Ask David Mullen ANYTHING

[David Mullen ASC]

Hi David, thanks for being available for interaction with us mere mortals ;op

Do you still flash film every once in a while or do you find it unnecessary with modern stocks/post?

What's your opinion on the aesthetic benefits/differences of using a large (let's say 12'x12') silk for diffusion versus a smaller one. I know that you'd be able to place it further away and get a similar effect to a smaller one closer to the subject, etc... but please give me some practical example(s) of what a larger diffuser would allow you to achieve (i.e. more natural ambiance fill, etc...)

It's become less necessary now with wide latitude stocks combined with D.I. capabilities to alter color and contrast. I'd probably now only use it for something finishing photochemically where I had to bake the look into the original.

Larger diffusers (if you are talking about diffusing the sun overhead) obviously cover more area.

If you are talking about soft lighting techniques involving large frames of diffusion, the size of the diffusion (assuming it is filled evenly from edge to edge with light) plus the distance to the subject -- i.e. the relative size of the diffusion from the perspective of the subject -- determines the softness of the light. Heavier and lighter diffusion just affects your ability to fill the frame evenly or leave some hotter spot in the center, or how they light spreads (silks tend to spread the light in a star-shape.)

So a 4'x4' diffusion up close may produce the same softness of shadow as a 20'x20' diffusion farther back, if to the subject, they both appear to be the same size in the field of view.

But the difference is fall-off -- the diffusion becomes the defacto light source, not the light behind it, and if the subject moves closer or farther to the diffusion frame, they can become much brighter or darker if the diffusion is very close to them. With a large diffusion farther off, the fall-off is more gradual so an actor can move around without getting much brighter or darker. But with less extreme fall-off, it also means that the foreground and background are also getting more of the soft light compared to a smaller frame up close with tends to just light the subject.

When diffusion is very close to the face, you can even see the rapid fall-off on the skin, where the cheek closer to the frame gets hotter compared to the opposite cheek. Sometimes that mentally tells the viewer that the light must be very close, which works if the light is simulating a close lampshade but not so good if the light is simulating a distant soft source like an overcast sky or a large picture window across the room.

 

[Tom Lowe]

I don't recall the shot well-enough to tell you whether or not they did a speed ramp during it to go to slow-motion, but it's possible.

Back when it was shot, where there tricks that could have been used to make 24fps film suddenly ramp and look like 48fps film? Obviously, these days, with digital, that is very simple to do on After Effects or Twixtor with frame interpolation.

 

[David Mullen ASC]

Back then you would have just ramped the speed of the camera motor and compensated with either an f-stop or shutter angle change to maintain the same exposure. People still do that -- it's hard to beat actually shooting at the correct speeds you want for a motion effect as opposed to doing it in post. The main problem with on-camera speed ramps is the timing of the ramp, getting it to start and stop at the exact moment you want it to, as opposed to picking the moment in post.

Some people nowadays shoot the whole shot slow-motion and correct the speed in post for the normal motion section of the speed ramp, but that tends to look choppy because of the shorter per frame shutter speed. In other words, taking 48 fps footage shot with a 180 degree shutter angle (so 1/96th per frame) and converting it to 24 fps tends to look like you shot at 24 fps with a 90 degree shutter angle (also 1/96th per frame). Unless you apply some post effect to add more motion blur back in.

 

[Tom Lowe]

That's probably how they did it, with an in-camera ramp. That's pretty ballsy, though, because couldn't the ramp cause problems if it wasn't pulled off perfectly? That would really suck to do that entire 5-minute (or whatever) steadicam move from the crane through the entire club with all the actors hitting their marks and then ruin it with a badly shot ramp at the very end.... hehe.

Would the f/stop change on the ramp, for example, be automated, or something the operator was doing manually?

 

[David Mullen ASC]

Usually you set-up an external motor that automatically adjusts the iris ring to match the speed change, or use a camera with an electronic shutter control that changes the shutter angle to match the frame rate. Just a question of whether the change in depth of field is more distracting than a change in motion blur, plus you are more limited in how far you can compensate exposure by using the shutter angle.

 

[Ted Chu]

Green Gel and Tungsten Light

Green Gel and Tungsten Light

Shot an insert with the Red. Used a tungsten light and a green gel pack on the other tungsten light. Could not get the Red to read the green, kept looking very blue. I know the clients can color correct that easily but you know how commercial clients can be! How could I get the Red to read the tungsten light and the green light? Didn't have any HMI's in my package. Thanks.

