Incident vs Reflective = different readings?

[Rudy M.]

I am just beginning to use some light meters I have acquired and am puzzled by the variaiton is readings. Just now I took readings of a baige wall in my office primarly lit with outside light at ISO 800.My 20D meter says 1/30 @ f/3.5 which is a reflected reading.My Sekonic L-208 indicates 1/30 @ f/3.5 incident reading but 1/30 @ f/8 reflected.My Sekonic L-308s indicates 1/30 @ f/3.6 for incident, and 1/30 @ f/8 reflected.Why do the reflected readings from the light meters not match the reflected reading from the camera? And why does the incident reading from the light meters match the reflected reading from the camera?

[DrPablo]

First, you need to hold a reflective meter towards the surface you want to photograph, and you need to hold an incident meter towards the camera from the surface you want to photograph. An incident meter asks the question 'how much light is falling on me'. A reflective meter asks the question 'how can I turn this reflected light into middle gray'.

If your wall is a stop or two brighter than middle gray, then if you're holding your meters correctly the incident meter will tell you how to make it look beige and the reflective meter will tell you how to make it look middle gray, and those values might be a stop or more apart. So with a reflective meter you need to add a stop or two to things that you know are brighter than middle gray (snow, sunlit sand, caucasian skin, etc).

You should always get the same value if you take an incident metering, then take a reflective metering off of a gray card in the same light.

[PhotosGuy]

Mt personal solution to a good starting point:
First set the f-stop & shutter speed you need. Then adjust the ISO. Need an exposure crutch? (http://photography-on-the.net/forum/showthread.php?t=89123)

[Rudy M.]

Anther question about this has come up since getting a new camera. This morning I have a nice tree with all yellow foliage with the sun hitting it just right. So I stopped in the driveway and took an incident reading right off the yellow leaves pointing the dome of the meter towards where the camera would be. Meter indicated 1/250 at f/11 ISO 200. I set my 20D to manual, 1/250 at f/11 and pretty much filled the frame with the yellow trees zooming my 30mm f/1.4 with my feet. I got an underexposed image--just about everything was center and left of center on the histogram and dark compared to what a P mode shot gave me. I open up to f/8--better. Goint to f/5.6 is very nice. I know the hand-held meter is trying to expose for 18% gray, so why does the camera see it two stops or so different--just about every time? What the heck is a light meter for if you can't rely on it? BTW--I get the same results with another meter as well as with other digital cameras--they just don't seem to agree. I thought my photos would improve by taking more accurate and personal exposure readings with a hand-held light meter, but that is not the case. Also, is there a way to tell if exposure in the camera itself isn't up to speed? Maybe Canon meters low and doesn't want you to expose very close to the right for fear of blowing out highlights?

[SkipD]

The meter reading really sounds about right, using the "Sunny-16 Rule". That rule says that you should use f/16 and 1/ISO for the shutter speed. That works out to ISO 200, f/16, 1/200 shutter speed. Your meter was quite close to that. 1/250th second shutter speed gets you a bit less light over time than 1/200, and f/11 gets you more light than f/16. However, the f/5.6 at 1/250 and ISO 200 is WAY off from the rule.

Could it be that your camera actually has a problem? It wouldn't be the first time.

Are you possibly using a dark filter such as a polarizing filter on the lens? That would definitely play havoc with a meter's reading if you didn't compensate for the filter.

[Rudy M.]

I don't know if camera has problem or not--but I can exchange for another if it does. I forgot the 40D has a spot meter function so I'll play with it over the weekend getting close and metering that way and comparing what I get, and find a gray card and see what happens. The 40D does seem to be under exposing somewhat on its own, but not like when I manually set it to what the external light meter indicates. Just a thought--I wonder if it is the evaluative metering that is causing the differences in readings--perhaps I need to switch to averaging for this evaluation? Maybe just use the camera and forget about the external meters.

[Wilt]

Do not try to compare readings unless you are using an 18% gray card to fill the area seen by the reflected light meter...the incident meter is trying to give a reading equivalent to the 18% gray level. If you use a reflected meter and it sees any other density than 18% gray, it will be biased by the reflectivity of the surface of the items is is seeing at the time of the reading! An Irish bride in white satin gown will read differently than a black man in a tuxedo, for example.

And even if you use a gray card for the reflected light reading, meters do not necessarily agree from brand to brand or model to model! Each manufacturer can choose to use their own 'constant' that goes into the equation which determines the calibration point of a meter, and that can result in different readings.

So use the Sunny 16 rule of thumb to see if your camera is in the ballpark, or wildly out of whack... aim at a clear blue sky mid day, aiming with the sun on either shoulder and not in your viewfinder, and see if ISO 250 results in 1/250 f/16 or thereabouts.