Forum Replies Created
-
Replies
-
Kurosawa shot deep focus with longer anamorphic lenses in the 1960s — it’s just a matter of stopping down the lens. The depth of field with anamorphic is just lower because the focal lengths are longer, but jumping up from Super-35 to FF35 is not a big difference — in terms of anamorphic, if the 35mm sensor area was 18mm tall and the FF35 area was 24mm tall, that’s a 1.33X difference so that is also how much you’d have to stop down to match depth of field once you matched field of view and distance. So getting more depth of field in anamorphic is not insurmountable either by using more light or a higher ISO.
With 35mm film, the advantage of anamorphic over cropping spherical was the larger negative area for less grain, better resolution (though at wide apertures, often spherical lenses are sharper) but that’s less of an issue with digital where the main reason to shoot anamorphic is the anamorphic look (flares, stretched bokeh, some barrel distortion.) If you want a shallow focus or deep focus look, then you can shoot either spherical or anamorphic.
Think of it this way, in the days before digital when film was printed, if we did a lighting contrast ratio test, we printed and projected the results. So we were picking a lighting ratio based on, or factoring in, the contrast of the print stock. The equivalent today would be the viewing display LUT.
Soft lighting was not uncommon in the Silent Era, between natural sunlight being used through cloth on outdoor stages to later when the Cooper-Hewitt lamps were being used. But first Cooper-Hewitts disappeared when sound came along (they were noisy) and to get enough exposure, multiple tungsten lamps were used which had larger bulbs in them (Mazda globes) so a push was made to make tungsten more “precise” in fresnel fixtures, less of a floodlit look was desired. And as film stocks got faster (about 32 ASA for Pan-X in the 1930s until 64 ASA Plus-X came along in 1938), one could do more careful projected lighting effects along with achieving more depth of field. So the classic studio style lighting evolved in the 1930s by choice, not because they didn’t know how to do soft lighting. And it worked well with b&w where you are trying to create depth and separation.
But soft lighting started to appear again in the 1960s on color films when the stock was 50 ASA. But it wasn’t easy to light larger spaces with soft light as opposed to just some close-ups. David Watkin famously lit the set for “Marat/Sade” with one big soft light window — which had 26 10Ks behind a frame of muslin. The African landscape set in “2001” was lit with hundreds of 1K globes. Even today if you ask for a couple hundred of space lights, tungsten or LED, to light a large set for overcast daylight, you will be fighting with the production manager to spend that much money.
Sure, there are differences in color rendition from using bi-color, RGB, RGBW, etc. LEDs and tungsten light — you can see that by looking at spectrum chart, the cheaper LEDs have a “spikey” RGB plot. As they add more color LEDs outside of RGB, they fill in some of the gaps and create a smoother range. On skin tones, you may see the effect of less saturation from LEDs or too much saturation in some wavelength range (some lean towards adding too much magenta to skin for example). When a source light is not continuous-spectrum enough, you lose some color complexity and richness — that can lead to a somewhat blander “band-aid tan/pink” rendition in Caucasian skin for example. But as I said, LEDs are getting better all the time in addressing this.
Practically-speaking, sure, LEDs are far too useful to be ignored and most of the time, the effect on skin tones is not noticed unless one does a side-by-side comparison with tungsten.
Tim Kang does a lot of research into this:
The show LUT is a personal choice. You need a basic conversion from log gamma to a display gamma (Rec.709 or P3) just to view the material with something close to a normal contrast.
But how far you tweak that from the standard is up to you, your taste, and the look desired for the project. Ultimately the LUT is just for monitoring on set and for generating dailies — you’re going to have total freedom to change things in the final grade if necessary.
Lighting should be a creative act more than a technical one. If you’re worried about working too close to the noise floor, then select a lower base ISO. If you want to light for more contrast, darker shadows, or a deeper stop, etc. then light for that because that’s the look you want, not because of some technical reason like wanting to stay above the noise floor. Your base ISO should keep you from getting too much noise unless you try lifting the shadows for more detail in post. So don’t do that, light for the amount of shadow detail you want.
