Sekonic Light Meter

[ajohnw]

It may be worth mentioning what could be simple fix on this photo for the OP. Sometimes it's possible to develop 2 exposures from raw. In this case one that shows the grass fairly well and another that shows the children well then merge using HDR software. The children one would need to show them in some shadow. In my case this is a bit hit and miss because I have to guess the outcome when they are merged. Might be more adjustable on the fly in others.

The other alternative is the same approach and blend the 2 layers using layer masks. Hand painting masks often isn't as difficult as it might seem to be if some one has never done it before. Sometimes useful masks can be obtained from colour channels that need less hand work. Different package used and for different purpose but masks are shades of grey.

https://patdavid.net/2013/02/getting...d-white50.html

That person writes some excellent tutorials. They sometimes deal with the nuts and bolts of pp rather than automated plugins which give less control.

Where merging will fall down is if there is too much clipping in the raw file but some off the subject can be tolerated.

John
-

[DanK]

I believe it's how it works.That value the light meter is using can be taken from just one measurement or the result of more measurements like a matrix metering. But what comes out is just one value that's used to calculate the camera settings so the amount of light is corrected for the sensors/film sensitivity. And that will be placed somewhere in the middle of the diagram. The trick with the black, gray and white cat.
George

This is an essential point. Any metering system reduces an array of data to a scalar--a single number that serves as a proxy for a complete measure of the distribution of luminance--that is, for the information displayed by the histogram. (Of course, with at least some cameras, the histogram is distorted by the camera's jpeg conversion, which is why the choice of picture style can be important for obtaining a reasonably accurate histogram. However, that is a topic for another day.)

Understanding HOW a given approach to metering reduces the data to that one number is the key to understanding how to meter complex situations. Take, for example, George's black and white cat. The similar example I use is a scene with a lot of snow. If the metering is taken from the snowy areas--or if you use an averaging or evaluative metering that gives a lot of weight to the snowy areas--then you get an underexposed image, because the the meter will give you a setting that turns the snow (or the mix including snow) neutral gray. One way to handle this is with exposure compensation. Another is to think in terms of the zone system, spot meter off the snow, and open up about two stops. Yet another is to meter off your palm (exposing it like the snow) and open up a stop or so. To a newbie, these can seem like a bewildering set of unrelated methods, but in fact, they are all simply methods to force the meter to place the histogram where you want it, not where it's algorithm would automatically put it because of the snow--that is, not too far to the left.

[xpatUSA]

What you measure comes in the middle of the histogram. Like the white cat.
I don't know how you will place the ligbt meters reading in the middle of the histogram.
Unless you mean something like EC.
George

Beyond terse one-line statements, it will be well worth reading Section 3.5 of:

dougkerr.net/Pumpkin/articles/Secret_Life_Exposure_Metering.pdf

There we can learn exactly what exposure will be recommended by an incident-light meter. It is t/N^2 = C/(Es.S)

where t=shutter time, N=f-number, C= calibration constant=340 on my Sekonic, Es=illuminance of the light in lux and S=ISO setting on the meter, not the camera.

So let's say those kids are sitting in partial shade at say 2000 lux and Evert sets say 400 ISO because the kids are moving a bit.

In it's tiny brain, the meter decides that t/N^2 is 4.25x10^-4. From that, in camera exposure terms, my meter will recommend about 11EV. Since some people have no idea what that means, my meter also has a handy-dandy circular slide-rule scale that sets shutter speed against aperture. If, like Bill I decide f/4, the meter's recommendation on that scale is 1/125 sec for f/4.

None of the above has anything whatsoever to do with the resulting histogram in the developed image which, as George implies, has to do with the reflectivity of the various parts of the scene (dare I say Zones?) and the photographer's adjustments to the meter's recommended exposure.

[William W]

There’s a lot of (useful) information in this conversation and that’s primarily because the conversation has wandered a tad; so just to return to the beginning.

In the wandering of the conversation, it is possible that the OP might be pondering what are the more relevant bits to address the original, primary question which was asked.

Specifically, the OP asked: “What am I doing wrong [when using a Sekonic 308 Hand Held Light Meter?”]
(Here is the OP, repeated for convenience):

. . . I come here looking for some help. . . when I take a meter reading [Sekonic 308]... Say... of my son sitting sitting on the lawn in the back yard. I put the settings in my Canon 6d and then take the picture I get the blinkies saying the the highlights are blown. What am I doing wrong?

I think it is difficult to nail an answer to that question, without, firstly, the answers to the questions asked of the OP in Post #16, (here is an extract repeated for convenience):

IF your Sekonic Hand Held Light Meter gave you: F/5 @ 1/250s @ ISO200,

> how did you use the meter? what I mean is:
> to what function was the meter set?
> exactly where did you hold it?
> exactly where did you point it?

