Jump to content


Photo

Jennifer's Body


  • Please log in to reply
143 replies to this topic

#1 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 08 March 2008 - 06:17 PM

This is the week before the last week of pre-production, and things are really busy. Next week consists of a day of wardrobe, hair & make-up tests, then two days of tech scouting with all the department heads and their assistants (I think it will be around 40 people total on these scouts) and then the production meeting.

This is a high-school horror-drama-comedy written by Diablo Cody (Juno) and produced by the Juno team for Fox Atomic. The director is Karyn Kusama (Girlfight, Eon Flux).

It's interesting to note that before I interviewed for this job, I had just interviewed for a feature version of the "Ramona the Pest" children's books, to be directed by another woman, Elizabeth Allen (Aquamarine). But literally just as I was finishing that job interview, the director got a call from the studio that they were putting the project on hold, just after greenlighting it the week before and rushing it into production.

This was as the strike was going on and there was no word from "Big Love". So I went back to looking for projects the next week. I had gotten the script for "Jennifer's Body" but was told by my agent that they had already hired a DP. Then after the "Ramona" interview, my agent told me that they wanted to interview me afterall for "Jennifer's Body". By coincidence, I ran into Patrick Cady around that time, who shot "Girlfight" for Karyn, and I asked him if he was up for this project. He said he had interviewed for it, but before the studio could make a decision, he was offered the job of shooting a 20-mil. remake of "The Stepfather" for Sony by an associate he knew.

Anyway, it seems they hired Sam McCurdy, who had shot "The Descent" (and also "The Hills Have Eyes 2" for Fox Atomic) but due to a family emergency, he had to exit the project, opening the door for me to get hired. And it seems that even though the "Juno" producers were concerned that Fox Atomic might want another "horror DP", it turned out that they were familiar with my work on the Polish Brothers movies and had no objections.

So there was some time lost in prep and I arrived in Vancouver, BC with just five weeks of prep before shooting. Now five weeks may seem like plenty to some folks, but it's barely enough really to thoroughly plan an entire feature shot-by-shot. Partly because huge chunks of my time were devoted to scouting for locations in a van, but also because the last two weeks of any feature prep are just insane for the director, due to the cast arriving, etc. So my time with the director really is the three weeks before the last two weeks of prep, and as I said, a lot of that time was spent scouting.

The director, Karyn, had already storyboarded three or four major sequences before I was hired, so I concentrated first on the big sequences that hadn't been storyboarded. Finishing those, now we're getting through all the little scenes between the big scenes. We have about two weeks of shotlists finished plus all the major scenes storyboarded; our goal is to shotlist the entire movie, though it probably will require working together on our weekends off during the shoot.

Like I said in another post, the 1.85 format was chosen before I was hired and I didn't really disagree with that. For one thing, the director wants a lot of low-angle shots where we see the dark trees and sky in the background so a more vertical frame makes sense. And also, I haven't done a 1.85 feature in awhile (though "Big Love" was 1.78) so it seemed like a good opportunity to work in that format again. And I get to shoot in 3-perf Super-1.85, which is nice.

Since so much of this movie is at night, I decided to shoot most of this on the new Kodak 5219 500T stock. I shot my stock tests yesterday (5219, 5217, and 5212) and will see them next week.

We are shooting Panavision, with Primo lenses. Technicolor Vancouver labs. Dailies will be transferred to 4:4:4 HDCAM-SR (a first for me) and SD. Technicolor uses a SANS to temporarily store dailies transfers on and then does all sound-syncing and then downconversions to different tape formats from that.

I have a couple of shots where I'm planning on using the Phantom HD to be able to shoot at 1,000 fps -- I tested that on Friday as well.

My Gaffer is John Dekker and my Key Grip is David Askey, both "Juno" veterens, as is the operator John Clothier (and the AD staff, and many others.) I interviewed a bunch of people for these positions though, so it wasn't like I was asked to hire back the same people. DP Karl Herrmann, who worked on "Kyle XY" I believe, plus 2nd unit on "Into the West", has agreed to be B-Camera operator, so it will be great having a talented DP on the camera crew to pick up shots.

