Light meters with reflective, incident and spot metering

[Craig75]

Here's a pointless addendum just to stir the pot up even more

Will the safety factor of the film not just account for flare and/or basic shutter / aperture discrepancies?

I'm assuming film has a stop of safety built in but i never really understood all of that and its not something i see mentioned much

 

[138S]

I'm assuming film has a stop of safety built in but i never really understood all of that and its not something i see mentioned much

Film had 1 stop safety factor before 1960 when it was removed. That year speeds on the boxes doubled without any change in the manufacturing. Now we have no safety factor, still you may add a persnal safety to your exposure to make sure your shadows are ok.

Please see this graph... Light in the Hor axis (Log scale) density in the vertical axis (Log scale)

When you enter the film ISO in the meter, an spot reading +/-0 will be situated in the red vertical line (green square), before 1960 the meter aimed it was in the n point.

m is the "speed point", the exposure at what film starts building density. ISO norm considers places that exposure point where density has increased 0.1D over base + fog. There is another requisite for the ISO calibration, development should be nailed to increase density (over the density in the speed point) by 0.8D if exposure is increased by an x20 factor, whick is 4.3 stops. But from the meter point to the speed point we always have (in the ISO calibration conditions) 3.33 stops. Meter point to the speed point = 3.33 Stops.

For this reason, nominally, you always have 3+1/3 stops latitude. If you shot manual and use spot mode, an spot reading -3.33 will be in the speed point, mostly with texture destroyed.

Of course some developers will add or loss some speed (1/3 stop?), a reference Full Speed developer was D-76, but today a manufacturer may use the developer he wants to make the ISO calibration, he should say what one. For example Foma uses an speed increase developer to say the speed rating, what can be a bit missleading.

_______

For advanced metering, you always keep in mind where the m point is, what image quality you have there (at -3.3) and how it improves at say -2.3 or -1, depending on film and processing. Each spot in the scene has it's particular local exposure, you may read it if you want.

The more accurate you meter and expose (shutter tested?) then the more you can play in the underexposure Zones having a controlled result, and having to overexpose less highlights. So you may want to need a lower personal safety factor, you get that from consistency in the workflow, which is quite appreciated for LF: You shot way less and want all your shots on spot if possible.

 

[Craig75]

I would say treating flare as a safety factor is not a good idea. Refer to the diagram I posted. Even if we assumed the flare factor can be accurately predicted, flare is not simply uniform extra exposure you can offset by underexposing. Flare compresses contrast/detail, and it does this most at the low end of the exposure scale. The last thing you want to do is destroy even more shadow contrast by moving those compressed shadows further down the characteristic curve where the film itself has less contrast.

in that example though you would just meter at say 50 for a 400 speed film and try and eliminate the toe altogether if shadow contrast is important to you.

if you cant shoot that slow for other reasons (you need a specific shutter / aperture combo, thinnest negative, stop a highlight going on to the shoulder, want to block up shadows etc) then you dont have much choice in the matter but to measure the scene at a higher iso but there is still the option to dump some garbage into the safety factor of the film.

i always thought film when from a two stop safety factor to a one stop safety. is that not the case?

 

[BrianShaw]

Here's a pointless addendum just to stir the pot up even more

And in that spirit... what about the k-factor (constant) of the meter. Are all meters really telling the truth?

 

[138S]

I would say treating flare as an additional safety factor is not a good idea. Refer to the diagram I posted. Even if we assumed the flare factor can be accurately predicted, flare is not simply uniform extra exposure you can offset by underexposing. Flare compresses contrast/detail, and it does this most at the low end of the exposure scale. The last thing you want to do is destroy even more shadow contrast by moving those compressed shadows further down the characteristic curve where the film itself has less contrast.

A solution is a compendium shade + multicoating, many times flare can be reduced to 1/30 of the original level...



If not we may even end in an SCIM masking mess, to not have avoided flare when in the field.

And in that spirit... what about the k-factor (constant) of the meter. Are all meters really telling the truth?

Meters may have a 1/6 stop missmatch, from the constants used in the formula...

Still IMO we have an additional missmatch from different spectral sensitivities in different meters, at least this is important when our subject is color saturated.

