I'm wondering if anyone has used the new Low Light UltraCon filter from Tiffen, and what there thoughts about it are. Does it help pull up the shadows in low light for video formats, or did it not perform to your expectations? I've been thinking about getting one for my XL-1, but I'd like to know what you guys thought first.
Post Production Artist
The Low Light Ultra Con is exactly for that -- really low-light scenes that don't have enough ambient light to react to a normal strength Ultra Con (like night exteriors with large black areas). In most lighting situations, the Low Light Ultra Con produces results that are too milky-looking, too hazy. In more typical shooting scenarios, you would use the normal Ultra Cons (#1 through #5).
Ultra Cons are great when you want to lift up your shadows without softening definition or getting halation. But since a lot of video shooters actually want a little softening of definition (and a little halation), they are more likely to use something like ProMists, or even regular Low Con filters instead of Ultra Cons.
I find Ultra Cons useful mainly for day exterior work, when you want to reduce the contrast in harsh sunlight conditions, but not create a diffused look. I haven't used them much in HD yet, though.
Cinematographer / L.A.
>I find Ultra Cons useful mainly for day exterior work, when you want to reduce the >contrast in harsh sunlight conditions,
I know this question is a bit like how long is a piece of string but... Assuming a back/crosslight daylight exterior wide shot on a bright sunny day with a 24mm lens (35mm format) what strength Ultra Cons would you use ? And are there any dangers or traps using these filters ?
Tom Gleeson D.O.P.
The ability of being able to lift detail out of the shadows in film (flashing) as well as in video with a filter that introduces halation is not a phenomenon that is unique to image capture. A while ago there was a story on NPR about an MIT (I think) scientist who found if you add a hum to a recording of unintelligible low sounds you could lift up the hidden bits of sound frequencies that would "borrow" the same bits of frequency from the wash of the hum. He didn't know that there was a similar ability in photography (and videography).
David Mullen says "But since a lot of video shooters actually want a bit of softening of definition (and a little halation)..."
To me, halation is the last thing you want to add to your video image. That makes your $60,000 camera look like a $5,000 industrial model. Halation bad. Crisp clean blacks good.
Having said that, I hardly ever shoot video with out expensive french panty hose stretched behind the lens. That adds halation.
Edwin Myers, Atlanta dp
>To me, halation is the last thing you want to add to your video image. That makes >your $60,000 camera look like a $5,000 industrial model. Halation bad. Crisp clean >blacks good.
I don't know about that -- halation is just a stylistic trick, popular ever since the beginning of photography. Look at all the stuff shot through nets or stretched gauze in the mid 1920's. Robert Richardson has built a career out of halation -- so did Geoffrey Unsworth and Vilmos Zsigmond. I can hardly think of a shot where Allen Daviau did not use some form of ProMist diffusion...
Whether or not you want a little kick or glint or glow off of a hot object is a matter of personal taste. I don't really care if I make a $60,000 camera look like a $5,000 camera; I'm making movies, not selling cameras. If the story calls for me to knock down the quality of a Cooke S4 lens by throwing a diffusion filter over it, I will in a heartbeat.
Besides, with digital color-correction, you can get black blacks AND leave the halation around bright objects. In fact, I wish in film there was a filter that caused halation but NO contrast-reduction (as opposed to these new filters designed NOT to halate. To me, the halation is the most fun part of diffusion!)
Cinematographer / L.A.
>back/crosslight daylight exterior wide shot on a bright sunny day with a 24mm lens >(35mm format) what strength Ultra Cons would you use ? And are there any dangers >or traps using these filters . . .
Assuming you have some hours behind the camera's viewfinder, then I'd suggest simply judging by eye. If you are used to looking through the lens to light, then you should be able to make the same calls on the filters "by inspection." My rule of thumb is, if you're shooting for TK/transfer, then step over the line and go one grade up when making the judgment. If you're shooting for print, call it on the side of caution and stay one grade less than what you like through the finder. You can find the effect of a single filter varying, depending on your taking stop. Filter acts differently shooting wide open as opposed to stopped down. Using any image-affecting filter (beyond simple CC) is a bit of a craft and an art, I think many of us reach the point where we just "know" and don't scientifically quanitize what we're doing, because you can't. Part of the black magic!
One of the dangers is stray or strong ambient light falling on the face of the filter, and I don't mean the common "kick" reflection showing up in the glass. You may have to use top and bottom eyebrows, hard mattes, flying nuns, Duvatyne on the ground, etc., for control purposes. In short, the Ultra Cons are NOT a documentary tool!
