william_frucht

ISO isn't an attribute of exposure,

Correct, Andrew, but I never said ISO is an attribute of exposure. I said that correct exposure happens in the camera and that ISO is one of a camera's basic controls. (If you're shooting film, it's an attribute of the film you've chosen.) A beginning photographer needs to understand how aperture, shutter speed, and ISO interact--along with many other things, but those are good ones to start with. What your comparisons have succeeded in demonstrating is the truth of the OP's original complaint: when he tries to brighten his images in post, because they're underexposed, he gets added noise. He wants to know what to do about it. My solution is that he needs to expose his images properly in the first place, because the farther he drifts from correct exposure, the harder it is to get a usable photograph out of what he's shot. Your solution is--well, frankly, I don't know what your solution is. Perhaps it would be helpful if you would stop arguing with everyone else and tell the OP what you think he should do. (The pronoun is another assumption. Apologies, KoolM1, if I've gotten that wrong.)

 

As for your comment that it's overexposing the highlights that causes blown-out highlights, not adjusting your settings so that shadow detail is properly exposed, as far as I can tell this is a distinction without a difference. Most people--certainly most beginning photographers--are not going to spend time worrying about their sensors' response curves while they're out shooting. Maybe, as on my X-Pro 2, they'll darken or lighten the image by a stop or three, but that's the extent of it. Most photographers in the field will stick to the camera's basic controls, as they should, and concentrate on the important things: composition, mood, color, framing and other elements of making a good image.. If faced with a difficult lighting situation, they'll bracket their exposures--another thing KoolM1 should be doing.

 

Anyway, reply however you want. I'm done with you. Which is not to say I won't respond to other posts. Just not to yours.

Ed, I agree, and as I noted, exposing for the shadows with a digital camera can give you blown-out highlights. It's also true that good cameras today, which claim 14 stops of dynamic range, capture a lot of shadow detail that older cameras may not. I still think Adams's advice is worth heeding: get the information onto the sensor--even if, as you point out, we now have to worry more about the other end of the histogram. Point taken, and truth be told, I'm still working mostly with film.

KoolM1, your original question seems to be what to do if your images are underexposed. And I'm guessing from the lack of detail you've provided that you're something of a beginner at this. Let me answer you from the viewpoint of someone who's closer to a beginner level than Andrew Rodney's level.

 

First, correct exposure happens in the camera. You need to know how your camera's basic controls work: aperture, shutter speed, ISO. You may want to take your camera off automatic mode and shoot manually until you have a sense of how these functions interact, and how they relate to movement and depth of field. You may want to invest in a good tripod--make sure it's both light enough to carry around and sturdy enough to support whatever camera you're using. Back when everyone who wanted to learn photography seriously passed through a stage of shooting Kodak Tri-X 400 with a fully manual camera (because that's all there was) and developing it themselves, we were told "expose for the shadows, develop for the highlights." This is still good advice if you're shooting b/w film, but with many digital cameras it can lead to blown-out highlights--less so than it used to, though, as cameras' dynamic range improves. There are ways to bracket exposures and merge them; I won't get into that here. In brief, you can do a lot with a properly exposed image. With a poorly exposed image, you can do much less, and you may never get it to look good.

 

Your post-processing software offers you a lot of different controls, many of which can help if used gently. Raising the "exposure" (whatever that actually does) can help some. Bumping up the brightness of the shadows may help some. There's a noise control function that you'll want to use, because any brightening of an image can increase the noise level. Again, "gently" is key: a little bit of adjustment goes a long way, and unless you know what you're doing, if you find yourself making large adjustments you may be making things worse rather than better. The poster who suggested you use curves was offering good advice: at a certain point, you may find yourself doing almost all your tonal adjustments in curves, because it's the most precise way to control tonality. In Photoshop, which I use along with Camera Raw, you can make all of these adjustments on selected parts of the image, so that different areas have not only different brightness but different contrast levels that you can set just for that part of the image. But note that Ansel Adams's advice from a half-century ago still applies: "Get the information onto the negative." If there's no information in your shadows, no amount of brightening will put it there. All you'll get is more noise. So, to update Adams, get the information onto your sensor. If it's there, you can work with it in any number of ways. If it's not there, you may need to throw away the image and work on a different one.

