Guide Number vs Light Meter

[bvy]

I have various old flashes of which I wanted to verify the guide numbers. For instance, I have two Olympus flashes for the XA cameras -- the A11 and A16 (those are also the guide numbers in meters at ISO 100). So I set my light meter to ISO 100 and fired the flash at it (at full power). Both flashes registered one stop over what the guide number suggests. For example, if the guide number suggests f/11, the meter says f/8. I tried it with an accessory flash (Sunpak 383) and it's the same thing. At least it's consistent, but I have to doubt that all the flashes are off. The meter is a Sekonic L-308B. I have the dome over the sensor, and I'm shooting from one meter away. The meter is dead on, in my experience, for my large format portrait work -- even with the IP instant film I'm using that has very little latitude. So I'm not sure it's the meter either. Everything has fresh batteries. Is my methodology flawed?

 

[MattKing]

All of the guide number calculations and auto functions on flash include assumptions about reflections from ceilings, walls and floors. If there are no such surfaces involved in your tests, then you need to adjust.
I would also suggest checking your flash meter - it too may be calibrated based on similar assumptions.

 

[wiltw]

All of the guide number calculations and auto functions on flash include assumptions about reflections from ceilings, walls and floors. If there are no such surfaces involved in your tests, then you need to adjust.
I would also suggest checking your flash meter - it too may be calibrated based on similar assumptions.

^^
A number of us on a different photo forum (a 'di*ital' one) have tested our speedlight flash units with our flash meters over the years, and we have found -- regardless of brand/model of flash and regardless of brand/model of flashmeter -- that flashmeter readings are very typically -1EV lower in exposure than suggested by the manufacturer Guide Number!

I just tested my Metz 45CL4 and rated GN148 means f/14.8 at 10', and the calculator dial says about that same value. But the Minolta Autometer Vf says f/8.0 + 0.66EV or about GN100.

 

[AgX]

All of the guide number calculations and auto functions on flash include assumptions about reflections from ceilings, walls and floors. If there are no such surfaces involved in your tests, then you need to adjust.

But thus without walls and a ceiling or with them but with direct metering without dome (but calculating for the lacking opacity of it) one would likely get underexposure. (Reflections are part of the lighting that makes up the GN.)

The OP however gets overexposure.

 

[Sirius Glass]

The dome is used for incident light readings. Try RTFM before posting.

 

[cowanw]

I would have thought firing the flash at the metres was not the correct methodology. guide numbers are after all based on reflection of the light off the subject.

 

[AgX]

The dome is used for incident light readings. Try RTFM before posting.

I know, but nevertheless one can use a meter for incident metering without dome if one, as I said, takes the lack of its opacity into account. And in same cases such metering makes sense.
But when establishing a guide number one should use a dome.

 

[AgX]

I would have thought firing the flash at the metres was not the correct methodology. guide numbers are after all based on reflection of the light off the subject.

You either meter:

-) in reflected mode using the plain meter and a neutral grey card.

-) in incident mode using the meter with a dome.

The latter method is the better method.

 

[bvy]

Sigh. Segal's Law.

 

[cowanw]

You either meter:

-) in reflected mode using the plain meter and a neutral grey card.

-) in incident mode using the meter with a dome.

The latter method is the better method.

My error!

 

[MattKing]

But thus without walls and a ceiling or with them but with direct metering without dome (but calculating for the lacking opacity of it) one would likely get underexposure. (Reflections are part of the lighting that makes up the GN.)

The OP however gets overexposure.

If the meter is being used at the subject plane, pointed back to the flash, the incident dome should be in place.
The guide number and auto functions are calibrated for medium distances and in situations where reflection off surfaces is assumed.
The auto function, in particular, depends on the subject reflecting a fair percentage of the light back to the flash sensor.
The OP's test using a one meter distance most likely differs a lot from the "typical" - the conditions used to calibrate both guide numbers and auto functions. All I am saying is that both systems need to be checked in the relatively unusual circumstances that the OP is working in.
By the way, with respect to the auto functions built into the flash, over-exposure is not uncommon if the subject is rather small and there aren't a lot of nearby surfaces to reflect light back to the sensor.
Sort of like those shots of the wedding couple and their first dance at the darkened hotel ballroom.

 

[wiltw]

As I said earlier, I did a test today: Metz 45CL4 and rated GN148 means f/14.8 at 10', and the calculator dial says about that same value. But the Minolta Autometer Vf -- with white hemisphere in place -- says f/8.0 + 0.66EV or about GN100....IOW my testing -- at the identical distance listed for published manufacturer specs for GuideNumbers (10' or 3 meters) reflects what the OP observed, and what a lot of others have confirmed in testing of their own.

 

[Sirius Glass]

The dome is used for incident light readings. Try RTFM before posting.

Since we're talking about the Sekonic 308, strictly speaking, there is an accessory attachment for use with flash metering, as the regular incident dome is not really the correct thing to use for the purpose.

Which is what I said. RTFM and follow the directions. If one thinks that they know more than the manufacturer, then they are just wrong.

 

[John Koehrer]

Manufacturers ratings are usually ASA 100 @ 10' or 3 meters.
I believe it's based on a typical ceiling height and an "average" room. As everyone else says.
I don't think I've ever seen a GN and a flash meter agree.

