Lighting Tutorial - Video Samples, Diagram, Camera Settings

cardmaverick

cardmaverick

Well-known member
A few days ago I posted a teaser tutorial for my developing lighting website. Here is a link:

http://www.dvxuser.com/V6/showthread.php?t=93238


The teaser used a DSLR image as an example of a very important principal in digital cinematography, "exposing to the right".
Many found the tutorial helpful, however some wanted to see an example done with video, as well as see the lighting set up.

Here it is!

I recreated the photograph using an HVX-200 and M2 adapter. I shot with a Nikon 50mm lens. The format was DVCPRO HD 720p 24p. And yes, it stars me again, LOL :cheesy:

Final Clip:

http://www.chrisnuzzaco.com/final.mov

RAW Clip:

http://www.chrisnuzzaco.com/raw.mov

Comparison Clip:

http://www.chrisnuzzaco.com/comparison.mov

Lighting Setup:

The first thing I do with every setup is ask myself this simple question:

"Where in the world is the light supposed to be coming from?"

For this, I decided it should come from the side, and be a soft light that gradually falls off into a dramatic shadow.
Knowing that I would be using an M2 I automatically grabbed my biggest light, an Arri 2k. I set it up a few feet away from my
sitting position, and put grid cloth a few feet in front of it.

Next, I setup a light stand in the spot that I would later occupy and put a neutral colored hat on top. This hat became my subject
to light.

NOTE: I'm about to dive into some exposure settings... I have many methods for figuring it out, this only represents one of many! I also do not get into the technique, thats
a subject for another day!

I took a spot meter reading off of the bright and dark sides of my hat. It read about 4 stops difference. definantly a little too much for what I was going to do in post.
I then set up a reflector on the opposite side. I had a new reading of about 3 2/3 stops. Not much improvement, but I knew that other lights
would be coming into the equation and changing the ratio, so I let it got.

Next were my back lights. I had two. One for my face (650 watt fresnel), the other for my shoulder (300 watt fresnel). I set these up to act as rim lights on my darker side.
They read about 1 stop brighter than my key light reflectively. By this time my shadow side was only 3 stops darker than my key reflectively. Just right. I set my exposure via a
reflective reading from a patch lit by my main key. By the way, flag 2 in the diagram is keep light from the 2 K off of my lens.

Camera Settings:

Pedestal +15
detail level 0
V detail level 0
detail coring -2
chroma level +7
chroma phase 0
color temp 0
gama cinelike D
matrix cine-like
skin tone detail 0ff
operation Film Cam
180 degree shutter
24 FPS
camera f/stop 2.4
Nikon 50mm f/stop 4.5
white balance preset 3200 K

The WHY?

Why +15 on the pedestal and +7 on the chroma level? Post. Thats why. Flexibility. Thats why. The camera compresses my color, so I better start off with as much as I can get right? Right.
Because I can always desaturate without hurting the look of the footage. As for the pedestal, it's all about exposure flexibility, and pixel quality. A bright recording is cleaner than a dark one.
So I chose to get as much information as I could.

Post:

All posting done in After Effects 7.0

Simple curves, and color balance. I also tweaked the blue in my shirt. I would get more into this, but I'm gonna hold off and save it. Plus it also takes a while to explain if you
have never used After Effects before. By the way, I did mask out the flags and diffusion, not enough room in my room....

Here is a simple diagram of the lighting setup:


lightsetup.jpg




Even though this tutorial was a bit quick and dirty, I hope you enjoyed it and learned something from it! The future ones will be much better and feature some on set photos. One of my goals is to create come quicktime VR's of the sets so you can "walk around" the set up. I'm also planning on pulling a video capture from my desktop for the posting work, complete with audio description of whats happening.

Thanks for reading!
 
Last edited:
Nice work Chris. I'd love to see those tutorials once you have them working.
I also liked the idea of having the camera giving you as much information as possible and then fixing it all in post. though I'm sure it requires a lot of testing to get the fear of not exposing right on set. knowing that it will look amazing on post.
 
siniarch said:
Nice work Chris. I'd love to see those tutorials once you have them working.
I also liked the idea of having the camera giving you as much information as possible and then fixing it all in post. though I'm sure it requires a lot of testing to get the fear of not exposing right on set. knowing that it will look amazing on post.
Click to expand...