 

[MattKing]

Paging Dr. Benskin...

Somehow I could see Stephen carrying a pager...

 

[Bill Burk]

When I mentioned you can decrease exposure 1/3 to a half stop, I didn't mean dial in 640 for your 400 film as a matter of course (but you can). I just meant you can get away with it when you need to. I still dial in 250 for 400 film because after all the studying I have done, I have decided for myself that I want better negatives for less difficulty in the darkroom, and I believe that a slightly greater exposure will give me that.

Here's the 1960 Nelson paper:

http://beefalobill.com/benskin/content/safety factors in Camera Exposure.pdf

 

[Bill Burk]

Paging Dr. Benskin...

Here's Stephen's paper on K

http://beefalobill.com/benskin/content/Defining K.pdf

 

[BrianShaw]

Thanks Bill.

 

[grat]

In average lighting, flare brings exposures that you think are at the speed point up in density as if they had received 0.4 greater exposure (about 1 1/3 stop).
This is "previewable" on a spotmeter. You can also think it through, walk up to a car on the street and meter the light underneath. Then walk away from the car and meter the same light underneath the car. From where you are now, the new reading will be higher due to flare.

It could be argued that your video demonstrates a change in ambient lighting, more than flare-- I assume, in order to get closer to the dog, you were literally closer-- so you were blocking reflected light from the pavement (even in shadow, you being next to the dog alters the reflected light around the dog).

Even so, as you point out, that's a heavily back-lit scene-- unless you want the garish, washed-out background, your metering should incorporate the highlight areas as well. Personally, I would meter the dog's fur, as it appears to be the darkest area, "scan" the shadows between me and the dog, and under the table, and also meter the area behind the dog (the brightly lit background).

Your example of metering under the car while next to it, is irrelevant for metering a photograph taken some distance from the car-- all you're really demonstrating is that you need to meter the scene you're actually going to photograph. The "flare" as you put it, is going to affect the film as well as the metering, so it should be part of your measurements.

 

[grat]

And in that spirit... what about the k-factor (constant) of the meter. Are all meters really telling the truth?

No. But that's OK, the film isn't recording the truth either.

 

[138S]

When I mentioned you can decrease exposure 1/3 to a half stop, I didn't mean dial in 640 for your 400 film as a matter of course (but you can). I just meant you can get away with it when you need to. I still dial in 250 for 400 film because after all the studying I have done, I have decided for myself that I want better negatives for less difficulty in the darkroom, and I believe that a slightly greater exposure will give me that.
Here's the 1960 Nelson paper:
http://beefalobill.com/benskin/content/safety factors in Camera Exposure.pdf

Bill, this Jan. 1960 paper is quite interesting... it just paves the way for the 1960 Film Speed Change, when box speeds were doubled but contained the same film.

Interestingly, it considers that speed point to meter point was H = 1.57, this is 4.7 stops, and not 4.33




IMO that paper was to justify the Speed Change that was in course that year... and probably that change was aimed for the consumer segment.


Of course consumers obeyed and dialed the new speeds, exposing the hallf in 1961 than in 1959. Instead Pros and Artists mostly ignored the new stamps on the boxes. Suposedly they had their workflow well tunned, and it would be LOL thinking they were to change how they worked because of an stamp.

Still the 1960 x2 speed new stamp still generates controversy today, 60 years later.

In the ZS some some say that meter has to aim Z-4 instead Z-5. Others rate speed to the half (250 to 400 is nearly the half), and some say that very deep shadows have to be placed in Z-3 instead Z-2. Well, it's all the same... this is exposing like in 1959 before the speed change, and just before that paper was published.

_______

Another interesting factor is how Nelson defines Safety Factor: as the distance from speed point to "shadows"... but each shot has a particular shadow depth compared to metering.

IMO, that speed change had to be favoured for consumers, it was mostly benefical, but that paper sports some conceptual flaws when generalizing about what is the "typical shot".

Anyway it was aimed to consumer segment... Artists/Pros of the era laughed a lot about this matter.