Stephen Maier will be A-Camera 1st AC. He's one of the top local guys here.

So I'm excited about the crew we've put together.

We start out the first week in a high school, shooting perhaps some of the most straightforward stuff, walking & talking in hallways, classrooms, etc. The school is somewhat windowless and fluorescent-lit unfortunately (compared to the wonderful old-world high school I shot in New Jersey for "Assassination of a High School President".) To help me out, the production designer Arv Grewal ( http://www.imdb.com/name/nm0340490/ ) -- a very talented man, we've had many fascinating discussions on art and movies -- will put a window flat plug at the ends of two hallways so I can create that daytime backlit sheen on the lockers that works so well in high school hallways.

The science classroom is tricky too because it has two little windows at each end but nothing in the middle stretch of wall, so it's hard not to end up using the overhead lights. I'm hoping to create a fake windowlight effect in the middle of the room when I'm not looking at that wall.

The basic look is what I would call "moody naturalism" -- hopefully the gloomy Vancouver weather will help in that regards, just as it did in "Jennifer Eight" for Conrad Hall. But with so much night exterior work in this movie, on deserted roads in the woods, etc. obviously I have to plan on artificially lighting those scenes in some manner. I feel that the setting, a small town in Minnesota surrounded by the forest, lends itself to something of a Grimm's fairy tale feeling at night, at least a hint of that. Creepy moonlit, etc.

We're also using the photos of Todd Hido as a reference:
http://www.toddhido.com/

That's all for now. I don't know how much I'll be able to reveal about the production, so my posts may be rather technical and/or non-descript sometimes.

So the week after I started, the WGA strike ended, and my phone started ringing with work, including "Big Love" and a new Polish Brothers movie. So I may be going on to shooting something immediately once I finish work on this movie in early-mid May.
  • 0

#2 Brian Rose

Brian Rose
  • Basic Members
  • PipPipPipPip
  • 896 posts
  • Student
  • Kansas City area

Posted 08 March 2008 - 07:53 PM

Man, I'm so jealous. If I EVER get to DP a film like that with that kind of talent, I could die I happy man!

Brian R.
  • 0

#3 Kevin Zanit

Kevin Zanit
  • Sustaining Members
  • 1223 posts
  • Cinematographer
  • LA

Posted 08 March 2008 - 09:36 PM

Good luck with the shoot, sounds like you are off to a great start with a top notch crew!

If you have any questions about the Phantom, you are on the right forum, I know at the very least I have shot a lot with the camera, Elhanan Matos is on here, one of the best techs and of course Mitch from Abel is on here as well.

Kevin Zanit
  • 0

#4 Eric Steelberg ASC

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

Posted 08 March 2008 - 09:48 PM

Good luck and enjoy my crew! They were all wonderful. You're in very good and capable hands with Dave and John. Clothier has a deft touch with the camera. Please tell them Hi for me. I believe Steve Maier also came out a day or two. Very good.
  • 0

#5 Dan Goulder

Dan Goulder
  • Sustaining Members
  • 1259 posts
  • Cinematographer

Posted 08 March 2008 - 10:01 PM

We are shooting Panavision, with Primo lenses. Technicolor Vancouver labs. Dailies will be transferred to 4:4:4 HDCAM-SR (a first for me) and SD. Technicolor uses a SANS to temporarily store dailies transfers on and then does all sound-syncing and then downconversions to different tape formats from that.

Do you know at this point if the HDCAM SR transfers may be used for the film out master, or are the plans finalized for a 2K DI (or optical) finish? Thanks. (and good luck with the project)
  • 0

#6 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 08 March 2008 - 10:22 PM

Do you know at this point if the HDCAM SR transfers may be used for the film out master, or are the plans finalized for a 2K DI (or optical) finish? Thanks. (and good luck with the project)


That was something I brought up to them because of what I'm going through right now on "Assassination of a High School President". On that film, the movie got into Sundance and they spent the money (not in the budget) to transfer all the negative to HD for the screening. They asked me and the director of what we thought about doing the D.I. using HDCAM-SR and based on a poor comparison test, we told them that we didn't want to and that they should just transfer to 4:2:2 for the screening.

