Two Tools For Flashing Film
Copyright 1997 By Mark Woods  Director of Photography

Over the last year or so I wanted to conduct a sensotometric test with Arriflex’s Varicon and Panavision’s PanaFlasher. I thought the results between the two tools would vary more than they did, but that’s getting ahead of the story. First I’d like to explain what each tool is, how to use it, and different things to be aware of when you use each tool.
 

Four ways to Flash!

Most of us are familiar with the concept of flashing film. In fact, many of us may still believe that the lab will provide that service since many directors of photography have used this technique in the past. I approached a lab about post flashing a film I was preparing to shoot, and was told there was no way they would flash the film – post or pre flash. I knew from my days when I shot fine art black and white still photography, I could flash the film myself by shooting an evenly lighted gray card with the lens set at infinity. This was not a practical solution. (Heaven forbid I expose the film and the camera is one perf off the frame I exposed when it’s threaded for the production shot!) That left the last two options, Panavision’s PanaFlasher and Arri’s Varicon. I’m not going to go deeply into the use and construction of these tools, since the manufacturers do that very well. What I am going to cover is the similar results these tools achieve, but accomplish in two very different and unique ways.
 

Panavision’s PanaFlasher

What is the PanaFlasher?

The PanaFlasher is a small device that emits a calibrated amount of light on the negative as it travels through the camera body. This is accomplished by placing the unit in one of the magazine ports. The film is given an even exposure as it passes the unit. A meter is built into the unit and there are instructions on how to use the unit to get various Flash percentages on the film (e.g., 5% or 10%).

PanaFlasher Testing Methodology

The methodology I used to evaluate the unit took some thought. Usually I vary the T/stop, camera speed, and the shutter angle when I expose the gray card. Obviously, changing speed changes the duration the film is exposed to the PanaFlasher. In fact, Panavision has included notes to compensate the EV on the PanaFlasher unit when the cinematographer changes the frames per second. In evaluating the unit, all exposures were filmed at a constant 24 fps.

Things to think about when using the PanaFlasher

First is to remember to change the EV when the camera speed is changed from 24 fps. The second is to consider what type of film you use with the unit. The color balance of the unit is set for 3200°K. A filter is provided with a filter factor for a "cool" appearance. I didn’t test a daylight film with the PanaFlasher, but if one were to use a daylight film, this filter, or other color compensating filter, would be necessary if the cinematographer didn’t want warmish shadows. If s/he did desire that look, then no compensation would be necessary. For each filter used, a filter factor must be entered into the meter for accurate results. Overall, the unit is compact and fits in well within the Panavision system. It can easily be used in the hand held mode, and plugs into one of the power ports on the camera. If you shoot a test with a PanaFlasher, Panavision recommends noting the serial number for future use.
 

Arriflex’s Varicon

What is the Arri Varicon?

The easiest way to think of the Varicon is to think of a piece of glass that glows in front of the lens. If you haven’t seen the Varicon in action, it’s very unusual in that the lens is usually one of the darkest areas on the set. But when the Varicon is turned on, the filters in the matte box glow. I’ve put red, green, and other colors into the Varicon’s filter slot, and had actors comment about something unusual going on in the camera lens. The history of the Varicon goes back to the Lightflex first used on "Young Winston." Freddie Francis used it on "Glory," and it’s been used on a multitude of projects, including many of my own.

Varicon Testing Methodology

Since the light from the Varicon comes through the lens during exposure, I had to consider the testing method. With the Varicon I couldn’t change the len’s T/stop since that would directly affect the amount of light, or flash, the film would receive. Consequently I had to expose my gray card at a constant T/stop, but I could change the amount of light, the camera speed, and the shutter to achieve the desired T/stop.

Things to think about when using the Varicon

First, the Varicon only works in the "new" style MB-14  6.6 x 6.6 Matt Box. For hand held work, this can be cumbersome. Second is the power supply. The Varicon works with an AC 12/24v power supply. It also can be powered with 12v and 24v batteries. This is in addition to the batteries needed to power the camera. Again, in the hand held mode, it can be a little difficult, although I’ve done many hand held shots with the Varicon. On the plus side, it’s extremely easy to use color filters in the Varicon, and if you use the meter to set the amount of light, the meter will compensate for any filter. Also, the cinematographer can see the effect of the flash on the shot. If s/he is working with color filters for effect, this is very helpful.
 

Some Background on "Flashing" or Pre-Exposing Film

Why would one want to flash film?