 

[Neil Poulsen]

I have a Sekonic 508 which has all three, plus more. Includes variable spot (between 1 deg and 5 deg) incident and reflective meter readings. Also includes broad 30 deg incident readings with a retractable dome. And, it can do all this for either ambient or flash meter readings.

But for spot, I miss being able to read EV's through the viewerfinder. So instead, I can take a 1 deg spot reading through the viewer, but then I have to take my eye away from the viewerfinder to see the reading on the external digital display. For this reason, for spot readings, I prefer using my Minolat II digital spot meter that displays either f-stops or EV's in the viewfinder.

 

[bonk]

I just found a nice little device that can do incident metering, reflective metering and reflective spot metering all in one device and it is extremely small because it attaches to the phone.

https://lu.mu/

Did anyone use it? Is it any good?

 

[138S]

I just found a nice little device that can do incident metering, reflective metering and reflective spot metering all in one device

Probably an Android App in the phone will deliver very matching results for $2, or for free





You can check the App reading with the spot metering from a SLR/DSLR on a (Kodak) grey card.

The incident reading from the App using the front camera reads only the wide field seen by the camera, but it should be enough.

Only suggesting that first you may check (with spot metering on tha grey card) if the App reading is good enough for you...

 

[Alan Edward Klein]

Probably an Android App in the phone will deliver very matching results for $2, or for free


View attachment 263836


You can check the App reading with the spot metering from a SLR/DSLR on a (Kodak) grey card.

The incident reading from the App using the front camera reads only the wide field seen by the camera, but it should be enough.

Only suggesting that first you may check (with spot metering on tha grey card) if the App reading is good enough for you...

The APP can only be as accurate as the phone's light sensor. Are there published charts of tests for this?

 

[138S]

The APP can only be as accurate as the phone's light sensor. Are there published charts of tests for this?

Alan, mine it is, and it is a cheap ($250 ?) Xiaomi Note 8 Pro, just tested it right now to show it to you, see bellow.

The App may use the light sensor for screen auto-bright or the front camera. If it uses the light sensor then reading may differ because that sensor is usually very directional.

Still, best is purchasing a (cheap) lux meter and also comparing to spot reding on a grey card to make sure both the phone and the App are working ok. The great thing is that if it ends working OK then the investment consists in a some $20 lux meter that may also be used in the darkroom for film calibrations.

 

[bonk]

Please note that I was talking about the Luma sensor for the phone. So there is indeed a separate sensor that is advertised as being extremely accurate.

But today I learned that the external sensor is (of course) only used for incident metering. For reflective metering (spot and „normal“), the phone‘s camera is used.

 

[138S]

Please note that I was talking about the Luma sensor for the phone. So there is indeed a separate sensor that is advertised as being extremely accurate.
But today I learned that the external sensor is (of course) only used for incident metering. For reflective metering (spot and „normal“), the phone‘s camera is used.

Yes, of course. My intention was to suggest that you may save $200 to $400, having also very extremly accurate readings, without having to carry an accessory, but pointing that it has to be checked if the phone-app combination is accurate, so it was not a direct answer... only telling an idea I judged it could be useful and cheap.

 

[JBrunner]

yup. I agree. There is no need for a spot meter. Ever.

Funny, a spot meter is all I use...it's hard to get a mile across the canyon with my meter and then run and climb a mile back to the shady side to make the exposure...guess I'm just lazy.

Kidding aside, if I was stuck with one meter, I'd choose a spot meter. I find it far more informative about the parts of the exposure that actually matter, and with the application of some common sense, its very easy to use it as an incident meter.

 

[Alan Edward Klein]

Aren't there issues with cellphone meter readinss of either incident or reflective?

 

[Donald Qualls]

The metering app I use on my phone (original Pixel) seems accurate; it's either averaging or spot, and shows a circle on the screen when using spot. Pretty big spot, but when I need real spot, I take along my Pentax 21/1/

 

[138S]

Aren't there issues with cellphone meter readinss of either incident or reflective?

Alan, sensitometry for digital sensors is well standarized, today ISO 12232:2019 is in force. That standarization allows an smooth integration of software vs hardware, as many teams are involved in development of those gadgets.

