Lunasix F. Sekonic Twinmate. How accurate is "accurate"?

[elerion]

I've got an old Lunasix F, which works fairly well.
I just compared its readings with a new little Sekonic Twinmate.
The Sekonic usually reads about 1/3 or 1/2 stop higher.
They are so close, that I don't know how to decide which one is really spot on.

Digital images taken with both reading look perfect. Histogram could help. But digital cameras are not usually ISO precise.

Analog images are great too, as expected, but developing adds some little unavoidable offsets in density, so I cannot use them to determine which reading is "the good one".

Checking the light readings against an old EOS 500 (only reflected mode, of course), it matches those of the old Lunasix F.

I thought that a new meter should always work better than an old one.
Is it feasible that the new Sekonic is not as precise as the old Lunasix F?
or is it most likely that the old meter needs some recalibration?

 

[AgX]

I thought that a new meter should always work better than an old one.

It depends on the circuit and the used types of electronic elements. In some circuits only elements are used that hardly age. The resting troublemaker then could be the contacts.

 

[bdial]

a 1/3 - 1/2 stop difference isn't significant.

There is no real standard that light meters are calibrated to manufacturers get to choose their own, so a small difference between a Sekonic and a Gossen isn't surprising.

 

[benjiboy]

I think the minor discrepancy an the exposure readings can be accounted for by the the difference in the two meters acceptance anges, the Lunasix F.s being 30° and the Sekonic Twinmate being 33° and so small to be insignificant.

 

[AgX]

There is no real standard that light meters are calibrated to manufacturers get to choose their own.

Not true.

In the past there was DIN 19010 in 3 editions.
Now there is ISO 2720:1974-08.

 

[elerion]

I thought about the acceptance angles, and even though they are almost identical, I took measurements against a much bigger uniform surface.
Which lightmeter should I take as more accurate? A new low-end like the twinmate, or a better build but old one? (Lunasix F is already silicon based and uses a regular 9V battery, not like older CdS ones, which I think could lose some accuracy easier).
Any experience?

 

[Chan Tran]

I would pick the new low end one for accuracy. The technology in sensors and amplification have improved a lot so the new ones are more likely to be more accurate.

 

[guangong]

Does it really matter? Even with a meter the brain must be engaged for judgement. Minuscule differences of a 1/3 or 1/2 stop is only significant in cinematography or commercial still pictures Nader controlled lighting conditions. Since both meters give good results can you really see any difference? My Gossen Luna pro f, Weston ranger 9, Gossen pilot and sekonic twinmate all give similar but not exact readings. I just made these comparisons because just bought and checked out a newly purchased Nikon dp-1 meter. It was also within the range of reasonableness. The meter is just one of a whole chain of numerical values which may not be exactly what the numbers indicate and only practice can reveal.

 

[benjiboy]

Confucius he say " a man with one exposure meter knows the correct exposure, one with many is never sure" .
I have three modern Silicon cell digital exposure meters a Sekonic L358, a Kenko KFM 2100, and a Gossen Digipro F they all give slightly different exposures when taking a reading off a Kodak 18% Grey Card in the same light, yet each when used individually with a camera and film give perfectly acceptable exposures, this is because the parameters the manufacturers used when designing them were different.

 

[Pioneer]

I jump from Sunny Sixteen, to Weston, to Voigtlander, to Sekonic, sometimes all on the same roll of film. Once the roll has been developed I can't tell which meter was used for any given photograph.

Heck, I sometimes even use different lenses on the same roll of film.

Once in awhile I will even use the same roll of film in more than one camera...



I still can't tell the difference.

 

[elerion]

Thank you!

 

[Chan Tran]

Does it really matter? Even with a meter the brain must be engaged for judgement. Minuscule differences of a 1/3 or 1/2 stop is only significant in cinematography or commercial still pictures Nader controlled lighting conditions. Since both meters give good results can you really see any difference? My Gossen Luna pro f, Weston ranger 9, Gossen pilot and sekonic twinmate all give similar but not exact readings. I just made these comparisons because just bought and checked out a newly purchased Nikon dp-1 meter. It was also within the range of reasonableness. The meter is just one of a whole chain of numerical values which may not be exactly what the numbers indicate and only practice can reveal.

If it doesn't matter then I don't need a meter. When I use the meter it's better be accurate.

 

[elerion]

Well, that's the reason why I got them. Specially for zone system testing. If meters are 1/3 or 1/2 stop away, then it's impossible to get an exact zone V. But if many people find readings to be just a little bit off between models, then that seems to be the best we can get. This is a science, but not an exact science.

 

[MattKing]

Well, that's the reason why I got them. Specially for zone system testing. If meters are 1/3 or 1/2 stop away, then it's impossible to get an exact zone V. But if many people find readings to be just a little bit off between models, then that seems to be the best we can get. This is a science, but not an exact science.

What is important is a combination of consistency with at least reasonable accuracy.
You want a meter that will give repeatable results that are reasonably close to a specified standard.
It is the consistency that allows you to first calibrate your procedures, and then obtain repeatable and useful results when you actually take photographs.
Both meters mentioned will give you sufficiently accurate results if they are working normally and you use them consistently. You may get inaccurate results if you switch back and forth between the two.

 

[Bill Burk]

Depends on what you have got for test purposes, I could imagine a test to find which meter is right.

I'd setup a target and pick a camera with a good shutter. I'd meter the target and set the manual f/stop and shutter speed recommended by the exposure meter and shoot a test for that meter. Do the same with the other meter. To get precise exposures, a pair of ND filters come in handy: 0.1 ND and 0.2 ND so you can set the camera in third-stop increments to match the meter reading to the nearest third-stop the meter reads.

Expose a step wedge onto the film as well. Develop to the best of your ability to the ASA gradient. Then read all the targets with a densitometer. Graph the results and look for the point that falls closest to the density corresponding to "ten times the speed point".

 

[elerion]

Develop to the best of your ability to the ASA gradient. Then read all the targets with a densitometer. Graph the results and look for the point that falls closest to the density corresponding to "ten times the speed point".

Good point.
I currently don't need to get that precise. But the test seems to be rigorous enough.

What would happend if developing pushes/pulls a little bit? While I get very close ISO for some film/dev combinations, absolutely exact development is very difficult to achieve. If developing the step wedge exposure along with the other exposures, wouldn't it be irrelevant the development strength up to a reasonable point?