Home of Professional Cinematography since 1996

style="margin-bottom: 0">Incident / Reflective Readings

Published : 9th January 2004


I know that the incident and reflective readings on a gray card should be the same. But every time I compare the two readings when I am attempting to gray card a particular scene my reflective reading is usually at least a half stop hotter than my incident reading. I try to light the card evenly -- usually a tungsten light full flood with heavy diffusion at a good distance (more than 10 feet from the card. This happens on brand new cards as well -- ones that have not yet been ruined by my oily finger prints. The meter I use is Sekonic L-508. Any suggestions for improving my gray card exposing technique?

Thanks.

Nick Anthony



Nick Anthony wrote :

>-------------But every time I compare the two readings when I am >attempting to gray card a particular scene my reflective reading is >usually at least a half stop hotter than my incident reading. I try to light >the card evenly ----------

IMHO I think that it is most likely the glare, however minute from a mat gray card, that is causing it to be hotter. I just tried your experiment with my 508c and Pentax Digital spotmeter and all both readings to pretty much make sense, with a definitely not new card. Just nudging the angle of the card ever so slightly induces different reflected readings both under and over. Don't worry too much, a film set is not a processing lab.

Daniel Villeneuve, c.s.c.
Directeur-Photo/Director of Photography
Montréal, Canada
demo à / at : http://pages.infinit.net/davil



Nick Anthony wrote :


>I know that the incident and reflective readings on a gray card should be >the same.

It gets very tricky to read a flat surface with a spot meter and compare the reading to a domed incident meter. You may be lighting the card evenly but you may not have the card angled properly. Try taking the card and angling it back and forth as well as up and down to see how much variation in stop you read. Or adjusting your light up and down. Sometimes the card picks up the reflection of the diffusion just enough to alter the true reading. You don't want the reflection of the diffusion on the light onto the grey card.

Another thing to try is a flat disc on the incident meter and see how that translates to the spot reading. I've never found all my meters to be right in line with each other but I would think the Sekonic, being integrated spot and incident meter, wouldn't have such an issue. My meters are usually within 1/3 of a stop and, if not, I'll adjust them to be in unison and check my numbers from dailies. I usually compare with the gaffer's meter to be sure we're on the same page.

Hope this helps.

Jim Sofranko
NY/DP



Metering with either a spot or incident meter is as much art as it is science with just a bit of voodoo thrown in. And everyone has their own method, developed through experience, successes and mistakes along the road that works for them. And as Jim said, reading a flat gray card with a spot meter can be a little tricky. A small angle change toward or away from a light source will change the reading.

I wouldn't worry about the difference in the two readings if I were you. I'd concentrate on a repeatable metering method with one meter or the other that delivers the neg you want and the neg you thought at the time you were getting. Then add the other meter to your exposure evaluation. That way you'll be able to translate what your meter is telling you and make informed adjustments.

I'm not recommending that you lock the "other" meter away and not use it at all, because there are times when an incident meter just can't do what a spot meter can and vice versa. If you can't afford to shoot tests with negative motion film then buy some 35mm slide film and practice with it. Reversal film is generally unforgiving when it comes to exposure error and you'll be able to see what affect a 1/3 or 1/2 stop change will make. Be sure to include someone or something in your tests that you can use as a reference. Then you'll be on your way to developing a consistent method to use either meter where appropriate.

Of course if your meters are broken and/or out of calibration then that's the first place to start, at the meter repair shop.

Best,
Randy Miller, DP in LA



Jim Sofranko writes:

>It gets very tricky to read a flat surface with a spot meter and compare >the reading to a domed incident meter...

Right!

The flat disc is intended mainly for metering flat surfaces (walls, flats, flat art on easels, etc.), and the dome for metering three-dimensional forms (foreground subjects) that are more significantly influenced by sidelight and backlight.

So don't expect the disc and the dome to read identically. They just don't integrate light in the same way. Ideally, the disc reading should come closer to the gray-card reading, assuming they're both held at the same angle in the
same illumination.

