Incident Light Meter - where to point it?

[xpatUSA]

I've been playing with a Sekonic lately and am quite impressed with the ease of use and how it so often "gets it right" in one shot.

The manual and most tutorials say to aim the meter at the camera for incident light readings. On the other hand, this gentleman is fond of pointing it at the source of the light:

http://jimdoty.com/learn/exp101/exp_...inc_meter.html

I'm not fond of rules where you always have to do this or that but the literature is quite vehement and Jim Doty is equally so, as far as I can tell.

Would like to hear some informed opinions, please.

[pnodrog]

At the source of light - you may have set your camera up in a dark corner or have it set up under it's own spotlight.

P.S. I should clarify. Point it at the source of light that is illuminating the reflective faces that the camera is viewing. If the source of light is from behind the "subject" you will need to decide whether the subject is in silhouette (under exposed) or to be correctly exposed.

[Manfred M]

Correct answer is "it depends". The two camps that vehemently defend either at the camera or at the light source are wrong as I can show examples where either technique will give you the wrong reading.

1. A single (frontal) light source, it doesn't much matter. The measurement dome gathers light at 180° around in front of it and whether you point at the camera or the light source, the reading should be pretty well identical.

2. If you are shooting multiple light source images and the light from some of the sources (hair light, rim light, background light) are falling behind the subject (from the camera's viewpoint), you have to point the meter at the light source. I can't think of any other way of setting up the lighting ratios correctly. I will take a final reading after I have set up all my lights pointing at either the key light or the camera and the reading will be identical.

3. If I am trying some funky lighting, like side lighting, I will definitely point the light at the light source, but will also take a reading on the other side of the subject to see what the lighting ratio is.


Just to confuse things a bit more, there are times where I use both incident and reflective metering when I shoot flash.

As one example, I will get an incident reading of the whole scene when I shoot outdoors with fill light. I will meter the scene to understand the ambient light input to the shot and will meter the flash separately. That way I can get the background lit by the ambient light and use the flash on the subject (I will tend to underexpose the background by a stop or two).

As I said previously, I've asked the folks who tell me to point at the camera, how they would measure a rim light or background light and I have yet to get a satisfactory answer. (Correct answer is meter toward the light)

Likewise, I ask the folks who say to point the meter at the light source, I ask how they get a correct reading when shooting outdoors with the sun being behind the subject (backlight). Again, lots of head scratching there too. (Correct answer meter towards the camera).

[Dave Humphries]

First let me say that I haven't read every word of that page Ted, but I skimmed through to the bits that seemed relevant, so I can hopefully summarise.

What he suggests ("point the meter at the light source, not the camera") could result in an amount of under exposure and this will differ depending upon how much off axis the sun (or other light source) is relative to the line between subject and camera.

He covers the reasoning for continually saying "point the meter at the light source" in the paragraphs below the pictures of large red boulders - he is protecting the highlights.

So the argument stands up if you don't want to blow the highlights, but if metering reflected, I'd allow for that and dial in some EC (to protect the highlights), just as you could similarly offset the incident reading with an 'EC' to accommodate the lower reading if the meter was pointed back at the camera and hence the dome only being half illuminated.

Clearly he shoot mainly landscapes, so he's found a method that works for him outdoors - and like many of us, he thinks his way is the right way, so preaches it.

But it isn't the only way.

Cheers, Dave

[Manfred M]

and like many of us, he thinks his way is the right way, so preaches it.

I agree 100% Dave. Most people who write these articles either do no know enough about the subject to understand that they are working to a certain set of assumptions that are very specific to the conditions that they are shooting under and do not recognize that this is in fact something they have not explained to the reader.

I'm reading a book right now, that was written by a very well known photographer / photographic educator who is a firm proponent of shooting 100% manually (including focusing the camera that way). While I don't wholly disagree with where he is coming from, his photographic work flow was developed in using manual film cameras decades ago, and even though he has "gone digital", he still shoots like a film shooter and writes that way. The fact that he distrusts automation (rather than understanding the limits of automation) is unfortunately being passed on to his readers.

[xpatUSA]

Thanks all for your responses so far.

Interesting challenge to visualize the 3D cardioid response and it's relationship to surroundings.

I shot the street yesterday. Two shots: one where I metered the overcast sky (pointed it vertically upward) and one where I pointed it down the street, just to see (I know, not recommended in the literature). Results were predictable:



Currently the Sekonic is set to 40 ASA because it over-reads by at least 1 EV most of the time - the camera remains at 100 ISO of course.

[Pixl8tor]

Is there an iPhone app that reads incident light or must I buy a light meter?

[Manfred M]

Is there an iPhone app that reads incident light or must I buy a light meter?

No the camera in any phone is far too directional to be used as an incident light meter. Incident light meters are generally equipped with diffusers to ensure that they get a large sample area. I remember seeing a Kickstarter diffuser attachment or an iPhone that as supposed to do this, but I did not follow up as I have no interest in that type of attachment.

My Sekonic L-356, works just fine for the work I do. It's an incident light flash meter but can also be used to measure ambient light.

[Dave Humphries]

Hi Bill,

Is there an iPhone app that reads incident light?

Probably (but I haven't looked)

... or must I buy a light meter? (to read incident light)

YES - If you want to do it accurately enough for it to be of any benefit over the skilled use of the camera's light meter (reflective) and applying Exposure Compensation (to account for the tones in the scene).


For the reasons Manfred states.

Dave

[dennybeall]

Why not use the light meter in the camera? You can set it for reading a small point to averaging the entire scene like the "LIGHT METER does.
You can set it to read a point - lock in that reading and then compose the scene differently.
I haven't used my light meter since I got a Photomic Head for my Nikon "F". But then that's just me.