WW

[george013]

Beyond terse one-line statements, it will be well worth reading Section 3.5 of:

dougkerr.net/Pumpkin/articles/Secret_Life_Exposure_Metering.pdf

There we can learn exactly what exposure will be recommended by an incident-light meter. It is t/N^2 = C/(Es.S)

where t=shutter time, N=f-number, C= calibration constant=340 on my Sekonic, Es=illuminance of the light in lux and S=ISO setting on the meter, not the camera.

So let's say those kids are sitting in partial shade at say 2000 lux and Evert sets say 400 ISO because the kids are moving a bit.

In it's tiny brain, the meter decides that t/N^2 is 4.25x10^-4. From that, in camera exposure terms, my meter will recommend about 11EV. Since some people have no idea what that means, my meter also has a handy-dandy circular slide-rule scale that sets shutter speed against aperture. If, like Bill I decide f/4, the meter's recommendation on that scale is 1/125 sec for f/4.

None of the above has anything whatsoever to do with the resulting histogram in the developed image which, as George implies, has to do with the reflectivity of the various parts of the scene (dare I say Zones?) and the photographer's adjustments to the meter's recommended exposure.

The kids in your example will be in the middle of the histogram. Where they shouldn't be I think.
The result of your calculation will be in the middle of the histogram.
All the cats, grey,black and white, will be in the middle of the histogram if you spot metered them.

George

[xpatUSA]

The kids in your example will be in the middle of the histogram. <>.
The result of your calculation will be in the middle of the histogram.

The incident light meter, used as such, does not know that the kids are even there and it has no idea of their reflectivity. So your predictions are false, sorry.

All the cats, grey,black and white, will be in the middle of the histogram if you spot metered them.

Spot-metered? Are you still talking about reflected light metering? I am only talking about incident light metering.

Furthermore, if you had actually read this: dougkerr.net/Pumpkin/articles/Secret_Life_Exposure_Metering.pdf you would also know by now that your continual use of "the middle of the histogram" is incorrect (it's too high).

Perhaps you meant to say 99/255 (saturation-based method per ISO) or maybe 118/255 (standard output sensitivity method per C.I.P.A.) - both methods mentioned in the above link? . . . or perhaps not

Pardon my pedantry . . .

[pnodrog]

All the cats, grey,black and white, will be in the middle of the histogram if you spot metered them.

George

More or less. The reading simply indicates the exposure required to render the measured surface grey. It is up to the photographer to use this information to determine the correct overall exposure. If the photographer cannot do that then the exposure meter is of no use to them with the possible exception they use a grey card. Using a grey card will still be of limited value if they do not have at least a basic grasp of tonal relationships.

[xpatUSA]

Originally Posted by george013 All the cats, grey,black and white, will be in the middle of the histogram if you spot metered them

Absolutely. The reading simply indicates the exposure required to render the measured surface grey. It is up to the photographer to use this information to determine the correct overall exposure. If the photographer cannot do that then the exposure meter is of no use to them with the possible exception they use a grey card. Using a grey card will still be of limited value if they do not have at least a basic grasp of tonal relationships.

I fold.

It is quite clear to me that the OP was using his Sekonic as an incident light-meter whereas you gentlemen appear to be now talking about reflected light-metering, ("spot meter", "measured surface"), i.e. the old saw that reflected light meters try to turn any scene to an average of mid-gray.

One last try: with incident light metering, black cats will be black and white cats will be white and definitely NOT in the "middle" of the histogram.

[pnodrog]

I fold.

It is clear to me that the OP was using his Sekonic as an incident light-meter whereas you gentlemen appear to talking about reflected light-metering, ("spot meter", "measured surface"), i.e. the old saw that reflected light meters try to turn any scene to an average of mid-gray.

One last try: with incident light metering, black cats will be black and white cats will be white - definitely NOT in the "middle" of the histogram.

I was simply responding to George's comment which was obviously in regard to reflectance measurements. Even with incident light measuring, the overall exposure may well need adjustments based on whether the reading was taken in the tree shadow or out in the open...

With my luck the black cat would be lurking in the shadows and the white cat basking in the sun and not a grey cat to be seen.....

[pnodrog]

The approach nowadays with exposure (excluding SOOC jpeg) has very little to do with placing a tone in it's "correct" position. The exposure should be primarily based on capturing as wider range of tones as possible at the lowest noise while protecting either the highlights or shadow detail in accordance to their importance in the final image. Once captured the placement and relationship of the tones can be adjusted in PP. Hence ETTR etc. The approach has changed as we no longer have to worry about film characteristics but now sensor response..

[William W]

. . . It is quite clear to me that the OP was using his Sekonic as an incident light-meter. . .

I think that it is a reasonable bet the OP was intending to use the Sekonic 308 for an incident light meter reading.

But to be sure of that as fact: to ensure the use was correct: to more clearly identify other matters we have discussed here: (asking again), the questions in Post#16.

> how did you use the meter? what I mean is:
> to what function was the meter set?
> exactly where did you hold it?
> exactly where did you point it?

***

To be fair, Ted, Paul has an applicable and relevant point tucked in Post #49 –

Even with incident light measuring, the overall exposure may well need adjustments based on whether the reading was taken in the tree shadow or out in the open...

