Is it possible to train oneself to accurately interpret light intensity?

[kintatsu]

I don't use a gray card for metering. I spot meter and use those values to place things where I want. I meter several times, though. I meter the areas to determine the basic light, then pass the meter over the entire scene which, with the Starlite 2, gives me the full range of values in the scene expressed as little pips over f/stops, then meter for my main placement. Of course, that's just me.

I discovered that metering just one way, especially using the little pips as your only guide, or using the average mode, gives me decent results. I prefer to do it the other way, as my results are better. I don't worry about k or c, and seem to get good results.

 

[markbarendt]

The weird thing is that you "know" that if you point at the sky your reading won't be the same as if you just aim and shoot straight ahead. Even though you know that the two pictures require the same exposure to look good together (for example in a slide show or diptych).

But... If you deliberately aim slightly down to reduce the influence of the sky... you are messing with the intended calibration. You can deliberately aim down all the time. But then you would have to recalibrate to account for your new unusual metering method.

This is one of the reasons I appreciate manual cameras more and more these days. I don't even use built-in meters any more, and remove batteries if the camera will operate without them, because I want to choose the exposure myself.

Another interesting way to use this concept is to turn away from your subject 180 and using a somewhat wide angle metering method like center weighted, meter the sky that is lighting your subject.

With a bit of experience you can find very reliable reference points/areas all around you.

This method turns a camera's reflective meter into a crude incident meter.

 

[markbarendt]

Cameras with built-in meters can rule out a couple variables that a hand-held meter has to estimate. For example, with Through-the-lens metering, a camera can directly measure the flare from the lens, because the meter is looking through the taking lens.

A good feature of ttl.

An important side note to this is that focus should not be changed while checking these readings especially when short DOF is being used.

A subject will meter very differently when it looks sharp, than when it looks fuzzy.

 

[benjiboy]

This is why I recommend to our students not to look at their prints just as they walk out of the darkroom, and instead wait several seconds for their eyes to adjust to the light. If a student looks at his/her print as they exit the darkroom, their pupils are still dialated and that first impression is made. They then tend to add a little more time as the prints will seem a little light -- even though their eyes adjust and they continue to look at the print. Sort of a psychological dry-down effect.

Very true, and why they train soldiers in the army to keep the eye they use for shooting closed as much as they can when on night operations to preserve their night vision .

 

[DREW WILEY]

I always use the same or perfectly matched spotmeters for everything. TTL metering introduces certain extraneous variables, though what is most important is to simply be well accustomed to your specific chosen method. And gray cards? Ha! I've experimentally gone thru stacks of them and measured
the full visible spectral response on a continuous tone spectrophotometer. Not only were none of them actually 18% gray, but they varied widely with regard to the point on the spectrum where they even approached it. Even cards from the same manufacturer varied significantly - enough to spoil a critical chrome exposure, for example. An unfaded Macbeath Color Checker chart with both color and gray patches is a much better option, with respect to quality control. When we calibrate spectrophotometers, it's always with a special ceramic tile, which won't fade or absorb stains. My color vision is highly trained, and I've gotten good enough at estimating outdoor exposures from sheer experience in analogous situations that on a couple of mtn trips when I accidentally dunked my meter in icy streams, even my 4x5 chromes came out perfectly exposed. But that formula probably wouldn't work very well in unfamiliar circumstances. It's really a function of memory, from thousands of analogous shots, and not of any innate physiological ability to properly judge luminance value. So with regard to the original question - I'd give a strong "NO". Use a light meter for anything critical. And leave that "film latitude" excuse
to Aunt Maude with Kodak Gold in her Holga.

 

[benjiboy]

What if the most reasonable answer IS to read the palm of your hand? I'm not a person of color, but I believe human palms are universally Zone VI.

Same can't be said of the 18% gray card. It's not necessarily Zone V, and if you treat it like that, then you have discrepancies to frustrate you. People who calibrate their Zone System to camera tests with gray card as Zone V rate their EI differently than box speed. Certainly it will not agree with the incident meter at that point.

These frustrations can interfere with your desire to "accurately interpret light intensity". You'll ask yourself "Why don't I get the same results when incident reading compared to when I meter the gray card?" Haa, I think you did ask that.

I have a graycard from Sekonic that has several patches of gray. Trying to remember which one of those patches reads same as incident reading in same light has distracted me before. It's somewhere near the 12.7% patch. But remembering to pick the right patch - or using White paper and remembering how many stops that is from gray, is beyond my ability when taking pictures.

Placing my palm on Zone VI is not that hard. And on top of that... I find it pretty well matches my incident readings.

No Bill Black palms are Zone 111 or IV depending on the shade of the skin tone.

 

[Stephen Benskin]

Cameras with built-in meters can rule out a couple variables that a hand-held meter has to estimate. For example, with Through-the-lens metering, a camera can directly measure the flare from the lens, because the meter is looking through the taking lens.

TTL meters also give a real value for q and not just an average.

 

[Stephen Benskin]

I remember reading that Canon and Nikon use 12.5 for their K, which means with c=250, they come out at around 15.7%.

Metering an exterior scene using a disk diffuser requires averaging two measurements. This means using it's value of C is inappropriate to determine a value for average reflectance. You need to use the semihemispheric diffuser. There's a detail explanation in the thread "Is the K factor relevant to me or should I cancel it out?"