Incident light meters question

wclark5179 said:

Today people sure like those photos I made about 50 years ago!

Click to expand...

In other words: you rule! We really admire you for that. Show us the way, please!

wclark5179 said:

And now on the other side I've got that lovely beautiful graph called the histogram that does give some information. And I also shoot in RAW. Smiles & Fun!

Click to expand...

OMG, you sure did discover some new tech out here! Wow, you rule! Thanks for sharing all this with us ...

dfoo said:

Where i find things tricky is when the subject is partially in shade, partially in the light. With an incident meter only what do you in this case?

Click to expand...

The rule of thumb is expose for the shadows develop for your highlights. That is if you are using the entire role or most of it on the same subject matter.

A wealth of information in this thread, thank you guys & gals.

Galah said:

This is a great thread on the subject and an excellent review of the topic. Thanks to the OP for posting it, and to all who contributed.

Something like this, every now and again, is great "revison".

Click to expand...

My thoughts, exactly. Thanks to everyone for the refresher!

Looks like a lot of incident meter users. So how are you all calibrating your meter? That is, how are you getting your exposure index?

Those of you that regularly use the meters know that there are a number of issues to be addressed in the calibration. For example, as pointed out, the dome is very sensitive to the angle of incidence of the light source. Also, one has to estimate the number of stops below the midpoint reading given by the meter that will contain detail.

For example, copy artwork will have the shadows about 3 stops below a middle gray, whereas a complex outdoor scene could have shadow detail 5 stops below a middle gray.

Personally I set mine to get about 5 or 6 stops of detail below a middle gray, knowing that I will be giving more exposure than needed in situations of limited scene brightness range. That is with B&W negative film.

With B&W reversal cine film, I like the reading to give an exposure that straddles the midpoint of the H&D curve. (That is a technical way of saying I don't want the projected movies to be too light or too dark )

ic-racer said:

That is, how are you getting your exposure index?

Click to expand...

Film testing, even just informal testing.

ic-racer said:

Those of you that regularly use the meters know that there are a number of issues to be addressed in the calibration. For example, as pointed out, the dome is very sensitive to the angle of incidence of the light source.

Click to expand...

For a basic no frills reading, the meter is held in the same light as the subject, the dome is pointed squarely at the camera, the button is pushed, the meter reading is dialed into the camera.

There's nothing fancy or tricky or hard here.

For a basic no frills reading, the meter is held in the same light as the subject, the dome is pointed squarely at the camera, the button is pushed, the meter reading is dialed into the camera.
There's nothing fancy or tricky or hard here.

Click to expand...

Except filter factors. Too many people make the same mistake over and over: putting a red, yellow or blue filter on, or even a POL, making a reading and then the exposure — only then realising he/she didn't adjust the meter for the filter factor!

Baseline calibration is still best set up based on a lot of critical analysis experience of the finished image be it neg or reversal (reversal is less forgiving). Either baseline (calibration (offset from meter defaults) or additive compensation (after the reading) have their uses.

Poisson Du Jour said:

Except filter factors. Too many people make the same mistake over and over: putting a red, yellow or blue filter on, or even a POL, making a reading and then the exposure — only then realising he/she didn't adjust the meter for the filter factor!

Baseline calibration is still best set up based on a lot of critical analysis experience of the finished image be it neg or reversal (reversal is less forgiving). Either baseline (calibration (offset from meter defaults) or additive compensation (after the reading) have their uses.

Click to expand...

This is a fact.

And one of the reasons I like TTL metering of an SLR.

Steve

Poisson Du Jour said:

Except filter factors. Too many people make the same mistake over and over: putting a red, yellow or blue filter on, or even a POL, making a reading and then the exposure — only then realising he/she didn't adjust the meter for the filter factor!

Click to expand...

But then, too many people make the same mistake over and over again: they put a filter on the lens, then use TTL metering to meter through it.

Contrast filters are used to selectively darken bits of the image. You should let the filter do just that. If you meter through it, the effect is 'averaged', with the suggested reading overexposing both the unaffected bits and the affected bits.