But they transferred to 4:4:4 instead and now they are pretty much forcing me to use that HDCAM-SR tape for the final D.I. To top things off, the whole thing was transferred in LIN, not LOG (have no idea what they mean by "linear" though in this case) and in Rec 709, but the post house doing the D.I. keeps telling me that it doesn't make any difference if you do a D.I. in Rec 709 or LIN instead of LOG. They tell me this is how they did the D.I. on their last project for the same company, shot on the Genesis. This is why I'm not sure what they mean by "linear" because they said that the Genesis movie was also not shot in LOG. I said "but the Genesis uses Panalog" and they said "yeah, but that's not really LOG." They tell me "we just do a LIN to LOG conversion for the film-out and it looks fine."

What I find odd is that the people doing the D.I. (never heard of the company) seem to think that there's no reason to not use Rec 709 color space or linear gamma for D.I. work. That strikes me as being very "broadcast video post" centric.

Anyway, it's all very depressing because, though there isn't much difference from the 2K version that this post house did as a test (in Rec 709 they tell me), I can still see some extra noise in the whites that isn't there in the 2K version. But I can't seem to convince them that a little extra quality is worth spending the money on scanning the movie at 2K. Apparently EFILM took a shot at doing a test using the HDCAM-SR transfer vs. a 2K scan and the HD version didn't look very good. And they didn't show me that test. So now the distributor has found this new post house that made a test where the HDCAM-SR and 2K version look much closer, but as I told the director, how do we know that the 2K version is the best that it can be? How do we know that the 2K wasn't just brought closer to the quality of the HDCAM-SR instead of the other way around?

But it's all pointless because the D.I. is almost finished now. The whole experience almost made me want to swear off doing D.I.'s for awhile. It's too bad because "Shadowboxer" also went through a low-budget HDCAM-SR D.I. and both that and "Assassination" have some of my nicest work. It was worse with "Shadowboxer" though because that was an anamorphic movie so a 2K or 4K scan would have had more vertical resolution. But on "Assassination", since it was shot in 3-perf to be cropped to 2.40, the final film-out will be from a 1920 x 800 area of the HD transfer.

So when Fox told me that they wanted to do dailies to 4:4:4 HDCAM-SR for "Jennifer's Body", probably so they can create HD preview screening versions, I asked them if there was a chance they'd end up using these transfers for the final D.I. and luckily they said "Why would we do that??? That's the low-budget way of doing D.I.'s..." or something to that effect, which was a relief to hear.

I'm finding that there seems to be two types of companies going into D.I. work -- those with a mostly broadcast video background and those with some film post background. Obviously I feel a little more comfortable with people who have some grounding in film.
  • 0

#7 Matthew Bennett

Matthew Bennett
  • Basic Members
  • PipPip
  • 38 posts
  • Director

Posted 11 March 2008 - 10:08 AM

Thanks for the interesting 'behind the deal' information, fascinating to hear about. Love those Todd Hido photographs, all the wet, misty atmospheres and single color schemes. I love fog and obstruction, distortion.
The Hido portrait work is great, too, is that an inspiration or is it just the landscapes/interiors?
  • 0

#8 Jon Rosenbloom

Jon Rosenbloom
  • Basic Members
  • PipPipPipPip
  • 713 posts
  • Cinematographer

Posted 11 March 2008 - 02:26 PM

If the movie is being done 1.85:1 is there a need to do a DI? Can't you just print it?
How does making a 2k scan compare to adding the generations of photo-chemical internegative to get to a release print?

As far as "Assassination ... " goes, isn't that the nature of "work for hire?" You give me the producers the options, but they make the final decisions, usually to our artistic chagrin.
  • 0

#9 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 11 March 2008 - 08:25 PM

Of course you could do a 1.85 movie without a D.I. but it is a horror movie and certain moments of stylization will be called for, darkening corners of the frame, pulling out some colors, etc.
  • 0

#10 Max Jacoby

Max Jacoby
  • Sustaining Members
  • 2955 posts
  • Other

Posted 11 March 2008 - 09:28 PM

We're also using the photos of Todd Hido as a reference:
http://www.toddhido.com/

For my birthday I was given a book of his photographs, very moody stuff. Love the houses at night. Interesting that one can always feel the light just out of frame, but it is motivated in these cases and could be a streetlight.
  • 0

#11 John Holland

John Holland
  • Basic Members
  • PipPipPipPip
  • 2248 posts
  • Cinematographer
  • London England

Posted 12 March 2008 - 12:54 PM

And if you are shooting 3 perf Super 1-85 you have to go down the DI route and less i am wrong about that ?
  • 0

#12 Michael Nash

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

Posted 12 March 2008 - 02:49 PM

And if you are shooting 3 perf Super 1-85 you have to go down the DI route and less i am wrong about that ?


Well yes, 3 perf = scan only, no contact print. But it's more likely the decision to shoot 3 perf was made after the choice to go DI, rather than the other way around ;).
  • 0

#13 M Joel W

M Joel W
  • Basic Members
  • PipPipPipPip
  • 268 posts
  • Student

Posted 12 March 2008 - 05:31 PM

What approach are you taking toward interiors to make them scary? No fill light and crushed blacks in DI? Hard light rather than soft lights? Any tricks? Color gels? Shaky cam or steadicam? I'm curious since I'm shooting something in a similar genre....on an infinitely smaller scale.
  • 0

#14 Jon Rosenbloom

Jon Rosenbloom
  • Basic Members
  • PipPipPipPip
  • 713 posts
  • Cinematographer

Posted 14 March 2008 - 10:02 AM

"We're also using the photos of Todd Hido as a reference:
[url="http://www.toddhido.com/""]http://www.toddhido.com/"[/url]

I just saw an Avis ad that looks like it has the same reference.
  • 0

#15 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 14 March 2008 - 07:23 PM

Well yes, 3 perf = scan only, no contact print. But it's more likely the decision to shoot 3 perf was made after the choice to go DI, rather than the other way around ;).


Yes.
  • 0

#16 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 14 March 2008 - 08:09 PM

What approach are you taking toward interiors to make them scary? No fill light and crushed blacks in DI? Hard light rather than soft lights? Any tricks? Color gels? Shaky cam or steadicam? I'm curious since I'm shooting something in a similar genre....on an infinitely smaller scale.


Good question.

The general approach is natural, realistic, but whenever you're talking about scenes set in moonlight, you have a lot of leeway for stylization -- hard light with dancing shadows, soft dim light, high contrast, etc. Half the movie is set at night, which makes it harder to shoot but easier to create a mood.

I want the blacks and contrast to be rich, to create depth and sharpness, but for the most part, I don't see going with some odd skip-bleached look.

We start out the first week in a high school doing daytime scenes, so I just hope that whoever is watching dailies realizes that these will be the most "ordinary" of scenes, lighting-wise. Mainly in those scenes, I'm looking for ways to transition from a warmer sunnier look in early scenes to a cooler, greyer, overcast look in later scenes.

I just watched my camera tests this week. I mainly tested how the three stocks (5212, 5217, and 5219) handled being shot in 3200K light and then in 5500K daylight (HMI) with an LLD filter and then with no filter, since I'll have some low-light daylight interiors no doubt (as I did in New Jersey shooting "Assassination of a High School President" on Fuji Eterna stocks.) I also tested some different moonlight approaches (hard, soft, frontal, side, back, etc.) at different degrees of underexposure. And I tested wall colors and costumes, some make-up, and a few key elements from scenes.

