Timing A First Feature

My first feature is at the negative cutter and in early January I'll travel to LA to time the first print. Something that I've never done on a feature (obviously -- this being my first).

I have many questions, probably too lengthy for the list, and would like to know if any of the more experienced feature DPs here on the list would voluteer to answer my questions off the list.

Jim Dollarhide Director of Photography

Jim,

I suspect most CMLers will be most intrested in this discussion. Sounds far more intresting than some other threads !!! Go public, sell you story, bare your soul and enthrall us all with the "First Feature Advice" The movie !!!

Tom Gleeson

Jim,

To hell with how long it is, break it into sections if you can, but ask it here.

My god! a thread to do with cinematography

Geoff

Well great. Thank you Geoff. Here goes part one...

This is regard to my first feature as a DP and my inexperience at timing the print on a feature film, which I'll get to do in a few weeks.

Print stocks. For the past 15 years, nearly all my work has gone to telecine. Before that, I printed my 16mm films at DuArt and Arthur Nalven would make recommendations to me and I would just accept his judgement. I seldom sat in on the timing of the first answerprint, I would just get it, screen it, and make notes and ask for a second, corrected print. (and so on, it usually took three prints to get close to what I wanted).

So I know virtually nothing about prints stocks. What is available? What are the attributes of each? What are the downfalls/problems? Contrast is/will be a big issue for me.

I shot the picture on 93 and 98, so I don't even know if Vision prints stocks are an option. It is being done at CFI and I don't even know anyone there except some people in telecine.

End of first question.

Jim Dollarhide Director of Photography

By default the lab will probably print on Kodak Vision (2383). This is "normal" contrast, somewhat higher than the stock it replaced last year, although it seems to have settled back a little. Vision Premier is the Kodak High Contrast stock, which will block up your shadows with really really deep blacks, and give you slightly more saturated colours.

Fuji also make normal and HC stocks, with slightly different dye sets and therefore different colorimetry. I find the yellows warmer, more orangey-brown: other people comment about the saturation of the green tones.

I haven't looked at the Agfa stock recently.

I beleive CFI normally use a version of the technique that Jeff Kreines described, printing sample frames from each shot on the cut negative. (Though not for a separate pilot test neg as he described). That way, although it IS the original neg running through the printer, it only uses 100ft or so of stock each time, and you can shuttle to and fro on the still projector in the lab's review room, to correct each shot, if needed.

With a good grader (timer) and a well-calibrated Colormaster (or Hazeltine), (I take it for granted they have a good negative to work with), they should be able to get what you are aiming for very accurately with a minimum of reprints.

Listen to the timer (grader), and take their advice. Grading (timing) is what they do all day every day: it's as much a specialised craft as all the others on a film crew, and as long as you explain the intent of the film, they should deliver the right feel without too much intervention.

Dominic Case

Group Technology & Services Manager

Atlab Australia

www.atlab.com.au

P.S Thanks for what is not only a genuine on-topic question, but one that's on MY topic.

Dominic, what sort of printer does this? Is it a modernized Cinex machine or Sensitester, with a Model C and shutter attached? Or is it a modern printer (BHP Modular Panel, say) with a mechanism added that lets one lift the print stock away from the neg, while the neg advances to the next shot? Does it cause any additional wear on the neg?

Jeff "curious Jeff" Kreines

This is exactly the sort of thing we should be discussing. I've timed several features but have always felt there is more to know about this process. On one hand, "Hey, it's just three numbers", and on the other, "Hey, this is more complicated than I thought".

Fortunately, for me, the timers I have had the privileged of working with have been very conscientious. There have been occasions though where decisions taken during production have not worked their way through the process as planed.

It seems obvious but watching the picture several times with the timer is the most important part. I find a first screening of the workprint, or off-line video, very productive and a real time saver.

Interesting topic.

David Perrault

So now I know there are basically only two choices for print stocks. 2383 for normal contrast and 2393 for higher contrast - deeper blacks. And the compriable Fuji produts. Having been told early on that I couldn't count on any special print process to deepen the blacks, I lit to the contast that I wanted -- knowing that I couldn't likely experiment in printing. So a special printing process ENR, etc., isn't in the cards.

My next question is what will happen at CFI? I assume that I'll screen the Avid low-res picture lock with the timer. And if the producer will go for this technique of sample frame printing that Dominic and Jeff have described -- I certainly would start with that.

But then what happens -- after the screening of the picture lock with the timer. Does he go into the Hazeltine room alone? Without me?

Or do I get to look over his shoulder? And if I do, what sort of basic interpretation do I have to make.

For instance, I assume that the video display on a Hazeltine can not accurately reproduce the constrast range of a print. (or can it? Is it some sort of high-resolution display capable of more constast range than a standard monitor in a telecine room?).

After years of taking my negative to telecine, I'm accustomed to -- to use a Macintosh attribute -- WYSIWYG. What you see is what you get.

I'm sure once I get to CFI they'll be able to answer these questions, but it is nice to get some feedback from people who have done it, prior to me going into the lab.

I assume that once they have timed the negative, that overnight there will be a first trial answer print and that I'll screen that with the timer. Both projected, and then on a bench for the ability to stop and discuss, and make notes together.

