Question for Mr Mullen about Twin Falls Idaho

[F Bulgarelli]

Hello Mr Mullen,

 

I just watched Twin Falls Idaho and I have to say I really liked the look of the film.

You probably have already written about this film but I wanted to ask you a few questions, if you don't mind.

What stock did you use? and what was your approach to exposure and processing.

I really like the variety of approaches in the lighting as far as contrast ratios and the use of hard and soft light, there is certain softness in the overall lighting that works perfectly for the subject matter, can you talk about your lighting approach?

 

Thanks a lot,

 

Francisco

[David Mullen ASC]

here are parts from an early draft of an article I wrote about it:

 

We discussed the look of the film many times during the long period when the Polish Brothers were looking for financing. The dominant thought that Michael had was that it should be painterly and static, with an emphasis on natural light. He had in mind the works of Vermeer. Then we thought of Edward Hopper, because of his modern images of urban alienation, of loneliness, and his use of natural and artificial light sources. We also discussed Edvard Munch and his psychological use of colors ? there was this sort of sickly green that he sometimes used for the color of background walls that inspired us. Green has many ?meanings? from decay to disease, to repose and rejuvenation. We felt that the hotel should have this odd, timeless quality and that certain shades of green would help create this strange tone that was neither positive or negative. Green also has the advantage of neither being a ?warm? or ?cold? color but can become either in certain lighting, allowing us to change the mood of the room by altering the color bias of the lighting as the story dictated.

 

At some point, we also discussed whether the film should actually be shot in black and white, which we had used in our first project together. There was a certain ?Eraserhead? tone to the project and black and white photography would help create a non-descript time period for the story, that timeless quality that we were seeking. There were a number of downsides to shooting in black and white, beyond the inevitable resistance from any future distributor. For one thing, the use of color symbolism would have been impossible; there was also the fear that the project was already ?artsy? enough and that black and white was too obvious a choice.

 

The solution was to shoot in color but desaturate the image enough that it had some of the mood and tonality of black and white photography. Art direction would be important in accomplishing this, but we also decided to test the skip-bleach (or bleach-bypass) method of printing in combination with flashing the negative.

 

--

 

When the project finally got funding, we shot extensive tests of the Fuji film stocks that we would be using, in combination with flashing (using the Panaflasher) and skip-bleach processing. I tested a variety of lighting contrast ratios, colored lighting and colored flashing, flash percentages, etc. We decided to use Deluxe Lab?s CCE process for the printing, which is a slight variation of straight skip-bleach processing.

 

The effect of negative flashing was: (1) a decrease in apparent contrast, mostly due to the fogging of the blacks; (2) a very slight increase in shadow detail; (3) a decrease in color saturation due to the mixing of white light into the colors (I tend to think in terms of painting); and (4) a slight softening of apparent definition, again because of the lowering of contrast rather than a true loss of resolution.

 

The effect of the CCE printing was: (1) a large increase in contrast, mostly due to an increase in the density of the blacks; (2) a decrease in shadow detail for the same reason; (3) a slight decrease in color saturation due to black silver being left in the colors, also making them darker in tone; (4) an increase in apparent defintion due to the increase in contrast; and (5) a slightly grainy quality due to the retained silver ? but the image had the gritty, silvery quality of a black and white print, so it was not as objectionable to my eyes as normal color dye grain.

 

Deluxe Labs also has another silver retention process called ?ACE?, similar to Technicolor Lab?s ENR process, which allows for variable degrees of silver to be left in the print. We only tested the CCE process, feeling that if it ended up being too strong in the later final product, we could always switch to the ACE process and control the degree of silver retention to our liking. We wanted to see the strongest possible effect for our tests.

 

You can see from the above descriptions that the silver retention process had almost the opposite effect of negative flashing, effectively canceling each other out except in one critical area: color saturation. Since both approaches lowered color saturation in different ways, when combined the final result was noticeably desaturated ? with the more pastel colors like in fleshtones being visibly effected faster than the more saturated colors in the frame. The slightly monochromatic image also had some of the silvery, gritty quality that one associates with black and white prints, which was another look that we wanted.

 

Having never used flashing before, I wanted to play it safe; from the tests, it looked like a 15% flash was the best level to use when combined with the CCE printing. Any flashing higher than 20 to 25% started to visibly lighten the blacks even with all the silver retained using CCE.

