Hand held metering - spot verses incedent - help

[rosswordhouse.com]

The situation I was in this morning:

10am - overcast sky - clear cut opening in forest shooting fall

colors with small pines and small foliage trees.

 

Tech Info:

Shooting a Fuji G617 pano, 105mm, Velvia 100f, metering with a

Sekonic l-508.

 

Problem:

When I metered using incident at f22 it read 2. Then when I spot

metered the lightest elements in the shot it gave me a reading of

2 seconds. Dark area's spotted at 4-8 seconds. It was very flat

light and all my spot readings fell within a 5 stop range.

 

But, which one do I use? The spot or incident. There's a big

difference between 2 (incident) and 2 seconds (spot).

 

I'll be getting the film back later so soon I'll know, but just thought

someone might have a suggestion to help me out.

 

Maybe I should just bracket the heck out of it. But shooting 617

means I only get 4 images per roll of 120. So I want to be able to

hit it pretty much dead on.

 

Any thoughts would be welcomed

[achilles]

I think your should read Ansel Adams's "The Negative", or any of the Zone System books you can find in your bookstore. But, in the meantime, the incident measure should get you close to the right answer, provided that: 1) you place the meter right in front of the subject, facing the camera and on the lens axis, and 2)you have an even source of light, like the overcast sky you describe.

 

The difference between highlights and shadows that you noted (only 1 or 2 stops) is consistent with the flat light produced by an overcast sky, so you'll probably be fine.

 

Bracketing is not practical with medium and large format. If you get to understand the Zone System, you will not need bracketing and, more importantly, you'll see that it takes away a big percentage of the fun of photography.

[guytal]

The Zone System is a good suggestion, but needs some modification for your purposes. There's a huge difference between B&W film (for which the Zone System was developed) and transparency film.<br>

With B&W film you would normally expose for the shadows, and you will likely have enough lattitude to keep the highlights. This is NOT the case for transparency film in general, and a contrasty film like Velvia in particular. 5 stops is reallys tretching it for transparencies. A low contrast film like Astia might handle it, but with Velvia you'll likely lose the highlights or shadows. In this situation my preference is to keep the highlights and set the exposure 1 stop longer than the reading for the brightest area. I find blocked up shadow are more acceptable than blown highlights, and some judicious dodging may even get you more shadow detail (whereas blown highlights are a total loss).

<br><br>

Guy<br>

<a href="http://scenicwild.com">Scenic Wild</a>

[rosswordhouse.com]

Thanks for your comments!

 

Just got my film back - incident was right on the money. Spot

was over exposed.

 

Lesson learned.

[achilles]

I'm happy to hear the good news about your film. I agree with Ross that, once you understand the Zone System, you need to adjust to the logic of slides, both B&W and color, where the main problem is to preserve detail in the highlighs instead of the shadows, as it is the case with negatives.

[lex_jenkins]

Unless you're willing to devote yourself to a lot of reading, film exposure tests and development tests, all of which will prove that incident metering is perfectly appropriate 90% of the time, just stick with incident metering.

 

I love these discussions. They're invariably circuitous.

[david_littleboy__tokyo__ja]

"They're invariably circuitous."

 

The only inevitiability is that people who don't understand spot metering conclude that spot metering doesn't work...

 

Ross: Spot metering is really simple; a spot reading tells you how to place that part of the image at zone V. You use that reading to compute the exposure that places that part of the image at the zone you want.

 

With slide films, the only zones you have are III to VII; spot meter readings allow you to find an exposure that places the important parts of the scene in those zones. Spot readings also tell you what parts of the scene are going to be lost. Those are things that incident readings can't do.

[david_henderson]

"Lesson learned."

 

But what lesson? If it's that incident metering is a more accurate

way of determining exposure than spot-metering then I'd

suggest that you are likely to be disappointed when faced with

more demanding circumstances in future, especially if using

contrasty slide film like Velvia. Basically if your scene contrast

is way below what even Velvia can handle, then pretty much any

metering technique can give you the right answer. The fact that

your spotmeter gave different answers may well indicate a meter

problem or a technique issue because the results you obtained

shouldn't happen, and indeed with a Sekonic 508 ( which I've

been using for 5 years myself) there's usually more of a danger

that the incident mode will underestimate the light necessary a

little because of stray light entering via the spotmeter

receptor/eyepiece during incident measurement.