<p>Jake, I often do the same thing--scan b/w film as if it were color--but I generally don't worry about the color cast that comes out of the scan. Much more important is that the histogram be reasonably balanced. If your histogram is pushed way over to the right, it means you're losing information. Instead of 256 gradations from pure black to pure white, you may have only 150 or fewer gradations that you can actually use. When you correct the exposure in Photoshop (which, by the way, you're probably better off doing in Camera Raw, before you even get to Photoshop), if the histogram looks jagged with lots of gaps, it means you've lost information. The image may look OK, but you may have trouble recovering shadow or highlight detail, and your midtones may not be as smooth as you'd like. (In your excellent flickr images, for instance, I'd be unhappy with the tones in that very pretty girl's blouse.) I have Vuescan but don't use it much--I prefer Silverfast--but both programs give you an option to select your film type and correct the exposure in the scan. The only thing I really care about in a b/w scan is that the tones be a balanced representation of what I was trying to photograph. Ansel Adams's advice--"Get the information onto the negative"--applies here as well: get the information into the scan. The more information you preserve from the negative, the more flexibility you have in post-processing.<br>

As for the color cast in the scan, I pretty much ignore it because there are so many ways to fix it later. If you open the image in Camera Raw before starting to work on it in Photoshop, a bad way to correct the cast is to pull the saturation all the way down or select "convert to grayscale" in the HSL/Grayscale menu. I do something that's probably just as bad: when I first open the image in Photoshop, I select "Image/Mode" and click on "grayscale," let it convert, and then change it back to "RGB color." This gives me an image that is a perfect neutral grey, but with RGB channels that will allow me to open it in Silver Efex Pro. After that, I don't even think about toning the image until I get to the print dialogue--although, of course, there are plenty of ways to tone it before that point.<br>

Scanning color is a whole other ball game: There you really do need to worry about the color you're capturing in the scan. </p>

 

<p>John-Paul, I have the same setup--Opticscan 120 run from Silverfast--and I tend to be very intuitive about the settings. Keep playing with them until the image looks good. Ansel Adams's advice still holds: "get the information onto the negative," or in this case, into the scan. I actually prefer a somewhat overexposed, low-contrast scan because it's easier to darken it later than to brighten it, as brightening can introduce noise in the shadows. I agree with Charles that you should make sure you've selected the correct film type. But if you have, and the scans are still overexposed, you can use the sliders to lower the exposure or adjust the histogram until it looks better. Remember, too, that once you start working with the image, you can always lower the exposure in Camera Raw before you put it into Lightroom or Photoshop--where you can also darken it. <br>

The detail that concerns me more than the brightness is that you have the resolution at just 300 ppi. Unless you have a severe problem with storage space on your hard drive, the resolution should be up around 2650 or even 5,300. (There's also an option for 10,800 but that's only for fanatics.) Otherwise, if you need to crop or make a large print, the image will look soft. </p>

<p>I bought the OpticFilm 120 about a year ago to use with my iMac and always had problems with the Silverfast software. The biggest problem was that often the scanner wouldn't run at all, even though the software was clearly detecting it. After about half an hour of restarting over and over it would finally run. At the suggestion of a Plustek tech support person I detached everything else--my pen tablet, my backup hard drive, my printer--and this got me down to about 10 tries before the thing would fire up. It was like starting an old car on a winter morning. Then software functions gradually began disappearing: the histogram wouldn't show up, the grain reduction wouldn't show up. Sometimes all the functions were greyed out and the scans came out blank. Eventually, after about six months, the software just stopped working. Uninstalling and reinstalling didn't help. So I got Vuescan, which works fine. A little hard to use, but I follow what Ansel Adams said about exposing film: "Just get the information onto the negative." Just get the information into the scan and you can always adjust it into a good print. </p>