To set your f-stop (and even to get your set lighting uniform, if that's what you're doing), incident readings should be made with the meter located where your subject will be, and pointing exactly at the camera. The chief reason for incident readings is to integrate those light sources whose effects are seen from the point of view of the camera. The other main reason is to keep skin tones consistent from scene to scene, regardless of the changing reflectivity of the subjects' surroundings.

In my own experience, incident (or greycard) readings are usually most useful for setting one's basic exposure, and spot readings for managing the extremes of the tonal scale.

Dan Drasin
Producer/DP
Marin County, CA



This whole problem, by the way, points up the difference between measuring light intensity in objective scientific units -- e.g., foot-candles -- and translating those units into a derived number like an f-stop. The accuracy of the derived f-stop is only as good as the accuracy of the translating mechanism. The most accurate meter reading, I think, will be had using an incident meter with a flat-disk diffuser, reading foot-candles. Analog meters, like the classic Sekonic, don't always calculate an f-stop accurately from the initial foot-candle reading. You're much better off, I believe, following an illuminance chart (showing the relations among film speed, aperture, and foot-candles) to set your aperture. I think there's a chart in the ASC manual.

The general rule, derived from sensitometry, is "with EI-100 film, and a key light at 100 foot-candles, a normal exposure will be had at f-2.8." Interior, I like to shoot at a consistent f-stop, so when I shoot a gray-card at the head of a roll, I pre-set the aperture to 2.8, light the gray-card to 100 fc (incident meter with flat disk, pointing at the light source), and shoot. If the gray-card is not angled absolutely perfectly, it doesn't matter: you've still got 100fc on it, regardless. I only use the hemisphere diffuser exterior day (where light is bouncing around everywhere -- off of clothing, the ground, etc.) and I want an average reading. I'll use the hemisphere interior only if the quality of the light is extremely soft and if it's seemingly coming from everywhere:
floor, ceiling, walls, etc.

Peter Corey
Cinematographer, NYC



For accurate measurements, take readings from the camera position(or same direction as the camera). Be sure the card is illuminated by the same light as your subject being photographed. Don't allow shadows or glaring on the card.

Positioning the card in the following manner(in front of and as close to the subject as possible) :

Aim the surface of the card toward a point 1/3 of the compound angle between your camera and the main light.(Artificial, or sun) So if the main light(key) is 30 degrees to the right and 45 degrees above from the camera -to- subject axis, place the card 10 degrees to the side(1/3rd of the angle) and 15 degrees up (1/3rd of the angle) You can also read the card near the camera if you mimic the subject's illumination and orient the card correctly.

Sometimes I fish around for a similar gray in long landscapes or crowds to get a better idea, often mixing incident and refected readings if necessary to come to a decision.

Someone on the list mentioned a little voodoo on the side...lol

John F. Babl
Miami



>I only use the hemisphere diffuser exterior day (where light is bouncing >around everywhere -- off of clothing, the ground, etc.) and I want an >average reading.

I did a test a while back that shook my faith in the incident meter. I held it in a beam of light such that the dome was half lit. On the other side was nothing but blackness. The meter told me to open up exactly one stop to compensate. (The average of 100 fc and 0 fc is 50 fc... or one stop open.)

From then on it became very apparent that the incident meter would be useful for reading the cumulative effects of a number of light sources, or reading a base source with a little bit of edge light tossed in for good measure sneaking around the back of the dome. I used a reflected meter for a couple of years to determine my exposures.

Recently I shot a road trip where I didn't have time to use a spot meter in my usual manner. I decided to drag out the incident meter again. I discovered that, if I held it facing the lens at exactly the right angle, it would tell me the same exposure that I'd determined with my spot meter. I decided to trust it somewhat, and things worked out more or less just fine. I could stick it out the window

Whenever I was shooting in half light, however, I just aimed it into the sun and then opened up a stop... because I knew that if I held the dome in half light I'd get the same reading. (Average of X footcandles and 0 footcandles is X/2 footcandles. The error that occurs when you don't have 0 footcandles is effectively meaningless if you've got a difference of two stops or more.)

