply to this message
Attachments:
Download 'Chest36x24x18-Leg7.jpg' (61 KB)
Download 'Chest36x24x18-Leg8.jpg' (62 KB)
Preview of image 'Chest36x24x18-Leg7.jpg'
![Chest36x24x18-Leg7.jpg](/povray.binaries.images/attachment/%3C41c52c47%40news.povray.org%3E/Chest36x24x18-Leg7.jpg?preview=1)
Preview of image 'Chest36x24x18-Leg8.jpg'
![Chest36x24x18-Leg8.jpg](/povray.binaries.images/attachment/%3C41c52c47%40news.povray.org%3E/Chest36x24x18-Leg8.jpg?preview=1)
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Stephen nous apporta ses lumieres ainsi en ce 2004-12-19 02:22... :
>Hi and thanks for the info.
>
>In rating myself, I know I'm a LONG way from some of the experts here, given
>what I have seen (lots of positive challenges ahead), but my strong math and
>programming skills suggest that I'm a reasonably capable newbie.
>
>Where I'm having some troubles is in fine-tuning things such as lighting.
>Therefore I appreciate your comments that .... "you'd do this, this & that"
>as a matter of *your* standards. In other words it works for you and it's
>worth sharing the experience. YMMV, as they say.
>
>I just re-lighted what was to be my last image of the night, and now the
>last and second-last are attached for your consideration.
>
>I just changed the Z component without regard to the direction of the
>lights, so the cabinet is lit differently in each picture. I didn't mean for
>this to become a comparison, but, the lights in picture 8 are further away
>are at z*-300 and intensity of 1.0
>
>light_source { < -30, 10, -300> color rgb <1.0,1.0,1.0> }
>light_source { < 0, 20, -300> color rgb <1.0,1.0,1.0> }
>
>while the lights in picture 7 are close to the chest at z*-30 and intensity
>2.0:
>light_source { < -30, 10, -30> color rgb <2.0,2.0,2.0> }
>light_source { < 0, 20, -30> color rgb <2.0,2.0,2.0> }
>
>Does anyone typically use an intensity (light colour) greater than 1.0? I
>just tried this to get more light into the picture. Both scenes are lit with
>the above lights and a sky sphere. This is getting better, although I do
>like the highlight on the front left leg from the close-in light.
>
>
>I have been tending to put lights reasonably close but quiet happy to move
>them to 500-1000 units away based on a 30x30x30 object (or a 3x3x3 as you
>suggested). I'm wondering if the wood-textures I'm using are part of my
>issue. Usually I use {Green} when I'm prototyping or {Red & Blue & Yellow}
>and these colours tend to show up well.
>
>I don't intend to go so far as the photorealistic lighting i.e. falloff,
>etc, unless I have a specific project in mind, but thanks for pointing me in
>that direction for future reference.
>
>I didn't mean to turn this conversation into such a long one, but I do
>appreciate suggestions such as you have provided.
>
>best regards
>
>Stephen
>
>
>
What diffuse value do you use? To low will darken your object, to high
can wash out the colours.
I sometimes use large values for light(s) like 100+. In deffining the
light, rgb 1 = rgb<1.0,1.0,1.0>.
Alain
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
> There appears to be a small misunderstanding on how light sources
> work here. <snip>
I see where I went wrong here. Thanks for clarifying that! :)
~Mike
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Thanks for the update.
I tried something new which I thought I'd share. Typically I move lights
around until it "looks right", but now that they're farther away than I had
been placing them, it's a bit harder to guess how far they should be moved
to make a significant difference in terms of the vector of the light and the
vector of the camera.
So..... Let's find out where the lights actually are and make them visible
for testing:
1) define the light's position
#declare LightSource0 = < 0, 0, 0 >; // assume this is the
center of your scene or an important element:
#declare LightSource1 = < 300, 200, -300 >;
#declare LightSource2 = <-300, 100, -300 >;
#declare LightSource3 = < 0 , 70, -300 >;
2) define the lights at that position
light_source { LightSource1 color rgb <1.0,1.0,1.0> }
light_source { LightSource2 color rgb <1.0,1.0,1.0> }
light_source { LightSource3 color rgb <1.0,1.0,1.0> }
3) build a cylinder from the "origin" towards the lightsource, but not all
the way to the light because it then blocks the light
Perhaps there's other ways to do this, but this works.
cylinder { VWith_Len(vnormalize(LightSource1), 120), LightSource0, 1.0
pigment {Red} }
cylinder { VWith_Len(vnormalize(LightSource2), 40), LightSource0, 1.0
pigment {Green} }
cylinder { VWith_Len(vnormalize(LightSource3), 40), LightSource0, 1.0
pigment {Blue} }
Now that I know where the light is pointing from, I can see why I have
shadows in certain places. Works for me.
--
Stephen
and
Kathie
(S+K)
news:41c525eb@news.povray.org...
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Mike Thorn wrote:
>
> | First of all, a default light source has no falloff. The light rays
> will
> | be equally bright whether they're 100 units away or 10,000 units. The
> | only difference is how many rays hit the object (since a default light
> | source emits light in all directions, placing the light source very
> far
> | away spreads the rays out more, thereby making an object *appear*
> | darker). In reality, it's just receiving fewer beams.
> |
> There appears to be a small misunderstanding on how light sources
> work here. For regular light sources (ie not area lights), there is
> exactly one ray of light for each point of your object no matter what
> the distance to the light source. If you don't use anything special
> like falloff or semi-transparent objects, that light ray will have
> the exact intensity specified in the light source definition. As a
> result, the object will have about the same brightness no matter how
> far the light source is situated. However the brightness is affected
> by the angle between the light ray and the surface: it is maximal if
> the ray hits the surface straight on and diminishes to zero when the
> ray is nearly parallel to the surface. This is why placing the light
> source farther away makes for more uniform illumination: the farther
> away, the closer to parallel the rays.
>
> Incidentally, your description ("rays spread out more") is more or
> less what happens in real life and the reason why real light sources
> get dimmer with distance.
>
> | There are ways to make the light sources behave realistically. POV-Ray
> | supports light falloff and fading, which, when set properly (an area I
> | can't advise on, unfortunately) can produce quite realistic
> results. If
> | you want to look these up, look for the 'falloff' and 'fade_distance'
> | keywords in the Help file.
> |
> To get realistic results, you should use "fade_power 2" and set the
> fade_distance to the size of your light bulb, then give a *very* high
> value for light intensity (you do know that intensity is not
> constrained to the 0-1 range, don't you? In fact you can even create
> negative light sources wich allows for some interresting special
> effects...). 'falloff' has nothing to do with it and has meaning only
> with spotlights and cylindrical lights: it defines how much light you
> get as you get further from the center of the beam.
>
> BTW, if you intend to use cylindrical lights, read the doc
> carefully, they probably don't do what you think. To get the right
> effect, you'll have to look at the 'parallel' keyword :p
>
> Jerome
>
> - --
> ******************************
> * Jerome M. Berger *
> * mailto:jbe### [at] ifrance com *
> * http://jeberger.free.fr/ *
> ******************************
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.4 (GNU/Linux)
>
> iD8DBQFBxSXpqIYJdJhyixIRAtD6AJ9/Pr7e2Fq16s0OIHA9exblcJ9NHgCeKBO0
> AYq6+ef9Ev9ahbfnsBNmEBQ=
> =DYOL
> -----END PGP SIGNATURE-----
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |