doing another D.I.

[David Mullen ASC]

The supernatural thriller I shot in New Orleans, "Solstice", is going through a D.I. right now over at Technicolor (TDI). Just like when the D.I. was done for "Astronaut Farmer" four months ago, I'm still shooting "Big Love" so I had to go in this Sunday to go through the timing, but unlike the four days over two weekends that I put in before for "Astronaut Farmer", today was my only shot at color-correcting the movie since it has to be delivered in the middle of next week. But since this movie has a much smaller budget, I'm not surprised that the D.I. time is limited to about 8 days total (they started last Monday.)

 

The 3-perf 35mm negative was scanned on a 4K Spirit (just like "Astronaut Farmer" was over at Warner MPI) but only at 2K unfortunately (standard procedure over at TDI unless you pay more I guess.) But it looked pretty good projected on a medium-sized theater screen over there, using a 2K DLP projector. They use a DaVinci color-corrector over there.

 

So I had about seven hours to go through the whole movie and make timing adjustments, which meant I couldn't really go through every shot make changes -- a few scenes I just made an overall shift to every shot. Luckily what the colorist had done over the week was pretty close to finished.

 

I used the new Fuji Eterna 250D stock for most of the day interior and exterior scenes, and Eterna 500T for the night work, all rated 2/3's of a stop slower than normal. The colorist was impressed with how fine-grained the image was -- it looked very clean. The low-con nature of the Eterna stocks are well-suited for D.I. work too, having a lot of information at both ends of the exposure scale. I shot in 3-perf using Panavision Primo lenses, mostly the primes, which looked really sharp in general. There were a couple of shots where I used a little diffusion on some close-ups, but not much, since the cast was young and I wasn't sure how well Super-35 would blow-up to anamorphic, this being my first Super-35 movie. I guess I'll see the film-out next month sometime.

 

The original opening scene, a Christmas party from the year before the main story, was cut-up and spread throughout the movie as a flashaback. I had used a 1/8 ProMist for that scene, but now that it was being used for brief flashbacks, I decided to try something new (for me) -- digital diffusion. I gave the highlights in those shots a sort of gaussian blur that was quite lovely.

 

Most of the exterior scenes in the movie were shot in available light as much as possible, so I had to spend a little time correcting for weather mismatching, since the humid weather in New Orleans meant that I had a lot of hazy clouds moving past the sun.

 

My only disappointment was that the budget & schedule never allowed me to go out and shoot a lot of beauty shots of the swamps and the city of New Orleans -- we have a few swamp shots that I shot, or my B-cam operator Theo Pingarelli shot on the side -- but otherwise they opted to buy some stock footage, aerial shots of New Orleans and the swamps, which unfortunately was delivered in HD tape and uprezzed to 2K. It looks a little video-ish and edge-enhanced, even though the original footage appears to have been shot on film. "Akeelah and the Bee" also had HD stock footage, as well as HD footage shot by the director in Washington D.C. I guess HD stock footage for low-budget movies is just an emerging reality. I suppose in the old days I would have been working with a dupe of a dupe for all I know, now I get uprezzed 4:2:2 HD.

 

Anyway, the quality of 3-perf Super-35, at least when projected on a 2K DLP projector in 2.35, is pretty impressive. I still think that anamorphic is superior for achieving 2.35, but since this movie had so much night exterior work shot at T/2, anamorphic seemed more like a liability than an advantage.

 

The director will get a chance this week to see the timing and I'm curious to get his feedback because when timing night exterior scenes lit by "moonlight" it's always unclear just how dark to make them -- timed too bright and they look "lit" and timed too dark and you're straining to see the actors' expressions.

[Mark Henderson]

David:

 

It's great to see you were impressed with the Super 35, 3-perf. Did you shoot this format going in believing it would be a theatrical release or was it planned to go straight to TV and/or DVD? Did the extra run time on the mags seem of a great benefit? Were there any special problems in creating a DI of it or is that in the past because of all the TV work?

