Hello,

I've just got a Sekonic 758DR Light Meter and I'm testing it (without uploading any exposure profile, so I'm "factory default")

I set up a Gray 18% Exposure Chart and I get different reflected and incident readings.

It doesn't matter what ISO is in the light meter. The incident reading is about almost 1 stop above.

Is this normal??? How could I fix it?

Thanks!!!

Perfectly normal. Reflected readings should be taken from the grey card (or your most important subject in your frame, assuming skin tone) while incidental readings should be taken from subject position pointing back at your flash or hot lights giving you the reading of the light that is falling "incidentally" on your subject. If you are putting the cover on and trying to read light reflecting off your grey card, you are introducing a neutral density in front of the light meter's sensor.

uh, not the way I'm reading this--reflected or spot readings off an 18% gray card and incident reading of the light falling on the card should read the same. That meter is a dual-function spot and incident mete. Which cover you are referring to, Shaun?

If you were taking the spot reading off caucasian skin tone, for instance, I would indeed expect it to read a full stop differently than the incident reading. But the gray card should deliver the same reading.

Charles, unless I am missing something, the only way to get a reflected and incident reading that's the same is with a mirror. Or are you saying he should change modes on the meter which somehow compensates for the expected light loss?

This is why I don't fool with reflected readings. Too confusing. Read the light, set the camera, and roll.

Oh, and Shaun, I thought you were supposed to point the bulb toward the CAMERA on incident readings, not the source. Which would give grossly inaccurate readings depending on where the light source was coming from. If you have 3/4 light and pointed the meter toward the source instead of the camera, the face would be underexposed several stops.

Which cover you are referring to, Shaun?

The white half dome slide-over cover.

PS. I ONLY use incidental readings for key-fill-back ratios, not for exposure in video. That's what zebra and my eyes are for.

Oh, and Shaun, I thought you were supposed to point the bulb toward the CAMERA on incident readings, not the source.

Incidental readings should be aimed at the light source(s)

The intent of the spherical collector is, I believe, to integrate all the light incident on the object - hence you should include all sources as Shaun says if there are more than one. A flat collector will also read differently than a spherical one

I'd be very surprised if I got exactly the same reading from incident and reflected - the angle of the gray card will influence its reflectance.

And by the way, nothing is calibrated to 18% gray - 12% or 13% would be more like it.

Good article

Meters Don't See 18% Gray by Thom Hogan (http://www.bythom.com/graycards.htm)

Thank you!

Jim: Your link is very interesting.

Shaun: I didn't explain very well. I know how to use a light meter. In fact I was testing this new meter to know how reliable it is.

These were the steps followed:

-Set the Gray Chart and only one soft light lighting uniformly

-Take the reflected reading from camera position (Sony EX1, automatic iris, zoom closed to the chart)

-The camera auto-adjusts itself for 50% Grey. The histogram is correct. The iris is 4.0 (for example) The light meter is 4.0 at 320 ISO (same shutter, etc...) (You can adjust the camera iris manually with a waveform monitor and the result is the same)

-They're both the same and it seems the light meter is calibrated for that camera

-If I take an incident reading from the chart pointing to the light I get a different reading 5.6 (for example)

-Depending on the target wanted you could point the sensor to the camera instead of the lighting source. To take a global reading of the scene, for example, etc... (it depends on the cinematographer)

-As I have only 1 frontal soft lighting source I get the same reading pointing to the camera or to the source


As I think the meter should give the same readings for the same luminance value I don't know what is the good one (or the better one) Reflected or Incident???

This meter doesn't have a cover for the sensor. It has 2 different inputs. The incident input is a retractable "lumisphere" (for hiding from inconvenient sources) and the reflected input is a spot lens (1º)

In my experience shooting video I think the light meters are very useful not only for the lighting ratios. Working video is more critical than film and you need to fight for every pixel. There is no way to get a calibrated monitor in all the situations with the same dinamic range (or gamma curve) as the camera sensor can capture. That's my opinion of course.

Thanks again!

I've been thinking about all this again - thanks for raising this issue as it has me really going back and sorting through all these things in my mind.

Maybe I've never really worried too much about matching the various meter readings because in my mind incident and reflected and reflected spot and internal camera meters are each in their own ways trying to tell you something different and I've always just dealt with them as different animals - or rather different tools.

The incident light meter is just trying to tell you how much total light is falling on your scene. It has no idea if your scene is the proverbial black cat on a pile of coal, or a white rabbit on a white blanket. If you shot something on a cloudy day and were happy with the result, then the incident light meter would tell you how to compensate if you came back to reshoot on a sunny day (I know there's more to it depending on the components of the scene). And it's absolutely great for checking how much light is falling on various parts of your scene - like a green screen

The various flavors of reflected light meters tell you how much reflected light is reaching the camera position. Here's where the black cat and the white rabbit get different results.

And the internal meters tell you something else altogether - namely how much light is making it through your lens to the "film" plane. When I have my 5 x 7 view camera focused in very close the bellows is extended quite dramatically - and the amount of light falling on the film plane is much less than the amount falling on the lens due to light fall-off inside the camera. Even on regular cameras, we used to have to make all kinds of compensation when focusing in the macro range for the same reasons. It was really great when they started coming out with metering arms that you could move around the ground glass on the big cameras to eliminate most of this math.

Anyhow, these things are just tools to provide the information that the photographer/videographer has to interpret and integrate with his/her artistic intent to set the exposure. As long as each tool in the set is self-consistent, I'm happy!

Victor, thanks for the description of your setup. I still stand by my assertion that the incident and spot readings should theoretically be the same (I understand that there are variables in the way one reads from a grey card depending on angle). I don't have that Sekonic myself although I know a lot of people that do, maybe I'll run this past them (I use a Spectra IV for incident and a Minolta for spot metering).

All of these are indeed interpretive tools and as such different people have different ways of using them. Jim--I myself much prefer to use a spot meter for checking linearity on a green screen as I can stand near camera and take readings much more easily than walking around the backing and ducking lights, especially for the bits above my head! As seen within this thread, folks are using meters in entirely different ways for similar purposes.

I personally will only use my meter when shooting video for setting a base stop and then work from the monitor and waveform or histogram from there. Depending on what settings are being applied on a given camera, I feel like values can shift around too much for me to know exactly what a 3-stop overexposed window will look like just from the meter reading as opposed to using the monitor. I do only use properly calibrated broadcast monitors however--I did a job recently that required minimal gear due to airline travel and had to make do with an onboard LCD that was obviously not accurate, so I went back to "film" mode mentally and trusted my eye more than anything. The meter wouldn't have really helped me there.

Charles,

Of course I agree that using a spot meter is easier than wandering all over in a complex set up with an incident light meter.

Only thing I sometimes dislike about the spot meter is that it reads such a small area that you can over-react to a tiny difference that won't really matter in the final shoot. That's where the experience and human interpretation comse in!

I actually just unearthed my real favorite SEI spot meter - sadly it no longer seems to be working. It reads a somewhat less than 1/2 degree spot and it's an extinction photometer - there are two crossed ND wedges and a gear mechanism that drives them in opposing directions, and a small spot in the middle that reflects a calibrated reference bulb.

Probably weighs more than most handycams! Because the reading depends on the human eye matching the spot to the surrounding area in the scope you could do things like take a reading on an electric line hanging between two light poles. Very interesting gadget indeed. Probably more targeted at lighting engineers than photographers.

Found a link to an article about it - great piece of British engineering and I'm going to try to get it repaired.

Shutterbug: The SEI Photometer (http://shutterbug.com/techniques/lighting/1002sb_thesei/)