Yes... we may have some discrepances because sensors have particular spectral responses and some colors may influence more than others... but this also happens with classic meters, see the spectral sensitivity of the Pentax V in the manual... it may not match our film's spectral sensitivity and saturated colors in particular may have a missmatch imn the reading vs predicted density.

For BW we many times use filters... a Red Filter will have a different "correction factor" depending on subject's color and on film red sensitivity, depending on how deep is the red of the filter different films behave different: we have Pan and Ortho, but may OrthoPan mid points...

The smartphone always nail the shot bacause the Preview on screen is used in the calculations, and (today) a lot of Artificial Intelligence is dedicated to nail exposure and HDR compressions. Still probably those Apps take the (ISO calibrated) raw image straight from sensor, so a precise evaluation can be delivered.

Funny, a spot meter is all I use...it's hard to get a mile across the canyon with my meter and then run and climb a mile back to the shady side to make the exposure...guess I'm just lazy.

Kidding aside, if I was stuck with one meter, I'd choose a spot meter. I find it far more informative about the parts of the exposure that actually matter, and with the application of some common sense, its very easy to use it as an incident meter.

+1, no doubt spot metering it is the most informative and the most suitable way when scene is challenging and we want a good prediction of the result.


Still IMO incident can be very suitable for studio portraiture, we may be moving a lot key and fill illumination... and the skin tone may influence the spot reading by two stops when considering dark skin, a full stop difference in caucasian skins alone... also glares may jam the spot reading... so in that situation IMO incident may be superior.

Also when we shot slides with our subject evenly illuminated , and having the possibility to meter the light reaching the subject. In that situation incident may provide a more consistent reading than spot, if we don't find gray references in the subject. Still we always can use a grey card to have the incident metering with the spot meter...

Perhaps we may agree that the spot meter assited (when required) with a grey card is to provide total information. Next step would be adding an Spectromaster or similar gadgets used in Hollywood movies by colorists, but IMO this is another war.

 

[Alan Edward Klein]

Alan, sensitometry for digital sensors is well standarized, today ISO 12232:2019 is in force. That standarization allows an smooth integration of software vs hardware, as many teams are involved in development of those gadgets.

Yes... we may have some discrepances because sensors have particular spectral responses and some colors may influence more than others... but this also happens with classic meters, see the spectral sensitivity of the Pentax V in the manual... it may not match our film's spectral sensitivity and saturated colors in particular may have a missmatch imn the reading vs predicted density.

For BW we many times use filters... a Red Filter will have a different "correction factor" depending on subject's color and on film red sensitivity, depending on how deep is the red of the filter different films behave different: we have Pan and Ortho, but may OrthoPan mid points...

The smartphone always nail the shot bacause the Preview on screen is used in the calculations, and (today) a lot of Artificial Intelligence is dedicated to nail exposure and HDR compressions. Still probably those Apps take the (ISO calibrated) raw image straight from sensor, so a precise evaluation can be delivered.




+1, no doubt spot metering it is the most informative and the most suitable way when scene is challenging and we want a good prediction of the result.


Still IMO incident can be very suitable for studio portraiture, we may be moving a lot key and fill illumination... and the skin tone may influence the spot reading by two stops when considering dark skin, a full stop difference in caucasian skins alone... also glares may jam the spot reading... so in that situation IMO incident may be superior.

Also when we shot slides with our subject evenly illuminated , and having the possibility to meter the light reaching the subject. In that situation incident may provide a more consistent reading than spot, if we don't find gray references in the subject. Still we always can use a grey card to have the incident metering with the spot meter...

Perhaps we may agree that the spot meter assited (when required) with a grey card is to provide total information. Next step would be adding an Spectromaster or similar gadgets used in Hollywood movies by colorists, but IMO this is another war.

View attachment 263843

Isn't there an issue with incident readings since a cell phone has no "globe"?

 

[Bob S]

Isn't there an issue with incident readings since a cell phone has no "globe"?

Many incident meters used both a flat plate and a dome. The dome for metering 3 dimensionally, such as scenes. Flat plate when taking readings of 2 dimensional objects, like copy work.