 

Mark

[George Lekovic]

David,

 

thank you for the thorough report of your Super35 DI experience. I recently shot my first 3-perf feature on an Aaton 35III. Being a low budget one it did not allow for a DI and for "rough cut" purposes we went to DigiBeta, semi-supervised, and it looks surpisingly good. As a test I did a 3-perf to 4-perf blow up (or blow down?) via 2K to see what results to expect if we ever go that route. I will see that screened tomorrow.

 

Your posts are always insightful, helpful and much appreciated.

 

Thanks.

[Patrick Cooper]

Good to read about the progress of this film, David. And I do like supernatural thrillers! Also good to hear that you are pleased with Fuji Eterna 250D as this will be the same film stock that I will be using for a project of mine soon - though I'll be shooting it in 16mm rather than 35mm - (first time I will be exposing negative movie film.) Pity you were not able to go out and film more 'beauty shots' around the location. That sounds like an opportunity to be really creative and let loose and not be restricted to a rigidly set shot list.

 

"...since the humid weather in New Orleans meant that I had a lot of hazy clouds moving past the sun."

 

I hope that the weather over there is not too humid to affect camera gear.....

 

In Australia, I live in a relatively dry state but over in Queensland, it's not uncommon to discover fungus growing in lenses.

[Mark Henderson]

David:

 

Why did you decide on Fuji stock instead of Kodak? I use the kodak Vision II a lot and love it. Great grain structure. Just wondering. I used Fuji exclusively as a studio still photographer for years and loved what I believed to be a "warmer" look.

 

Back then in still photography (4x5 work), before Photoshop, you had to do all your color adjustments with Kodak wratten (?) filters just to get the film close to being usable; one box might need a 30 Magenta filter, the next a 20 Blue.

 

Thanks,

Mark

[David Mullen ASC]

I figured the movie might get a theatrical since the producers last horror movie, "Stay Alive", did (also shot in New Orleans). But it doesn't matter because I always try to shoot as if a movie were going to be shown on the big screen, just in case.

 

No, in this case, I wanted to shoot the movie in 2.35 because of the large cast, the locations, and because I think it often suggests that the budget is higher than it really is. And I wanted to do a D.I. because it's a supernatural thriller and I might want to play around with the color & contrast.

 

But since it had so much night work, often people running around the woods in a rainstorm, anamorphic just seemed too unwieldly, with too much chance for missed focus. I had just gone through some big night scenes at a fairground for "The Astronaut Farmer" using anamorphic at T/2.8 and pushing the film a stop, and the focus-pulling was tough. It gets a bit uncomfortable worrying all the time whether the focus puller got the take or not.

 

In retrospect, getting to a T/2.8 in the woods at night wasn't as hard as I thought, even rating the Eterna 500T stock at 320 ASA. On the other hand, with all the Steadicam work we did, I would have been forced to use Panavision C-Series primes probably, which aren't that good at T/2.8.

 

So I picked 3-perf over 4-perf Super-35 because I feel that there's not much reason to shoot 4-perf Super-35 if you are framing for 1.85 or 2.35. It just makes sense -- there's no quality difference between 3 and 4-perf if the final aspect ratio is 1.78 or wider, and there's a 25% cost savings plus you get 25% more running time on the mag.

 

The only problem I had was the lack of ability to screen a 3-perf contact print in New Orleans. I had this problem with two shots with soft focus in the center of a wide-angle wide shot, where only the edges of the frame were sharp, on different cameras with different lenses. This showed up in DVD dailies and we struck a print and it showed up in the print according to the editor who screened it in Los Angeles. So I had Panavision come out and check the lenses and the cameras, etc. but they found nothing. Someone suggested on the CML that perhaps the film had been over-dried by the lab and had curled, and to re-wash and dry it to fix it. I decided to put that off until we did the D.I. -- but when those two shots came up in the D.I. theater, there was nothing wrong with them! So who knows what the problem was. Maybe the film negative camera roll flattened itself out over the months sitting in the vault.

 

I've also learned since then, dealing with DVD dailies on "Big Love", that they often seem to go in and out of focus as if there was a registration problem, but it's due to the MPEG compression being used.

 

The grain of the new Fuji Eterna line-up (so far, only 250T, 250D, 400T, and 500T) is as good as the Vision-2 line-up, if not slightly better for the same speed stocks (i.e. I think Eterna 500T is slightly finer-grained than 5218). However, Eterna has a slightly more pastel, low-con look compared to Vision-2 -- more like the look of Kodak Expression 500T in terms of color and contrast.

 

I like Kodak Vision2 as well (I shot "Akeelah and the Bee" on those stocks -- '12, '05, '18, and '29). If I weren't doing a D.I./telecine and I didn't was as low-con or pastel a look, I'd probably use Vision-2 instead of Eterna, unless I could guarantee being able to print on Vision Premier or Fuji XD stock.

 

But when you do any sort of electronic color-correction, what you mainly need is a low-grain stock with wide exposure latitude -- you can make it more contrasty and saturated in color-correcting.

 

And Fuji is cheaper. Sometimes I've asked for Kodak and been told that the company wants to use Fuji to save money, and since I don't have a problem with Fuji, I usually say OK. But it just depends on how the project will be posted. Kodak is a little snappier and edge-sharp, which can be useful in doing things like an optical printer blow-up, let's say.

[Chayse Irvin ASC, CSC]

Sounds like a tight shoot. How many set ups perday were you doing on that show David?

[David Mullen ASC]

It was a 25-day shoot, which was very tough with so many large night exteriors to light with not the most experienced crew, working in swamps, bayous, around lakes, etc. We got condors stuck in mud in a couple of locations.

 

I don't really ever pay attention to the number of set-ups we do a day. Either I feel we got the scene and the necessary coverage, in the time allotted, or we didn't -- it doesn't matter if it was 20 set-ups or 40 set-ups to do that.

 

Plus now that I do so much more 2-camera shooting, it's harder to keep track of what's considered a set-up.

 

I do remember that I was shocked to hear that on one day for "Akeelah and the Bee" shooting the final spelling bee that we did some 75 set-ups in under 12 hours.

[Chayse Irvin ASC, CSC]

It was a 25-day shoot, which was very tough with so many large night exteriors to light with not the most experienced crew, working in swamps, bayous, around lakes, etc. We got condors stuck in mud in a couple of locations.

 

I don't really ever pay attention to the number of set-ups we do a day. Either I feel we got the scene and the necessary coverage, in the time allotted, or we didn't -- it doesn't matter if it was 20 set-ups or 40 set-ups to do that.

 

Plus now that I do so much more 2-camera shooting, it's harder to keep track of what's considered a set-up.

 

I do remember that I was shocked to hear that on one day for "Akeelah and the Bee" shooting the final spelling bee that we did some 75 set-ups in under 12 hours.

 

Do you find that some days so much pressure is put on you to complete 75 set ups that your lighting suffers because of it? How do you prepare for such a day without knowing how many set ups there are? Do you fall back on your experience to get it done?

[David Mullen ASC]

Of course when you work fast and shoot a lot of shots, the quality will generally suffer, but that's just the nature of making movies. You can't make everything look like a David Lean movie -- you have to pick your battles, and for the rest, you try and hide the compromises as best you can.

 

Obviously on a complex day of action, whatever, you plan a shot list -- but that's often shorter than the final number of set-ups. In other words, if you show up with a shot list of 20 items, you'll probably shoot 30 by the end of the day. But if you had shown up with a list of 30 set-ups needed, it probably meant that you really needed 50 set-ups. You always find that you need some extra shot here and there, or have to break something up more than you planned, or put a second camera on some action just to grab it.

 

I got to draw storyboards for the final spelling bee in "Akeelah and the Bee", based on what the director wanted, and we pasted all the boards on a big board and checked off what we got each day.

 

Doing dialogue scenes, you are more likely to block and rehearse with the actors and then decide on the coverage, and then just keep track of the time.

[Bryan Jacobson]

Thanks guy's this was fun thread to read.

 

Hi David,

What lab are you using to burn back to film? Does TDI do that? I heard they do digital

distribution there for digital cinema, I would be interested in hearing about the compression

quality they are using. I'm sure it's been discussed here before but do you know what theaters are

doing that now and what kind of projectors they are using? Thanks for sharing about your experience over there.

[Max Jacoby]

David

 

Did you have a chat with TDI about that typical TDI look that we've all become aware over the years?

