Light Metering

[Dinarius]

10 seconds with the camera's spot meter will answer that question easily.

Not true. If you place a sheet of foamcore (or something similarly neutral and uniform in color) where the artwork will be and shoot a "polaroid" shot, if the lighting is even the histogram will exhibit a very narrow spike. If the lighting is uneven or the lens is vignetting, the histogram will display a wider hump rather than a spike.

If you're happy to remove the camera from the tripod having set up the shot and the lighting, then yes. But, I wouldn't be. Secondly, that technique works if the highlights and shadows are easily identifiable  or pointed. Less so, when you are more concerned about "areas" within the image.

Of course. But, if there isn't a narrow spike, the only way of knowing which part of the surface is unevenly lit is to take incident readings over the surface (easy) or use your preferred techinque of spotmeter and grey card (a pain!) And as for bringing a large sheet of foamcore on location and working out how to hang it in front of, or in place of, an equally large painting, when all I need is my trusty Sekonic, don't start me......! ;-)

D. ;-)

[dlashier]

> I am not sure that "expose to the right" is the essence of the Zone System.

Howard, I agree with you. For example with these shots I deliberately let them blow, and often do when conditions dictate. And imo even when conditions do not dictate (eg flat lighting) I still don't usually "shoot to the right" simply because with a compressed shot I'm most likely going to be setting BP anyway reducing reducing the bit depth issue rather than aggravating it.

The zone system is really about producing the optimum tonal gradient for the scene in question within the constraints of the medium, and there's no blanket formula for this outside of the context of the scene itself.

- DL

[howard smith]

Similar to reducing the developing time with b&w film?  Shadows relatively unaffected but the highlights are depressed.  Before I learned to use the Zone System, the vogue was to overexpose (increase shadow detail) and underdevelop (lower highlights).  Flatten the film's curve.

[Ray]

 But I just don't understand why you think a 12 stop dynamic range will solve exposure problems.

Howard,
Well, it depends on what you mean by 12 stops. If you have in mind just one stop more than the 11 stop range described by Ansel Adams in the zone system, then it would be an improvement, but wouldn't solve all exposure problems.  

The zone system describes 6 stops with full texture. 10 stops with full texture would do the trick, at least for 99% of all shots.

Eric,
I'm completely unaware of that 3 emulsion film you referred to but I can understand if it created problems and hassles elsewhere in the processing chain or compromised quality elswhere, it would not be popular.

Before I bought my first DSLR, I discovered Royal Gold 25 which quickly became my favourite film. The colours were lovely, it was fine grained and very sharp and had the high DR that most colour negative films have.

It was discontinued. It was so slow, I guess few people were using it. Great pity, I thought.

[howardbatt]

"I'd like the camera to be an extension of my fingers as a panit brush is to a painter. No fussing with an external meter and the possibility of blown highlights or intractable shadows. No worrying about flashing histograms and decisions to autobracket or not. I just take the shot and the pre-visualisation takes the form of what I can do with it in Photoshop later.  This, I would suggest, frees me up to concentrate on purely artisitc matters whilst on the job.  "

You had me going for a while but now I see you jest.  If the paint brush is simply an extension of the artist's fingers, then we should similarly be asking the paint-by-numbers companies to get digitized so we can specify a "Reubens" or "Rembrandt" or other masterly stroke.  And depth of paint.  That way, we could simply click the button and get a masterpiece.  But whose creation?  Easier is to simply go to the gallery and admire - without the hassle (and expense) of "creating" the work.

I wonder - if you are serious - which "artistic matters" are left after you've had the camera take care of all the other aspects of making the image.  What's left besides deciding in which direction to point the camera?  Composition?  Crop and clone in photoshop.  

I think you overlook an important point - "Hernandez" and all his portraits in and of Yosemite are admired as his artistic creations.  The comment is not "It's fortunate that Ansel Adams was present with a mechanical device to accurately record the lovely scene God presented at 4 in the afternoon of a strange October day in 1941."  I think I read somewhere that he developed the zone system specifically so he could create what was in his mind - that which he wanted to show others.  If he had a yellow filter on his lens, it seems clear he had been doing something with his camera other than "accurately" capturing scenes.  Sort of like using a polarizing filter or a haze filter or photoshop curves.  The result is not what was actually there - it's what the artist sees in his or her mind.

n'est ce pas?

[howardbatt]

I apologize if this has been answered but I looked and couldn't find it.