 

It's unlikely that any metering technique that only involves one

reading will be really helpful when scene contrast approaches or

exceeds the capabilities of your film. To decide to recompose;

use a grad or a polariser, decide that for you the shot won't work,

requires you to understand the extent of brightness variation in

your scene. Incident doesn't do that. Further as others here

have said, incident works only when your subject and the meter

are in the same light. There are lots of situations where this

simply isn't the case-eg you're in full sun but your subject has

thin cloud reducing the light reaching it. You're in shade

photographing a subject in direct sunlight , and so on. Incident

metering is best used when you can meter from within or right

next to your subject. Added to all this there are technique

complexities with incident metering in certain light conditions -

for example in backlighting. Finally the use of Incident metering

tends to produce a result whereby what you produce is a scene

that looks as it did to the eye. Sometimes you'll actively want it to

be different - and indeed all those zone system masters followed

that route in an attempt to control what their images looked like

rather than simply take what's there.

 

What I'm saying is that incident metering can be very accurate

sometimes, but that it isn't tremendously versatile. There are

many circumstances in which it won't work as well as it did on

your shot, and for those you'll need to have mastered a form of

multiple point metering. On the other hand multiple spotmetering

is a versatile technique and can be used in close to all

landscape/ urban landscapes and scenics. In short if you use a

508 and major on incident metering you'll need to master

multiple spotmetering as well. If you concentrate on building

your spotmetering technique it's likely to be the only one you

need.

[bob_finley]

David wrote - "the incident mode will underestimate the light necessary a little because of stray light entering via the spotmeter receptor/eyepiece during incident measurement."

 

Are you really having this problem with your 508? Unless something is wrong with your unit I don't understand what you are saying as I can't see how this can happen.

[photo5]

I used a spot meter with color film though later I sold it and use my incident meter and my

Nikon F100 if I want to double check. The spot is good if you want to place a zone, or

check the range of a scene. If you want the shadow behind the horse on the barn in a

zone 3, usually defined as the minimum shadow density, you may find the highlight on the

horse ends up 5 stops above that reading, leaving it in zone 8 which is pretty hot and will

look too bright. This is when you "expose for the shadows and develop for the highlights"

and tell your lab this roll is N-1 development. And you will compress the scene down a

stop. You can do the reverse if the light is flat.

 

Or you can "expose for the secrets, develop for the surprises!"

[david_henderson]

Bob. On photo.net you'll find a number of threads regarding the

tendency to under-expose whilst using Sekonic meters. This

appears to refer pretty specifically to incident mode and the

tendency for stray light to get in . Depending on conditions

(amount and direction of light) this can alter the value of a

reading on my meter by anything from zero to about 0.3/0.4

stop. You can avoid the problem by covering the spot eyepiece

or ameliorate it by swivelling the dome so strong light doesn't hit

the eyepiece. You'll be aware that much the same can happen

with cameras, therefore viewfinder blinds.

 

I probably made a mistake mentioning this since it's only a small

point and couldn't account for the big difference in spot/incident

values this poster has experienced. I'm certainly not a member

of the "Sekonics are inaccurate" crew. I do think that the

difference in readings Ross has experienced points to a

problem with *his* meter or a technique issue.

[lex_jenkins]

David, you're jumping to conclusions. I've been a photographer for 30+ years, including a stint as a professional photojournalist. I was taught the Zone System by age 13 (I was very lucky to have attended an advanced arts magnet program). I thoroughly understand metering.

 

And spotmetering is overrated for most 35mm and rollfilm use. You want to enjoy the real benefits of total exposure control, including spotmetering and N+/- development, with all that entails? Shoot LF and sheet film.

 

Rollfilm and 35mm invariably necessitates compromises to get an exposure that's close enough for a skillful printer to do the rest.

 

Perhaps some of you weren't listening when Ross responded that his incident metered exposures were fine. Maybe you'd rather hear yourselves pontificate than help someone with the kind of help he/she wants and at the level of his/her ability and interest.

 

Yeh, there are exceptions. Yeh, spotmetering can help nail exposures more precisely in, say, 10% of all circumstances. Ross and others in his position will know when they're ready. Since his rollfilm use very closely approximates the conditions of sheet film use, with only 4 exposures per roll, I suspect he'll be ready soon.

 

As I said before, these "discussions" tend to become rather circuitous.