[David Mullen ASC]

Did you have a chat with TDI about that typical TDI look that we've all become aware over the years?

 

Yes, politely (diplomatically) of course. Let's just say that there have been some strong disagreements among the colorists there regarding the proper use and degree of grain reduction... But one of the problems is that some of the big DP / clients insist on the grain reduction being cranked up.

 

I had a good chat with my colorist on all the issues and he readily pointed out some grain reduction artifacts in their D.I.'s that slipped by me even. We also agreed that anamorphic photography was great for D.I. work because of the larger negative area.

 

I think many D.I. houses are getting a little tired of dealing with grainy underexposed Super-35 photography. I had a similar discussion at Warner MPI, which just finished the D.I. for some holiday release which had to be grain reduced because the director and DP had shot it underexposed for a dark and gritty look when suddenly at the end of post, the studio decided it was a comedy and needed a more cheerful, brighter look (and if I told you the name of the title, you'd be surprised that the filmmakers even thought it wasn't a holiday comedy --- maybe they were in denial.)

[K Borowski]

But one of the problems is that some of the big DP / clients insist on the grain reduction being cranked up.

 

:blink:

 

Do they want us all to just assume the projector is out of focus with this approach? If so, they're doing an excellent job. . .

[Michel Hafner]

Yes, politely (diplomatically) of course. Let's just say that there have been some strong disagreements among the colorists there regarding the proper use and degree of grain reduction... But one of the problems is that some of the big DP / clients insist on the grain reduction being cranked up.

If these big DOPs dislike grain so much why do they shoot their films so that a lot of grain has to be dealt with? Can't they see all the artifacts the filtering creates?

Michel Hafner

Edited February 4, 2007 by Michel Hafner

[David Mullen ASC]

Because in the real world, we pick and expose high-speed stocks based on the light level we want to be working at, and sometimes it is low for either creative reasons or logistical reasons -- plus sometimes we simply make mistakes, or the 2nd Unit DP makes them, etc. So we may want a fine-grained look but not be able to shoot slower-speed stocks, or shoot anamorphic, or overexpose the stocks, etc.

 

I do agree though that there is a limit to how far you should push an image to become something it wasn't originally, unless you don't mind the artifacts.

 

I had one 2nd Unit shot in "Astronaut Farmer" that had to be grain-reduced because it was underexposed too much. It was a hallway shot of someone silhouette against the daylight at the far end of the hall, which is tricky to know where to set the exposure. I shot one scene in that light and got the right exposure, but a 2nd Unit pick-up of another shot in the same lighting set-up was a couple of stops too dark (they didn't ask me what stop I used.) It comes down to the fact that I always err on the side of more exposure, figuring that the film can handle it. It's easy to get fooled if you point a spot meter at the bright end of the hallway, figuring that you will of course then silhouette the foreground. But the problem is that the person in the foreground isn't silhouette against the windows, but the glow of those windows along the walls, so you need enough brightness on the walls to create a solid silhouette, otherwise you get a dark figure against a dim wall. It's too bad the shot was underexposed because there was plenty of light in there. I think I shot around an f/4 so I'm guessing the second unit shot was made around an f/8.

[Alessandro Machi]

Because in the real world, we pick and expose high-speed stocks based on the light level we want to be working at, and sometimes it is low for either creative reasons or logistical reasons -- plus sometimes we simply make mistakes, or the 2nd Unit DP makes them, etc. So we may want a fine-grained look but not be able to shoot slower-speed stocks, or shoot anamorphic, or overexpose the stocks, etc.

 

I do agree though that there is a limit to how far you should push an image to become something it wasn't originally, unless you don't mind the artifacts.

 

I had one 2nd Unit shot in "Astronaut Farmer" that had to be grain-reduced because it was underexposed too much. It was a hallway shot of someone silhouette against the daylight at the far end of the hall, which is tricky to know where to set the exposure. I shot one scene in that light and got the right exposure, but a 2nd Unit pick-up of another shot in the same lighting set-up was a couple of stops too dark (they didn't ask me what stop I used.) It comes down to the fact that I always err on the side of more exposure, figuring that the film can handle it. It's easy to get fooled if you point a spot meter at the bright end of the hallway, figuring that you will of course then silhouette the foreground. But the problem is that the person in the foreground isn't silhouette against the windows, but the glow of those windows along the walls, so you need enough brightness on the walls to create a solid silhouette, otherwise you get a dark figure against a dim wall. It's too bad the shot was underexposed because there was plenty of light in there. I think I shot around an f/4 so I'm guessing the second unit shot was made around an f/8.