Jim Furrer, Director of Photography
Dark Street Films / VGG Systems, Inc.
class="c533">I'm just curious, can I double up two UltraCon filters and get an even greater contrast reduction effect, or will it do some wierd thing to the image that isn't really desireable. I did a couple of experiments with an UltraCon #5 and found that I could get a bit over 1/2 a stop of dynamic range increase (at the expense of deep blacks. Actually I was expecting a bit more, I'm wondering if I'm doing something wrong?). If I'm going to be going into color correction after the shoot anyways (where I can correct to get some good black back in the image, but where I want it instead of where the camera will clip it), can I double up the filters and maybe get up to a stop in the blacks, or am I missing something? Thanks.
Virginia Beach, VA
>I'm just curious, can I double up two UltraCon filters and get an even greater >contrast reduction effect I did a couple of experiments with an UltraCon #5 and found >that I could get a bit over 1/2 a stop of dynamic range increase
Yes, stacking two UltraCons should increase the effect, although you have to deal with the typical problems of using two pieces of glass together.
However, since a filter is not a substitute for fill lighting, you should expect a law of diminishing returns -- once you sort of lift up detail to a range where it can be seen by using an UltraCon, any heavier filtration doesn't produce even greater levels of shadow detail. You'd be just adding a fog level to the blacks that you're just going to be removing later, with no net gain.
Look at flashing -- once you reach a flash level that actually improves shadow detail by bringing it up to the visible range, anymore flash is just milking out the blacks. So once you get a little bit of dynamic range extention for the UltraCon, you aren't going to keep getting MORE dynamic range by using heavier filters, you are just getting weaker and weaker blacks. Which even if you fix in post later, isn't necessarily going to give you any advantage. In fact, as you know, you reach a point where the stronger correction you have to make in post, the more likely some artifact (like noise) will be introduced. The point of a contrast-lowering filter is to allow you to record more information within the limited range that video allows: milky blacks is worthless picture information so you reach a point where you're just wasting data on something other than image detail.
It's like cooking -- if a little salt makes something taste better, then twice as much salt won't make it twice as good.
Cinematographer / L.A.
You didn«t mention with which camera you are going to shoot. Generally i never would use double UC-filtration - you will loose some amount of image detail. On my projects I never used more than an UC 3- because it did affect the look to much. I would rather change the camera-setup. First of all you could raise the black-level a little bit. Then you should use a softer gamma-curve. Finally you could use the black gamma (or blackstretch) -feature to raise the shadows a bit. Using these electronic "helps" will surely give you the missing half stop - without affecting the image detail so much. If you are shooting with the F900 (and it is setup quite well) - you can trust the camera a lot in the shadows... But as it was already mentioned, the best way to get along with hard contrasts is to add a little fill light.
>First of all you could raise the black-level a little bit. Then you should use a softer >gamma-curve. Finally you could use the blackgamma (or blackstretch)-feature to >raise the shadows a bit.
I humbly agree with all that... but raising the black level bugs me. I really like to leave the black level alone, and instruct my engineer to do so unless there's a drastic emergency and he tells me about it first.
A picture without a real black somewhere in the shot just doesn't have the same "punch". I'd go with more fill light first.
There's a local engineer (a VERY good one) who always sets the black level "by eye". He finds something in the shot that looks black to his eye and he makes that black on tape. In some cases that may allow you to cheat things up a little higher than they might be otherwise, just as long as you have that black reference in there.
I think this is the same principle as what happens when you have a shot without a real highlight in it: it doesn't pack as much punch. You need to show the eye as much range as possible, when possible, in order to please it.
I'm curious if you could pull this off on a shot-by-shot basis with an ultracon. I wonder if, in a high contrast situation, you bring the shadows and blacks way up with an ultracon, and then bring them down so that you have a black reference somewhere in the shot, if you end up with more shadow detail. Or are you just defeating yourself by bringing everything up and then pulling it down again?
(I may be shooting a multi-camera cooking show outdoors and on location this next summer so I have a vested interest in finding out the answer to that question before then.)
-Art Adams, S.O.C.
>. . . or are you just defeating yourself by bringing everything up and then pulling it >down again . . .
No, you're not defeating yourself at all. I know a shooter in Boston who swears by the technique for studio video shoots. If you're willing to reset the black levels on a shot-by-shot basis, then doing it while using an UltraCon will definitely increase your dynamic range, all else being equal. It's not a tremendous increase, but very similar to shooting for print with a LowCon filter and then printing down the scene later.
Or, as with the ancient tube-imager video cameras, which used a tiny lamp inside the optical path to artificially lift the black levels with photons and then cranked the blacks back down again electronically to compensate . . . it was called a "bias light" but disappeared with the advent of CCD imagers.
Jim Furrer, Director of Photography
Dark Street Films / VGG Systems, Inc.
>I humbly agree with all that... but raising the black level bugs me. I really like to >leave the black level alone, and instruct my engineer to do so unless there's a >drastic emergency and he tells me about it first.
If you are planning on doing a color-correction later, it's OK to shoot with the blacks slightly lifted and then reset them later. Roland House in Washington D.C. once had a demo of this technique -- sort of creating a flattened look (like a "digital negative") with weaker blacks and slightly underexposed highlights, and then restoring the blacks & contrast in post. This seemed to create less of a harsh "video" look in the end result.