I wouldn't quite call it fixing in post. It's more like finishing in post. I tell all of my directors and producers (everyone actually!) that lighting the image on set is only 50% of the work. The image isn't finished till post. As for getting past the fear of exposing wrong, well, I can't say I've ever been to nervous, but then again, I've always shot with light meters and such. I suppose it will take time for people who traditionally crush their blacks in camera (I've always been very scarred of doing that actually!). Funny you mention it though. One director was a little wary of what I was doing during production, but after color correction, he was like "Holy crap!".

The first tutorial is going to be pretty killer man. The example is a varicam shoot I have this month. I'll be teaching how you deal with large interior shots that have cielings visible in the frame. When I say big, I mean like 60 feet long big....:Drogar-Love(DBG):
 
I can understand not crushing the blacks into oblivion in-camera, but what would happen if you tried to get the image as close as possible, in-camera, to that final image? Wouldn't that give you the image of the highest possible quality?

Would you recommend this technique for an SD camera like the DVX100? It seems that with the pedestal set so high the blacks would be so close to the mid/white levels that you would have to stretch the range out quite a bit, which would create an image with a lot of noise.

Is this technique only good for really high contrast shots like the example? Would you set the pedestal lower for a more normal look?
 
agcohn said:
I can understand not crushing the blacks into oblivion in-camera, but what would happen if you tried to get the image as close as possible, in-camera, to that final image? Wouldn't that give you the image of the highest possible quality?

Would you recommend this technique for an SD camera like the DVX100? It seems that with the pedestal set so high the blacks would be so close to the mid/white levels that you would have to stretch the range out quite a bit, which would create an image with a lot of noise.

Is this technique only good for really high contrast shots like the example? Would you set the pedestal lower for a more normal look?
Click to expand...
It all depends, thats a wide open question. As for setting it too high, yes you could. The risk is accidentally recording seperate tones as the same tone. When you shoot this way, you need to light for it properly. Also keep in mind, the technique provides greater flexibility in post, where you have vastly greater control over the image. Just for the record, I don't always shoot with the pedestal up at +15. Every image I make is custom, so are my approaches. I do however recommend setting the cameras colors in camera as close as you can. When I can get around to it, I'll post some high res screen grabs.

NOTE!

Many people forget about project bit settings....If you color correct at 8 bits, you will see noise pop up, but turn AE into a 16 bit project, and I think you'll be pleasantly suprised, especially if you work with DVCPRO HD. It's the details that kill... Also color correctin technique can have a HUGE effect on noise when you stretch the luminance out, same for color, other programs do it better, and even within AE 7.0 some techniques are FAR better than others when it comes to not accetuating image noise.
 
Last edited:
cardmaverick said:
It all depends, thats a wide open question. As for setting it too high, yes you could. The risk is accidentally recording seperate tones as the same tone. When you shoot this way, you need to light for it properly. Also keep in mind, the technique provides greater flexibility in post, where you have vastly greater control over the image. Just for the record, I don't always shoot with the pedestal up at +15. Every image I make is custom, so are my approaches. I do however recommend setting the cameras colors in camera as close as you can. When I can get around to it, I'll post some high res screen grabs.

NOTE!

Many people forget about project bit settings....If you color correct at 8 bits, you will see noise pop up, but turn AE into a 16 bit project, and I think you'll be pleasantly suprised, especially if you work with DVCPRO HD. It's the details that kill... Also color correctin technique can have a HUGE effect on noise when you stretch the luminance out, same for color, other programs do it better, and even within AE 7.0 some techniques are FAR better than others when it comes to not accetuating image noise.
Click to expand...
That's why I asked about SD, because there really is not a lot of room to stretch out the exposure in post. DVCPRO HD is over 2.5x the resolution of DV25 footage, so of course you'll be able to stretch out the levels more without causing a significant increase in grain or noise, and HDCAM is 6x the resolution.

