Jump to content


Photo

Soft Light - Technique


  • Please log in to reply
30 replies to this topic

#1 Charles DeRosa

Charles DeRosa
  • Basic Members
  • PipPip
  • 13 posts
  • Cinematographer
  • los angeles, ca

Posted 21 September 2005 - 09:43 PM

Two questions about soft light-

1) I just heard first hand info about a big time DP lighting Nicole Kidman. He ran a big light, bounced it off a card, then through 3 different frames before it reached her.
My question - Did those frames make a difference in the quality? Aren't they simply cutting the light? If the main factors are the size of the source and its proximity to the subject, why all the extra frames of diffucsion? I don't know if the frames got progressively larger or not - I'll find out. By the way, Kidman looked fantastic after all those frames.

2)Drop off- We constantly hear about light drop off...but doesn't all light comply with the inverse square law? Why does soft light seem to have a faster drop off?

thanks,
Chuck DeRosa
Los Angeles
  • 0

#2 Michael Nash

Michael Nash
  • Sustaining Members
  • 3330 posts
  • Cinematographer
  • Pasadena, CA

Posted 21 September 2005 - 10:50 PM

Two questions about soft light-

1)  I just heard first hand info about a big time DP lighting Nicole Kidman.  He ran a big light, bounced it off a card, then through 3 different frames before it reached her.
My question - Did those frames make a difference in the quality?  Aren't they simply cutting the light?  If the main factors are the size of the source and its proximity to the subject, why all the extra frames of diffucsion?  I don't know if the frames got progressively larger or not - I'll find out.  By the way, Kidman looked fantastic after all those frames.

<{POST_SNAPBACK}>


If you use successively lighter densities of diffusion, then the diffusion patern of the light can become more "complex;" that is, it can retain some of the spread and falloff of from each layer of diffusion, but all coming from the same direction. But there comes a point where too many layers of diffusion or too dense a frame can just obliterate the effect the other frames are creating.

Think of it this way; put up a hard light and aim it at a subject. now put a frame of lightweight diffusion, like a china silk, close to the subject. Look at the quality of light and the shadow pattern on the subject -- you'll notice that there's a hard shadow "punching through" the silk, while there is a softer quality of light wrapping around the subject and falling off quickly at the same time.

The trick is that each layer of diffusion has to let through a significant amount of the light from the layer before it, or else it just wipes it out. You could put three layers of hampshire or Opal frost (with a bit of space between them) and get a lovely and complex depth to the light; but if the last layer was something heavy like full grid cloth you'd be softening all the subtle quality of the other layers.


2)Drop off-  We constantly hear about light drop off...but doesn't all light comply with the inverse square law?  Why does soft light seem to have a faster drop off? 

thanks,
Chuck DeRosa
Los Angeles

<{POST_SNAPBACK}>


The inverse square law applies only to hard light. What happens with soft light is that the light rays are going in multiple directions at once, not just in a straight line perpendicular to the source. The light is spreading up, down, sideways, and at every angle away from the soft source. So you really can't calculate the falloff in only one direction away from the source, because only some of the the light is going that direction.
  • 0

#3 Chris Cooke

Chris Cooke
  • Sustaining Members
  • 246 posts
  • Cinematographer
  • Lethbridge, AB Canada

Posted 21 September 2005 - 11:15 PM

In my experience, those extra frames of diffusion will help diffuse the light even more. There comes a point though where light becomes so scattered that it is no longer posible to make a visible difference (except to knock the light level down) by adding more diffusion or bouncing it one more time.
Like Michael said, the the more you diffuse the light, the less the inverse square law will have to play in light falloff. Specular sources fall off less rapidly because their rays are parallel.
  • 0

#4 Dimitrios Koukas

Dimitrios Koukas
  • Sustaining Members
  • 569 posts
  • Cinematographer
  • Athens, Greece, London UK

Posted 22 September 2005 - 12:10 AM

Two questions about soft light-

1)  I just heard first hand info about a big time DP lighting Nicole Kidman.  He ran a big light, bounced it off a card, then through 3 different frames before it reached her.
My question - Did those frames make a difference in the quality?  Aren't they simply cutting the light?  If the main factors are the size of the source and its proximity to the subject, why all the extra frames of diffucsion?  I don't know if the frames got progressively larger or not - I'll find out.  By the way, Kidman looked fantastic after all those frames.