 

I decided to shoot the majority of the project on Fuji?s medium speed film stocks, 8551 (F-250) and 8561 (F-250D) ? I didn?t want any graininess from the negative stock to add to the silvery grain that would be visible in the print. I could have just used the tungsten-balanced 8551 for all my interiors, but I had decided that it would look more interesting to shoot the day scenes earlier in the story using tungsten-balanced stock in uncorrected daylight, then switch later to daylight-balanced stock (8561) maybe even using coral filters to increase the warmth of the image. I felt that this approach of having an overly blue negative in the early scenes and a warmer image later on the negative would have the effect of making the fleshtones more desaturated, monochromatic, and cold at first, due to the lack of density in the red end of the spectrum on the negative. Later in the film, we would have something closer to normal saturation in the fleshtones, with maybe a corresponding loss of saturation in the blue colors.

 

The only variation in this overall arc from cold to warm occurs near the end of the movie: the twins end up in a hospital, which I would be lighting in very blue tones, knowing that when combined with the CCE printing would produce more of a blue-grey tone rather than a super-saturated blue color. Then one of the twins dies in the hospital and the films cuts to a dream sequence shot in Super8 Kodachrome, which would obviously be very different in tone and saturation.

Finally, the last scenes of the film, almost an epilogue, would be edited onto its own projection reel so that I could process the print normally, with no special silver retention process. The color saturation would return to ?normal?, coinciding with a change in story location from the city to the countryside. For these scenes, I planned on using the new Fuji F-64D (8522) which I had tested so many months ago.

 

My lighting package during production included a 12K HMI Fresnel, a 4K HMI PAR, and a 4K Xenon as my large units for creating daylight effects on the location; this was augmented by a selection of KinoFlo fixtures, including the 10-tube Wall-O-Light. We also carried a variety of smaller tungsten units for the night interior scenes.

 

We began our shoot at Lacy Street Studios; a former textile factory, its rooms have been converted into shooting spaces ? but with real windows facing a street and an alleyway. We had to treat the rooms as if we were on a real location, combing natural daylight with my own HMI?s and KinoFlos (as opposed to being on a real soundstage with an all-tungsten package for recreating daylight on sets.)

 

The first days of production were spent shooting the hotel room scenes. We began in the small bathroom set; since the scenes here were so short, I thought that it would be a good chance to do something visually bold without it becoming annoying. I wanted to ?sell? the green color symbolism by bathing the room in green light. I had the art department dress the single window with a green plastic curtain, outside, I covered the window with diffusion paper and green gel. The room became flooded with a very soft but saturated green light (I knew that the color would be muted a little by the later skip-bleach printing). It was a rather strong artistic choice to make considering that our first dailies shown to the producers would contain all this deep green footage! When shooting the close-ups, I cheated the angle of the light by blacking-out the window and using the KinoFlo Wall-O-Light instead with green gel on it (luckily this light, which produced a nice, large softlight effect, took up very little space in the small bathroom space.) For a night scene in the bathroom, we hung a 3?x3? softbox from the ceiling with tungsten bulbs. We also put it on a flicker box so that we could simulate the effect of a randomly flickering fluorescent fixture. All of this footage was shot on Fuji F-250T (8551) using the Panaflasher set for a 15% flash ? this combination of stock & flashing was probably the most common approach that we used throughout the shoot.

 

As we made the film, I tried to stick to more ?neutral? focal-length lenses, working mostly in the 35mm to 75mm range, to avoid distortion and maintain a classical, painterly quality to the compositions. For a few shots, we went to a 27mm Primo when needed to get a wide shot in the small rooms, or when we wanted to make a dolly move a little more dramatic. However, I suspect that over 50% of the feature was shot on the 75mm Primo. We also didn?t carry a zoom lens on this show; I rarely use them anyway and I needed to shave some costs from our camera package rental. Michael didn?t see any place for zooming in the film anyway. (I?ve always felt that working within a limited range of prime lenses gives a movie a visual consistency and a sense of precision.)

 

I used Kinoflos quite a bit during production as my main source of light. We were filming during a summer heat wave in small rooms with poor ventilation so using fluorescents helped keep the temperature down. For one love scene, I wanted a soft backlight over the bed; I had first considered Chinese lanterns, but these are so hard to flag off the walls. I ended up hanging 4-bank KinoFlos off of a rail that was cantilevered over the bed. We could easily raise or lower the light until it was just above the frameline, creating a very nice backlight over the bed with minimal spill.

