Forum Replies Created
-
Replies
-
A shadow is just another tonal value in the greyscale. If this were a scene shot in bright sunlight and a person was wearing a black coat, let’s say it was Zone I (-4 stops) in the Zone System, would you be worried about noise in the black coat compared to a scene where a dark wall was in Zone I? They are the same tone so the noise would be the same, for whatever ISO value you set and for whatever viewing gamma (LUT) you pick.
So if you tested the camera at 800 ISO and were OK with the noise in the lower tonal values, you can light a scene as dark as you’d like, the noise shouldn’t change just because the scene was very dim or dark or the actor was wearing black or standing against a black curtain. The only problem is when you change your mind in post and try to manipulate the shadows, push the color in the shadows or lift them, then noise might start to appear.
I think mercury vapor lamps vary somewhat in color just as Cool White fluorescents do, some have a lot more green in them. But you have to factor in how film stocks of the past reacted to that narrow wavelength in the cyan range versus how a digital sensor would react today.
It gets to the point where you might as well just pick a degree of cyan you like — at the saturation level you like — when using an RGB LED rather than worry about the accuracy of recreating a lamp of the past (unless you are trying to match the LED to a real mercury vapor lamp in the shot.)
Here’s a shot from a movie I did 15 years ago which had a real mercury vapor lamp on the corner of the trailer, and one on a streetlamp, shot on Fuji 500T pushed one-stop. But the pool of light over the phone booth was a tungsten parcan gelled with cyan:
From the internet: “The F-stop is also known as the f-number or f-ratio. In photography, the f-number of a camera lens is the ratio of the system’s focal length or capacity to bend light to the diameter of the entrance pupil of the lens. It is also known as the focal ratio.”
In other words, the f-stop is not simply the size of the aperture, it’s a ratio between the focal length and the diameter of the entrance pupil (which is not necessarily the aperture.) So as the focal length changes in a zoom, the ratio will vary — it takes some engineering to design a zoom lens to compensate for this to achieve a consistent f-stop throughout a zoom (and some big zooms still get darker at the most telephoto end even if in general they keep the f-stop consistent.). So you pay more for a constant f-stop zoom, and often the zoom lens might be bigger as well.
The ISO chosen (i.e. your exposure) determines the dynamic range captured, not the recording format if it’s LogC versus Arriraw.
You are forgetting that if you select a higher ISO you are underexposing the sensor so there is no loss in highlight information recording ProRes compared to recording raw and underexposing, and then brightening the shot. You bake in noise and color temperature in ProRes Log C but there is no loss in dynamic range.
In theory in a controlled environment with no practical or natural lighting of its own illumination level, then doubling the light level should give you the same contrast, the brighter lamp bounces back brighter off of surfaces, etc.
However, visually when you shoot at wide apertures and the background goes soft, it can feel lower in contrast because the highlights and shadows in the distance blur over each other — you can see that on a waveform monitor just pointed at a 11-step grey card, if you throw the image out of focus, most of the waveform signal piles into the middle grey zone even though the signal edges still hit the far targets. The in-focus subject doesn’t change, contrast-wise, but with the blurry background softening contrast back there, the contrast can feel less harsh overall.
Depends on your day exterior situation — most of the time, the headroom on the Alexa at ISO 800 is fine. But I suppose you may have a scene where everyone is dressed in white on a sunny day or the camera is pointed into the ocean with hot sunlight glaring off of it where you may want a bit more headroom, but keep in mind that higher ISO means more noise.
I took this vertical photo of mine and cropped it to a horizontal aspect ratio, so the widescreen version has the same horizontal space as the vertical photo. So in this case you can’t say that the widescreen frame has more space in it:
I don’t think any aspect ratio “reduces screen space” overall, you just trade vertical space over horizontal space depending on the aspect ratio. If you see “Oppenheimer” in IMAX film, I don’t think you’d say that when the image expands vertically to 1.43 : 1 15-perf 70mm IMAX from 2.20 : 1 5-perf 70mm that a loss of screen space had occurred.
It’s hard to top Kurosawa for ‘scope compositions, especially of groups of people.
Leone and Spielberg are worth studying for that, 2.35 : 1 framing, as are Lean’s movies.
“Last Year at Marienbad”, “La Dolce Vita”.
Nicholas Ray and Anthony Mann movies.