So... I feel like I need to go to remedial metering school. After years of working with a spot meter exclusively I don't feel like I know how to use an incident meter anymore. I used to watch DP's hold them out and average a number of light sources and think to myself, "I couldn't do that. I want to know what everything's doing out there!" Then again... maybe that's not
necessary.

Art Adams, DP
Mountain View, California - "Silicon Valley"
http://www.artadams.net/



Art Adams writes :

>So... I feel like I need to go to remedial metering school. After years of >working with a spot meter exclusively I don't feel like I know how to use >an incident meter anymore.

I was taught to aim an incident meter, from the subjects position, pointing half way between the source and the subject.

It has seemed to work OK for the last 30+ years.

Cheers

Geoff Boyle FBKS
Director of Photography
EU Based
www.cinematography.net



It's a good test. More than showing the difference between incident and reflected methods of metering, it shows me the difference between the hemisphere/f-stop approach (where I would stand by the subject and point the hemisphere at the lens) and the flat-disk/foot-candle approach (where I would stand by the subject and point the flat disk at the light source).

Both of these are incident readings, but the second one avoids the averaging problem your test clearly demonstrates. If I'm shooting 7218 at EI500, I know that I need 20fc to shoot at 2.8. If I'm working with a half-key, I simply adjust the intensity until I've got 20fc on the subject (pointing the flat disk at the light source from subject position).

For me, foot-candles make it a lot easier to figure ratios, since I'm not limited to multiples of 2 as I would be with f-stops. If I want, for example, a 7:1 ratio, it's no problem: assuming 7218 and f-2.8, I could key at 21fc and fill at 3fc (or 35 key:5 fill, or 70 key:10 fill, etc., depending on how hot I want the key). Incident readings of foot-candles also make it a bit easier to light within the exposure range of the film, since many meters, even pricey digital ones, don't have a low enough f-stop range. If I'm shooting 7248 at EI100, I need 100fc to shoot at 2.8. If I want part of the scene to be 4 stops under key, I know that I need 6fc in that area. Any decent fc-meter (analog or digital) can read 6fc accurately. But many meters won't go below an f-1, and to read 4 stops under key, the scale would have to go down to an f-0.7.

Peter Corey
NYC



>Be sure the card is illuminated by the same light as your subject being >photographed.

That is, assuming that you want your color correction of the gray card to match the lighting you are using for your subject.

But let's say you want your key warmer and it is gelled to 2500K. Light the gray card or scale to 3200K so when the color timing is set the key will read warm. Personally I use a separate, clean, gray scale light measured at 3200K without any diffusion which will add warmth.

But I'm getting off topic as the original question was about matching the incident and reflective readings. Another thing perhaps worth mentioning is to be sure you are reflectively reading the 18% zone on the gray scale. A gray card is 18% but on the gray scale the large area is not 18% but the next zone lighter which is the zone for Caucasian skin tone.

Hope this helps.

Jim Sofranko
NY/DP



I used the flat disk approach for a while years ago and it drove me mad. My Spectra IV-A is quite directional with the flat disk on. I found I had to hold the meter in front of me with the disk aimed straight at the light to get any sort of reliable reading.

I don't work in ratios, either, because it occurred to me that ratios tell you how bright the brightest light is in relation to the dimmest, but since one can expose anywhere in between it didn't work for me as a way of thinking.

I find the "light by eye, then figure it out with a spot meter" approach to be the way I work best.

Geoff, a question on your method:

>I was taught to aim an incident meter, from the subjects position, >pointing half way between the source and the subject.

Do you mean "halfway between the source and the camera"?

Art Adams, DP
Mountain View, California - "Silicon Valley"



Art Adams writes :

> Do you mean "halfway between the source and the camera"?

Yes I do.

Cheers

Geoff Boyle FBKS
Director of Photography
EU Based



>The accuracy of the derived f-stop is only as good as the accuracy of the >translating mechanism.