 

For some day scenes, I used a technique of putting a frame of diffusion in front of an HMI and letting some hard light leak out by raising the frame ? this created a slash of hard, overexposed light that felt like sunlight on the lower half of the subject, with soft light falling on the upper half. Sometimes I would clip a piece of CTO gel to the bottom of the diffusion frame so that the hard light was warmer than the soft light (just as sunlight is warmer than skylight.) For other scenes, the 4K Xenon, bounced into a mirror board, was my ?sunlight? falling through the window. In one scene, I only used griflons and reflectors outside the window, bouncing real sunlight, without using much electrical lighting inside at all.

 

When we got to the hospital scenes near the end of the film, I switched to Fuji F-500 (8571). This was partially for practical reasons (I didn?t know until the last minute if I would be shooting in a real hospital or a set and had to be prepared for some low light levels) but I wasn?t adverse to having these scenes look a little grittier than the rest of the movie. I rated the stock at 320 ASA, however, instead of 500 (I also rated the 250 ASA stocks at 160 ASA ? I like a dense negative.) I lit most of the scenes with Kino55 tubes, letting them go blue on the tungsten-balanced stock. This gave the hospital a very antiseptic, cold look and really desaturated the fleshtones. I also used less fill light to give everything a starker quality.

 

A couple of weeks after the film had wrapped, a small group of us went out to an abandoned cliffside road over the ocean to shoot the dream sequence which appears right after the twins are wheeled into surgery at the hospital. In the dream, we see the twins, no longer conjoined, riding on bicycles until at one point, they take separate paths and separate forever. We shot this on Super8 Kodachrome 64; the weather was rather foggy with a hot, hazy background (this pleased Michael, since the scene is labeled ?overcast? in the script.) Later we transferred this footage to 35mm by refilming the image as it was being projected on a white screen (with no sync box ? we just let the image pulse a little.) The burned-out background really gave the images a surreal quality. I had sort of guessed when shooting with the Super8 camera where the 1.85 frame would be; luckily since we were re-photographing everything, I could adjust the framing for the correct headroom.

 

When we started answer-printing the film, we tested the first reel using the CCE process and started thinking that the contrast was now too high. Bob McMillan, our timer at Deluxe, suggested we switch to the ACE process, which even at full strength (100%) was a less strong effect than CCE. We tried that and were very happy with the results. The desaturation was not as extreme as I had wanted originally; I think the contrast problem I was having was the result of my overexposing the negatives ? as I increased the overexposure, I was weakening the effect of the 15% flash, which I probably should have raised to 20% given the way I was rating the negative stock. But as I said before, I felt safer playing things on the side of too little flashing (and hence higher contrast and deeper blacks) rather than too much. Anyway, the final ACE print looks great; when the last reel comes up (with no ACE), the change in effect was very subtle, almost like ?reality? had come back into the story, with very natural colors and contrast ? while the ACE reels have a brooding, dramatic quality with extremely rich blacks and darker tones overall.

[F Bulgarelli]

Whao! This is more info that anyone could ask for. Thank you.

 

One of the things I really like about your work in this film is the use of color, there is a certain structure to the color scheme that keeps things cohesive and interesting at the same time. After reading your article I can see how a great deal of preparation went into the work.

I understand what you are saying about how flashing and silver retention cancel each other out since they are almost the opposite of each other in a sense, how does the CCE alone (without the flashing) would have looked, based on your testing.

 

There is a scene I particularly enjoyed. It is toward the beginning of the film when she is giving one of them certain medicine, then there is a lapse of time emphasized really well by the hard sunlight coming through the window and edging the actress.

 

Nice work!

 

Francisco

[David Mullen ASC]

I understand what you are saying about how flashing and silver retention cancel each other out since they are almost the opposite of each other in a sense, how does the CCE alone (without the flashing) would have looked, based on your testing.

<{POST_SNAPBACK}>

Well, they both cause desaturation -- they have opposite effects on contrast and black level, so CCE alone, which is close to a full skip-bleach to a print, causes a big increase in contrast with very deep blacks, with some desaturation.

[Jeremy]

I originally saw this film in the theater and thought it was gorgeous. Thanks so much for all the information!

 

Jeremy

http://www.throughaglass.com

[Jonathan Bowerbank]

I was shot over to this old thread from another recent post about Twin Falls Idaho.

 

But I just had to mention one shot that has stuck in my mind. It's the one where one of the Polish brothers is looking down the hall through the peephole in his door, and there's this smooth, symmetrical and slow dolly shot towards the door that along with the music is just the most memorable moment in the film for me.

 

I can't give you enough kudos, David :)

[David Mullen ASC]

I seem to recall it was a dolly away from his face in the door peephole looking out. I wanted to dolly away down the hall but had to stop short because the set wall to the right of the door was missing, since we pulled it in order to shoot the bathroom scenes, and didn't have time to put it back (remember, this movie was shot in 17 days.)