Gordon Willis’ compositions in anamorphic movies such as “Klute”, “The Parallax View”, “The Paper Chase”, “Manhattan”, “Comes a Horseman”.
In terms of a looser style of widescreen framing, there are some of Altman’s movies worth looking at such as “McCabe and Mrs. Miller” and “The Long Goodbye” (both shot by Zsigmond.)
You might want to read Sidney Lumet’s book on directing; he’s the sort of director who designed a shooting plan with those sorts of structural ideas, such as in “The Hill” when the focal lengths get wider and the depth of field gets deeper as the story goes, or in “Murder on the Orient Express” where the flashbacks to the interviews with the suspects that Poirot then recalls in his summation speech were shot with wider-angle lenses in a more POV style than the original scenes were shot.
Storaro has talked about the camera style of “Agatha” where the movements and angles were more rigid and linear in the early scenes to suggest Agatha’s entrapment in her marriage, and looser and more fluid in later scenes.
At ISO 400 instead of ISO 800 on the Alexa, you still get the same dynamic range, about 14.5-stops, it’s just that you’ve added more one stop of shadow detail in exchange for losing one stop of highlight detail, plus there is less noise overall.
So the latitude you’ve added is all in the shadow region. But considering how good the Alexa is in terms of highlight detail, often you can get away with that 1-stop overexposure.
Raising the ISO for a non-raw recording format like ProRes 444 Log-C in the Alexa does not necessarily mean you will have more clipping compared to recording Arriraw with the high ISO just as metadata. It’s when your recording codec is in a display gamma like Rec.709 that you get in trouble with lopping off dynamic range.
The best way to control noise is to work at an ISO that gives you a noise level you like or can live with, and then expose consistently so there aren’t major adjustments in post (particularly to the shadows.)
Assuming your codec keeps all the dynamic range in log as a raw recording would, the problems you run into are more related to bit depth, compression artifacts and to color, which is now baked in. So in that case, there can be color channel noise issues if you’ve baked in a certain color bias that you later try to remove in post. With a debayered RGB recording, you have to work in post with the colors you see in the recording.
“Expose to the right” of a histogram is a still photography concept, similar to Ansel Adam’s old adage “expose for the shadows and print for the highlights”. It has very limited use in cinematography because generally the goal is to not expose every shot as if it were unique and stand-alone, but to expose a sequence of shots for a scene so that they fall within the same range of correction (in the old days, so the shots in a scene use printer light values that are similar.) In moviemaking, you have the issue of continuity and consistency of coverage across a sequence (if not the entire movie) which is less relevant in a single still photo.
Let’s say you had a wide master shot of a room where one person is supposed to be standing in a dim area (3-stops under) whereas another person is standing under a light of moderate brightness (normal exposure) while a third person opens a curtain and is hit with sunlight that is a number of stops over.
In coverage of singles, the person who gets hit by the very hot light is naturally “exposed” to the right side of the histogram. The person under the lamp of normal brightness might have to be overexposed by three stops (depending on tones) to get close to the right. The person in the darkness who is 3-stops under in the wide shot would have to be exposed maybe 6-stops brighter to get close to the right side of the histogram!
So first issue is how this looks on the set — who is going to “fix” these radically different exposures to look as they should for the scene? A DIT maybe, otherwise the director is going to wonder why this moody scene is all over the map in terms of exposure in the coverage. And will these corrections be applied to dailies?
Second issue is that the shots that have to be darkened a lot more to get back to the correct look will have less noise than the shots that didn’t need any adjustment. In other words, the person who already was nearly at the clip point by being in the bright light might have the noise of ISO 800, let’s say. But the person who was darkened by a number of stops in post to look correctly dark now has the noise of a low ISO setting.
So for consistency of noise across coverage, you should pick an ISO that works for the whole scene and then expose each set-up for the creative look you want. If you are having noise problems with the dark areas of a particular shot, then your base ISO is too high. Some people might choose to rate an Alexa at ISO 400, for example, instead of ISO 800, just to have a lower base noise level.
Now there is some flexibility here, you can choose to “play it safe” and when you start covering the person in very dim light, you may opt to give them a bit more exposure “just in case” — but I’m talking about a small adjustment, not many f-stops of overexposure to get the subject to the right side of the histogram. You also may choose to use a slightly lower ISO in scenes where noise might be a factor, like when shooting green screens. Or when shooting subjects and sets that are very dark in tonal value.