Another engineer, though, told me that a slight lifting of the blacks is OK, since it helps ensure that you haven't crushed any detail in them, but anymore than that is a waste of the limited bandwidth of video -- you want to be recording picture information, not grey blacks.
I usually shoot HD with the master black set at "1" instead of "0" in the HDW-F900, and then restore the blacks in the color-correction later. You do tend to see more noise in the shadows in very low-light situations, but once you correct the blacks in post, the noise goes down again. Anyway, I've color-corrected three HD features do far and this technique of shooting with the blacks very slightly lifted seems to help a little.
As for whether the slightly extra shadow detail from using an UltraCon, which lifts the blacks, is lost once you restore the blacks in post, I think there is some advantage from filtering before recording & compression -- i.e. you get extra information before it get recorded, giving you perhaps an edge in doing a color-correction later. However, no filter is a substiture for fill lighting -- the benefits in shadow detail from flashing or filtering or lifting the blacks will only be minor; after that and you're not really getting any extra information onto the tape.
Cinematographer / L.A.
>If you are planning on doing a color-correction later, it's OK to shoot with the blacks >slightly lifted and then reset them later.
Unfortunately I go into every shoot assuming that my footage will not be touched in any way, other than to perhaps screw it up. None of the projects I've been shooting recently budget for any tape-to-tape color correction time. Most are finished on non-linear editing systems. If anyone makes corrections it's usually the editor, and the corrections usually aren't very pleasant.
I saw Sony's very early "film look" card for the DVW-700 which was VERY low contrast. I was told it was intended to give you the maximum range possible so that you could go into post and tweak it to perfection. I'm sure it
worked perfectly but there was no way I could ever show a director a picture like that on a monitor and keep my job.
It is my experience in the video world that few producers see the need to bring the DP in for on-line color correction, if it happens at all.
I recently got a copy of my first HD project. It looks the same as it did when I shot it, and when shown in HD it always will. I just love that.
-Art Adams, S.O.C.
I think in the feature world, there is a greater expectation that the DP will be given a chance to color-correct the final product if he's available. When shooting film, often the video dailies are unsupervised & inaccurate, so many DP's know that they won't have a chance to color-correct more carefully until the time of the answer print (and the truth is that a number of producers and directors actually feel a little overwhelmed supervising the color-correction on an entire feature and are glad to have the DP there.) So with feature filmmaking, there is a general understanding of the idea of "finishing" the image in the final color timing later.
If you are trying to shoot HD and then later color-correct for an optimal transfer to film, it's very important that the DP be there to supervise the color-correction.
It would be a great shame to shoot a feature in HD with the idea to transfer it to 35mm and NOT count on having access to post-production color-correction tools like a 2K DaVinci with Power Windows. As I tell producers, what's the point of shooting HD instead of film if you don't take advantage of all the tools that digital color-correction allows over traditional film printing? And so far, I have not had any problems explaining to a director that how I'm shooting is optimized not to look the best on the on-set monitor, but for the final transfer to film (they just sort of nod their heads....) When I start a feature HD production, it is with a clear understanding that there will be a tape-to-tape color-correction and that I will be asked to supervise it.
Anyway, the amount of lifting that I employ for the blacks is quite slight and pretty basic, and even if I weren't there to supervise any color-correction, more than likely the colorist would reset the blacks to "0" anyway on their own, especially if I had warned him what I had done in-camera.
It's the same idea in film -- if someone like Geoff Boyle uses the lowest contrast stock out there like Fuji F-400, I'm sure that it is with the intention to color-correct it in post to the degree of contrast that he likes, using a flatter negative for greater flexibility in color-correction. This style of shooting requires the input of the DP in the color-correction session because the negative itself is not an indication of the final contrast desired. The same idea applies to the notion of getting a flatter "digital negative" in HD, a recording that allows maximum flexibility in post color-correction to control contrast, color, black level, etc.
Cinematographer / L.A.
I do absolutely agree with David. I did work for some years as a video-engineer and color-corrected multicamera-shows "on-set" with class-A monitor and a waveform. There I always tried to get as much of the final look as possible, since there was just a reduced colorcorrection in postproduction. But sometimes, when the D.P. changed his mind in the post. He wanted to have not as much contrast and reveal more shadow information.
There we limited ourselves by grading to much on set. Now I did extend my working-area to do both HD-color correction on set and final grading on the davinci 2K as a colorist. This changed my set-work a lot, because its so easy to put down the blacks and create the final look in postproduction if your camera wasn«t setup to "hard". So I keep the blacks between at 1-2 (with F900, gamma table 4)and try to keep the highlights as much as possible (using knee and switching knee-saturation "on") . If you carefully set up the iris and keep an eye on lightning-contrast, you can do pretty everything later on with the 2K.