Among the reasons a cinematographer flashes the negative is for contrast control. In a daylight situation where s/he doesn’t have the necessary equipment to light the deep shadows, flashing is an option. At the opposite extreme, in a night exterior where budget and time constraints limit the amount of lighting units available to illuminate the shot, flashing may be a good option. Beyond that, there are creative choices a director of photography may employ. For example, on "Evita" Darius Khondji used the Varicon in combination with other filters in different parts of the film for artistic effect. I spoke with Vilmos Zsigmond about flashing film, and he told me in the past that he used the lab to post flash "McCabe & Mrs. Miller" 10% to 25%. He varied the amount depending upon the desired effect. I asked him about the PanaFlasher, and he told me he used it on the film "Intersection," among other films. He set it at 10% -- 15%. (Panavision has a print illustrating the PanaFlasher at various percentages. It’s extremely interesting and helpful.) Some the results from flashing the film are slight desaturation, more information in the shadow detail, and in extreme uses, fogging, and tinting the film. On both the PanaFlasher and the Varicon, the cinematographer can introduce color into the light source. Any color can be used, but with the PanaFlasher, one has to figure in a filter factor to obtain accurate results.

What part of the characteristic curve is affected, and why?

Flashing only affects the toe of the curve. The reason for that is evident if one looks at the following table for a hypothetical flash of a negative film emulsion:

 

 

 

 

 

The top row indicates the stop above and below "normal." The second row indicates the amount of exposure necessary to achieve the density at each stop. If N-6 needs one foot candle (fc) of exposure, 2 fc would add one stop more density for N-5. Two times 2 fc is 4 fc, for N-4, etc. The third row indicates the overall "flash" necessary to equal the N-6 density. The forth row is the cumulative density of the flash added to each stop’s density (for clarity it’s indicated as foot candles). As you can see, the effect is greater when there is less light, or density, involved. The 1 fc necessary to achieve the N-6 density immediately lifts N-6 to N-5. N-5 is lifted half way to N-4 for 50% added density. N-4 adds 20% density while N-3 only adds about 11%. N-2 adds only about 6% density, and so on. I think this clearly demonstrates why a flash only affects the toe of the negative’s characteristic curve. If you look at the characteristic curves I plotted with the Varicon and the PanaFlasher, you will note that they fall fairly close to the hypothetical example above. Also, I only plotted the toe of the curves, i.e., Base Fog to "Normal." If you look at the densities from "Normal" and above, you’ll see there is very little divergence from the LAD Reference densities.

 

Some notes about the characteristic curve charts.

Dmin is the base fog of the film. If the film isn’t fogged, this is the clearest, most transparent part of the film. Obviously if the film is Flashed (i.e., fogged), the potential Dmin remains constant, but now we have to measure the new density results as Base Fog, not Dmin. It is upon the Base Fog that our images are built. If we start at a true Dmin, there is a greater contrast range. If we flash the film, we decrease the contrast range.

On the charts, I don’t list Dmin separately since I feel it would be confusing. Dmin hasn‘t changed, but what I did to the film has changed the Base Fog density. The 5274 Reference column indicates Dmin in the Base Fog row. The numbers to the right indicate the resulting Base Fog from the indicated flasher tool. Finally, to keep things simple, I’ve only indicated the green layer since we’re only talking density and the shapes of the characteristic curves. You can imagine what a mess these charts would be with the RGB lines of only two characteristic curves overlaid! On the note of color temperature and emulsion tracking, both the PanaFlasher and the Varicon tracked true at 3200°K.
 
 

What is a 10% "Flash?"

In the simplest terms, a 10% Flash is actually .1 over Dmin. If one looks at Dmin, and the density of 1 stop over Dmin, or even 2 stops over Dmin, there is no correlation to a 10% Flash. A 10% Flash on the Reference column would equal:

.56+.1=.66 N-4=.68

It’s a matter of convenience that has evolved over time to call .1 over Dmin a 10% "flash." A 5% "flash" should be .05 over Dmin, and a 15% "flash" should be .15 over Dmin. Probably what happened is someone noticed the decimal point in front of the one and decided it "must mean" 10%. This is another film misnomer like MOS. Anyway, look at the charts. The 5% Flash of the PanaFlasher lines up almost perfectly as a text book 10% Flash. Dmin is .56 and the 5% Flash totals at .67, the 10% Flash density is .69, .13 over Dmin or equal to a "13% Flash, and the 15% Flash density is .80, or .24 over Dmin. As Panavision cautions, you should shoot a test to make sure you use the equipment correctly to achieve the desired results. The reference numbers are only that. The "look" is what matters, and if that "look" is achieved with a "10% Flash" that’s .13 over Dmin, it doesn’t really matter. I plotted the densities to track the sensitometery and note the lifts on the toe of the negative and to see if there is a different response in the two tools.