Far too much emphasis is placed on LUTs — 99% of the look of Roger’s cinematography is composition and lighting, and 1% is the LUT.
Study how he composes and lights his scenes and stop thinking about his LUT.
If you’ve ever spent time shooting scenes against blue and greenscreens, you can understand why most cinematographers would rather have anything else! LED screens, grey screens, a painted drop, etc. It’s very hard to balance the light in a day interior — and get the ambience right — relative to the windows… when there is no view. You lose a lot of the reflectivity around the room with light bouncing around and reflecting off of surfaces when you have a big bluescreen out the window.
A 1000W LED set to full warm (below 3200K) for sunset might be enough, it really depends on how dark it is outside and how far away you need to put the light out the window. If you can darken the view beyond with, let’s say, a Double Net scrim on a frame, that might help the balance. Or wait until dusk or for heavier overcast weather. If it is a bright overcast day then the 1000W LED might not overpower the natural daylight enough. It’s hard to say, it’s a bit like asking if a 5K tungsten fresnel outside a window is bright enough for a sunset effect — half the time, it is.
I just by coincidence was looking at this shot in “Barry Lyndon”, which used I think a 10K tungsten fresnel outside the window for a sunset effect (the light is barely visible for a few frames as the camera pans 180 degrees.)
That BTS clip you put up doesn’t show a poor-man’s process, it shows a bluescreen process shot. Poor-man’s process is when you fake driving with no vfx, just a black background at night, or outside with wind machines blowing dust so you can’t tell the car is static, etc. And there is LED work in “Disclaimer”, like the backyard view from her kitchen.
Personally I’d rather use LED screens rather than bluescreens for driving work and get everything “in-camera”.
A lot of that depends on the art department plans for the stage space, it’s often restricted. If they aren’t giving you enough space for the backing and lights, you can see about pushing the set back but often there’s a reason they’ve squeezed you, like another set is in the way, the fire lanes, etc. But I would hope that I’d have at least fifteen feet back for the backing and pipe for the 12Ks. Twenty feet would be even better but the further you push everything back, the higher the lights have to go, the larger the backing has to be, etc. You often run into the limitation of the height of the stage ceiling and greenbeds.
I think the question is too abstract — I mean, I’d place the hard light a city block away to get a sharp pattern like sunlight from it, with the correct fall-off, but that’s hardly practical and the intensity would be useless.
You basically will try and get the hard light as far back as is practical or possible while still maintaining the intensity you want. You also have to factor in whether you are using multiple hard lights for a row of windows and if you need to separate the beams.
I don’t think there is anything wrong with a hard top light in this case, especially since it is not hitting the face. I think if you want the eye to go to the person, the top light has to be hotter on them than the ground to frame left. I tried to play with it below.
In terms of the face, if your bounce card was further upstage so the face was more 1/4 lit by it, not as frontal, I think that would give you some light in the eyes but still keep some darkness.
There may be a very slight advantage with a lower contrast lens in terms of preserving shadow detail in a high contrast situation, but on the other hand, it’s also harder to fix an image from a low-contrast lens with a lot of flaring in a situation where there is too much lens flare and contrast loss – plus there may be resolution loss too in that situation that is harder to fix.
So I tend to agree that a lens with a softer contrast isn’t absolutely necessary if you want to get that look in digital. On the other hand, if the lens gives you the look you want “out of the box” without any adjustments to the LUT, etc. then some cinematographers will prefer that. To me, contrast is a bit down the list of priorities since it can be adjusted in color-correction, as opposed to sharpness, the flare characteristics or the shape of the bokeh – though it all may be tied together. If one wants a softer image from an older lens, it may be that the contrast loss is just something that is part of the equation that the cinematographer accepts as a way of avoiding diffusion filters.
And the problem with post-diffusion is that if the soft look isn’t baked into the original, or at least baked into dailies, it leaves the door open for the studio to change their minds against the will of the director and cinematographer.
