Knowing how to determine how powerful of a light you need

morgan_moore said:
Great paul but..

How many skypanels for a 100X200ft warehouse?

If the shoot is too dark ill fire you. If there are unused but paid for lights on the truck - Ill fire you.
Click to expand...

Maybe the former, Sam, but not likely the latter. Sure, if the package was exorbitant (which SKypanels in quantity are) and you go way overboard on the overuse, that's not ideal. But in reality, especially with Skypanels, you rig them all up and then dial them down if you have too much light. I have yet to have a producer ask me what my dimmer board settings are. They have other fish to fry.

It's not just about maximum firepower, sometimes it's how many units do you need to create a big enough source. 10 spread out Skypanels blasting at full power is a different looking source than 25 of them at 40%. For a permanent set that I lit last year on an NBC series, we installed 4x4 Quasars into fluorescent housings in the ceiling of the set as well as modified "china hat" hanging fixtures with LED sticks mounted in to them. Most of the time we ran both of those units under 25%, but the point was that I had two extra stops available to me if we needed to go high speed or skinny shutter or something to that effect. In a situation where that kind of thinking runs afoul of the budget, I will point out that it leaves us vulnerable for those kinds of situations.
 
I'm thinking back to the "Les Mis" sketch we did for Key & Peele on the Universal backlot. I talk about it in this thread at length, but the specific part I was thinking about was the 20x20 of diffusion we rigged with 5 2K open face (blondes) shooting down through it with a deep cyan gel. In a perfect world I would have dry run this on a stage by building the frame vertically and mounting the lights on stands so we could spot the optimal distance between heads and diffusion (and what type of diffusion) to help us learn how to build the eventual rig, and what the output through the color and diffusion would end up being. Unfortunately, no time or space to do so. So it was pure instinct that led me to the design of that unit and the choice and number of heads, and it wasn't until it landed in position up in the air and I took a meter to it that I realized we were exactly where we needed to be stop-wise. In that instance, the photometrics of the blondes coupled with the transmission percentage of the gel would have been a stepping off point but the diffusion could only be calculated to a certain point because of the distance to the ground--the falloff would have not been as mathematical to calculate in this real world situation.
 
CharlesPapert said:
Maybe the former, Sam, but not likely the latter. Sure, if the package was exorbitant (which SKypanels in quantity are) and you go way overboard on the overuse, that's not ideal. But in reality, especially with Skypanels, you rig them all up and then dial them down if you have too much light. I have yet to have a producer ask me what my dimmer board settings are. They have other fish to fry.

It's not just about maximum firepower, sometimes it's how many units do you need to create a big enough source. 10 spread out Skypanels blasting at full power is a different looking source than 25 of them at 40%. For a permanent set that I lit last year on an NBC series, we installed 4x4 Quasars into fluorescent housings in the ceiling of the set as well as modified "china hat" hanging fixtures with LED sticks mounted in to them. Most of the time we ran both of those units under 25%, but the point was that I had two extra stops available to me if we needed to go high speed or skinny shutter or something to that effect. In a situation where that kind of thinking runs afoul of the budget, I will point out that it leaves us vulnerable for those kinds of situations.
Click to expand...

Indeed.

The overall point Im trying to make is that it is probably worth using some maths before embarking on a larger setup that is beyond ones norms (your norm being much larger than my norm of 4 led lights)

I did the big shed wharehouse through a 12.12, and also the night ext of a stately home, both beyond my norms. The wharehous we clearly could not fight the sun, the stately, worked but Id have liked the fixtures higher and further back.

I have coming up a (rickshaw/movi?) move through a large shop or supermarket, the camera will open on a crowd and push in onto a single individual. Not only will a large area need lighting but also the lights will have to be hidden or hung not to be seen in the opening wide. It is therefore important that I spec a big enough package and dont have to dink too much on the day :)
 
morgan_moore said:
I think it is a hard question and the main answer is experience.

One of my weaknesses as a DP would be exactly that.. how powerful are big lights that I dont own and have not used?
Click to expand...