[lex_jenkins]

BTW, I was addressing David Littleboy, not David Henderson.

[david_littleboy__tokyo__ja]

Lex, you're the one jumping to conclusions: I was referring the original poster when I said "people who don't understand spot metering conclude that spot metering doesn't work". The original poster was clearly having troubles with his spot meter readings.

 

FWIW, I find incident metering singularly useless for landscape photography: one can rarely get to one's subject to incident meter, and even if you could, there are usually enough different types of lighting (full sun, open shade, deep shade) in any one scene that incident metering is hopeless.

 

And the question was about slide film, where there's no need for additional film exposure tests and developments tests beyond what one would do for incident metering to use a spotmeter.

 

"Maybe you'd rather hear yourselves pontificate than help someone ..."

 

That's why I described the basics of spotmetering. From the original post, it's clear that something is seriously wrong. If the scene covered 5 stops, and incident metering read 1/2 second at f/22 (which we now know to be a correct reading), then a spot reading of a mid-gray tone should read 1/2 second at f/22 as well. A zone VII tone should read 1/8 second at f/22, and a zone III tone should read 2 seconds at f/22. If he's getting anything different, then something's wrong.

 

For example, at ISO 100, my Sekonic L-208 reads EV 7 in incident mode pointed at my desk lamp from my document stand, and my Pentax spotmeter reads EV 5 to 9 for a Kodak grayscale on said document stand and EV 7 for a gray card. Ross should verify that he gets similar results with his meters.

 

Also note that the original note is inconsistent: he claims he saw a 5-stop range with the spotmeter but got 2 seconds for the light areas and 4-8 seconds for the dark area: a 2 to 3 stop range. There's clearly some confusion.

 

Ross: appologies for accusing you of being confused, but I do suspect that that's what happened. Before you go out with the camera, you should verify that you can get reliable readings in both spot and incident modes.

[lex_jenkins]

Then I misunderstood you, David - please accept my apologies for snappishness. (I'll blame insomnia and plead diminished caffeine level.)

 

I think the main difficult many folks have with spotmetering is seeing how to place a tone in difficult lighting. In bright enough light an 18% gray card or tone can look quite white in contrast against a darker area. Makes it difficult to know how to place one's zones if we don't tote an 18% gray card everywhere.

 

I tend now to use a spotmeter to evaluate an entire scene to determine what, if anything, I'll have to compromise in recording a scene, and to note in my notebook the EV range. Usually if I tote my spotmeter I'll also have the incident meter so I'll compare that reading against the range of EVs from the spotmeter.

 

Later, I'll decide how to proceed with development based on the most important exposures on the roll. This is typically easier with rollfilm - fewer decisions to make with a 12-exposure 120 roll - and more difficult with a 36-exposure 35mm roll.

 

Since Ross is dealing with a panoramic camera and, presumably, has some opportunities to expose all four frames under similar, if not identical, lighting conditions, he might often be able to apply the Zone System verbatim if he chooses.

 

OTOH, if his exposures on a single roll are separated by many hours or even days under all kinds of lighting conditions he's typically going to get better, if compromised, results using incident metering.

 

And now I'm the one pontificating. So I'll shaddap.

[rosswordhouse.com]

Thanks all for your posts - I had fun reading all of your advice and

banter.

 

I guess the real problem is not with my meter, it's with my lack of

experience using it. Looking back at my slide I see that the

yellow fern that i spot metered is in a hue of 18% grey. So infact

the spot meter was dead on, I just didn't move the fern into it's

correct zone. Shooting the fern and the rest of the scene at the

ferns reading made it all over exposed. I understand this now,

thanks!

 

The lesson I learned was if I'm shooting subject mater in that I

can walk up to then use the incident reading and trust it. If it's a

landscape distance shot then find a 18% grey hue in the sky and

spot meter that.

 

I guess it's not a "verses" thing but an application thing. Use the

right metering for the right situation.

 

I totally agree about metering for the highlights and bracketing for

shadows.

 

I'll get the zone system book and see what I can learn. heck, I've

only been doing this for 2 years, it's a hoot and a kick in the pants

all at the same time.

 

Thanks again,

[david_littleboy__tokyo__ja]

My turn to pontificate.

 

"I think the main difficult many folks have with spotmetering is seeing how to place a tone in difficult lighting. In bright enough light an 18% gray card or tone can look quite white in contrast against a darker area. Makes it difficult to know how to place one's zones if we don't tote an 18% gray card everywhere."

