Home > Ted’s Commentaries > About Photography

About Photography

About Photography

What’s the key to taking great pictures?

Heck if I know! I make no claims to greatness, but you can look at the pictures on this Web site and form your own opinion about whether they are any good (and I certainly welcome your comments or opinions).

But... I can offer some practical advice that might make your pictures better. To start with, a simple principle: Open your eyes, look, and think. Your eyes and your mind are really the most important tools you have, far more important in fact than what camera or film you use. Come to think of it, they’re useful for many things besides photography.

1. Look and Think

Photography, like any other artistic or creative endeavor, requires some effort to make it work well. A few seconds of looking and thinking may make all the difference between a snapshot that’s meaningless or incomprehensible without the memory you attach to the picture, and an image that others find as exciting as you do.

So when you’re about to take a picture, stop and really look through the viewfinder or LCD. Look at the whole scene, including the edges of the picture. What exactly is interesting about the scene that makes you want to photograph it? What are you trying to say?

After you’ve looked, then it’s time to think. Ask yourself some questions like these:

Those are just some possible questions to focus your thinking (as it were).

Only after you’ve looked and thought is it time to press the shutter release. Then think about alternative compositions: Perhaps horizontal instead of a vertical. Or move around to get a different angle, which may offer a whole cascade of new ideas.

With each new possibility, look and think before shooting. But also keep in mind that memory cards are cheap. It’s better to try many different alternatives than to just make one shot and pack up. Don’t be afraid to spend as much time as it takes to do all the looking and thinking.

Still, it ultimately all comes down to your instinct and inherent aesthetic sensibility: You’ll know whether what you see in your viewfinder “grabs” you. Looking and thinking are tools that help you make that happen. My illustrated article on Transcending the Familiar describes an exercise and some ideas that can help you look and think.

(Also have a look at an article by writer/photographer Ken Rockwell that cleverly summarizes “look and think” with an audacious mnemonic you won’t forget. I disagree with a number of things Mr. Rockwell says elsewhere, but this time he’s right on the money.)

2. Learn

Taking classes can help a lot, particularly if you have a good teacher. You’ll learn the basics of exposure and lens choices, as well as lighting and filters. You’ll also learn the classic principles of composition, including the infamous “Rule of Thirds.” (In a nutshell: Imagine a tic-tac-toe grid that divides the viewfinder or LCD into three equal sections, horizontally and vertically. Place the main subject in one of the four “crash points” where the horizontal and vertical lines intersect, never in the center. Align the horizon with the upper or lower horizontal line, again never in the center.)

You can get valuable basic tools and techniques from classes. But they should supplement and support, rather than supplant, instinct and aesthetic sense. They are good, solid starting points for your own adventures in looking and thinking. Also, even the best teachers can sometimes do unintended harm. I had one otherwise excellent instructor who mercilessly (but quite validly) criticized another student’s “glamour” shots of women. His demonstration of all the ways an errant shadow or ungraceful arm position can devastate the female form has left me permanently hesitant to photograph women.

Contrary to what elderly camera club competition judges might tell you, the rule that a picture must have a single, strong center of interest located at or near a “crash point” is not cast in concrete. Moses did not bring it down from Mount Sinai along with the Tablets of the Law. Jesus did not preach it from the Mount. Nor did the Archangel Gabriel reveal it to Mohammad along with the Noble Qur’an. It’s just a time-tested guideline that has produced a great many aesthetically-pleasing pictures, and will surely produce a great many more.

A slavish adherence to this rule may well produce a composition that pleases competition judges— maybe because they can “read” it quickly through their cataracts?— but is banal and lacking in individual vision. Still, the “Rule of Thirds” is an extremely valuable reference and starting point for thinking about a scenic composition. There’s a good reason why many compact cameras provide an option to overlay their LCDs with a “Rule of Thirds” grid. (I have that option permanently set on my Canon S100.) Photoshop provides an optional grid for its cropping tool.

The rule that the “golden hours” just after sunrise and just before sunset are the only time for photography is also not carved in stone. Yes, that’s usually the best and most interesting light. Indeed, shooting early in the morning and late in the afternoon would probably qualify as “the key to taking great pictures,” if such a “key” actually existed. But there are exceptions! A subject’s location may place it in a shadow at those times. And a canyon floor, or a cityscape with skyscrapers, may photograph best at mid-day, normally the “forbidden time.” It’s possible to make interesting images in any light, as long as you look and think about how best to use it.

It is important to know the rules, so you can decide whether to follow them or break them deliberately to produce the effect you’re looking for. For example, I tend to like compositions that competition judges regard as excessively “busy,” without a single obvious center of interest. I find that when this works right, I get an interesting and thought-provoking picture that invites the viewer to spend time looking at it— exactly because it isn’t immediately “readable” or “restful to the eye.”

I very often discard pictures that don’t work. I’m not ashamed to admit it, and you shouldn’t be either! It’s important to experiment, and to avoid letting the the solemn pronouncements of “experts” spook or stifle you. If something you try doesn’t work, nobody has to know about it.