2)Drop off-  We constantly hear about light drop off...but doesn't all light comply with the inverse square law?  Why does soft light seem to have a faster drop off? 

thanks,
Chuck DeRosa
Los Angeles

<{POST_SNAPBACK}>


Chuck,
I believe both replies covered your questions, the only think that I have to add, is that it's the subject that orders your lighting, the specific DP at the specific shot believed that he needed more diffusion for her face so he added some more.
A shadowless light source was probabaly his approach.
Do u have a frame of it? I am curius to see the results.
Sometimes we come up with weird things in the set that pisses gaffers off, but it's the result u want that counts first.
Dimitrios Koukas
  • 0

#5 Eric Steelberg ASC

Eric Steelberg ASC
  • Basic Members
  • PipPipPipPip
  • 538 posts
  • Cinematographer
  • Los Angeles, CA

Posted 22 September 2005 - 12:58 AM

The inverse square law would not be a law unless it was universal.
  • 0

#6 Chris Keth

Chris Keth
  • Sustaining Members
  • 4427 posts
  • 1st Assistant Camera
  • Los Angeles

Posted 22 September 2005 - 10:40 AM

The inverse square law would not be a law unless it was universal.

<{POST_SNAPBACK}>



Completely untrue, many laws in science have conditions under which they are true. The condition for the inverse square law is that it only works with a point source. The larger the source in relation to the subject, the more inaccurate the law will be. Test it.
  • 0

#7 John Thomas

John Thomas
  • Sustaining Members
  • 116 posts
  • Cinematographer
  • Akron, Ohio USA

Posted 22 September 2005 - 10:53 AM

Sometimes it is important to let the actress know that we are working very hard to make her look beautiful.
  • 0

#8 Chris Keth

Chris Keth
  • Sustaining Members
  • 4427 posts
  • 1st Assistant Camera
  • Los Angeles

Posted 22 September 2005 - 11:07 AM

Sometimes it is important to let the actress know that we are working very hard to make her look beautiful.

<{POST_SNAPBACK}>



Very true, I remember hearing of a shoot with a very famous actress and the crew ALWAYS used a rig they built specially for her (essentially very soft frontal fill from just below eye level, I believe it was) and just dimmed it when it wasn't called for, but left it there to make her feel better.

Edited by Christopher D. Keth, 22 September 2005 - 11:07 AM.

  • 0

#9 Wendell_Greene

Wendell_Greene
  • Sustaining Members
  • 545 posts
  • Other

Posted 22 September 2005 - 03:10 PM

Sometimes it is important to let the actress know that we are working very hard to make her look beautiful.

<{POST_SNAPBACK}>


And sometimes they'll let you know they appreciate it:

Jack Cardiff sits at his kitchen table surrounded by mementos of an illustrious past. Huge, exquisite photographs of Marilyn Monroe, Ava Gardner, Sophia Loren and Marlene Dietrich in their youthful prime gaze down on the diminutive cinematographer, each one autographed with messages of delirious, mischievous and occasionally heartbreaking intimacy.

Marilyn Monroe, for example, has written: "Dear Jack, If only I could be the way you have created me! I love you, Marilyn."
  • 0

#10 Dimitrios Koukas

Dimitrios Koukas
  • Sustaining Members
  • 569 posts
  • Cinematographer
  • Athens, Greece, London UK

Posted 22 September 2005 - 03:42 PM

Sometimes it is important to let the actress know that we are working very hard to make her look beautiful.

<{POST_SNAPBACK}>



he-he,
A good point there Thomas
Dimitrios Koukas
  • 0

#11 Adam Frisch FSF

Adam Frisch FSF
  • Sustaining Members
  • 2009 posts
  • Cinematographer
  • Los Angeles, USA

Posted 22 September 2005 - 06:20 PM

I'd argue that relative size to object is all that counts. Diffusion isn't even necessary to make a source soft, in fact. I've written a little piece about the theories together with David Mullen - hopefully we'll have it in the FAQ at some point.
  • 0

#12 Adam Frisch FSF

Adam Frisch FSF
  • Sustaining Members
  • 2009 posts
  • Cinematographer
  • Los Angeles, USA

