Jump to content




Photo

Tungsten vs Daylight Film Stock


  • Please log in to reply
9 replies to this topic

#1 Sebastien Scandiuzzi

Sebastien Scandiuzzi
  • Basic Members
  • PipPip
  • 35 posts
  • Cinematographer
  • Seattle

Posted 27 April 2016 - 12:57 AM

Always wondered, why is film stock daylight and tungsten balanced in the first place? Why isn't there a universal negative for any condition, similar to how our eyes/brain perceives our environment (meaning, our eyes/brain doesn't adjust for daylight and then for tungsten)? 

 

Thanks!


  • 0




#2 Brian Drysdale

Brian Drysdale
  • Basic Members
  • PipPipPipPip
  • 4743 posts
  • Cinematographer

Posted 27 April 2016 - 02:00 AM

The brain does a lot of processing, when perceiving the world it makes adjustments, corrections and assumptions all the time. There is some latancy in all this, which is why they use a starting pistol for starting the sprint rather than a light.

 

Film and video sensors collect images based on their native colour sensitivity, so if they tend towards being blue or red sensitive some from of correction needs to be put in place for other colour sources. Some video cameras have automatic colour conrectiion (acting a bit like the brain), but these can be fooled by mixture of  colour temperature sources. This means you have to put your brain into the loop by applying corrections either in/on the camera or in post.


Edited by Brian Drysdale, 27 April 2016 - 02:01 AM.

  • 0

#3 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 27 April 2016 - 09:55 AM

Our eyes adjust all the time for color temperature -- if we are standing in a space that is lit with daylight-balanced lights and in the distance we see a tungsten-lit room, we can see that the distant room is warm in comparison. Film stock can't do that sort of "smart" processing of information. A video camera can re-balance RBG levels electronically in order to adjust for different color temperatures.

You couldn't make a film stock adjust its RBG balance photochemically unless you could develop each layer individually from each other to adjust density for each color record separately.
  • 0

#4 Bruce Greene

Bruce Greene
  • Basic Members
  • PipPipPipPip
  • 393 posts
  • Cinematographer
  • Los Angeles

Posted 27 April 2016 - 10:34 AM

Film, as with digital capture, has a limited range of light that it can record.  This dynamic range is maximized when the blue range equals the red range equals the green range.  In tungsten light, tungsten balanced film roughly equalizes this and creates neutral whites and blacks.  Same with daylight film in daylight.  (or tungsten film in daylight with an #85 orange filter on the camera).

 

When the color of the light is not matched to the the film stock, one or more colors is either over or underexposed compared to the others leaving either the highlights or shadows to roll off to a tinted color.  In a chemical/optical finish, this can be difficult to hide with color timing in the lab.  And what doesn't show up in the first generation can show up later after the negative is duplicated a few times before the theatrical prints are made as some dynamic range is lost with each duplication.  This can be hidden with digital color correction, but some dynamic range will be lost as this requires clipping one or more of the colors to match the least exposed color.

 

This also holds true for digital cameras.  The last time I tested this, my camera maximized dynamic range at a color temp of around 4400K.  Shooting in daylight lost a little bit of dynamic range, and shooting in tungsten lost a bit more.  If I were really picky, I would add a slight orange filter when shooting in daylight, and a blue filter when shooting in tungsten light.  And this is what one would need to do if shooting a "one size fits all" color negative film as well.

 

In practice, many of todays digital cameras record enough dynamic range in each color to provide a good image between tungsten and daylight, even when some colors might clip before the others.  So we don't bother with the colored filters.  When shooting film, for the most adaptable (color wise) result, it's best to choose the film that matches best the color of the light that one is using.


  • 1

#5 Sebastien Scandiuzzi

Sebastien Scandiuzzi
  • Basic Members
  • PipPip
  • 35 posts
  • Cinematographer
  • Seattle

Posted 27 April 2016 - 01:30 PM

Thanks Bruce! And everyone else, really appreciate the detailed answers. Very interesting that dynamic range is tied to color temperature as well with digital sensors. 

 

I'm oversimplifying but it seems to me that the reason for different color temperature film stocks is to mimic how our brains interpret daylight and tungsten lighting conditions versus sensors/film inability to create a universal negative for all conditions? 

 

Going down the rabbit hole here but, regarding our brains making adjustments depending on lighting conditions, it would be interesting to see if there is a difference when looking at a daylight picture inside a tungsten environment and then looking at the same picture outside in daylight. If our brains make adjustments depending on the environment we should be able to 'see' the difference, a shift in color temperature when viewing the picture?  