<p>Adrian, bending of film in the scanner is much more of a problem with MF than with 35mm, and if some areas of your images are sharp but others are soft, this could be what's happening. Even a barely detectable warping will throw parts of the image out of focus. When I had my late, lamented Nikon 9000, I had to use a glass negative holder. Let me suggest a simple test that will tell you if the problem is in the camera or in the scans. When you magnify your images in Photoshop (or Lightroom, Aperture or whatever), look for the grain structure. In the sharp areas, of course, both the image and the grain will be sharp. In the soft areas, if the grain is still sharp then the problem is in the camera. If the grain is soft (or not visible), the problem is in the scans: the negative is not being held flat enough. I'm currently using a Plustek Opticscan 120, which has other issues, but the negative holder is quite good.</p>

<p>I recently bought a Plustek 120 after my excellent Nikon 9000 died. With the Nikon, scanning 120 b/w film, I always used the glass negative holder as Nikon's regular holder never kept the film flat enough. The Plustek holder is more solid and I have not had any problem: the scans are in perfect focus throughout. Not having to use a glass holder saves me a lot of time in spotting, and the Plustek is also faster than the Nikon. The Silverfast software bundled with the Plustek is a little more complex than I need (I'm basically just looking for a good histogram and do all my fine tuning in Camera Raw and then Photoshop), and it requires some workflow adjustments. But it's certainly useable, and the scans are superb. </p>

<div></div>

<p>I know exactly zero about wedding photography, but looking at these strictly in terms of mood and photographic elements I notice the following:<br>

--These photographers are very, very attentive to exposure, especially skin tones, and they don't mind blown-out highlights. In fact these are almost a signature. On the other hand there is not one image here with lost shadow detail. Everything is very bright, which fits the mood.<br>

--There are no standard poses here, nothing Hallmark-romantic or fuzzy-sentimental. Everything looks spontaneous and a bit quirky--they obviously encourage their subjects to have fun and even to clown a bit. This requires a very relaxed, confident, humorous approach while shooting. They know how to get people to let their guard down. I imagine being photographed by these folks must be a lot of fun.<br>

--There's a good range of distances here. They aren't afraid to get in people's faces to make surprising compositions--and their subjects clearly don't mind. <br>

--As Ellis noted, their lenses don't seem wildly exotic but they are very attentive to focusing. A lot of wide-aperture, short depth of field shots that pick out specific details. They have an eye for odd details, which these shots bring out. They obviously own a macro lens.<br>

--It's hard to tell how much post-processing they do, but it could be quite a bit. For instance, the colors are generally a bit under-saturated--except when they're not. And there are those beautiful skin tones. They are clearly controlling their palette quite deliberately, and they do it subtly and well. If I were doing this I'd find myself making selections, setting the overall tone and contrast of each selection in Curves and then using Hue/Saturation. But they may do it differently.<br>

These folks are really good, and they have a LOT of personality. </p>

You'll find that certain attributes of a Rollei make street photography more difficult than with, say, a Leica: winding the

film takes longer (and two hands); you need to change film every 12 exposures; you don't get the depth of field with an

80 mm Planar or Xenotar that you do with a 35 mm summilux, let alone a 28 or 21. (Don't know where the Rolleiflex

Wide fits in.) In all, you can't be as spontaneous. Which doesn't mean you can't do stunning work--but it's a bit different.

I think you have to be more studied, more composed: less like Robert Frank or Cartier-Bresson and more like Walker

Evans. (Personally I'd give my eyeteeth to be 1% as good as any of them.) Whatever Rollei you buy, though, it would

be well worth your while to get a Maxwell focusing screen put in. You'll need the superior brightness in the changing

conditions you'll encounter on the street. Harry Fleenor at Oceanside Camera (www.rolleirepairs.com) keeps them in

stock; no doubt any number of others do too.

People use all different kinds of monitors and produce wonderful photographs. The

important thing to remember is that what you see on the monitor is not the final product. It's

only an approximation of the final product, which is the print. (This may mark me as old-

fashioned.) You need to have a sense of how what you see on your own monitor will

translate into the print. Obviously the closer the match, the more confidently you can work

(which is why good calibration is essential). But what comes out of your printer will never

exactly match what you see on screen. The real calibration is in your head.