Posted 22 September 2005 - 06:36 PM

Noticed that it has been posted under FAQ In Progress already.
  • 0

#13 Chris Cooke

Chris Cooke
  • Sustaining Members
  • 246 posts
  • Cinematographer
  • Lethbridge, AB Canada

Posted 22 September 2005 - 10:16 PM

I love it Adam. I've never looked at the FAQ's before but I'm really impressed. David, Laurent and you have done a great job so far. Soft light is often missunderstood. I'd recommend that everyone read this FAQ (even though it's not finished).
  • 0

#14 Glenn Hanns

Glenn Hanns
  • Sustaining Members
  • 160 posts
  • Cinematographer

Posted 22 September 2005 - 10:47 PM

The inverse square law applies only to hard light. What happens with soft light is that the light rays are going in multiple directions at once, not just in a straight line perpendicular to the source. The light is spreading up, down, sideways, and at every angle away from the soft source. So you really can't calculate the falloff in only one direction away from the source, because only some of the the light is going that direction.

<{POST_SNAPBACK}>


Sorry Michael have to disagree with you, I spoke to Peter James ASC ACS about this same issue a little while ago. All light obeys the inverse Square law. You have to take into account the POINT of the source. When you add diffusion the DIFFUSION now becomes the source of the light and such is a very large point source. Falloff does reduce with different sources (focused Vs diffused) but the energy twice as far from the source is still one quarter.
  • 0

#15 Luke Prendergast

Luke Prendergast
  • Sustaining Members
  • 491 posts
  • Cinematographer
  • Victoria Australia

Posted 23 September 2005 - 12:02 AM

'Very large point source' is a contradiction of itself. The inverse square law is for photons originating from a point, not a plane. Not a focussed beam. Not a soft source.
  • 0

#16 Glenn Hanns

Glenn Hanns
  • Sustaining Members
  • 160 posts
  • Cinematographer

Posted 23 September 2005 - 01:33 AM

'Very large point source' is a contradiction of itself. The inverse square law is for photons originating from a point, not a plane. Not a focussed beam. Not a soft source.

<{POST_SNAPBACK}>


By point source they-re referring the source of the light. Not size. The suns a whopper of a point source isnt it!
Point source as in an origin of light opposed to a reflection.
  • 0

#17 Glenn Hanns

Glenn Hanns
  • Sustaining Members
  • 160 posts
  • Cinematographer

Posted 23 September 2005 - 01:46 AM

'Very large point source' is a contradiction of itself. The inverse square law is for photons originating from a point, not a plane. Not a focussed beam. Not a soft source.

<{POST_SNAPBACK}>


From a photography mag:

All light sources* follow this rule, which is why light from a camera?s flash unit tends to drop off in intensity pretty rapidly. It also explains why you don?t necessarily gain much more flash range when you buy a moderately more powerful flash unit, and why foreground objects are much more brightly illuminated by your camera-mounted flash unit than distant objects.

* the one exception being collimated light of the type produced by lasers. Collimated light has light waves which are precisely parallel and do not spread out - and so such light does not follow the inverse square law.
  • 0