The big lights aren't quite as mysterious once you get your head around the numbers. There's nothing fundamentally all that different between a big light and a small one. In the case of Arrimaxes or Fresnels the relationship between the output of the fixture, and its wattage, is pretty consistent (you can check this with photometric calculators).

And since the lumens-per-watt emitted by tungsten and HMI sources stays reasonably consistent, as your wattage doubles, so does your output. You can basically just plan around each step up in wattage, providing an additional stop of light.

So a 2500w HMI is going to give you a stop more than a 1200w. A 4000w is going to give a bit less than a full stop over a 2500w. A 9000w is going to give you a bit more than a full stop over a 4000w, and an 18000w is going to give you a stop more than a 9000w.

So an 18kw Arrimax, is basically giving you about 3 stops more than a little 1800w M18.

That's all there is to it.

Now when you start thinking of it that way, the whole thing becomes much less daunting, right?
 
Indeed. But with my own package I can pretty much do without thinking.

So for example my supermarket. Lets guess the photographed set is 30ft by 30ft, the ceiling 15ft high and white and I want 800 5.6.
I plan four lights in a 35ft by 35ft square bashed off the ceiling. to create the general ambiencen.

What shall I order 300d? 2.5 hmi 4k hmi
 
This is indeed cool.. http://calc.arri.de/calculator

Given my 30ft square set 4 arri 60s bashed off the 'perfect mirror' roof, shoould be about T3.5

Id want a stop more and some safety so Id say 8 required.

We wont have the budget for 8 skypanels.. but at least I did the super rough math.

2 M18s however look like T8 which is much more in the park.

S
 
morgan_moore said:
With 10 fixtures you might get the emotion of unhappiness at being fired for lighting the scene to T0.5/400ISO and the whole shoot being under exposed or needing to be shot at ISO 10000.
Click to expand...

...or you might get hired by HBO... ;)
 
morgan_moore said:
This is indeed cool.. http://calc.arri.de/calculator

Given my 30ft square set 4 arri 60s bashed off the 'perfect mirror' roof, shoould be about T3.5

Id want a stop more and some safety so Id say 8 required.

We wont have the budget for 8 skypanels.. but at least I did the super rough math.

2 M18s however look like T8 which is much more in the park.

S
Click to expand...

And that is the "danger" of going off photometrics. With the situation as you described it Sam, I don't feel that two M18's would get you to an uniform 5.6 across a 30x30 space. You wouldn't be punching them straight up into the ceiling to get even spread as it would fall off towards the center, you'd want to angle them so that the beam spreads out further, and I'd probably spot them in about 1/3 to 1/2 of the way so you get more output further from the light (leveraging the inverse square law). To be truly uniform you'd be better off with four units vs two. Not knowing what the ambient base is in the space already (did you say if you were augmenting existing lighting or turning it off? I'll assume turning them off) and the nature of the ceiling itself (it is surely broken up with built in fixtures, and how white is it, is it dirty etc), I'd be more inclined to spec 4 M40's to be safe than 4 M18's. That may well be overkill, but that is exactly what we are discussing here--you don't want to be squeaking for that last half a stop.

By the way this is an instance where I think the M series HMI's are better suited than Skypanels, because of that ability to push the beam further which spreads it out more. 4 Skypanels in the corners of the space would make a hot ring around the perimeter with falloff in the middle.

So this is all to say--photometrics are a start but the real-life variables of bouncing of a ceiling that is not a pure and calculable entity means number crunching will only get you so far.
 
Besides everything here one thing that intimidates me is distance light is from whatever it’s hitting. I understand inverse square law but unless you know roughly where your lights will be (not an issue I guess if there’s a grid or small room etc) seems like you could get yourself in a lot of trouble if you end up having to move your lights 3 or 4 times further away than you initially thought.
 
Josh Bass said:
Besides everything here one thing that intimidates me is distance light is from whatever it’s hitting. I understand inverse square law but unless you know roughly where your lights will be (not an issue I guess if there’s a grid or small room etc) seems like you could get yourself in a lot of trouble if you end up having to move your lights 3 or 4 times further away than you initially thought.
Click to expand...