I agree with everything Craig said and would add that unless your printing service is

extremely good, you're likely to find their results less than satisfying. I scan my mf film on

a Nikon 9000 and then process the images in photoshop and am much more pleased with

the results than I ever could be if I let someone else do this.

 

One point about the 9000 that could be important if you're planning to enlarge your

images a lot: it's supposed to scan at 4000 ppi but I'm told this is only true of a standard

35mm frame. For a 6 x 6, which is about three times the area, the resolution is about 1/3

as much, or 1666 ppi. If you set your scanner at 4000 it will scan at 1666 and interpolate

to add pixels. So you might as well set the resolution lower and save your computer's

memory.

 

This is what I'm told, anyway. Personally, I ignore this advice, but then I can only print 13

x 19 paper on my printer, so it's not an issue. But for large prints you might want to

consider having them drum scanned.<div></div>

Taken with a Rolleiflex D, 2.8 Planar on T-Max 100<div></div>

And I didn't mean to be overly critical, Fred--my apologies. Rather than go into my diatribe

about animal cognition, though, let me try to return the conversation (if anybody's still

participating) to photography. There's a strong presumption in photography that what you're

capturing with your camera is "reality." It's an important part of the aesthetics and even, in

some genres, the ethics of the art.

 

Well, I started to ask a question but realized I've gotten onto a very well-trodden path.

I wouldn't be too smug about that swan's aesthetic capabilities. Have a look at a new book

by James and Carol Gould (he's a Princeton ecologist) called Animal Architects, which makes

clear that at least some birds (for instance, bowerbirds) have something like an aesthetic

sense. It requires mind to perceive beauty, but it doesn't necessarily have to be a human

mind.

 

On the other hand, the swan's concept of beauty might be entirely inaccessible to us. As

Wittgenstein said, "If a lion could speak. we would not understand him."

There need not be "unalterable ideals and standards of beauty" for the question to make

sense. On the other hand, it's not a hard question to answer. Without a mind to perceive it,

there is no beauty, just as there is no color without eyes and brains to perceive it. There are

only different wavelengths and intensities of electromagnetic radiation. Color is a mental

phenomenon, as you can easily see when you consider that color perception is not constant

but varies with conditions. (Hence the need for tungsten filters to help us capture the colors

we thought we saw.) Beauty is also a mental phenomenon. Which is precisely why it can

never be subject to unalterable standards.

Spectra on Park Avenue South between 28th and 29th charges $6 a roll for b/w processing.

Turnaround time of about a day. I don't know what a contact sheet costs there, but it's very

easy to make one in a darkroom.

Folks, what Eisner was using was not a photographic scanner but a scanning electron

microscope. This is a sophisticated scientific instrument that does indeed produce very

interesting images, but it takes up an entire room, and sample preparation is difficult. You

can get a used one for about $35,000 at the low end. There are a bunch of sites featuring

this kind of microphotography, which you can find by Googling "scanning electron

microscope."

My new motto is, if less is more, just think how much more more is.

There is room for much thought in photography, but often it happens after the

fact. You cannot do good street photography if you are constantly checking

with an internal censor. You have to shoot first and ask questions later. (I

often shoot with the camera away from my eye specifically to make accidents

happen--unexpected framings and odd compositions I wouldn't consciously

choose through the rangefinder.) But then you look at the negatives and

decide what to print, or print something and decide whether to show it. I once

took a picture of a girl half-asleep on the subway, very close, and it came out

beautifully, in the technical sense. But she looked drugged, stupid, ugly--a

stereotype of the ghetto person wasting her life. It's a cliche that the camera

doesn't lie, but there was nothing true about the picture. I didn't know

anything about this girl. The voyeurism, or exploitation, or injustice--whatever

you want to call it--would have been in showing such an image, not in taking

it. "It is not that which goes into a man that defiles him, but that which comes

out of him."