I have an old Combi II -- reading FC and using a chart or interpolated F-stops are the same thing. It's a mechanical ring.

Are you saying some digital meters don't do the math for the display right ? Which ones ?

I don't currently have a digital incident meter - I do it all with a Sekonic spot, (with above Spectra as backup/spare) but I have no reason to believe there's a discrepancy between its reading candellas/sq. meter and displayed F stops.

Sam Wells



> Are you saying some digital meters don't do the math for the display >right ?

Many years ago I had a "scientific calculator" made by a company called Kingspoint. The manufacturers were kind enough to include a little piece of paper that said "Errata" telling the consumer which functions the sophisticated chip did incorrectly.

Incident meters -- analog or digital -- are constructed to read foot-candles. All other values are calculated. If you trust the chip, that's fine. Personally, I don't (not completely, anyway). I do trust, however, the foot-candle readings, the dLogE curve, and the sensitometry data relating foot-candles to density at any given aperture for any given exposure index. None of these are derived from the others; each is measured directly, so I know it's correct -- there's no element of "trust" involved (or, rather, it would then be a matter of whether or not I trust the lab technicians at Kodak and the DPs who put the chart together at the ASC).

I recently did an experiment with a class I teach. We were shooting 7218, rated at EI500, at f/4 (the purpose of the exercise was to mimic a sunset by using progressively warmer gels over the key; when we got to the purple/blue gels, talent would turn on a practical). The students use a Sekonic digital meter that only displays f/#s. It has a non-removable hemisphere diffuser that retracts, apparently for the purpose of reading a key directly. By retracting the hemisphere into the body of the meter, the key would be hitting it more in the centre, and less on the periphery; the centre of a hemisphere approaches a flat surface, so it more or less resembles a flat-disk diffuser. Anyway, to shoot at f/4, we needed 40fc. So I checked the key light with a Sekonic "Handy-Lumi," an analog meter that only has a floating fc dial, but which is very accurate. We had exactly 40 fc. Then a student checked the key with the Sekonic digital. It was set to EI500, so, in theory, by retracting the dome and pointing it at the key from talent position, the readout should say f/4. Instead, it said f/2.8.5, a half-stop difference!

I believe the difference comes from the fact that, by retracting the dome (which puts a shadow around the base of the dome), the meter must use less surface area of the photocell : the meter thinks there's less light than there really is, so it tells me to open up more than I really need to.

I realize that this is the fault of the geometry of the meter's construction, not the chip; but that's actually what I meant when I wrote about the "translating mechanism," which would include the entire apparatus for converting FCs into some other value, not just the chip alone.

Peter Corey
NYC



When it comes right down to it, all metering drives me mad, and all meters -- especially spot-meters -- are quite directional. For me, that's the sign of an accurate meter. Again, for me, it's no more trouble to hold the flat-disk straight at the light to get a true fc reading than it is to have someone else position a gray-card in just the position, at just the right angle, to get a true spotmeter reading. I agree that lighting by eye is the best method (especially if the emulsion is forgiving enough), using the meter -- any sort of meter with any reliable, repeatable metering technique -- as a check.

As for ratios, I always try to shoot interiors at a consistent aperture.

For various (arbitrary) reasons, I like 2.8 (fairly wide open, but not completely; limited depth of field). I keep my aperture at 2.8 and I don't change it (unless I must, but that's rare). If I have 120fc key and 40fc fill, I have a 3:1 ratio. If, for some reason, I absolutely must change the aperture to f/4, I know that the same 3:1 ratio now requires 240fc key and 80fc fill.

Peter Corey
NYC



>It's a valid, if perhaps conservative approach. I wonder if it doesn't reflect >(no pun) the aesthetics of studio shooting.


I hope so. I'd give my left canine-tooth to shoot like Gregg Toland or Russell Metty.

Peter Corey



>But I think there is a lot of filmmaking now where the matching of shots >is more "conceptual" than "technical", do you know what I mean ?


