Jump to content

Native ISO & Metering


Zahi Farah

Recommended Posts

Hello everyone,

 

As I consider buying my first light meter and delve deeper into the world of cinematography, there are a few questions I can't seem to find (satisfying) answers for, at least online. I'd really appreciate any help in these matters.

 

Question 1

Why is it so important to define a camera's true native ISO. Not the marketed one but the actual real one. I've learned how to do it it but I still don't understand why I'd do it.

 

As an example (hypothetical), let's say I'm using an Alexa and I come to the conclusion that its real ISO is 400 instead of the proclaimed 800. What am I supposed to do with that? Should I question the way the stops are supposedly distributed between shadows and highlights? Supposedly at ISO800 it's -7/+7 stops of range, did that range now move to 400? Like what's happening here?

 

Also, does it differ from camera to camera, even if it's the same model? Or can I trust the results from tests done from people I know I trust; whether online or otherwise, for that specific model?

 

I understand we must test, test and test. As a novice I can assure you this notion is drilled into our minds by almost every self respecting cinematographer, gaffer, DIT etc. And I am thankful for it. However can't we trust these manufacturers at least a little bit?

 

Question 2

I took a cinematography workshop not too long ago, my first, and was introduced to light meters and how useful they are. Now that I'm about to work on my first couple of projects I can't imagine myself not using one.

 

However. During that workshop several people had the same light meter (mostly Sekonic 858) and, surprise, not all readings were the same. From the same position on the same spot, whether reading incident light or spotting, people were getting (slightly to not-so-slightly) different readings.

 

So here's my question, how can I tell if I can trust my light meter? I live in a country where most people don't use light meters and we don't have any kind of support to send them off to calibration. Any suggestions?

 

In case it's essential to make sure that the light meter is reading accurately, does anyone know of someone or of a service center that does this anywhere not in the Americas? Also, has anyone ever dealt with Sekonic's service center, the one in Oman?

 

Many thanks. Don't judge me too harshly on my trust issues :)

 

 

 

Link to comment
Share on other sites

  • Premium Member

It's not critical you know the "real" ISO, it's just useful to explain certain image characteristics. You pick an ISO that gives you what you need in terms of noise vs. highlight information, and dynamic range, etc.

 

Meter information has to be interpreted, it's rarely as clear-cut as a flat-lit color chart. What matters more than accuracy (assuming the meter is internally consistent) compared to every other person holding a meter is you being consistent in how you use it and interpret it. Over time you start to find a metering technique that gives you consistent results that you can trust. Doesn't matter if it doesn't work for someone else.

 

Of course, your meter should be accurate when pointed at something that requires no creative interpretation. Usually when you compare meters in those situations, everyone is within 2/10ths of a stop from each other.

 

One reason why I usually rate film negative 1/3-stop slower isn't so much to get a denser negative but 1/3-stop is within a margin of error for most people metering a shot and I'd rather err on the side of more exposure. If I definitely want a denser negative consistently, I have to start out with a 2/3-stop slower rating.

 

It was helpful back in the day when movie negative was printed for evaluation because printer lights gave you some sort of feedback in terms of your exposure technique, so if your prints were coming back in the high 20's and you really wanted them more in the low 30's, you could adjust your ISO rating or exposure technique.

Link to comment
Share on other sites

The difficulty of being a novice light meter user and testing cameras for an "optimum" ISO setting is that there are many variables. If you are shooting LOG for example, how will you be de-LOGing the footage? Will you be using a manufacturer's LOG to 709 LUT? Something else you've designed? It all effects the final result. And, every camera might be different as well.

 

I think first you need to learn to use your lightmeter. And I will suggest here that you use a 35mm still camera and shoot reversal/slide film. Why? Because slide films are very unforgiving to improper exposure and the results are easy to see. And, because the development of slide film is standardized there will be very few variables to contend with.

 

I would suggest first shooting a couple rolls of slide film and viewing the results on a light table with a magnifying loupe. Put the slides on the light table so that you can view them side by side and see what you've done.

 

Next you'll need to design a test. I hope by this point you really have read about and understand how to use an incident meter vs a spot meter.

What I've done for testing is to place a white towel on the wall. Set up your camera with the slide film on a tripod and fill the frame with the towel. You should light the towel evenly with a light that matches the color temperature of the film you're using. With slide film this is almost always daylight. Next, expose the towel 1st at 5 stops under the reading provided by the spot meter, at the ISO on the film box. Increase the exposure on 1/2 stop intervals until you've exposed the towel at +5 stops. Keep in mind that the spot meter reading should expose the towel as middle gray, and not white.

 

Place the developed slides on the light table in order of darkest to lightest. (note: it would have been helpful to place a little piece of paper on the towel that says the exposure setting for each frame!) If all has been done correctly, the frame exposed at the meter reading should look close to an 18% grey card. Study the results on the light table and carefully look to see where the towel details become hard to see due to over/and/or under exposure. Keep mental notes of how bright is 1/2 stop over and 2 stops under etc.

 

Armed with this knowledge, shoot a roll of slide film using only the spot meter. For example, take a reading off someone's face and decide how bright you would like the face in the final image. Something like reading a white skinned face and deciding that you would like it 1 stop lighter than 18% grey. Since the spot meter always suggests an exposure for 18% grey, increase the exposure by 1 stop to expose the face at +1 stop.