My camera is an EOS 20D.  It has, of course, all manner of metering options.  Lately I've read about the "advantage" of incident light metering over (in camera) reflective metering and the argument for incident makes a lot of sense.  Especially, I suppose, with flash or other artificial light.  On the other hand, it surely is convenient to meter with the camera.

Is the expensive light meter really that valuable with today's cameras?  The EOS 20D will bracket in less than a second, will spot meter, will average and so forth.  So it really comes down to whether incident metering is "better" than reflective.  Or does it?

Thanks for any help

Howard Batt

[Dinarius]

As has been implied by some respondents, for critical studio work a meter is essential, in my view. It is not enough to simply look at the histogram. That only tells you that nothing is "blown out". It will not tell you the difference between the highlight and shadow in a studio set - merely that they are all within the compass of the camera's range.

I shoot a lot of artwork and taking incident readings is the only way to be sure that the entire picture plane is evenly lit and not simply within the histogram's range. Moving the camera around the picture is not enough...unless you use a grey card, which is a pain. One could think of many similar situations.

In summary, I think that most pros would agree that, for critical work, incident readings are essential.

But, it really depends on what type of photography you do.

D.

[Jonathan Wienke]

What bracketing *won't* do is tell you whether the highlight and the shadow are within, say, five stops of one another, if that's what you desire.

10 seconds with the camera's spot meter will answer that question easily.

Even more importantly, no amount of bracketing will tell you if every part of the picture area is reading the same, such as when you are copying fine art. Only an incident meter will tell you this quickly and effectively.

Not true. If you place a sheet of foamcore (or something similarly neutral and uniform in color) where the artwork will be and shoot a "polaroid" shot, if the lighting is even the histogram will exhibit a very narrow spike. If the lighting is uneven or the lens is vignetting, the histogram will display a wider hump rather than a spike.

[Ray]

Of course this is precisely the reason most of us desire a camera with a wide dynamic range, say a full 12 stops. We wouldn't then have to stuff around trying to get that zone system right at the time of the shot or even bother with incident light meter readings. Just take the shot with an 'evaluative' automatic exposure and decide later what the balance between highlights and shadows will be.

There are many situations where current DSLRs simply don't have sufficient dynamic range for the scene. Don's surfing scenes are a good example. Other examples would be indoor scenes with a view out of a window. Without fill-in flash, or bright studio lights or bracketed exposures, there's no way you could fully capture the entire range with a single exposure.

One should also bear in mind that greyscale blown highlights (or close to greyscale), such as clouds in an overcast sky or surf as in Don's shots lend themselves to quite spectacular highlight recovery in ACR (and C1 of course). Exposing beyond the right with a flashing histogram could be considered normal, recommended practice in such situations.

[howard smith]

DL, I don't know how you did it, but the examples were certainly "exposed right."

Right again.  The Zone System exposes the film to a proper tone and then adjusts the development to get the desired highlight tone and fit the negatives contrast to a particular paper grade, say a #2.

"Expose to the right" or close to it merely places the highlight at the edge of the cameras dynamic range and lets the rest fall where it might.

Does the "Curves" and "Brightness/contrast" in Photoshop adjust the image much the same as variable development time?

[Ray]

Yes. A stop is a stop for selection of shutter speed, aperture and ISO. I'm really referring to Jonathan's idea of devising a photodetector with a variable sensitivity that would kick in when the well is, say, 2/3rds or 3/4trs full. This would have the effect of compressing the brighter values in terms of the range of voltage readings, thus allowing the sensor to capture a higher dynamic range and reducing the advantage or need to expose to the right.

[howard smith]

The full range of b&w film (black to white) is about 10 stops per Ansel Adams.  Some can get a bit more or less.  The dynamic range is about 8 stops, and the textural range about 6 stops.  The dynamic range of printing papers is about 5 or a little more stops - less than film.

The usual way of doing the Zone System is to set a printing paper and vary contrast only if needed.

I would say pure black and white are to hide poor quality detail, but rather add important information by showing there isn't any important detail to see.  Personally, I find a full range print - pure black to pure white - to be more pleasing usually.

The exposure to get Zone 1 to show as Zone 1 is the correct film speed.  Varying development time does little or nothing to change that Zone.  That is why once exposed, there isn't much that can be done, especially for shadow detail.

I would think that for digital, making the dynamic range of the camera greater would not allow more exposure error.  According to Fred Picker, who quotes Todd and Zakia, states there is "virtually no latitude in film exposure for optimum print quality."  Exposure errors can be tolerated only at the expense of print quailty.

[howard smith]

"Often one simply doesn't have the time to get everything right before pressing that shutter. The changing light conditions wait for no-one."