 

[David Mullen ASC]

That's odd... because if anything, the RED is sometimes too sensitive to any green bias or tint in a light, not the other way around. So it seems weird that the green gel would not read. Maybe the TINT setting in the camera was adding a magenta shift that was cancelling the green gel effect. Maybe in your settings for output, the correction for 3200K was adding magenta / reducing green (to reduce yellow). I don't know.

 

[Ted Chu]

Green and 3200K

Green and 3200K

I had the RED set on 3200K and all the color options were on default, 0 or 1.0. I assumed the Red did not like the green or could not resolve the large difference between the 3200K (actually 3150K) light and green. I'll post a JPEG later. Thanks.

 

[Michel Hafner]

In ACM October 2008 Lubezki laments when we was given HD Dailies only for "Burn After Reading": In HD you can't... judge focus.
Why does he say that? Do we need 4K now to judge focus?

 

[Sarah C.]

Hi miha,
One problem with DVD dailies is the screens(monitors/hdtv's) they are viewed on... some add sharpening to the picture. Also, some don't have the correct contrast and can add an apparant sharpness. (I'm sure Mr. Mullen/others here could write from greater experience but this is what I saw on the recent feature I worked on)
~Sarah

 

[David Mullen ASC]

In ACM October 2008 Lubezki laments when we was given HD Dailies only for "Burn After Reading": In HD you can't... judge focus.
Why does he say that? Do we need 4K now to judge focus?

He may mean that compared to projected 35mm prints on a screen, it's hard to judge focus in HD dailies viewed on a monitor. 1080P projected on a large screen should be sharp enough to see most problems but even then, some things may be slipping by.

 

[Michael Lindsay]

Shot an insert with the Red. Used a tungsten light and a green gel pack on the other tungsten light. Could not get the Red to read the green, kept looking very blue. I know the clients can color correct that easily but you know how commercial clients can be! How could I get the Red to read the tungsten light and the green light? Didn't have any HMI's in my package. Thanks.

Despite this seeming impossible with Reds green sensitivity. I've seen this. I suspect Red has an uneven green sensitivity .

Despite having seen a green screen end up with a blue tinge it still keyed a treat (better than anything else I've used).

Michael

 

[Sarah C.]

Does this green issue have to do with the in-camera interpretation of the Bayer-Pattern Imaging Sensor?

~Sarah

 

[Deanan]

Despite this seeming impossible with Reds green sensitivity. I've seen this. I suspect Red has an uneven green sensitivity .

Despite having seen a green screen end up with a blue tinge it still keyed a treat (better than anything else I've used).

Michael

I've seen a couple of cases of customer footage regarding this and in each case it was because WB was set too high or too low. White balancing to something white in the shot fixed the balance.

 

[Seth Iliff]

Mr. Mullen,
I was reading earlier you're explanation of Roger Deakins' night lighting in No Country. You suggested to just not putting yourself in that situation, but what are you're thoughts on day for night.... the most recent example I can think of being, The Proposition. I really enjoyed those huge nighttime landscapes. Do you like this effect?

 

[David Mullen ASC]

I grew up in the desert and when you are away from any other light sources, a moonlit landscape can seem like a day-for-night effect in a movie, other than the dark sky and stars. It's just that a lot of viewers don't have that personal experience of seeing large areas of desert illuminated by a full moon.

I posted these before, but here are some real night shots I took with my Nikon under moonlight, shot with a tungsten preset for a blue cast, underexposed by about two or three stops:

And this was exposed longer in the same moonlight:

So you see that daylight and moonlight are rather similar.

The main problem with day-for-night, besides the dark night sky, is if any other light sources appear in the frame, like car headlights, a flashlight, campfire, etc.

 

[Matthew Greene]

That last one is pretty awesome in making your point.

 

[Charles Angus]

I have noticed that before myself - odd how our perception of night is so different.

 

[Dj Joofa]

odd how our perception of night is so different.

Some of the color shift at night happens because of the Purkinje effect:

http://en.wikipedia.org/wiki/Purkinje_effect

 

[Matthew Greene]

Thanks joofa, that's new to me. I've had pretty good results at lighting realistic ext. nights although I've never lit large areas, just small ones. All this info is good to know to have the know how when it comes to making decisions.