Ask David Mullen ANYTHING

[MichaelHalsell]

David, could you list a couple of films shot with Panavision Primo primes? Also, in your opinion, what lens set matches the Primos in terms of contrast and flare management? Thanks.

 

[David Mullen ASC]

There are too many Primo lensed shoots to mention, almost anything shot on a Panaflex is probably using Primo lenses. The "Pirates of the Caribbean" trilogy is one big-budget example.

I think the Zeiss Ultra Primes are similar.

 

[MichaelHalsell]

Thank you.

 

[Christopher Grant Harvey]

Christopher: we have been discussing this in another thread about full frame and 645 epic (David has made a number of posts over there as well.) At 9K the 645 Epic is about the same size and resolution as actual IMAX film and wouldn't require any multicamera rigs and could use existing medium format still lenses or Imax lenses.

Epic X, Epic FF35 and IMAX... thread:

http://reduser.net/forum/showthread.php?t=38767

Thanks for this.

 

[Christopher Grant Harvey]

One of the strengths and weaknesses of the 3-camera Cinerama was that each third of the final 2.66 frame had its own vanishing point (each camera had a 27mm lens on it by the way, for a total view of 144 degrees.) This created a lot of distortion, particular when the camera was not dead-on to the horizon line. On the other hand, there was no barrel distortion as a super wide-angle lens with a 144 degree view might have. But it had the odd effect on the big screen that no matter where you stared at the image, you seemed to be staring into the center of it.

Thanks for the explanation David.

It seems it is better to wait for a bigger digital sensor then.

 

[Brad Webb]

What is the difference between a zoom and a variable prime? Is there even a difference?

 

[David Mullen ASC]

Not really a difference, though in theory a variable prime is not designed to be zoomed during the shot because artifacts that happen in motion (shifting, ramping, etc.) don't have to be designed out if it's just going to be used as a variable prime. However, those Zeiss Variable Primes worked fine as short zooms.

 

[Colin Edelman]

Sony XDCAM EX3 workflow

Sony XDCAM EX3 workflow

Hi David,
I was linked off the Rubber Monkey Software page. For a RED workflow, I'm doing just fine, but the Sony EX3 camera is where I run into turn. In FCP 6.0.5, I have XDCAM EX 1080p24 quicktimes with timecode created by the camera.

I was hoping you could help me out regarding exporting a still image sequence (tiff or DPX) out of FCP that will preserve the timecode. I'd take this image sequence to a color correct, so they can apply the grade and I'd receive a D5 with same timecode from my dailies so that the conform can use an EDL from FCP.

The alternate workflow I am using now is to make 5 second handles on each cut before making an image sequence and at the conform the flame artist sets in and out point of 5 seconds for each side of the cut. It comes down to eye matching over the exactness of an EDL.

Is there a program equivalent to Monkey Extract that can read the original .mp4 file from an FCP XML file and create timecode embedded DPX sequences or any another workaround?

Thank you,
Colin

 

[Stephen Williams]

What is the difference between a zoom and a variable prime? Is there even a difference?

Hi,

AFAIK the Cooke 20-60 was the first Variable prime, it was of equal quality to Prime lenses at the time it was introduced, the same is true of the Zeiss Variable prime which is as good as ultra primes.

Stephen

 

[Joe G.]

David,

I have a pretty big question. What should the role of tungsten lighting be with the next generation of digital cinema cameras?

I know that you have written how you like the look of tungsten on faces. If the sensors and processing are up to the task, then are we going to want to see more tungsten than daylight? It seems all the classics have that pink/red facial lighting that really stands out.

How do you decide about tungsten vs. daylight bulbs when you shoot digital?

 

[David Mullen ASC]

Not only is tungsten more attractive on skintones, but it is more reliable and consistent. HMI's are finicky and sometimes noisy on a quiet set. LED's and fluorescents have slightly odd color tints to them still.

But the main reason to use tungsten lights is that most night interiors are still lit with 3200K fixtures. Plus a tungsten-lit set can be run through a dimmer board.

So I don't see a lot of good replacements still for tungsten, which means I hope digital cameras will improve how they handle 3200K (if not lower -- after all, candlelight and sodium streetlamps are even lower than 3200K...) I don't see this as a huge hurdle though, already many digital cameras do pretty well in tungsten lighting, they are just even cleaner in daylight balance.