If you don't get it right before pressing the shutter, there usually isn't a lot you can do later.  An overexposed image is hard to fix the highlights.  An underexposed image is hard to coax out shadows.  If you want both shadow and highlight details, the exposure had better be pretty much right on.  You simply cannot add details that aren't there.  Not every scene can be photographed the way you see it in your mind in the time allowed.

[howard smith]

"The camera is a mechanical device and is designed to be as accurate as possible in its role of capturing (or representing) what was actually there."

Probably true.  But in actuality, the modern camera, be it digital or film, fails pretty miserably at capturing what is actually there.  The human eye can see differences in luminocity of perhaps a million to one.  An actual scene can be much greater than that.  Printed photogrpahs about 5 stops.  Even your dream 12 stop dynamic range digital wonder camera fails.

Unless you are color blind, Hernadez was in living color.  Adams photographed it in black, white and shades of gray.  I don't know for sure, but I suppose the sky was some shade of blue, not pure black.  I have no idea what bit depth of color a human can see, but it may be better than most cameras.

My point is Ray, that Hernandez at 4:05pm on October 31, 1941, didn't look like Adams' famous photograph.  And I would also guess that a very faithful to reality, ultra high resolution, 20 stop dynamic range 64 bit color image would look pretty ordinary by comparison.


Ray, you say that it is difficult to add organization to an image, but randomness is easy.  That is pure, hang on, science called thermodynamics.  Seems to contradict evolution theory.

[howard smith]

OK Ray, where I get lost is, regardless of the dynamic range, there will be pure black stops just below the blackest zone with texture.  Underexpose by one stop, and that zone turns to pure black.  Same arguement for the other end and over exposed image.  Or are you saying that if you had a very large dynamic range, you wouldn't care if you lost a zone or three by messing up the exposure?

[Ray]

My question to you Ray is, just what the heck is photography anyway?

(1) Capturing the moment.

(2) Reproducing it on a print.

[Jonathan Wienke]

If you're willing to take a "polaroid" test shot and evaluate the histogram (which takes no longer than taking a reading with a handheld meter) you can do anything you could accomplish wth an external meter, and then some. I really don't see any reason to buy one.

[Dinarius]

I understand the difference between incident and reflected metering but I don't understand why bracketing, for example, (with not just three exposures but five or ten) won't yield the same results.

Bracketing will work just fine and, indeed, it is perfectly suited to digital since you can use the histogram to avoid blowouts and then just bracket in 1/3 stops to your heart's content.

What bracketing *won't* do is tell you whether the highlight and the shadow are within, say, five stops of one another, if that's what you desire.

Even more importantly, no amount of bracketing will tell you if every part of the picture area is reading the same, such as when you are copying fine art. Only an incident meter will tell you this quickly and effectively.

The principle and usefulness of incident readings hasn't gone away just because we're using chips instead of celluloid.

D.

[Dinarius]

So I do it like I used to and then explain that what people are looking at is not bad exposure but, rather, my artistic interpretation of the scene.  Add a little scowl as I say that and I get away with it.

Howard

Great response! ;-)

As long as you enjoy it, that's all that matters.

I totally agree that Adam's is heavy going. He would have argued, correctly, that you could have the shadow detail AND the correctly exposed highlight, in black and white.

Whether or not such latitude is available to those using digital, I don't know. I haven't looked into it.

Does anyone know if Zone Sytem techniques have been adapted to digital photograpy?

D.

[Graham Welland]

I totally agree that Adam's is heavy going. He would have argued, correctly, that you could have the shadow detail AND the correctly exposed highlight, in black and white.

Whether or not such latitude is available to those using digital, I don't know. I haven't looked into it.

Does anyone know if Zone Sytem techniques have been adapted to digital photograpy?

The basic principles of the zone system absolutely apply to digital photography. Selecting exposure to place selected brightness zones into the available dynamic range of a digital camera is just the same as it is with film, arguably more so for highlight protection.

If you are 'exposing to the right' then you're pretty much doing it unconsciously anyway. As regards all the post-exposure zone stuff, that's a different issue and not really applicable to digital post-processing although nothing is stopping you from adopting the goals and applying them as certain photoshop techniques for expanding or compressing dynamic range in the image. You're doing this with levels, curves and much more flexibly using selections and ranges.

As regards Ansel's approach to dynamic range, I'm sure he'd be one of the first to leverage multiple exposures & post-processing with digital to capture the full available brightness range. At the end of the day the camera is just the means to the end so do whayou need  get the pictures you want.

(I'm still a dinosaur with handheld meter too ... )