#18 Greg Gross

Greg Gross
  • Sustaining Members
  • 869 posts
  • Harrisburg,PA

Posted 23 September 2005 - 02:10 AM

If you increase the distance between the subject and a light source,the level of
light illuminating the subject decreases. When you double the distance from the
light to the subject,the amount of light falling on the subject is about 25% of its
original intensity. This is from my experience in the camera room doing portrait-
ure. Now this change of light that I'm talking about is called fall-off. This applies to
the "inverse square law". Inverse Square Law- When a surface is illuminated by a
point source of light the intensity of the light at the surface is inversely proportional
to the square of its distance from the light source. Lets say we take a meter reading
of Nicole Kidman at f16,the key is one meter from her. At two meters from Nicole
the reading should be f8 and at four meters from her the reading should be f4. Note
what is happening as the key is moved away from Nicole. It doesn't matter what the
light source is! These rules still apply and will not change. Fall-off is not a problem if
using the sun(direct sunlight) as all objects are equal distance from the sun. If you
are using window light,reflected light,artificial light then you must make some consid-
erations. Understanding light has a helluva lot to do with being a successful cinema-
tographer. Probably what you'll see with subjects at different distances from the light
source will be uneven illumination. Choose a key light,lets just say that its window
light. Take two actors and place them at one and two meters from the window. Then
try them three and four meters from the window. What happens to the brightness of
the two people as the light source becomes further away from them? It doesn't matter what intensity of light is falling on the subject,different levels of light will be
reflected from the subject,the basis for your reflected light reading. Light reflecting
from your subject is called subject reflectance. The level of this light can be influ-
enced by color,texture,angle of light to the subject. White shirt reflects more than a
black dress. A dark wooden wall will reflect less than a mirror. The level of reflected
light is directly proportional to the viewpoint of the camera. If the viewpoint of the
camera is equal to the angle of the light to the subject,the reflected light level will
be greater.The level of reflected light is determined by - reflectance of subject,in-
tensity of light source,angle of viewpoint and light to subject,distance of the light.
The cinematographer you are speaking of is simply using the laws/caracteristics of
light. If you place certain materials between the light source and the subject you
can create diffused light. The size of the light source is increased when you diffuse
it. Why? Because you spread the light over a greater area. The further the light
source from the diffusion material,the larger the light appears to be. So,what happ-
ens? Shadows are softened, shadow detail is increased and the light falling on the
subject decreases. I'm not being critical of the cinematographer and I'm sure he's
good at what he does,to be working with such a fine actress. There are probably
a hundred ways to achieve the same lighting and some of them are simpler. I was
just thinking here about flat digital lighting panels. In my camera room I have dist-
ances marked off on the floor from the subject. When the key is setting at one of
these marks I know how much light is falling on the subject. I can shoot without a
meter reading although at times I need to make a reading. at times I may want to
take an incident reading. Light from a point source is what I call hard light. Is he
lighting Nicole Kidman straight on key,facing her key light? The direction of the key
is important as it will determine where the shadows fall. I have a hunch that he is
lighting her facing her key,he knows this light and how it will fall on her. This prob-
ably is producing a beautiful soft modeling of her face. The chracteristic of this light
I believe would eliminate shadows. The wonderful art of photographing beautiful
women. I really liked what Michael Ballhaus ASC did with Diane Keaton in Some-
thing's Gotta Give.

Greg Gross
  • 0

#19 Chris Cooke

Chris Cooke
  • Sustaining Members
  • 246 posts
  • Cinematographer
  • Lethbridge, AB Canada

Posted 23 September 2005 - 10:28 AM

I believe would eliminate shadows.

<{POST_SNAPBACK}>


Depending on placement of the key, this would eliminate shadows but hopefully create some shade. I imagine that he keyed her about 20 degrees off camera axis and a couple feet above camera height. This would create shade on the fill side of the face in order to model it a little and bring out some of her features.
  • 0

#20 Dimitrios Koukas

Dimitrios Koukas
  • Sustaining Members
  • 569 posts
  • Cinematographer
  • Athens, Greece, London UK

Posted 23 September 2005 - 01:40 PM

It doesn't matter what the
light source is!
Greg Gross

<{POST_SNAPBACK}>

[/quote]

Greg,
I have to dissagree with you, Fall off and the inverse square law is excactly applied when your lights are difused or bounced.
Not for openface lights or Fresnels nor cinepars.
Do a test , u will see that the output has no connection at all with this law.
Dimitrios Koukas
  • 0


Ritter Battery

Visual Products

Tai Audio

CineLab

Wooden Camera

Media Blackout - Custom Cables and AKS

Technodolly

Paralinx LLC

Glidecam

Metropolis Post

Broadcast Solutions Inc

Rig Wheels Passport

FJS International, LLC

CineTape

Aerial Filmworks

Opal

Willys Widgets

The Slider

Abel Cine

rebotnix Technologies

Gamma Ray Digital Inc

Metropolis Post

Gamma Ray Digital Inc

Media Blackout - Custom Cables and AKS

CineLab

CineTape

Aerial Filmworks

Technodolly

The Slider

Tai Audio

Visual Products

Ritter Battery

Abel Cine

Wooden Camera

rebotnix Technologies

Paralinx LLC

Opal

Broadcast Solutions Inc

Glidecam

Rig Wheels Passport

FJS International, LLC

Willys Widgets