3. Experiment Fearlessly. Winnow Aggressively.

A trashcan (real or virtual) is perhaps the most valuable tool for any creative activity!

If you you’ve taken a set of pictures, and you’ve applied all the techniques of looking, thinking, and experimenting with alternatives, you’ll probably have a lot of camera files, or possibly slides or prints. Now comes the other half of the artistic process: Choosing the images that are worth keeping.

Scrutinize each picture carefully. Look and think yet again: Look closely and think about what works and what doesn’t, from artistic as well as technical standpoints. If something doesn’t work, don’t be afraid to discard the picture.

This is how I winnow my pictures. There are numerous other ways to do it, but this is what works for me.

Since I’m an American, and I live in a society that worships quantitative measures and statistics, let’s look at what this process means in terms of numbers. If, hypothetically, I take 100 pictures, I might initially flag 25 to 30 of them for the Second Look. After the Second Look, I’d have maybe 15 to 20, of which I’d then select 12 to 15 for a full Photoshop workup. Of those, you might see 10 on this Web site. That’s a 10% yield, a figure that has been fairly typical for me, for as long as I’ve been preparing pictures for this Web site.

I’m not embarrassed about that “low yield,” or even afraid to admit it. We’re not talking about a manufacturing process that seeks to maximize efficiency and quarterly profit. Quality and distinctiveness, not quantity or yield, is what matters. After all, nobody needs to know anything about all those discarded pictures.

Being very critical and selective about the pictures you let other people see may not be the “key to taking great pictures.” But it definitely helps if you show only the pictures you consider your best! (And by the way, do keep all your camera files, slides, or negatives. Years later, you may discover something of value in one or more of the pictures that didn’t make the Initial Assessment when you took them.)

Commentary Contents

tedsimages.com Virtual Light Table Home Page

What kind of equipment to use?

Many people worry too much about their camera gear. In reality, how you use a camera makes far more difference to your pictures than the camera itself. As long as your camera has decently sharp optics and a reasonably accurate exposure system, you can take excellent pictures. Many inexpensive 35mm or digital “point-and-shoot” cameras will do just fine. It’s unfortunate that small, convenient auto-focus cameras have the somewhat pejorative name of “point-and-shoot.” Some of them have excellent lenses and provide a lot of features in a small and light package.

Commentary Contents

tedsimages.com Virtual Light Table Home Page

Take Control with an SLR

So what do you get for the money you spend on a fancy Single-Lens Reflex (SLR) “system” (a marketeer’s term for cameras, lenses, and accessories that go together)? In a word, control. The all-automatic point-and-shoot camera gives you very little of it. You can’t, for example, choose a fast shutter speed to stop action, or a large aperture to blur a distracting background. Nor can you choose what parts of your picture will be in focus, except for the spot where the camera auto-focuses. You have a limited selection of lenses (or no selection at all, if it isn’t a zoom). Whether film or digital, the SLR gives you control over all this, and more.

You can take very good pictures without that much control, as long as you use your eyes and brain to find and compose your pictures. Sometimes you can take better pictures if you just let the camera worry about exposure and focus so you can concentrate on composition or following the action. That’s why most fancy SLRs have a “program” or “full-automation” mode that turns them into expensive point-and-shoots. But having the ability to control the variables is what makes photography interesting (and, at times, infuriating).

Commentary Contents

tedsimages.com Virtual Light Table Home Page

Bells, Whistles, and Useful Features

Today’s computerized SLRs are loaded with bells and whistles. They can seem daunting. Camera makers feel compelled to keep adding more features for the same reason that each version of Microsoft Office is more bloated than the previous one. If the marketeers couldn’t tout a host of new (but possibly dubious) features, nobody would feel compelled to upgrade and they’d rapidly go out of business.

Essential Functions

Fortunately, you don’t have to use all the fancy features. You really only need the basic exposure modes:


If you’re accustomed to manual focusing (a few manual-focus film SLRs are still sold, believe it or not), auto-focus may seem like a frill. But it really isn’t. The auto-focus in today’s cameras will usually do a better, faster, and more accurate job than your eye squinting through the viewfinder. But even though newer auto-focus systems work faster under more conditions than ever before, there are still times when auto-focus flounders. That’s when you need to switch to manual.

I like the Canon EOS system because it offers a unique “Depth” (DEP) exposure mode. Auto-focus first on the closest subject and then on the farthest (or infinity). The camera then selects an aperture and distance that will keep the entire range in focus, provided the lighting conditions permit it. DEP provides simple but very powerful control over depth of field (the range of distances within the picture that are in focus). It automates and makes very easy what used to require tables and tedious adjustments.