 

 

What are you looking at to determine an error was made? Is there any chance the lab made an error in printing (assuming you made a workprint), or do you actually make check the original negative?

[David Mullen ASC]

What are you looking at to determine an error was made? Is there any chance the lab made an error in printing (assuming you made a workprint), or do you actually make check the original negative?

 

We had video dailies from a telecine transfer of the negative -- the shot was dark in dailies but I assumed it was just transferred that way (besides, there was no chance of reshooting it anyway). Second unit grabbed a lot of unscripted shots, so half of them weren't going to end up in the movie anyway, but this one did.

 

The D.I. was done by scanning the negative at 4K on a Spirit and then downrezzing to 2K. So the shot was obviously underexposed once we looked at the scan and tried color-correcting it. If I had made a print, I could determine if the shot was underexposed by the printer lights used.

[Alessandro Machi]

We had video dailies from a telecine transfer of the negative -- the shot was dark in dailies but I assumed it was just transferred that way (besides, there was no chance of reshooting it anyway). Second unit grabbed a lot of unscripted shots, so half of them weren't going to end up in the movie anyway, but this one did.

 

The D.I. was done by scanning the negative at 4K on a Spirit and then downrezzing to 2K. So the shot was obviously underexposed once we looked at the scan and tried color-correcting it. If I had made a print, I could determine if the shot was underexposed by the printer lights used.

 

In the 80's and 90's, using Polaroid stills to take test shots was still in vogue to some degree. I have a very brief explanation about it on my website. Polaroids Stills

 

With the advent of digital stills, haven't offset calibrations been created so that second unit can point a digital still camera that mimics the film stock being used? Even if it's off a stop, the negative would easily handle it. The few times I have used polaroid black and white stills I matched my super-8 reversal footage exactly, which is probably the most difficult type of exposure to match.

 

It seems like a good rule to have second unit verify their shots via some type of digital or polaroid stills, especially potential "money shots", shots that take extra time to set up and aren't "quickies".

[David Mullen ASC]

Time, time, time, time...

 

We work very fast on a typical movie shoot. Some of us haven't found the time to develop any system of previz despite the nice idea. We set-up a shot, grab a meter reading, shoot, move on to the next shot. I'm lucky if I have time to shoot a grey scale at the head of the day.

[Angeliki Makraki]

So I had about seven hours to go through the whole movie and make timing adjustments, which meant I couldn't really go through every shot make changes -- a few scenes I just made an overall shift to every shot. Luckily what the colorist had done over the week was pretty close to finished.

 

Anyway, the quality of 3-perf Super-35, at least when projected on a 2K DLP projector in 2.35, is pretty impressive. I still think that anamorphic is superior for achieving 2.35, but since this movie had so much night exterior work shot at T/2, anamorphic seemed more like a liability than an advantage.

 

The director will get a chance this week to see the timing and I'm curious to get his feedback because when timing night exterior scenes lit by "moonlight" it's always unclear just how dark to make them -- timed too bright and they look "lit" and timed too dark and you're straining to see the actors' expressions.

 

 

 

 

What do you mean by timing adjustments ? Are these timed stills ?

Please explain. Thanks.

Edited February 6, 2007 by Angeliki Makraki

[John Holland]

Timing mean s grading each scene for exposure or colour correction .

[Angeliki Makraki]

Timing mean s grading each scene for exposure or colour correction .

 

Is that the same as a timed still ?

[John Holland]

Never heard of the term [timed still] but if you mean getting the perfect picture yes .

[David Mullen ASC]

"Timing" or "grading" (a U.K. term) refers to correcting / balancing each shot for color and brightness, and in the case of digital color-correction, also contrast, black level, saturation, etc.