This has happened to all of us. As Clarkson says, "Power!"
 
A single high speed scanning laser placed near the camera lens can light any scene with complete control (diffuse surfaces). Real world ray tracing :)

Coupled with LIDAR you also get per pixel 3D depth- green screen with a green screen or key (generalized background subtraction), 3D scene relighting in post, etc.
 
Josh Bass said:
Youre saying get 4x more than you think youll need in the first place? In a perfect world, sure.
Click to expand...

One of the first things I ask when we scout a location of the director is: how wide will be seeing in a given shot? and/or what are the overall sight lines (what is "fourth wall"). The more we need to see at once, or with minimal relight, generally speaking the bigger the units due to distance from subject. You can balance or overpower a closeup with a hot window in the background with a fairly minimal sized unit, but if you need to go head to toe, that will dictate probably 4x the output to cover the spread and distance.

An inexperienced director may be convinced they will only see X and then on the day want to see Y, so it is good to be able to explain ahead of time what the limitations will be so they don't get frustrated when you can't deliver to their whims on the day

To be clear, I'm well aware that many people here are not in a position to order the big guns or numerous units for a given job due to budget. Even at my level I rarely get the size package that I want without compromise. The key is managing expectations and having discussions ahead of time--if a line producer forces my package to be cut down, I will go to the director and explain what the ramifications will be to our creative plan, and sometimes they will take up the battle with the producer. If you say "we'll just see what happens", it is more likely to end up making you look bad on the day.
 
CharlesPapert said:
One of the first things I ask when we scout a location of the director is: how wide will be seeing in a given shot? and/or what are the overall sight lines (what is "fourth wall"). The more we need to see at once, or with minimal relight, generally speaking the bigger the units due to distance from subject. You can balance or overpower a closeup with a hot window in the background with a fairly minimal sized unit, but if you need to go head to toe, that will dictate probably 4x the output to cover the spread and distance.

An inexperienced director may be convinced they will only see X and then on the day want to see Y, so it is good to be able to explain ahead of time what the limitations will be so they don't get frustrated when you can't deliver to their whims on the day

To be clear, I'm well aware that many people here are not in a position to order the big guns or numerous units for a given job due to budget. Even at my level I rarely get the size package that I want without compromise. The key is managing expectations and having discussions ahead of time--if a line producer forces my package to be cut down, I will go to the director and explain what the ramifications will be to our creative plan, and sometimes they will take up the battle with the producer. If you say "we'll just see what happens", it is more likely to end up making you look bad on the day.
Click to expand...

To add on to what Charles is saying, if I may, part of being a good DP is, "say it all together" managing expectations. It's simple in a way, you can only light what the budget and resources you are given will allow you to do. When I get clients who are unrealistic about what they want to shoot and how they want to shoot, it's my job to educate them, not to be a naysayer, to be a reality check. You as DP, working with your producer, have to be able to manage expectations. If the budget can't support the director's vision, your worth as a DP comes in when you can figure out creative ways to get around limitations. That's the beauty of production, there are soooooo many ways to tell a story. Often, a director needs to be shown some viable alternatives that will still let them tell the story they want to tell, but visually, often, in a different way than they may have imagined. I can't tell you how many big epic scale scenes I have lit and shot where in retrospect, a lot of it ends up on the edit room floor or is barely used or just CUs or mediums are used and we killed ourselves to get that big, grand shot the director envisioned. It's not our job to tell the director what to do obviously, but it is our job to help your director with different ways to do things, different choices for them to choose from.

This is when you are working with a director. When you are working with corporate clients, ad agencies or music labels, it's easier in a way because they aren't always the creatives. When I have ad agency or studio people start talking about VR, shooting this with an Arri instead of a C300MKII, lighting it this way, why don't we have a Technocrane, it's simple. "Write the check and I'll have it on set tomorrow. It costs this much. Yes or no?". Usually when confronted with the hard numbers of what a specific piece or pieces of gear will cost them, it's simple. "Pay for the gear you think we need or we won't have it." It's different than working with a director whose vision your trying to fulfill.
 