What would be a safe level to set the pedestal at when using a camera like the DVX?
 
agcohn said:
That's why I asked about SD, because there really is not a lot of room to stretch out the exposure in post. DVCPRO HD is over 2.5x the resolution of DV25 footage, so of course you'll be able to stretch out the levels more without causing a significant increase in grain or noise, and HDCAM is 6x the resolution.

What would be a safe level to set the pedestal at when using a camera like the DVX?
Click to expand...

I think what you really mean is dynamic range, not resolution. And by the way, I haven't even scratched the surface of my posting techniques yet....I'm not really stretching all the levels out per se.... And remember, I've modified my lighting approach to fit this kind of workflow as well. As for DV, I'd have to do some tests before answering, it's been a long time since I worked with it, and when I was, I hadn't employed this technique. Also remember, I'm not only changing the pedestal, I'm also changing the knee, when you do that, you lighten up the footage, but also remap tonal values, causing them to separate more. Off the cuff, be careful with DV and do tests. As for how I handle the footage in post, well, I'll reveal my tricks later.

One of the big benefits of my approach are the options it can provide.
 
hey Chris,
Now you've really peaked my interest. I thank you for letting us know about your techniques and methods of working. I for one, love to learn about this stuff and I'm always grateful to have someone teach me. :)
By the way, how old are you? you seem to know a lot for being so young. and your website is great. I really enjoyed the photography and the videos.
Good work. (inspiring!!!)
Thanks you very much.
Sincerely,
Luis
 
agcohn said:
I can understand not crushing the blacks into oblivion in-camera, but what would happen if you tried to get the image as close as possible, in-camera, to that final image? Wouldn't that give you the image of the highest possible quality?
Click to expand...

That depends on what you mean by quality. If you have only (3x)8 bit data then you usually wouldn't want to waste a bit. Setting pedestal very high is like leaving a lot of headroom in audio. You can afford a lot if you're at 24 bit, you can afford less if at 16 bit and you can hardly afford any if you're at 8 bit.

But you can also lose quality by over- or underexposing so that clipping appears where you wouldn't want it.

So it's trading tonal range for flexibility.
 
TheAnimal said:
That depends on what you mean by quality. If you have only (3x)8 bit data then you usually wouldn't want to waste a bit. Setting pedestal very high is like leaving a lot of headroom in audio. You can afford a lot if you're at 24 bit, you can afford less if at 16 bit and you can hardly afford any if you're at 8 bit.

But you can also lose quality by over- or underexposing so that clipping appears where you wouldn't want it.

So it's trading tonal range for flexibility.
Click to expand...

You hit it just right! It is a compromise, but every approach is isn't it?

There are certain looks you just can't pull off in camera, or at least not with an HVX200, given the rather limited options for image control inside it. If I could have all of my Affter Effects tools inside the HVX, compete with actual scopes to look at (lets face it, -15 just looks like a number to my eyeballs), you can be sure I'd take advantage of that.

I've shot DVCPRO HD with lower pedestal settings and still been disappointed with the results though.

For HVX200 (DVCPRO HD) bottom line is this:

If you want the cleanest image possible, do it in camera, don't touch it in post.

Cons: There are only so many things you can control. Once you set the contrast and brightness, you're stuck with it (unless macroblocking and banding are your kind of thing..). High level of commitment. Little flexibility.

If you want the most options for your buck, do it in post.

Cons: Depending on how you post, and how extreme you do it, noise can pop up. Again, the amount is dependent on how extreme you go. I've shot with a raised pedestal and had footage that didn't look any more noisey after color correction than the raw footage, but it all depends on several variables.....

For DV:

Quite simply, Test it out. It's not like you can't raise it up, you just can't go as high up. As for going down, well its extremely noisey should you ever need to make any kind of adjustments.

I'm preping some more samples shot with an HVX200, stock lens. I'll be sure to dive more into depth (no pun intended) about bit depth and how it affects your exposure options and dynamic range. When I get the site going, and it won't be spread out over a bunch of threads, LOL. I probably should have touched on this more in the tutorials opening where I mentioned formats, think I'll go add this note....

Last but not least (and I know I'll say this again)...Test!
 
You must log in or register to reply here.
Back
Top