Metering for digital, use of digital test shots

I'm working on a project involving unusual light sources - namely an

old-school overhead projector with transparencies. I've been using an

incident meter and a digital camera to test the settings as I began

exploring, and today I'm going to be recording using film.

 

A friend has told me that the data from the digital files won't be

helpful in determining accurate exposures for film, since the sensor

plane is smaller. Is this true? Is there a standard compensation I can

apply to my meter (Sekonic L-508) to adjust for the different plane

sizes?

Your friend doesn't know what he is talking about. A given aperture/shutter/ISO combination

is the same exposure across any format.

 

That said, it does help to do a test beforehand because ISO sensitivities will vary a bit

between films and sensors.

Exposing for a digital sensor is similar to that of transparency film in that you don't want to overexpose because you will lose highlight detail.

 

This isn't the case for print film, where you need to ensure that you expose enough to capture shadow detail.

 

The sensor size is only relavent to the focal length differences and possibly depth of field.

Yea, but neither of you mention that different sensors treat light differently. I know for a

fact if I set my 20D for an Incident reading in the studio, that it will OVEREXPOSE from 1/3

to 2/3 of a stop. I doubt if it's the size of the sensor that makes the difference. Now if you

start adding different light sources to the mix, I wouldn't trust a digital for "poloroid type

tests". I'd shoot it on your film of choice and evaluate the results.

Bruce, that's what I meant by variances in ISO sensitivities of film vs digital sensors. The

ISO ratings on digital camera are merely equivalencies to film as rated by the

manufacturer; there isn't a standardization. Just as your 20D overexposes 1/3 to 2/3 of a

stop compared to an incident meter, many people report that their D70's underexpose.

 

That's why one needs to do tests to familiarize oneself with a digital camera's "true"

sensitivity, i.e, having to set an over/underexposure compensation amount. Just like when

working with film, you have to try to find your 'working' ISO, rather than using the one on

the film box.

 

But once the sensor/film sensitivity variances are taken care of, the point is that digital can

be a good way of 'Polaroiding' exposures because aperture/shutter/ISO combinations

work across any format.

Never forget that different lenses vary in their optical efficiency.

 

i.e., Lens A transmits 99% of the light that strikes the outer most element.

 

Lens B transmits 97% of the light that strikes the outer most element.

 

Cheap lens C transmits 90% of the light that strikes the outer most element.

 

The numbers were pulled out of thin air, but all lens elements reflect some light and contribute to light loss (otherwise they would be invisible ;o). Sadly this fact make the problem more complex rather than simpler.

 

my $0.02,

 

Sean

true in theory, relatively minor in practice. While it is best to do your test shots with the camera you'll be using, even a 20% difference would equate to less than 1/4 stop difference. Probably more differences in sensors(and between sensors) vs film than that

<p>Jim, </p>

 

<p>One thing to remember. Efficiency is linear, stop are logarithms base 2.</p>

 

<p>Hence we get the following table.</p>

 

<table border="2">

<tr>

<td></td>

<td></td>

</tr>

<tr>

<th>Stops Of Light Loss</th>

<th>Percentage of Light Lost</th>

<th>Optical Efficiency</th>

</tr>

<tr>

<td>1</td>

<td>50%</td>

<td>50%</td>

</tr>

<tr>

<td>1/2</td>

<td>25%</td>

<td>75%</td>

</tr>

<tr>

<td>1/4</td>

<td>12.5%</td>

<td>87.5%</td>

</tr>

</table>

 

<p>Hence, a lens with 90% optical efficiency would definitely affect exposure. Add in the supposition lack of lens coatings can cause a 1/4 stop of light loss (I do not have a pair of filters to run such a test) and we get the supposed fact that the quality of a manufacturers lens coatings can have a significant effect on exposure. </p>

 

<p>So how theoretical is this? I do not know. But since I have heard the difference between a coated and uncoated lens is significant but not huge, I would say the optical efficiency is a real factor. But I could be wrong.</p>

 

<p>my $0.02,</p>

 

<p>Sean</p>