Last edited:
Also, knowing your limits and when to defer to someone with more knowledge/experience. There have been times I've been asked to put something together that's a little larger than what I normally do and I'll recommend that we get a dedicated gaffer to bring in and then I tell them what we're doing/want to achieve and they handle the "big picture" from there and then we can tweak, if necessary after that.
 
I have lit entire sets purely by the numbers. Used a light meter reading in footcandles, predetermined my key/fill ratios, my stop, etc. I did this once for a B&W film that was replicating a noir detective film look. We were shooting on very slow stock which meant a high light level so you really couldn’t just trust your eyes. It was also very early in my career so I had to know what lights did what and how much I would need. I did a lot of reading and planning. Pre internet I had to rely on the photo metric tables in the Mole-Richardson and Lowel Lighting catalogs.

Knowing the math takes the creative art and turns it to craft. It’s important and valuable. Eventually you learn that a 4K HMI with stipples glass punched into a foam core bounce from 6’ to illuminate ambient light in a space 15’ away will be about 50fc, which is a T2.8 at 200ISO. You know because you do it a lot so you get used to it. But before that you read up in Lighting books and manufacturers websites and articles in American Cinematographer and you start to absorb stuff before you ever get to be on a set. Put in the time one way or another and you learn. Saying “experience” just means you don’t really know the numbers, just the general sense, and that can be okay. But when director Phil Joanou was making the doc U2: Rattle & Hum and veteran DP Jordan Cronenweth prelit a huge stadium during the day for a night shoot, Joanou asked how Jordan knew his lighting was right without seeing it. Cronenweth replies that he knew exactly what every light was going to do for him. And he was right. That’s craft.
 
That's a little contradictory, Mitch. I think we are probably saying the same thing but you are throwing me with the suggestion that "experience" means one never learned the numbers. I can only speak for myself, and yes I did all the book-smart studying wayyyy back too, but mostly what I'm carrying around is the real-world day-in and day-out stuff at this point. If I'm in a situation where I truly need to lock it down by numbers, we'll use whatever means are required, from photometrics to mockup tests.

We all have different yardsticks for the technical side of the craft. There are plenty of people on this board who can talk circles around me with the jargon of color science, codecs and the like. Meanwhile, some DP's who see the setup I use on set proclaim it as "too technical" for them, since they work purely film-style with a meter. I like to point out "you don't think metering 2.8 and 3/10ths and mentally subtracting a stop and a half to set your fill level is technical? that's literal number crunching, can't get much more technical than that".

Lighting to exact numbers and ratios works fine but (as you know), everything needs to be interpreted. A backlight at key on fine blonde hair is obviously going to be hotter than dark hair---but what about slicked black hair full of product vs curly black hair? Are you right at the kick angle or slightly offset? That's where the eye comes in, and the math goes out.

Last year I had to light an interior large day set at night while we were shooting in an adjacent room. Meant to go get my meter, but went as far as I could by eye and then got pulled back onto the first set until we made the company move. The cameras went up on that scene and it was good to go, exterior hallways were lit to believable daylight levels and we were right at the desired stop. Was actually comforting to know I could still do that, that the monitors hadn't made me soft! (was actually a little surprised).

I'm happy to encourage anyone to do it the "right" way--learn the numbers, read the articles, do the tests. Every craftsperson benefits from having a solid foundation. I
 
A side note on wides..

One of the reasons I think lens 'sets' like the canon cineprimes are weak and poorly conceptualised (50 .14, 20 1.4, 14 3.5) is the wide lens is exactly where a wide Tstop gets you out of jail if you are under lit. Having a 3.5 wide end is bad news effectively rendering 1.5 at the mid and tele useless.
On tight shots one can move the lights in but for the wide you cannot.
 
You must log in or register to reply here.
Back
Top