That has been one of my concerns from the beginning - the OP did ask “What am I doing wrong?”

Maybe the OP didn’t make an accurate Incident Reading or did make an accurate Incident Reading, but not the most appropriate incident reading, or not enough of them.

WW

[george013]

Ted,

Sorry I mixed up 2 different meterings.

While waiting for a reaction of OP, from post 1

when I take a meter reading... Say... of my son sitting sitting on the lawn in the back yard. I put the settings in my Canon 6d and then take the picture I get the blinkies saying the the highlights are blown

And from post 4

The meter was doing its job and giving me the correct exposure for my son and with out the use of flash to control the background I had to settle for what I got, Correct?

I think he is using it as a reflective meter.George

[xpatUSA]

I think that it is a reasonable bet the OP was intending to use the Sekonic 308 for an incident light meter reading.

To be fair, Ted, Paul has an applicable and relevant point tucked in Post #49 –

Even with incident light measuring, the overall exposure may well need adjustments based on whether the reading was taken in the tree shadow or out in the open...

Maybe the OP didn’t make an accurate Incident Reading or did make an accurate Incident Reading, but not the most appropriate incident reading, or not enough of them.

WW

Yes, Bill, I fully agree with Paul's quote and should have made it clearer that I wasn't disagreeing with it.

Now that George and consequently Paul have taken the opposite tack to ourselves, that the OP was using his Sekonic as a reflected light-meter, I guess we're dead in the water until the OP answers your repeated questions . .

[xpatUSA]

It may be worth mentioning what could be simple fix on this photo for the OP. Sometimes it's possible to develop 2 exposures from raw. In this case one that shows the grass fairly well and another that shows the children well then merge using HDR software. The children one would need to show them in some shadow. In my case this is a bit hit and miss because I have to guess the outcome when they are merged. Might be more adjustable on the fly in others.

Good point about making 2 exposures from raw. I use TuFuse and for an, exposure stack, one can play with weighting curves and levels while looking at a preview image before actually fusing the stack.

Apparently EnFuse does similar:

https://patdavid.net/2013/01/focus-s...os-enfuse.html

The other alternative is the same approach and blend the 2 layers using layer masks. Hand painting masks often isn't as difficult as it might seem to be if some one has never done it before. Sometimes useful masks can be obtained from colour channels that need less hand work. Different package used and for different purpose but masks are shades of grey.

https://patdavid.net/2013/02/getting...d-white50.html

That person writes some excellent tutorials. They sometimes deal with the nuts and bolts of pp rather than automated plugins which give less control.

Where merging will fall down is if there is too much clipping in the raw file but some off the subject can be tolerated.

John
-

Thanks for the link - I have the GIMP but use it rarely.

[William W]

"I fold" and "dead in the water" - metaphors from previous and that should be more often used: very visual and thus most suitable for a Forum of Photographers.

night night . . .

[pnodrog]

"Good point about making 2 exposures from raw"

Beats me as to any real benefit in the approach of separating the information then putting it back together again.
Maybe it makes a relatively easy process even easier to control but it seems a bit cumbersome to me.

P.S. I see John has covered it in his post #41. So it is really just a matter of preference but it is important to realise that if the information is not in the file it cannot be recovered either way.

[xpatUSA]

"Good point about making 2 exposures from raw"

Beats me as to any real benefit in the approach of separating the information then putting it back together again.

Indeed. By the same token, I don't see much point in splitting an image into layers and masks and messing around with brushes and stuff followed by blending and merging - but a lot of people do it, apparently. LOL

[pnodrog]

Indeed. By the same token, I don't see much point in splitting an image into layers and masks and messing around with brushes and stuff followed by blending and merging - but a lot of people do it, apparently. LOL

Control freaks..... I'm one of them.... (and I find it faster but that may well be because of a difference in familiarity)

[xpatUSA]

It may be worth mentioning what could be simple fix on this photo for the OP. Sometimes it's possible to develop 2 exposures from raw. In this case one that shows the grass fairly well and another that shows the children well then merge using HDR software. The children one would need to show them in some shadow. In my case this is a bit hit and miss because I have to guess the outcome when they are merged. Might be more adjustable on the fly in others.
-

Just had to give it a shot - it seems to work. I set up a scene outside with a chicken in part-shade and sunlit grass in the background:

SPP** review image:



bottoming and clipping warnings set at 1 and 254.

** Sigma Photo Pro V4.2 proprietary raw converter.

2 "exposures" from one raw file:



exposures fused with TuFuse:



No editing done to the above although it could do with a bit . .

This was a quick try to see if it works. I am certain that a dedicated PS expert could have made the original look much "better" than the above.

[Evertking]

I'm waaay in over my head in here. Lol..
"What am I doing wrong?"
When I asked that, I was looking for an answer to why when I use my light meter I am getting blown out highlights.
It was my inexperienced thinking, that if I use it and set the settings, I would get a perfect picture. I see now that something in the picture has to give and it was the sky behind my children.
I think...