I did 2/3's of the stage work on "Manure" with HMI's, using tungsten for orange lighting effects, and while that worked pretty well for the RED camera, it was a bit of a pain to deal with, sound-wise (noisy ballasts, singing heads) and reliability-wise (lights going out in midtake and needing a few minutes to re-strike.) Output-wise, they were great -- I have no problem getting a good stop when lighting interiors with HMI's.

Generally I light day interiors on location with daylight lighting and night interiors with tungtsen lighting. On soundstages, I'm more likely to use a large tungsten package, partly for cost reasons but also for sound issues, etc. I recently did a TV series though where we had two soundstages and I lit half of the second stage sets with HMI's, partly because I ran out of tungsten stage lighting and had to dip into my location package we used on the road, which includes HMI's. But also this was a large set with one entrance that went out the soundstage door into the street for interior/exterior views, so it made sense to light it to daylight balance.

 

[Jamie Havill]

David, whilst we're on this subject, what would be the best way to control the output of a tungsten light? I've checked up on a few lighting consoles, and they require a 16amp C form input.

Are there any options if your powering the lights of a locations mains supply?

 

[Matt W.]

On the subject of tungsten and skintones, I have a question about metamerism with high color temperature lights...

I'm color correcting a feature I shot in hd and we shot under mixed cloud cover, so not only do we have some soft overcast light and some hard daylight, we have vast differences in color temperature throughout single scenes (my guess: between 4800K and 8000K+)--all recorded to a 5600K whitebalance. When I color correct for this, even once I time out the blue tint of the overcast footage, skintones are much darker in overcast scenes (compared with overall exposure and with the green plants) and foliage is much brighter. Have you experienced this and do you have any tips to mitigate this, either in post or on set?

 

[David Mullen ASC]

David, whilst we're on this subject, what would be the best way to control the output of a tungsten light? I've checked up on a few lighting consoles, and they require a 16amp C form input.

Are there any options if your powering the lights of a locations mains supply?

You can get individual variac dimmers for lights up to 10K, plus there are scrims of course.

Scrims, nets, dimmers, spot-to-flood, using a smaller or bigger unit, heavier or lighter diffusion, backing up or moving in the light, and even ND gels can be used.

 

[David Mullen ASC]

I'm color correcting a feature I shot in hd and we shot under mixed cloud cover, so not only do we have some soft overcast light and some hard daylight, we have vast differences in color temperature throughout single scenes (my guess: between 4800K and 8000K+)--all recorded to a 5600K whitebalance. When I color correct for this, even once I time out the blue tint of the overcast footage, skintones are much darker in overcast scenes (compared with overall exposure and with the green plants) and foliage is much brighter. Have you experienced this and do you have any tips to mitigate this, either in post or on set?

Allen Daviau, in the book "Film Lighting" by Kris Malkiewicz, mentions something similar about timing out a fluorescent green cast causing an increase in contrast in faces.

It may just have something to do with the fact that a face mostly exists in the red channel, and you've overexposed the blue and green channels in relation to the red when you shoot under too much blue or cyan light (or you've underexposed the red information relative to the blue and green information.)

It may also be that all the soft overhead light is being reflected more easily by green foliage (which has a shiny surface) compared to hard sunlight, so it looks brighter.

Also, the toppiness of the overcast light may be causing more fall-off under the eyes, making the face seem darker, plus the sky may be brighter in comparison, being washed-out.

 

[Jamie Havill]

You can get individual variac dimmers for lights up to 10K, plus there are scrims of course.

Scrims, nets, dimmers, spot-to-flood, using a smaller or bigger unit, heavier or lighter diffusion, backing up or moving in the light, and even ND gels can be used.

David, am I correct in assuming the colour temprature of a tungsten light would decrease as you dimmed it?

 

[David Mullen ASC]

David, am I correct in assuming the colour temprature of a tungsten light would decrease as you dimmed it?

Yes.

 

[Jamie Havill]

Yes.

Thank you for your time David, I really appreciate it. That's the last of my silly obvious questions!

 

[Joe G.]

Hi David. Happy New Year.

What would you say about mixing tungsten and daylight in the same frame?

Would daylight work in the background (reducing noise in fill?) while using tungsten for the key? Or is this too drastic a difference?

 

[David Mullen ASC]

You create a lot of problems trying to be too clever by half... if you want the lighting to all be the same color, well, it has to all be the same color. Mixing basically orange and blue lighting together and then trying to time it all to look "white" is not a good idea. It can be fixed in post to some degree, but would require a lot of work pulling color keys, isolating areas, etc.

Now you could mix tungsten and daylight and gel everything halfway between 3200K and 5600K.