I had a great tour of the Technicolor facilities. Talking to Dave Armstrong, chief timer and lab manager, he asked me if I noticed that the printer lights of the newer stocks were odd in terms of density, that they were printing at lower numbers than you'd think was correct. I said I found that out to be true of the Fuji Eterna stocks, that due to the changes in color mask density and whatnot, in order to get less noise in scans, that you ended up with lower printer numbers even though the image looked dense-enough when printed.

I decided to rate the stocks like this: 1/3-stop overexposure for the 100T and 200T, and 2/3's of a stop overexposure for the 500T (so 320 ASA normally and 640 ASA with a one-stop push.)

The printer lights were interesting to compare (and at this lab, generally a normal negative should print in the 25's):

5212 100T (rated at 80 ASA)
3200K lighting R=32 G=37 B=32
5500K lighting with an LLD correction R=27 G=38 B=42
5500K lighting with no filter R=29 G=41 B=49

5217 200T (rated at 160 ASA)
3200K lighting R=28 G=34 B=29
5500K lighting with an LLD correction R=25 G=34 B=37
5500K lighting with no filter R=26 G=37 B=45

5219 500T (rated at 320 ASA)
3200K lighting R=27 G=35 B=24
5500K lighting with an LLD correction R=23 G=37 B=34
5500K lighting with no filter R=25 G=40 B=41

500T rated at 640 ASA and pushed one stop
3200K lighting R=28 G=36 B=24

So the new 500T stock overexposed by 2/3's of a stop prints at similar numbers as the 200T stock overexposed by 1/3 of a stop -- and the 100T stock seems the densist of all when rated 1/3 of a stop slower.

This would suggest that perhaps the 100T and 200T stock are really only 2/3's of a stop apart. Plus it's interesting but the 100T stock is not much less grainy, it's just sharper -- as if the extra sharpness was making its grain seem more like the 200T.

So is the new 500T stock really not 500 ASA? My guess is that it's just the change in masking and whatnot that is causing the lower printer light numbers, just as with the Fuji Eterna stocks. It certainly looks rich rated at 320 ASA.

While in the corrected prints, the differences between the shots made under tungsten light and under daylight with no filter and with the LLD filter all seem minor -- there is a hint of purple in the corrected daylight shots, unlike when you correct it digitally where there seems to be a brown bias -- the printer lights are interesting to look at.

The LLD seems to be suppressing the excess blue, even though oddly, the red information is one printer light point lower compared to the totally uncorrected daylight shot. You'd think the opposite would happen.

Anyway, this suggests to me that there is some benefit from using the LLD filter. I looked at one the other day at Panavision, next to a 1/8 Coral and an 812 filter -- the LLD is slightly paler and loses only 2/10ths of a stop rather than 1/3 of a stop. So I can understand why Tiffen says to just ignore the density of the filter.

There may be some benefit also in the D.I. session from not having an overly-dense blue layer. So I think I'll use the LLD filter when I can't use the normal 85 correction (which is most of the time indoors -- I'll need the stop.)

As for 100T versus 200T, I plan on using the 100T when there's enough light outdoors, plus the extra sharpness may be useful for establishing shots in daytime, but frankly, the 200T is so close in quality that I won't sweat switching to it for darker day exteriors. It has a "smoother" look in fleshtones compared to the 100T, which is crisper.

The new 500T has more red saturation, which becomes apparent quickly when push-processing it. The printer lights show that it prints almost exactly the same as the normally-processed version, but it looks more contrasty with stronger reds -- dare I say, it looks more Technicolor-ish? Grain hardly jumps with the one-stop push. I think if you liked the harder look of the 200T EXR stock 5293, the 5219 pushed one-stop may be what the doctor ordered...
  • 0

#17 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 15 March 2008 - 11:53 AM

Reposting some of this info on the CML, I realized why the LLD caused the red to drop one printer light point compared to no filter at all in 5500K light being corrected on tungsten stock -- I didn't use any filter factor for the LLD, since Tiffen suggests just ignoring the filter's density, but since the filter does actually cut about 2/10th's of a stop, this accounts for a few of the printer light points dropping.
  • 0

#18 M Joel W

M Joel W
  • Basic Members
  • PipPipPipPip
  • 268 posts
  • Student

Posted 15 March 2008 - 02:25 PM

Yikes, that's quite the careful analysis. It's interesting that the faster tungsten stocks are more sensitive to blue; this makes sense, particularly now that people who are shooting digital are noticing lots of noise in the blue channel in low light tungsten situations. I guess even film has trouble here. I assume you're only using the LLD filter with the 500T and not with the slower stocks?