 

[bvy]

Which is what I said. RTFM and follow the directions. If one thinks that they know more than the manufacturer, then they are just wrong.

No one is second guessing the manufacturer. I've read the manual. I have no idea what you're talking about.

 

[wiltw]

Since we're talking about the Sekonic 308, strictly speaking, there is an accessory attachment for use with flash metering, as the regular incident dome is not really the correct thing to use for the purpose.

Sorry, Sekonic documentation contradicts what you just said.
The hemisphere is used for ambient incident or flash incident readings,
the Lumidisc is only for flat art or to measure lighting contrast.
Flash exposure metering is done with the ordinary hemisphere

There is no optional accessory that you can buy to serve any purpose related to flash metering per se...the gray card is for either ambient or flash

 

[bvy]

Since we're talking about the Sekonic 308, strictly speaking, there is an accessory attachment for use with flash metering, as the regular incident dome is not really the correct thing to use for the purpose.

Yes, the lumidisc, which I don't have but which is described for metering flat surfaces or calculating lighting ratios. In this case, I'm after a reading which simulates the "average room" scenario as @John Koehrer describes above. The dome seems to be the best tool for this. Even then, though, this is starting to sound like a futile exercise.

 

[RalphLambrecht]

I have various old flashes of which I wanted to verify the guide numbers. For instance, I have two Olympus flashes for the XA cameras -- the A11 and A16 (those are also the guide numbers in meters at ISO 100). So I set my light meter to ISO 100 and fired the flash at it (at full power). Both flashes registered one stop over what the guide number suggests. For example, if the guide number suggests f/11, the meter says f/8. I tried it with an accessory flash (Sunpak 383) and it's the same thing. At least it's consistent, but I have to doubt that all the flashes are off. The meter is a Sekonic L-308B. I have the dome over the sensor, and I'm shooting from one meter away. The meter is dead on, in my experience, for my large format portrait work -- even with the IP instant film I'm using that has very little latitude. So I'm not sure it's the meter either. Everything has fresh batteries. Is my methodology flawed?

I would trust your meter.guide numbers can only be very rough estimates at best.They ignore the surroundings and thereby the bounce light from walls and ceiling but a flash meter measure the incident light actually arriving at the dome;much better approach.

 

[AgX]

I would trust your meter.guide numbers can only be very rough estimates at best.They ignore the surroundings and thereby the bounce light from walls and ceiling but a flash meter measure the incident light actually arriving at the dome;much better approach.

To the contrary, GN's include reflections.
BUT as the average user does not know the test-space they indeed will be vague to some extent.

 

[benjiboy]

Japanese flash guns always have optimistic guide numbers they must test them in rooms with white walls and ceilings. In my experience German flashguns like Metz have more accurate guide numbers because they have to conform to D.I.N. standards.

 

[wiltw]

Japanese flash guns always have optimistic guide numbers they must test them in rooms with white walls and ceilings. In my experience German flashguns like Metz have more accurate guide numbers because they have to conform to D.I.N. standards.

Ben, go back and read my post (4 or 13)...that was a METZ.

 

[RalphLambrecht]

As I said earlier, I did a test today: Metz 45CL4 and rated GN148 means f/14.8 at 10', and the calculator dial says about that same value. But the Minolta Autometer Vf -- with white hemisphere in place -- says f/8.0 + 0.66EV or about GN100....IOW my testing -- at the identical distance listed for published manufacturer specs for GuideNumbers (10' or 3 meters) reflects what the OP observed, and what a lot of others have confirmed in testing of their own.

the important part of your test is:which gave thebetter exposure,GN or flash meter?

 

[wiltw]

the important part of your test is:which gave thebetter exposure,GN or flash meter?

Ralph, I'll let you be the judge.

1. One exposed per Metz Guide Number 148' for 45CL4 -- actually slightly overexposed vs. GN (since GN would have f/14.8) and I exposed at f/14;
2. Two exposed per Minolta Autometer Vf reading (ISO 100, 1/100, f/8 + 0.6EV). No flash automation used. All postprocessing setting Zero'd, no adjustments.

Subject distance 10'.

BTW, density values on 18% gray card vs. 4th square bottom row of ColorChecker
GN exposure: 28%, 50%
Meter exposure: 43%, 65%

 

[Mainecoonmaniac]

When I shot weddings back in the film days, I used a camera without TTL flash. To this day, I still don't own a TTL camera. I used an automatic flash a Vivitar 285 which used a sensor that cut the flash off when there was enough light hitting the sensor. I mostly used it in manual mode. I read the focus distance of my subject then adjusted the F/stop according to the calculator. The guide number just worked. I also worked in a photo studio where the photographer just used a string that indicated predetermined f stops indicated with knots. Back in the 80's, flash meters were expensive so I didn't use one.

 

[MattKing]

There are too many intervening variables, but the exposure on the left appears better.

 

[wiltw]

There are too many intervening variables, but the exposure on the left appears better.

BTW the water pitcher inherent brightness of its white areas is within 0.1EV of the brightness of the white square on the ColorChecker