Now let’s look at the Varicon charts. For this test I set the amount of light with the electronic meter Arri provides with the unit. In practice, I usually set the unit by eye. Arri has further developed a method of setting the contrast with two small dark rectangular sections on the bottom frame of the Varicon. The Varicon isn’t gauged in percentages, Arri chose to use LC (Low Contrast) numbers. Again, the 5274 Reference densities show the potential Dmin and a comparison of densities at each setting. If we look at the Reference Dmin of .57 density, an equivalent 10% Flash would be .1 added density to .57 (Dmin) for a Base Fog density of .67. About LC 1½ would equal a 10% Flash. LC 2 is .13 over Dmin for about a 13% Flash. The other aspect to note is the shape of the characteristic curves. From a comparison of the two, it appears the PanaFlasher raises the Base Fog in the traditional manner, while the Varicon lifts the toe proportionately, grows in density to N-3, and rejoins the reference curve at about N-2. Since these tools work in completely different manners, that must be one of the resulting differences. This bulge, or non-linear aspect, is something I noticed when I did my initial tests on the Varicon in February 1994. I find this aspect of the tool most interesting since I can change just a portion of the imaging characteristics of the negative. If you remember the characteristic curves I published of the Fuji 500, I could use the Varicon with a Rosco Cal Color .60 Blue to shift that portion of the exposure so the emulsions track evenly.

Contrast Compression

In the new information Arri supplies with the Varicon, they mention that cinematographers can compress the highlights by using the tool in a particular manner. Arri supplies a N3 filter to cut the exposure 1 stop when the Varicon is set at a minimum setting of LC2. I know from my experience, I gained almost a stop of film speed with the Varicon set between LC1 and LC2 – my normal setting. One of the reasons I like the Varicon and the PanaFlasher, is the increase in film speed I can achieve. By increasing the EI by one stop, from EI 200 to EI 400, is a tremendous advantage. If one approached this issue from another point, s/he could gain film speed in the toe, in a way that it would lower the highlight density proportionately. That is, say I lighted a scene for EI 200, but exposed it at EI 400, the high lights would have 1 stop less density while the toe is lifted with the flashing tool. This is the thought behind the ND 3 that Arri provides with the Varicon. If a cinematographer approached a project with this in mind, the results could be stunning with rich shadow details while the highlights don’t block up. It’s sounds like a something to use with the slow daylight stocks. I’m sure this approach could be used with the PanaFlasher too, the key is to test and let your artistic sensibilities wander.

Why not let the lab flash the film?

Most labs are reluctant, or won’t, flash film because of the liability involved. Pre-flash liability includes scratching the raw stock, inconsistent density, the raw stock picking up dust or dirt, etc. The point of view of the lab is that they are in the business to develop negative and make prints. The creative aspects of flashing film really lie outside the area of what they want to take responsibility. Plus it’s a rather large inconvienence for them to perform this service. If the cinematographer wants to post flash the film, i.e., flash the film after exposure, the lab still doesn’t want the liability of possibly scratching, or in some other way, damaging the film.

Some thoughts about skip bleach process and negative flashing.

One technique I haven’t had the opportunity to use is the skip bleach process on the negative or one of the steps to the print, or the release print itself. I was talking with Mark Van Horn about this process after viewing the skip bleach test he oversaw at Foto-Kem. I proposed the idea that if the original camera negative were flashed to lift the toe and increase shadow detail, the contrast could be put back in a later stage. He agreed. In fact, he said on a couple of the projects, the prints were flashed to decrease contrast at that stage. There are many aspects I like about the skip bleach process, but one thing I don’t like is the loss of shadow detail. Flashing may provide a partial solution for that.

It’s a wrap & credits.

I’ve avoided a direct comparison of the two tools since I feel with the charts I’ve provided, you can make your own conclusions. I will say that I find it fascinating how two completely different approaches to Flashing have similar results. It is up to the individual to decide what is the best tool to use. The Varicon allows the cinematographer to view the results through the lens. It’s very comforting to actually see the results. The PanaFlasher works in a more traditional approach to Flashing. I know that in future projects I’ll use both tools.

Now while looking at all the charts, I realize the amount of help I had testing these tools. First, I’d like to thank my good friends, operator Rich Benda and 1st AC Michael Bratkowski; Mike Morelli, George Gush, and Don Henderson at Kodak for the new 5274 raw stock, Chuck Kemmer, and Bob Boyce at Foto-Kem. And last but not least the good people at Panavision where I shot the tests, Jim Rodebush particularly, and Franz Wieser at Arri.

Mark Woods