Jim Dollarhide

Director of Photography

Actually, the lower contrast Fuji print stock is comparable to the obsolete '86 Kodak stock, so now there is no Kodak stock that is comparable. The high-contrast Fuji is meant to be comparable to the standard VISION stock (2383), although now it is probably higher in contrast since Kodak lowered the contrast of VISION slightly.

>My next question is what will happen at CFI? I assume that I'll screen the Avid low->res picture lock with the timer.

Don't assume -- make sure you schedule some sort of screening of the AVID output with the timer if that's what you want to do.

>But then what happens -- after the screening of the picture lock with the timer. Does >he go into the Hazeltine room alone? Without me?

My experience is that DP's are not allowed to watch over a Hazeltine timing session; you get to see the results in the first answer print. This is why some type of proof print (select frames printed) or printing of outtakes can be helpful, although not required.

>I assume that once they have timed the negative, that overnight there will be a first >trial answer print and that I'll screen that with the timer. Both projected, and then on a >bench for the ability to stop and discuss, and make notes together.

Some labs have benches; others just watch the projected reel again if you need to make additional notes (while the film is screening, the timer makes notes based on the footage counter.) The system is not well-designed for close shot-by-shot examination, hence why getting a good timer is half the battle of getting a good print within a few answer prints. With a poor or inexperienced timer, you could be spotting mistimed shots well into the third or fourth answer print.

David Mullen ASC

Cinematographer / L.A.

Dominic Case [dominic_case@atlab.com.au] Sent: 14 December 1999 07:03

Jim, Firstly it looks as though it's time for a long chat with your contact person or the timer at CFI. I obviously can't speak for them, but here's my two penn'orth concerning a couple of your points. (And by the way, thanks for publishing these questions - this is really what CML is about.)

>Does he go into the Hazeltine room alone? Without me?

In the dark? Well, yes

Seriously, there is a very important concept to understand here: your next comment is a good starting point:-

>After years of taking my negative to telecine, I'm accustomed to -- to use a Macintosh >attribute -- WYSIWYG. What you see is what you get.

In telecine, what you see is indeed what you get - provided you watch a calibrated monitor: you are seeing a video image displayed on the sort of device it is intended for. However, in the film analyser (Hazeltine, or in CFI's case, and for that matter Atlab's case, Colormaster), what you see is a video image, typically an 8-bit log digital display, shown on a fairly average monitor, but what you (will) get is a film print: subtractive dyes not additive phosphors, projected with a much higher brightness range. Although the image and the monitor are calibrated to _simulate_ a film print as closely as possible, they will never look quite like the eventual print.

And so the final stage of calibration lies in the grader/timer's head - between the eyes and the brain. The timer/grader knows that if the Colormaster image looks like THIS, then the print will come out like THAT. THis is the fundamental difference between a grading/timing session and a telecine session. It relies not only on the grader/timer's experience in knowing how to get the desired colour, but also on their ability to interpret and mentally correct for different media. You, the client/filmmaker/DoP can't see inside the timer/grader's head, so you will have to trust them - until you see the print the next day.

And so, while not all labs agree on this, it is quite common for the grader/timer to go into the Hazeltine room quite alone, without the client, and get on with their job. Provided there has been good communication beforehand - not just at the level of "bluer"; "lighter"; "print that shot down at bit", but at the level of "I really want a lush green, natural feel in all the backyard scenes, in contrast with a hotter, claustrophobic feel in the dinner party", or "keep the housebreaking sequence light enough to see their faces", then I would expect you to be delighted even with your first print.

BTW - don't assume anything - talk to your lab liaison before you (or they) get to start on the job.

Good luck

Dominic Case

Group Technology & Services Manager

Atlab Australia

This timing conversation brings out one of the clear advantages of cutting the old fashioned way from workprint rather than from a transfer. The work print provides a frame of reference for conversations with the timer, it can serve as a 1st answer print and give you one leg up in the process.

Glenn Suprenard

www.matineepictures.com

>My experience is that DP's are not allowed to watch over a Hazeltine timing >session...

There's a good reason for that, since the Hazeltine rooms are not always cushioned client rooms.

I hope CFI doesn't read this...but...I've been able to sneak up to the Hazeltine Timers occassionally. There's one particular Colorist at CFI that is very experienced and extremely helpful (anonymous to protect their position). Sure, it helps to see the print, but it's also good to compare it to Fuji/Kodak LAD grays, and see what your lights are, and to get feedback from the Hazeltiner.

It's good to know how things work.

This also relates to the Fuji vs. Kodak question, since I found that when I tested both a month ago, the new Fuji 8572 likes to be rated at 500 ASA. And the Kodak '77 was OK at 500 ASA if you wanted it pastel (Vision print stock brings it back). Least that's what my meters and CFI's soup showed.

>For instance, I assume that the video display on a Hazeltine can not accurately >reproduce the constrast range of a print...

Hence, the Hazeltine Timer's skill in following the Print Timer's instructions and interpreting the image and DP's intentions.

And that's why we make prints to see the net result. It's nice to see your work projected. And sometimes a little scary.

Mark "wondering how the sets and smoke will look" Doering-Powell