  • 0

#6 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 27 April 2016 - 01:47 PM

You're oversimplying way too much, you can't design a film stock to mimic human brain processing (nothing at that level of complexity, but the image can be tweaked to look "natural" in terms of reproduction). There is no electronic image processing in a film stock except later in post.

(You should do some reading on subtractive and additive color...)

We can create a full color image from having the red, green, and blue information of the subject -- both film stock and digital sensors have basically monochrome light receptors (silver halide crystals in the case of film) with either red, green, or blue filters in front of them to limit that receptor to that narrow band of wavelengths.

I'm not sure why the concept that a stock is balanced for a particular color temperature seems odd to you, how would you magically make a stock change its own response to color to suit the environment? Other than with color filters on the camera... How could you make a stock less sensitive to blue wavelengths in daylight and suddenly more sensitive to blue wavelengths in tungsten light?
  • 1

#7 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 27 April 2016 - 02:03 PM

And even if you could make a stock that corrected for any overall color temperature bias it detected, it would be like a video camera constantly white balancing the frame. What if the actor walks from an orange-lit campfire area into a field lit blue for moonlight? How would the stock know the difference between a day interior you wanted to be neutral to a nightclub lit with daylight units for a blue color? How would you stop the stock from taking away the orange bias from a sunset exterior scene, thinking it was a tungsten interior? The constant adjusting for color, contrast, dynamic range, etc. that our eye-brain does could look very distracting if reproduced during in a dramatic scene because it wouldn't be the viewer's brain making those adjustments.
  • 1

#8 Sebastien Scandiuzzi

Sebastien Scandiuzzi
  • Basic Members
  • PipPip
  • 35 posts
  • Cinematographer
  • Seattle

Posted 27 April 2016 - 07:10 PM

You're oversimplying way too much, you can't design a film stock to mimic human brain processing (nothing at that level of complexity, but the image can be tweaked to look "natural" in terms of reproduction). There is no electronic image processing in a film stock except later in post.

(You should do some reading on subtractive and additive color...)

We can create a full color image from having the red, green, and blue information of the subject -- both film stock and digital sensors have basically monochrome light receptors (silver halide crystals in the case of film) with either red, green, or blue filters in front of them to limit that receptor to that narrow band of wavelengths.

I'm not sure why the concept that a stock is balanced for a particular color temperature seems odd to you, how would you magically make a stock change its own response to color to suit the environment? Other than with color filters on the camera... How could you make a stock less sensitive to blue wavelengths in daylight and suddenly more sensitive to blue wavelengths in tungsten light?

 

Thanks David! Really helpful. 

 

I think I may have not been clear. I'm not mystified how tungsten or daylight film stock work or why we choose them, I was curious as to why there were 2 to begin with. 'Why' more in terms of the brain's physiological response to light/color in comparison to the photochemical response of film to color. I was under the assumption that humans brains 'see' the world uniformly under any condition; that the brain doesn't make slight color corrections depending on the ambient light (and I'm not sure this isn't the case either, color perception isn't fully understood but I think that is for a different forum). But from what you wrote, our brains do make these adjustments and in order for a viewing audience to 'see' an indoor tungsten lit room 'correctly' one needs to use tungsten balanced film. Or you can go the other way as you did with Northfork to create a dramatic cooler look. 

 

Regarding how a universal stock would detect different lighting conditions and make adjustments, I have no idea. As I wrote above, I assumed our brains saw the world 'as it really is' and figured film stocks couldn't replicate both daylight and tungsten 'correctly' with one stock and therefor needed 2. 

 

Apologies for the confusion!


  • 0

#9 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 27 April 2016 - 08:03 PM

Our eye-brains constantly correct color balance -- have you ever worn a pair of yellowish sunglasses outside for a few hours and then taken them off? For a moment, everything seems really blue outdoors until your eye-brain adjusts again.

 

One thing that our eyes are less sensitive to are spikes in a non-continuous spectrum source -- we can see the green in a standard fluorescent tube or CFL, but it's not as strong as film stock or a digital camera might see that green spike.

 

And I think intellectually we still sense that a certain source has a color bias even as our eye-brains are correcting most of that bias -- like when in a candlelit room or by a campfire, we recognize a certain orange bias but it is not as strong as a camera might see it.

 

However, filmmaking is an art so often we take advantage of how a camera or film stock renders colors for artistic effect.

 

With the latitude of color negative stock today, Kodak could easily just make the stocks halfway between tungsten and daylight, like balanced for 4500K for example, though the blue layer would be a little grainier than it is in daylight-balanced stocks.  

