canon speedlight 580es and sekonic meter

drh681 wrote:

just out of curiosity, why are you trying to meter ETTL flashes?
the camera will do that for you.
the ratios are easy, each full step is ONE ƒstop. more and less.

The ratios set on the master flash don't translate into the expected results. Measuring the incident output of each flash is the only way to determine what the ratios are actually producing in terms on light output.

The steps in the Canon flash ratios do not indicate a full stop. 8:1 iindicates that the the A zone is 8x or 3-stops brighter than the B zone That is straight out of the 580EX manual.

There are two types of ratios with respect to lighting.

Incident = light hitting subject from each light, measured at the point the light hit the subject

and

Reflected = light bouncing off subject into camera.

When more than one light hits the same area (i.e., highlights) the sum of two incident values is reflected. That is why if you use two lights of equal strength at the same distance (1:1 INCIDENT ratio) with one on the camera axis (fill) and the other to the side (key) a highlight / shadow pattern is created. The brightness in the highlights is 1 unit key + 1 unit fill. The brightness in the shadows is only 1 unit fill. The REFLECTED or HIGHLIGHT:SHADOW ratio is 1+1:1.

If key is 2x (1-stop) brighter than fill the H:S ratio is 2+1:1 or 3:1
If key is 3x (1.5 stop) brighter the H:S ratio is 3+1:1 or 4:1
...
If key is 8x (3 stops brighter the H:S) ratio is 8+1:1 or 9:1

About the Canon ratio markings:

With the Canon flashes the on-camera light, which has the role of fill, must be the A zone which the off camera light is set to B. The resulting A:B, or Fill:Key notation is bassakwards from conventional H:S lighting ratio notation.

Next there's the question of what those ratio numbers actually mean. Used in a conventional portrait situation the neutral camera axis fill (A zone) would never be made brighter than the key (B zone) so all the ratios 8:1 - 2:1 are useless.

It's also not clear what 1:1 is meant to produce. Equal incident output from both A and B zones? How can the camera evaluate incident strength with a reflective reading if the two lights are hitting different parts of the subject?

Could it be meant as a refected H:S ratio? Not likely. The only way you could ever get a 1:1 reflected ratio would be if each light on opposite sides only illuminated exactly half the face or object. The two sides would appear equal in brightness the same as if fact lit with one light. Not much point in that is there?

I would be very interested in knowing the actual incident light meter readings for a dual 580EX configuation for each of the ratio settings in two different settings: first a subject in a white shirt, and then a subject in a dark shirt. Indentical oblique poses. Fill on camera, key 45 degrees off the nose of the subject (90 degrees off camera axis).

I know based on 30+ years of experience using manual flashes what H:S ratio two equal lights at equal distances produces; 2:1 If I want a 3:1 H:S ratio I just move the key light closer. I use these easy to remember distance combinations for 3:1 H:S - 16ft/11ft, 11/8, 8/5.6, 5.6/4 My benchmark exposure for the two flashes is f/8 @ 11ft. If I move in to 8ft, I move the key into 5.6 ft (my arm span) and close the lens to f/11. If I move into 5.6ft I move the key light to 4ft to maintain the 3:1 H:S ratio and stop down to f/16. Out to 16ft? Key to 11ft, f/stop to f/5.6. This simple to execute 4-stop, 16 - 5.6 ft range covers 99.9% of my flash shooting scenarios. It's relable, predicatable and only requires a pair of $90 Vivitar 285HV flashes.

I'd like to know a pair of 580EXs could do the job at least as well and predicably -- without Kentucky windage or a Owiji board to devine the meaning of the ratios - before shelling out $800.

CG

-
Tutorials listed here:
http://forums.dpreview.com/forums/read.asp?forum=1025&message=14317618