[xpatUSA]

Why not use the light meter in the camera? You can set it [from] reading a small point to averaging the entire scene like the "LIGHT METER" does. You can set it to read a point - lock in that reading and then compose the scene differently.

I haven't used my light meter since I got a Photomic Head for my Nikon "F". But then that's just me.

Denny, there's quite a big difference between using the camera metering (reflected light) and a separate light meter (in incident light mode per this thread's title). In incident light mode, the "LIGHT METER" does not "average the entire scene".

Looks like the Photomic head is a reflected light meter?

http://www.kenrockwell.com/nikon/f2.htm

Everybody loves Ken . .

[RBSinTo]

I've been playing with a Sekonic lately and am quite impressed with the ease of use and how it so often "gets it right" in one shot.

The manual and most tutorials say to aim the meter at the camera for incident light readings. On the other hand, this gentleman is fond of pointing it at the source of the light:

http://jimdoty.com/learn/exp101/exp_...inc_meter.html

I'm not fond of rules where you always have to do this or that but the literature is quite vehement and Jim Doty is equally so, as far as I can tell.

Would like to hear some informed opinions, please.

Ted,
From the manual for my Minolta Flashmeter III which I purchased a zillion years ago, and use virtually always in incident mode:
"Measuring incident light.
The reading is taken from the subject position with the spherical diffuser pointed toward the camera and the meter held so that it receives the same light as the subject."
This has almost always given me good results.
Robert

[tao2]

Folks,

just listen tae the oracle... Now where is my suit...

[george013]

But now the explanation.

George

[xpatUSA]

Ted,
From the manual for my Minolta Flashmeter III which I purchased a zillion years ago, and use virtually always in incident mode:
"Measuring incident light.
The reading is taken from the subject position with the spherical diffuser pointed toward the camera and the meter held so that it receives the same light as the subject."
This has almost always given me good results.
Robert

Thank you, my manual says that, too.

Folks,

just listen tae the oracle... Now where is my suit...

Thanks,

Interesting, the "substitute readings" bit where the ghostly figures are not pointing it at the camera but at the source of light relative to the side of the subjects that face the camera, and of course not actually at the sun which could be anyway because it is not shown.

[george013]

Thank you, my manual says that, too.




Thanks,

Interesting, the "substitute readings" bit where the ghostly figures are not pointing it at the camera but at the source of light relative to the side of the subjects that face the camera, and of course not actually at the sun which could be anyway because it is not shown.

Twed,

Maybe you should reread this thread "Analog" Lightmeter Sensitivity. and the link you gave in it http://dougkerr.net/Pumpkin/articles/Norwoods_Dome.pdf

If I understood it well, the dome has 2 functions: a light resistance, letting pass only 18% or something like that, and the form which let it meter light coming from the side too. This thread is about that last.

What I believe is that the dome is correcting the light not falling perpendicular on the meter. That's why it's said to aim to the camera. And that's why the grays in Roberts post are still ok: the've the right angle, the sun is far away.
I'll hear when I'm wrong.

George

[xpatUSA]

Twed,

Maybe you should reread this thread "Analog" Lightmeter Sensitivity. and the link you gave in it http://dougkerr.net/Pumpkin/articles/Norwoods_Dome.pdf

Gweorge, good advice, thank you.

If I understood it well, the dome has 2 functions: a light resistance, letting pass only 18% or something like that, and the form which let it meter light coming from the side too. This thread is about that last.

Yes, the resistance to light is so as to match the sensitivity of the photo-sensor, not necessarily your 18% transmission; also the combination should have the same spectral response as photopic human vision. The 3D spatial response is approximately that of a cardioid and can even pick up light beyond the side i.e. slightly to the rear.

What I believe is that the dome is correcting the light not falling perpendicular on the meter. That's why it's said to aim to the camera. And that's why the grays in Roberts post are still ok: they've the right angle, the sun is far away.
I'll hear when I'm wrong.

George

Sounds reasonable enough.

[george013]

Gweorge, good advice, thank you.



Yes, the resistance to light is so as to match the sensitivity of the photo-sensor, not necessarily your 18% transmission; also the combination should have the same spectral response as photopic human vision. The 3D spatial response is approximately that of a cardioid and can even pick up light beyond the side i.e. slightly to the rear.



Sounds reasonable enough.

When you want to translate an incident metering to a "reflective metering" you must make some assumption of what the average reflection must be. An often used value is 18%, for general purpose. The calibration of the meter comes in here, how to translate the measured light to camera settings.

Now you've an answer on your question in the other thread. It's not just subtracting stops for the angel between light source, object and camera is involved too.

All new for me too.

George

[rpcrowe]

Lots of photographers go crazy calculating lighting ratios by using all kinds of intricate formulas.

I can eyeball lighting ratios for portraiture when using a continuous light source.

When using a studio flash, I basically have a continuous light source to eyeball; since a decent studio flash will have a modeling light...

Of course, if you are limping along a la "Strobist" and using a hotshoe flash as a pseudo studio strobe, you do not have the advantage of modeling lights with which to work.

[xpatUSA]

When you want to translate an incident metering to a "reflective metering" you must make some assumption of what the average reflection must be. An often used value is 18%, for general purpose. The calibration of the meter comes in here, how to translate the measured light to camera settings.

George

I thought we were discussing the dome's light transmission, not "metering 101"?

You said "a light resistance, letting pass only 18% or something like that"

I said "the resistance to light is so as to match the sensitivity of the photo-sensor, not necessarily your 18% transmission"

George, is your post disagreeing with what I said?