 

Take a landscape photo and carefully think about how bright you want each part of the image. You'll know from your towel test how bright and/or dark you can go before you loose shadow or highlight details. You may find that you want detail in the forrest and in the clouds, but discover that you can't have both on your slide film. You'll need to choose! Or shoot it each way and keep careful notes of your exposures.

 

Once you've got this all mastered (and it will take some time!) you can then begin to start designing tests for digital camera optimum ISO settings. And also learning how to use the incident meter. Just keep in mind that the incident meter reading is the same as a spot reading off an 18% grey card (in the same lighting where you hold the incident meter)

 

As to calibrating the light meter, digital light meters tend to be very consistent. If you have a new digital light meter I wouldn't worry too much about calibration, but if you have doubts, find another light meter to compare it to.

  • Upvote 1
Link to comment
Share on other sites

Of course, your meter should be accurate when pointed at something that requires no creative interpretation. Usually when you compare meters in those situations, everyone is within 2/10ths of a stop from each other.

 

As to calibrating the light meter, digital light meters tend to be very consistent. If you have a new digital light meter I wouldn't worry too much about calibration, but if you have doubts, find another light meter to compare it to.

 

David, Bruce, thanks for taking the time to answering my concerns.

 

In a nutshell, I shouldn't obsess over the absolute accuracy of my meter as long as it's somewhat accurate -- up to 2/10ths of a stop -- and the best way to make sure of that is by comparing it to other light meters. Sounds about right?

Link to comment
Share on other sites

  • Premium Member

Yes.

 

Unless you are shooting charts, most of the time, there is a greater margin of error from how you interpret the data compared to how accurate your meter is, so it is more important that your meter is consistent so you have a reliable base to make decisions on, even if that base is consistently 2/10's of a stop off in one direction from the another meter.

 

Shooting reversal film is a good learning technique not just because your mistakes will be visible, but even if your metering and exposure is perfect, you will learn how real world objects sometimes need create exposure adjustments or you'll learn how the mood of a day exterior shot can change depending on how you expose it.

 

Of course, today, just lighting with a video camera shows you in real time how your meter results look and whether they need creative adjustments.

  • Upvote 1
Link to comment
Share on other sites

Yes.

 

Unless you are shooting charts, most of the time, there is a greater margin of error from how you interpret the data compared to how accurate your meter is, so it is more important that your meter is consistent so you have a reliable base to make decisions on, even if that base is consistently 2/10's of a stop off in one direction from the another meter.

 

Shooting reversal film is a good learning technique not just because your mistakes will be visible, but even if your metering and exposure is perfect, you will learn how real world objects sometimes need create exposure adjustments or you'll learn how the mood of a day exterior shot can change depending on how you expose it.

 

Of course, today, just lighting with a video camera shows you in real time how your meter results look and whether they need creative adjustments.

 

Thank you David.

Link to comment
Share on other sites

Question 1

Why is it so important to define a camera's true native ISO. Not the marketed one but the actual real one. I've learned how to do it it but I still don't understand why I'd do it.

 

However can't we trust these manufacturers at least a little bit?

 

In my opinion, the native ISO a manufacturer states carries a lot of weight because it's the ideal ISO they've determined their camera works best in regards to image quality. Is that quality good enough for you? That's where the testing comes in. You're figuring out if the manufacturer's definition of quality matches yours and if you can make the camera shoot in a way that matches your definition of "good".

 

Question 2

I took a cinematography workshop not too long ago, my first, and was introduced to light meters and how useful they are. Now that I'm about to work on my first couple of projects I can't imagine myself not using one.

 

However. During that workshop several people had the same light meter (mostly Sekonic 858) and, surprise, not all readings were the same. From the same position on the same spot, whether reading incident light or spotting, people were getting (slightly to not-so-slightly) different readings.

 

So here's my question, how can I tell if I can trust my light meter?

 

Every feature I shoot, I make my meter match the camera. Every camera see's middle grey differently (18% grey card, 50% IRE, whatever is in the middle). So, on the camera prep day I set up a grey card and color chart with a single source light on it. I use a waveform monitor with the LUT loaded (most likely REC709 one supplied by the manufacturer or developed by me and the colorist in prep), and determine where the grey card lands on 50% IRE. At that point, I use my meter and see what the meter says.

If it's over/under, I adjust the settings on my meter to make it match that specific camera (I use a Sekonic 758Cine). Now I know that when I do a meter reading, my meter reads 50% IRE the same way that specific camera does.

 

As an added bonus, I roll on the "correctly" exposed grey card for reference in the color grade. This gives me a good starting point and makes life a whole lot easier.

 

On set, with my gaffer, I have him/her match their meter to mine. That way now my gaffer's meter sees what the camera sees.

Link to comment
Share on other sites

  • Premium Member

It could be argued that the "native" ISO of an Alexa is 400 but ARRI recommends 800 as the optimal compromise between noise and highlight retention.

 

I think that a lot of digital stills cameras have a low native ISO but such minimal noise that higher ISO's look fine.

  • Upvote 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...