Lastly, I watched some recent horror movies and some new trailers, too, and developed a theory about lighting and I'm wondering if you could respond to it. In scenes where there is surprise or scenes that are building up to a jump scare that you don't know is coming, there seems to be unusually low key lighting because you don't want to see the monster or what's around the character who is being surprised. A lot of these scenes are handheld, too. In scenes that are more about suspense or dread, it seems like a lot of the time there's either higher key lighting or the "monster" or whatnot is lit visibly and the camera is much smoother. I suppose it's about range of narration: for surprise you want to know less, and for suspense you want to know more. But it's interesting that this manifests itself in lighting styles, too. Am I on to something or have I been taking theory classes too long?
  • 0

#19 Evan Winter

Evan Winter
  • Basic Members
  • PipPipPip
  • 202 posts
  • Director
  • Los Angeles

Posted 15 March 2008 - 02:30 PM

Thank you David. It feels like I'm in a Cinematography Master class at an Ivy League. :)
  • 0

#20 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 15 March 2008 - 02:58 PM

All tungsten stocks have to have a faster blue layer to compensate for the weakness of blue wavelengths in tungsten light -- that's why it gets so dense when shooting in daylight uncorrected. The question is whether you can correct it in post without artifacts, and generally, my experience is yes, you can, but you are limiting some range of corrections. For example, you can correct to normal but it would be hard to then time it to look warm on top of that. In printing, you may hit 50, the top of the scale, in the blue record unless you retrimmed the printer.

----

Obviously more shadows in the frame allows the audience to use its imagination and fill-in the missing information. But darkness can be just as suspenseful as a well-lit image can be...

Suspense is all about providing the audience with enough information that they anticipate something happening (usually something bad, but there can be happy suspense too -- "will she marry me?" "will I get the job?")

But you can provide that information in advance of a dark scene ("the monster is hiding in that dark warehouse" -- hey, it worked great in "I Am Legend") so that once the character enters the dark space, the audience is expecting something to happen -- that's suspense, but it also allows for surprises to happen.

For example, in "Alien" when Ripley, Ash, and the captain enter the darkened sick bay because the face hugger has disappeared, that's suspenseful because they don't know where it went but they know (and the audience knows) that it must be somewhere in the room. When Tom Skerritt knocks over that cannister and creates a loud bang, that's a surprise. When the face hugger (though dead) drops down from the ceiling and starts dangling behind Ripley's back, then dropping on her shoulder, that's suspense with a bit of surprise.

As for handheld versus smooth dolly moves, that's just an energy thing, what sort of emotional energy are you conveying. Handheld has a more nervous quality, more human -- it seems to react to things happening in the space. But all that movement can also make it harder to see things happening, which may or may not be good.
  • 0


Willys Widgets

Abel Cine

Glidecam

Technodolly

Media Blackout - Custom Cables and AKS

Paralinx LLC

Opal

FJS International, LLC

CineTape

Aerial Filmworks

Visual Products

Ritter Battery

Wooden Camera

rebotnix Technologies

Rig Wheels Passport

The Slider

Broadcast Solutions Inc

Gamma Ray Digital Inc

Tai Audio

Metropolis Post

CineLab

Rig Wheels Passport

Media Blackout - Custom Cables and AKS

Gamma Ray Digital Inc

Tai Audio

FJS International, LLC

Broadcast Solutions Inc

Wooden Camera

rebotnix Technologies

Abel Cine

Paralinx LLC

Metropolis Post

Technodolly

Aerial Filmworks

CineTape

Visual Products

The Slider

Glidecam

Opal

CineLab

Ritter Battery

Willys Widgets