You should meter without filter, apply the factor to the reading. Something much easier when not metering through the lens, of course. And something almost noone using TTL metering does.

markbarendt said:

Film testing, even just informal testing.

Click to expand...

So, you are photographing a calibrated reflection wedge, testing for a density of 0.1 on the darkest step, realizing it has only about 5 stops from white to black, and adding a couple of extra stops of exposure for non-copy photography?
If not, then what are you using for your x-axis on the 'film test' to get your exposure setting to calibrate the meter?

There's nothing fancy or tricky or hard here.

Click to expand...

Well, don't keep it a secret

You meter. You expose. You process. You look. You go "Hmmm". And you draw your conclusions.

Q.G. said:

You meter. You expose. You process. You look. You go "Hmmm". And you draw your conclusions.

Click to expand...

Yep, exactly that.

Q.G. said:

You meter. You expose. You process. You look. You go "Hmmm". And you draw your conclusions.

Click to expand...

Hmmm, that's not very scientific!

IC - I've checked filter factors by shooting a Macbeth Color Checker and then comparing the neg densitys of that neg with one taken without the filter.

But then to test an incident meter, there's no real way to test them without a calibrated illumination source.

Kirk Keyes said:

Hmmm, that's not very scientific!

Click to expand...

I know!

Yet i can't help but keep remembering that photography is a visual medium. So - i fool myself into believing - if it looks good...

The Scientific Method: you meter (i.e. you make an observation); you make an exposure (i.e. you draw a hypothesis); you process (i.e. you test your hypothesis); you look (i.e. you reach a conclusion based on the results of your test); you go "hmmm" (i.e. you formulate a theory based on your conclusions).

Sounds scientific enough for me.

~Joe

jmal said:

So, I know that incident meters measure the light falling on a subject, but I would like to know how they arrive at an exposure recommendation. I have always used center weighted reflected light meters in my cameras and know that I can measure an area of the subject and adjust upward or downward depending on whether I'm exposing for the highlights or the shadows. I am in a position where I need a meter to use with an unmetered camera and it seems that most affordable hand held meters are incident meters. So, what does the reading tell you? Does it simply tell you what exposure is needed for a middle gray based on the falling light? Also, any recommendations for something under $200, new or used, and not too large? Thanks.

Click to expand...

Upon what basis is it assumed that an incident meter (exclusively) is appropriate for the task? Most of the commonly available meters provide for incident, reflected, spot and multi-spot/multi-spot+average — these last three will provide you with much more information about the scene (e.g. the range of principal luminances) than a straight incident reading. This is where your real science should be, not in discussing the xth degree of calibration and x and y axes.

Kirk Keyes said:

Q.G. said:

You meter. You expose. You process. You look. You go "Hmmm". And you draw your conclusions.

Click to expand...

Hmmm, that's not very scientific! ...

Click to expand...

Yes, it's a bit of trial and error, but it doesn't mean that you can't do some tricks. First of all, you can do your own film speed tests and have a look at the results, at least with BW films(1). Once you do this, you know what fits the paper range and how things look under specific lighting conditions. With a bit of common sense, experience and fantasy, you can use elements of the zone system, even with roll films. And then there's the problem of proper metering. TTL is basically unreliable, incident is much better and spot meters are probably the best option, if you know how to use them and if you have the time to use them. Time can definitely be critical in some scenaria. Anyway, an incident meter can be substituted with TTL if you have a point of reference that is good enough for the task. I've used my palm as a point of reference (and adjust the reading) and I got nice results. Definitely better than following what the TTL meter says. Of course, finding the subject's brightness range requires a spot meter, but once you have a fair bit of experience you can guesstimate it, and using negative films makes the task a lot easier; you have the option to overexpose, that should do the trick. The worst thing that could happen would be to reach the film's shoulder and lose some highlight contrast(2).

Anyway, I don't pretend that I'm an expert, nor would I say that my method is foolproof. My gear imposes limitations and I have to take a "methodical" approach with what I have, in order to get good results without resorting to extreme tricks at the printing stage. But I can definitely say that my hit/miss ratio became far better once I understood how my equipment and materials work.

1 Barry Thornton's method is what I use.
2 Assuming that you use negatives, positives are a very different case.