 

Hmm, I disagree. The object of spot metering is not to replicate what an incident meter would have done, but to place the tones where you want them. Consider a cityscape lit by the setting sun. You want the deep red glow on the buldings at zone IV or so. An incident meter would put the building at whatever tone it would have been three hours earlier. Not the right thing at all. Similarly, if that 18% gray tone concrete (this is Tokyo, sigh) looks bright, then it should be bright, i.e. exposed at a zone on the film that matches your vision of the scene. (Not that you can get away with a lot of this sort of thing with Velvia 100F...)

 

(I lied, though. To play this spotmeter game, you really want to take lots of bracketed test shots to see how typical things look in different zones on the film you are using, so that when you see something in a scene you will know what it would look like placed at different zones.)

[blakley]

David L. is right on in the last post. Incident metering, done right, is accurate when you want an 18% reflective subject in the real world to look like an 18% reflective subject on your negative or transparency.

 

Spot metering is in a certain sense perfectly accurate by definition (barring a defective meter....); whatever you spot meter will turn out to be an 18% reflective subject on your negative or transparency. The hard part here is figuring out what to point the spotmeter at....

[q.g._de_bakker]

Bob,<br><br><i>"Incident metering, done right, is accurate when you want an 18% reflective subject in the real world to look like an 18% reflective subject on your negative or transparency.

Spot metering is in a certain sense perfectly accurate by definition (barring a defective meter....); whatever you spot meter will turn out to be an 18% reflective subject on your negative or transparency. The hard part here is figuring out what to point the spotmeter at...."</i><br><br>The fun and confusing bit is that when you have figured that out, you can apply the same insight to the incident light reading, 'bias' that, and get the same result. ;-)

[david_littleboy__tokyo__ja]

"whatever you spot meter will turn out to be an 18% reflective subject on your negative or transparency."

 

(Am I overreacting here? Were you just joking?)

 

No, the metered spot turns out whatever density you want it to turn out.

 

The spot meter measures the brightness of the spot; you pick the zone (density) you want and chose the exposure that achieves that density.

 

There's nothing sacred about zone V, other than that most spot meters do that calculation automatically.

 

"The hard part here is figuring out what to point the spotmeter at..."

 

You can point the spotmeter at anything, as long as you know (a) how you want that thing rendered (what zone you want it placed at), and (b) how to add and subtract f stops.

 

For slide film, it's usually easier to point it at something zone IV or zone VI, since those are more easily characterized than zone V. (Zone IV is things darker than mid-gray but still rich in detail, and Zone VI is things brighter than mid-gray but still rich in detail.) B&W is harder, since there are so many more zones.

[david_littleboy__tokyo__ja]

"you can apply the same insight to the incident light reading, 'bias' that, and get the same result."

 

No, you can't. To do that, you'd have to know the reflectivities of the significant parts of the subject to know how much to bias, but you don't know those. (Remember Lex's point: you can't recognize an 18% object as an 18% object in harsh lighting.)

 

Spot metering allows you to accurately, and _independently the properties of the subject and light_, go from a previsualization of the image to an exposure that achieves that previsualization.

 

Spot metering won't work (and incident metering will) for robots taking photographs to be read by Martians, but for people taking photographs to be seen by people, it's the right thing.

[lex_jenkins]

Ditto David L.'s explication of how to use a spotmeter to, essential, pre-determine how a scene will be recorded. I often do this, especially with city scenes.<p>

 

Early in the summer I captured a shot of an oft-disregarded view of an otherwise spectacular building. The white limestone was half in brilliant sun and half in shade. The challenge was to record the scene so that both halves would "read" on film without engaging in darkroom gymnastics during printing. Here is that photo:<p>

 

<a href=http://www.photo.net/photo/1605880&size=lg>"Bass Hall, in light and shadow"</a><p>

 

By incident and spotmetering the scene (through the orange filter) I could plan how to develop the entire roll for that one specific image.<p>

 

The downside, of course, is that later shots on the same roll were compromised because they were exposed under entirely different lighting. It meant the rest of the frames were pretty difficult to print.<p>

 

But film is cheap. Most of the time I can return to a site for another shot. And if I was really concerned enough I'd either start toting a MF camera with interchangeable backs or tote more than one camera.