 

Kodak did make a consumer Super-8 Ektachrome reversal stock "Type-G" that was balanced between the tungsten and daylight, for people who couldn't figure out how to use the 85 filter inside their Super-8 cameras, or had lost the key that switched it, but since reversal stocks were meant for direct projection, there was no color-correction step so you had to live with tungsten looking a little orange and daylight looking a little blue.  I used to play with that stock for creative effect now and then.

 

Traditionally most still color film has been daylight-balanced since it was often shot outdoors or indoors with a daylight-balanced flash, but most movie film has been tungsten-balanced since most movie sets were lit with tungsten units and the correction filter to use the stock in daylight, the 85B, only loses 2/3's of a stop -- whereas the correction filter to use daylight stock under tungsten lighting, the 80A, loses 2-stops of light, and you are more likely to need the extra speed indoors than outdoors in daylight.

 

Early 3-strip Technicolor and the first Kodak color negative stock were daylight-balanced, partly because they were so slow in sensitivity that sets had to be lit with powerful carbon arc lamps, which are usually daylight-balanced.  But Technicolor became tungsten-balanced by the late-1940's and Kodak's second color negative stock was tungsten-balanced (a 16D stock in 1950 was replaced by a 25T stock in 1952).


  • 1

#10 Sebastien Scandiuzzi

Sebastien Scandiuzzi
  • Basic Members
  • PipPip
  • 35 posts
  • Cinematographer
  • Seattle

Posted 29 April 2016 - 01:06 AM

Thanks

 

Our eye-brains constantly correct color balance -- have you ever worn a pair of yellowish sunglasses outside for a few hours and then taken them off? For a moment, everything seems really blue outdoors until your eye-brain adjusts again.

 

One thing that our eyes are less sensitive to are spikes in a non-continuous spectrum source -- we can see the green in a standard fluorescent tube or CFL, but it's not as strong as film stock or a digital camera might see that green spike.

 

And I think intellectually we still sense that a certain source has a color bias even as our eye-brains are correcting most of that bias -- like when in a candlelit room or by a campfire, we recognize a certain orange bias but it is not as strong as a camera might see it.

 

However, filmmaking is an art so often we take advantage of how a camera or film stock renders colors for artistic effect.

 

With the latitude of color negative stock today, Kodak could easily just make the stocks halfway between tungsten and daylight, like balanced for 4500K for example, though the blue layer would be a little grainier than it is in daylight-balanced stocks.  

 

Kodak did make a consumer Super-8 Ektachrome reversal stock "Type-G" that was balanced between the tungsten and daylight, for people who couldn't figure out how to use the 85 filter inside their Super-8 cameras, or had lost the key that switched it, but since reversal stocks were meant for direct projection, there was no color-correction step so you had to live with tungsten looking a little orange and daylight looking a little blue.  I used to play with that stock for creative effect now and then.

 

Traditionally most still color film has been daylight-balanced since it was often shot outdoors or indoors with a daylight-balanced flash, but most movie film has been tungsten-balanced since most movie sets were lit with tungsten units and the correction filter to use the stock in daylight, the 85B, only loses 2/3's of a stop -- whereas the correction filter to use daylight stock under tungsten lighting, the 80A, loses 2-stops of light, and you are more likely to need the extra speed indoors than outdoors in daylight.

 

Early 3-strip Technicolor and the first Kodak color negative stock were daylight-balanced, partly because they were so slow in sensitivity that sets had to be lit with powerful carbon arc lamps, which are usually daylight-balanced.  But Technicolor became tungsten-balanced by the late-1940's and Kodak's second color negative stock was tungsten-balanced (a 16D stock in 1950 was replaced by a 25T stock in 1952).

Thanks David, always a pleasure to read your posts and even more so when its a response to one's own question. You never, as far as I know, answer posts "yes" or "no" and I appreciate the extra time you put in when answering. Plus, I always learn something new- couldn't ask for much more then that!!

 

Sebastien


  • 0


Tai Audio

Abel Cine

Paralinx LLC

Pro 8mm

Aerial Filmworks

rebotnix Technologies

CineTape

CineLab

The Slider

Broadcast Solutions Inc

Ritter Battery

Zylight

Visual Products

Technodolly

Glidecam

Willys Widgets

Rig Wheels Passport

rebotnix Technologies

Technodolly

Visual Products

CineTape

Paralinx LLC

Zylight

Willys Widgets

The Slider

Ritter Battery

Abel Cine

CineLab

Tai Audio

Pro 8mm

Broadcast Solutions Inc

Rig Wheels Passport

Glidecam

Aerial Filmworks