[Jonathan Bowerbank]

You're probably right...it's been a while, but still that framing has stuck in my mind :)

[Christophe Collette]

After reading this thread this morning, I picked up a copy of the movie. I just finished watching it. It is really beautiful David, really... Fantastic work. The paleness of the skin tones, I loved it, it looks at times like you added soft blue gels to light the girl's face.... Is that possible or is it just your uncorrected tungsten balanced film?

 

I am tipping my hat to you David!

 

Christophe

[David Mullen ASC]

A lot of the "cold" scenes involved uncorrected daylight on tungsten stock (usually Fuji F-250T), with half the blue timed out in post, half left in.

[Frank Barrera]

I just saw Twin Falls Idaho for the first time. Beautiful work.

 

David, you discuss the rather complex approach to the release print but how do you feel the DVD reflects the work from big screen to little? I am trying to place all the information you provided about the print and translate it to my little tv. Are you happy with the transfer?

 

thanks

f

[David Mullen ASC]

I supervised the transfer a long time ago, to D1 NTSC and PAL, not HD. It was a Rank-Cintel transfer from the IP.

 

It looks fairly close to the theatrical print look, maybe a little more saturated.

 

Originally my show prints made for Sundance looked better, using 100 IRE level of ACE on the old Kodak print stock, but by the time the movie was released, the print stock was replaced by Vision 2383 and I had to use less ACE level to match, which was disappointing because I didn't get quite the same steely silver look in the colors.

 

If I did the transfer again today, I might pull back a little more on the saturation.

[Felipe Perez-Burchard]

Once again, Mr. Mullen, your generosity is vastly appreciated.

 

I saw this film some time ago and enjoyed it and thought it looked great.

I was curious; being an independent film shot in so little time (it amazes me that it was only 17 days) with probabaly not the biggest budget, was it a concern to use the ACE or CCE process for the producers, and where you allowed to have all the prints treated this way?

Or maybe a better question, is, was the ACE applied to the Master Positive and that way the Dupe neg has the effect and subsequentlly all the prints or is it done in the last printing step in the chain?

 

Thank you for the insight.

 

Best,

[David Mullen ASC]

I just got lucky that when a distributor bought the movie (Sony Pictures Classics in the case of "Twin Falls Idaho" and Paramount Classics in the case of "Northfork") they agreed to make ALL the release prints using the silver retention process.

 

In the case of "Twin Falls Idaho" it would have been possible to just use the higher-contrast Vision Premier 2393 print stock instead of ACE, to match the black levels -- but I didn't have to do that. And I would have gotten more saturation than I wanted.

 

In the case of "Northfork" the photographic approach was much more aggressive -- heavier flashing, diffusion, smoke, etc. -- so that it could only be printed with some sort of special process, otherwise it would just look washed-out and milky. So if the distributor wasn't going to approve of the skip-bleach print process, I was going to have to create the look in the IP. But again, when they bought the film, they promised it would be printed as we intended.

 

With the Polish Brothers, we have (so far) been able to shoot and post a movie more or less however we thought best, no matter what the budget was. This is mainly because they fight for it, with producers, with distributors, etc. They care a lot about the cinematography of their movies and they are willing to push more of the budget in that direction if necessary.

 

Now on our first studio-financed movie, we did have some heavy discussions with the studio about the format and I don't know what they would have said if I decided I wanted to flash everything, etc. They tried to talk me out of anamorphic and use Super-35 instead, but without a promise to do the blow-up digitally, and I wasn't going to allow myself to be stuck doing an optical blow-up, so we decided to use anamorphic just in case we did a photo-chemical post and release.

 

I was willing to use 3-perf so that I could do a D.I., but the studio didn't want me to shoot 3-perf because they didn't want to be committed to a D.I. -- and I wasn't going to use 4-perf Super-35 and be committed to an optical printer blow-up, so I stuck to what I knew worked best no matter how we posted the movie, anamorphic. And then later in post, the Polish Brothers got them to give us the D.I. and it turned out that anamorphic worked well for that too.

 

I also had to make a case for shooting on Fuji film with the studio and possibly was going to be arguing with them regarding the print stock (I needed a contrasty print stock to compensate for the Fuji Eterna lower-contrast look), but because we ended up doing a D.I., the normal Vision print stock worked fine for the look we wanted.

[Israel Yang]

Thank you for your generous sharing, I haven't seen the movie but now I will when I come by it in the future.