What do you mean?

Peter Corey



Art Adams wrote :

>I don't work in ratios, either, because it occurred to me that ratios tell you >how bright the brightest light is in relation to the dimmest, but since one >can expose anywhere in between it didn't work for me as a way of >thinking.

Well, really the key to key plus fill. Classically, working in ratios is a continuity thing. i.e. the people / things shot will have given reflectance values. With ratios, you would, in theory recreate your range of values then via known incident illumination.

It's a valid, if perhaps conservative approach. I wonder if it doesn't reflect (no pun) the aesthetics of studio shooting. I also wonder how much "eye knowledge" lies behind it anyway. I mean if you simply paint by numbers, it can look it; but there are plenty of examples where such is not the case.

But I think there is a lot of filmmaking now where the matching of shots is more "conceptual" than "technical", do you know what I mean ?

Sam Wells



>I usually compare with the gaffer's meter to be sure we're on the same >page.

I usually have my meter calibrated before a show and then calibrate the gaffers meter to mine. My meter is the one used for the emulsion tests. So that seems reasonable to be called the correct meter (Minolta Spot - which I hear they are discontinuing - Oh my gosh!!!). I also use The Permanent gray card which I feel gives a reliable reading at most angles. If you get an acute angle to the card it won't be a true reading. But for the most part it's pretty true to what I need to know.

Steven Poster ASC



Voodoo?

Interestingly enough I watched Connie Hall a couple of times wave an old Spectra around, sideways, upside down, here, there, everywhere until he came up with the number he wanted to have in the first place.

It's more intuition than voodoo. Trust your eyes.

Steven Poster ASC



> What do you mean?

I think films go deeper into shadows, go under, go over, highlights can be hotter, faces can play in more contrast --- not all of this is new, exactly.... sorry I brought this up, now I'll have to write a book

It's not like everyone has thrown the idea of matching densities out the window, but there is not so much a house style in evidence..

I hope I'm being clear, probably not, I decided to have a hangover today and avoid the rush Thursday...

-Sam Wells



>Classically, working in ratios is a continuity thing. i.e. the people / things >shot will have given reflectance values.

To my way of thinking, a 4:1 ratio doesn't tell me much. If the key is T4, and the fill is T2, that's a 4:1 ratio... but where's the exposure? If my shooting stop is T4 then all is clear, but if it's T2.8 or T3.5 does the 4:1 ratio lose all meaning?

Also, unless an entire scene is lit flatly, the ratio concept won't apply across the whole scene. A face may be technically 4:1, but is the background also 4:1? Or is it some other ratio, or some other completely different exposure?

When I was an assistant I worked with an episodic DP who apparently lit entirely by the numbers: 50fc key, 25 fc fill, and I can't remember what the back light was. It was fast and consistent but it wasn't a style I would have chosen.

Art Adams, DP
Mountain View, California - "Silicon Valley"



>I agree that lighting by eye is the best method (especially if the >emulsion is forgiving enough), using the meter -- any sort of meter with >any reliable, repeatable metering technique -- as a check.


I use a variant of a method Roy Wagner mentioned in an AC article a long time ago. I usually spot meter faces (key side and fill side) and then read a bright highlight and a dark shadow to make sure they are within range.

And that's about it.

I can very quickly drive myself crazy with a spot meter otherwise. And I hate doing footcandle math with a flat diffuser incident meter. I'm really focusing on that whole "intuitive" thing right now.

Art Adams, DP
Mountain View, California - "Silicon Valley"



Peter Corey writes :

>for me, it's no more trouble to hold the flat-disk straight at the light to get >a true fc reading

That will give you a true fc reading, but eventually your readings have to get translated into exposure settings on a camera that has a *particular point of view* relative to a scene illuminated by a number of light sources having different intensities, specularities, angles of incidence and so forth.

Pointing the incident meter at the camera from the subject's position won't give you accurate fc readings of particular sources, but (using a dome) it should integrate all the sources that matter, and should yield a useful *starting point*.

