Thanks Michael, I already got in contact with David, this was his fantastic reply (I also asked some other things, the last paragraph e.g. has some great insight on desaturation):
"One thing to keep in mind is that pre-digital color-correction for features, you mainly had a standardized contrast (gamma) & saturation using Kodak print stock and FCP processing. Black level could be affected by printer lights (which in turn affected contrast & saturation), but this was mitigated by the fact that most prints came from dupe negatives — the extra density in the original negative had been compensated for when making the color-corrected IP/IN so release prints used a normal set of printer lights (James Cameron got around this by asking the lab to make lighter IPs & denser INs for “Titanic”.).
You could affect the release prints by altering the IP or IN — plus there were silver retention processes for prints like ENR that increased the density of the blacks and the contrast in the shadows. There was also some darkening of colors and loss of saturation from leaving silver in the print.
All of this is to say that you worked backwards from your planned release print strategy in terms of the negative.
Flashing the negative increases the base fog level, lifting the blacks and bringing up some shadow detail normally buried in the contrast of the print, especially an ENR print which I believe “Fight Club” used (or Deluxe’s version called ACE.)
In terms of color, adding a weak amount of overall white light on a negative will lift the blacks and wash-out the colors, it’s like mixing a tiny bit of white paint into colored paints.
Leaving a little amount of black silver in a print will deepen the blacks and darken the colors, slightly softening them — it’s like mixing a tiny amount of black paint into colored paints.
It’s a bit harder to visualize the opposite — flashing a print or leaving silver in the negative.
Flashing a print with white light fogs the highlights with density (normally a brighter area in the image is denser on the negative, more clear on the print) so your whites get a bit greyed, less intense, lowering contrast in the highlights. Does that affect saturation? Maybe… but only because contrast, black level and saturation are all tied together, so softening the highlights and lowering the contrast will make the colors seem a bit softer.
Leaving black silver in the negative means the highlights in the subject will get denser as if overexposed (some people compensate by underexposing the stock). So there is an increase in contrast but mostly in the highlights. There is also a larger increase in visible graininess compared to leaving silver in the print because print stock has a very low ASA so the silver grains are quite small compared to those in camera negative. Again, does leaving black silver in the negative affect saturation? Sort of, it’s a bit like having a black & white negative layered on top of your color negative.
In the prints, the colors got darker and less saturated with ENR. Since usually this approach was decided in prep, costumes, props, lighting colors, etc. were tested to see if any compensation was needed. For example, blood effects in “Sleepy Hollow” were adjusted, using a bright red blood on set so that it would not look too black due to the ENR process. I can’t tell you the specific effect on each color shade but by the early 2000s, colorists had their own digital simulation of ENR for DIs, reducing the usage of ENR itself — though in truth, it was not the same thing because leaving silver in the print itself allows the blacks to be deeper than D-MAX, the max density a print can achieve, a sort of blacker-than-black. There were problems with this though, silver in the print (just as with true b&w prints) tends to absorb infrared heat from projector bulbs, reducing the life of the print.
One issue that always comes up with a desaturation process is that visually, pastel colors seem to drop in saturation faster than primary colors, so a subtle color like in skin tones will go monochromatic faster than a bright red stop sign will. This is why it is better to control saturation first with production design and costumes so as to preserve skin tone saturation relative to everything else in the frame. Of course, with digital color-correction, one can isolate objects and adjust saturation more selectively though that can be time-consuming.
In general, it’s always better to get the look you want first by what’s in front of the camera, second by how you film it with your camera, and third by post-production."
- David Mullen, ASC.
"One thing to keep in mind is that pre-digital color-correction for features, you mainly had a standardized contrast (gamma) & saturation using Kodak print stock and FCP processing. Black level could be affected by printer lights (which in turn affected contrast & saturation), but this was mitigated by the fact that most prints came from dupe negatives — the extra density in the original negative had been compensated for when making the color-corrected IP/IN so release prints used a normal set of printer lights (James Cameron got around this by asking the lab to make lighter IPs & denser INs for “Titanic”.).
You could affect the release prints by altering the IP or IN — plus there were silver retention processes for prints like ENR that increased the density of the blacks and the contrast in the shadows. There was also some darkening of colors and loss of saturation from leaving silver in the print.
All of this is to say that you worked backwards from your planned release print strategy in terms of the negative.
Flashing the negative increases the base fog level, lifting the blacks and bringing up some shadow detail normally buried in the contrast of the print, especially an ENR print which I believe “Fight Club” used (or Deluxe’s version called ACE.)
In terms of color, adding a weak amount of overall white light on a negative will lift the blacks and wash-out the colors, it’s like mixing a tiny bit of white paint into colored paints.
Leaving a little amount of black silver in a print will deepen the blacks and darken the colors, slightly softening them — it’s like mixing a tiny amount of black paint into colored paints.
It’s a bit harder to visualize the opposite — flashing a print or leaving silver in the negative.
Flashing a print with white light fogs the highlights with density (normally a brighter area in the image is denser on the negative, more clear on the print) so your whites get a bit greyed, less intense, lowering contrast in the highlights. Does that affect saturation? Maybe… but only because contrast, black level and saturation are all tied together, so softening the highlights and lowering the contrast will make the colors seem a bit softer.
Leaving black silver in the negative means the highlights in the subject will get denser as if overexposed (some people compensate by underexposing the stock). So there is an increase in contrast but mostly in the highlights. There is also a larger increase in visible graininess compared to leaving silver in the print because print stock has a very low ASA so the silver grains are quite small compared to those in camera negative. Again, does leaving black silver in the negative affect saturation? Sort of, it’s a bit like having a black & white negative layered on top of your color negative.
In the prints, the colors got darker and less saturated with ENR. Since usually this approach was decided in prep, costumes, props, lighting colors, etc. were tested to see if any compensation was needed. For example, blood effects in “Sleepy Hollow” were adjusted, using a bright red blood on set so that it would not look too black due to the ENR process. I can’t tell you the specific effect on each color shade but by the early 2000s, colorists had their own digital simulation of ENR for DIs, reducing the usage of ENR itself — though in truth, it was not the same thing because leaving silver in the print itself allows the blacks to be deeper than D-MAX, the max density a print can achieve, a sort of blacker-than-black. There were problems with this though, silver in the print (just as with true b&w prints) tends to absorb infrared heat from projector bulbs, reducing the life of the print.
One issue that always comes up with a desaturation process is that visually, pastel colors seem to drop in saturation faster than primary colors, so a subtle color like in skin tones will go monochromatic faster than a bright red stop sign will. This is why it is better to control saturation first with production design and costumes so as to preserve skin tone saturation relative to everything else in the frame. Of course, with digital color-correction, one can isolate objects and adjust saturation more selectively though that can be time-consuming.
In general, it’s always better to get the look you want first by what’s in front of the camera, second by how you film it with your camera, and third by post-production."
- David Mullen, ASC.