From there, one is probably best off tweaking things by eye, experience and, when possible, tests. "Correct exposure" means little if the entire chain of camera stock, filtration, processing, intermediates and print stocks aren't taken into account.. Not to mention the mood you wish to convey overall and in particular scenes.

Arrrghhh...

Dan "prefers a good video monitor" Drasin
Producer/DP
Marin County, CA



>I agree that lighting by eye is the best method (especially if the >emulsion is forgiving enough

The emulsions today are forgiving enough, thus so many workers have so many methods of achieving good exposures, as illustrated in this thread.

If you had learned your craft exposing color reversal Kodachrome many of these techniques would fall by the wayside!

First, to address the original question, very probably the reason for the difference in the incident reading and reflected reading from a gray card is due to the fact that the gray card theoretically should be a 12 per cent, not 18 per cent gray. Years ago, Kodak explained that the reason for choosing 18 per cent for the card was their determination that, while exterior scenes averaged 12 per cent reflectance, interior scenes averaged closer to 18 per cent or even higher, and that was what the card was designed to be used for. There is also the apparent fact that Ansel Adams prevailed upon them to make it 18, for reasons known only to him.

Anyway, that would make the reflected reading about one half stop higher than the incident reading, if everything is calibrated correctly. That's what I've noticed using a Spectra Combi II, although I haven't tried the same test with the IVa.

As far as incident meters and metering is concerned, a domed receptor isn't a true incident meter, it's a compensated incident meter. A true incident meter has a flat receptor, so that it reads the illumination incident to a flat surface. The purpose of the dome is to integrate sources from various directions all falling upon a three dimensional subject, and is really only intended for use in uncontrolled lighting. This is where you have a source that is at an angle to the subject and no ability to add fill, such as outdoors with the sun at a 45 degree angle or more. The first incident meters (flat disc) didn't handle this very well. In the temperate zones, the lighting ratio on a sunny day is around 7 to 1, which is pretty strong on reversal film. If the incident meter was pointed at the sun the shadows were almost 3 stops under. If pointed at the camera, the highlights blew out. If the angle of the sun wasn't too great, you could make a reading in both directions, then average the two for a compensated exposure that worked better. If the angle was too great, this was too much compensation.

So Don Norwood came up with the dome. It integrated the light from all frontal directions, providing some compensation for the shadows, when the dome was aimed directly at the camera from subject position, without overdoing it.

But for controlled lighting the dome can be a disadvantage. If you are interested in achieving a particular lighting ratio you don't want any compensation. That simply compromises what you've done and the effect is inconsistent as you meter around the set from different angles relative to the key angle. Thus, we shade the dome to read the lights independently.

At this point I want to take issue with Peter about ratios. First, I appreciate his determining that the recessed dome idea that has recently appeared on Sekonics is a faulty approach. I've never tried it, but it didn't make sense to me. It supposedly mimics the flat disc, but how can it? The flat disc will show the loss of reflectivity due to the cosine effect on flat surfaces, but no kind of dome could do that, it seems to me.

Anyway, if you read 120 fc from the key and 40 fc from the fill you don't have a 3:1 ratio, you have a 4:1 ratio, on any generally front lighted subject. The keylit side of the subject is receiving 120 from the key PLUS 40 from the fill, which makes it 160 fc to 40 fc. The true ratio is key plus fill to fill only. This assumes that you are reading the key and fill separately.

Of course, what matters is that you get what you're after, no one cares about the numbers. But when lighting for contrasty reversal it did matter. You lighted 3:1 for normal daytime situations and 4:1 for night scenes. It made a very obvious difference. If you wanted to be really dramatic on your night stuff, you stretched it all the way to 6:1! But in any case, to safeguard your highlights, you then used the dome pointed mostly toward the key light at such an angle as to achieve the highest reading and used that to determine exposure.

Nowadays, with negative, you have gobs of fudging room and there isn't much need to be so technically precise unless you are really pushing the range of the film, in which case you need a spot meter.

The whole idea of lighting ratios is really a matter of determining what the shadows are going to look like. Do you want them to be one, two, three stops underexposed or what? So you can simply read the key only for exposure and the fill only to set the shadows at the desired "darkness". If you want to know what ratio that gave you, read key plus fill and compare it to fill alone. The desired "darkness" of the shadows is what you have learned through experience to be the look you like for that subject.


And back before spot meters, you just learned by experience that white objects had to be scrimmed down a stop and dark objects, like mahogany panelling or navy drapes, had to be overlit by one stop.

Those were the days--and thankfully, were is the operative word!

Wade K. Ramsey, DP
Dept. of Cinema & Video Production
Bob Jones University
Greenville, SC 29614


Steven Poster ASC wrote:

>...I also use The Permanent gray card which I feel gives a reliable >reading at most angles....



If your Permanent gray card is the same as The Last Gray Card you used to use, I found it to be closer to 12 percent reflectance and have a very dead surface, which minimizes glare. A good choice.


Wade K. Ramsey, DP
Dept. of Cinema & Video Production
Bob Jones University
Greenville, SC 29614



> The emulsions today are forgiving enough...

Negative emulsions surely are; not necessarily reversal ones. My students regularly shoot Tri-X. In bracketing tests, we have about 2-1/2 stops under and over; that's a 6-stop exposure range. Compare that to 7218: the dLogE specs on the Kodak website show it to have a range of 13 stops.

I always thought 18% reflectance was chosen because it is perceptually (according to a logarithmic scale) halfway between black w/o detail and white w/o detail.

>Thus, we shade the dome to read the lights independently.


That's probably the idea behind the recessed dome on the digital Sekonics : by retracting the dome into the body of the meter, the sides of the meter-body "shade" the periphery of the dome, allowing light to fall more on the centre of the dome. Since there's less curvature of the dome being used, it more-or-less resembles a flat disk (I would say "less" rather than "more").

The problem is, if I shade the sides of the dome, I'm also using a smaller surface area of the photocell behind the dome to take the reading. The upshot has to be a lower f/# reading. By the way, perfectly acceptable digital illuminance meters can be bought for about US$150 from various industrial supply houses.

Most -- but not all -- have flat disk diffusers.

One pricey one used to be made by Minolta, called the T-1 Illuminance Meter. It was a beautiful meter, capable of reading down to 1/1000th of a foot-candle. For some reason, the diffuser was slightly curved. I don't know why.

>Anyway, if you read 120 fc from the key and 40 fc from the fill you don't >have a 3:1 ratio, you have a 4:1 ratio, on any generally front lighted >subject.

This assumes a traditional, classic portrait shot with fill from camera position. I certainly agree that in this situation a ratio has to be the comparison of the [key+fill]:fill. If we fill more from the side than the front, almost no soft light spills over to the key side, and the ratio becomes key:fill. Anyway, I use the term "ratio" to refer generally to a comparison between the brightest part of the composition and the darkest; it needn't be a portrait. If I hard-key talent in the foreground at 120fc and soft-light the background at 40fc, the lighting balance -- which is just another ratio -- is 3:1.

Peter Corey

=============================================

Art Adams wrote:

> I use a variant of a method Roy Wagner mentioned in an AC article a >long time ago.

This is what I do.

I remember the late Paul Beeson BSC said to me once, "People who are a slave to the meter never win Oscars." Interesting thought.

Mark Wiggins
DP/Operator/London
www.productionbase.co.uk/cv/scare



> I remember the late Paul Beeson BSC said to me once, "People who >are a slave to the meter never win Oscars."

That's certainly true. I've heard that John Alcott didn't even use a meter when he shot "Barry Lyndon," relying instead on Polaroid snapshots.

Many classic films had great photography, and they were shot before the invention of the meter: "Sunrise" (DP-?), "Since You Went Away" (DP-Stanley Cortez), "The Killers" (DP-Woody Bredell), etc.

Peter Corey
NYC