DEP has an annoying quirk with wide-angle lenses. The depth of field is inherently so great that the camera may choose a large aperture, possibly even wide open. Since a large aperture sacrifices overall sharpness, it’s a good idea to choose a smaller one whenever possible. So after the camera selects the distance, switch it to manual focus and aperture priority (without changing the focus), and stop down to a reasonably small aperture. This approach is useful with lenses of any focal length if the aperture the camera chooses seems unreasonably large. Remember that an SLR is all about control, so common sense must always prevail over automation.

For some unfathomable reason, Canon equips its digital SLRs with a “simplified” version of Depth mode called A-DEP. (Canon’s latest “professional” cameras don’t have DEP at all, supposedly because their firmware is so packed with other features that there isn’t room for it.) It works like DEP, except it automatically chooses the near and far elements from among the camera’s multiple focus points. This seems easier to use (which may be why Canon uses it), but it’s actually far less useful.

A-DEP works well enough if the near and far elements you want are each aligned with one of the camera’s focus points— and when the camera correctly identifies them. Unfortunately, too often that’s not the case. Making it work often requires a lot of “neck action,” rotating the camera to painstakingly align focus points with the near and far subjects until the system correctly identifies them. Then recompose the picture while carefully holding down the shutter release so the camera doesn’t lose the setting. Including an option in the firmware to select DEP or A-DEP should be technically simple. After all, the EOS 650 implemented DEP in 1987, using a feeble primitive microprocessor and the minimum amount of precious ROM. The daunting challenge probably is to convince the marketeers to allow it today.

A Few Words About Lenses

Avoid any zoom lens with a short focal length of 35mm (or its equivalent on digital cameras). Choose one that starts at 28mm, or better yet 24mm. That will allow the true wide-angle effects you can’t get at 35mm. Although most digital SLRs use lenses made for 35mm SLRs, their sensors are smaller than a 35mm frame. The resulting cropping “magnifies” the effective focal length of the lens. So true wide-angle lenses begin at 17mm or 18mm. A number of relatively affordable zoom lenses offer these focal lengths (and shorter). Most of them have a “reduced image circle” that covers a small digital sensor but not a full film frame.

Telephoto zoom lenses seem to acquire Freudian overtones for some people, more for prestige than practicality. Bigger isn’t always better. Don’t get anything longer than 200mm (or 125mm on a digital SLR) unless you also get— and consistently use— a sturdy tripod. Too often you can’t hold a lens that long steady enough for a sharp picture. Some lenses include an “image stabilization” or “vibration reduction” system that compensates for the inherent unsteadiness of hand-holding a camera. It’s not a magical cure, but it can allow sharp pictures at shutter speeds two or three stops slower (or sometimes more) than without it. I can’t prevent a blurred picture of a moving subject, but neither can a tripod.

28-105 (or 28-135) is a good, convenient range in one lens for travel; the latest versions are 24mm at the short end. There are a number of 28-200, and even 28-300 zooms available. They’re very convenient but make inevitable optical compromises in sharpness and distortion, particularly at their longest settings.

Commentary Contents

tedsimages.com Virtual Light Table Home Page

What I use (if that matters)

Film Cameras and Lenses

After all this discussion you probably want to know what I use. My first SLR was an Olympus OM-G, a basic manual-focus model, with a Tokina 28-80 zoom. I got a Canon EOS 650 in 1989. I used Canon’s long-obsolete 28-70 and 70-210 zoom lenses (the latter was a rather heavy model with a constant f/4 maximum aperture). For a time I had a Sigma 24-70 f/3.5-5.6 zoom, which gave better wide-angle coverage. But it was cheaply made, and it developed mechanical problems just after the warranty ran out.

In 1998 my 650 fell victim to the dreaded sticky shutter, a common problem with early EOS models. The little pieces of rubber that stop the shutter deteriorate into a gummy mess. Since fixing it would have been fairly expensive, I decided it was time to trade the 650 (and the 28-70 zoom) for an Elan II (it was called the EOS 50 outside the US). I also got the popular, lightweight, and quite sharp Canon 28-105 zoom as my one-size-fits-all travel lens, along with Canon’s small and light 22-55 super-wide-angle zoom (discontinued).

Canon makes two 28-105 lenses. The good one is the 28-105 f/3.5-4.5 II which costs around $230. It is optically excellent, well-made, and highly regarded by all who use it, including me. The new and not-so-good version is the “28-105mm f/4-5.6 USM,” which costs around $150. That one is slower, cheaply made, and has a plastic mount. It’s reportedly inferior optically, although a test in Popular Photography magazine gave it good marks. As always, it’s best to take reviews in mass-market magazines with at least a tablespoon of salt.

Canon’s discontinued 22-55 zoom was originally intended as a low-cost “kit lens” for their extinct APS SLRs (although its image circle covers a full 35mm frame). With that smaller format it provided the equivalent of a 35-80 “starter” zoom. Small, light, and inexpensive, it was often maligned because it’s so cheaply made. It has a plastic mount and no focusing scale. The front element rotates when focusing, making a polarizer difficult to use. For $150 (in 1998) you obviously can’t expect a professional-grade lens, particularly one this wide. But if you recognize and work within its limitations, this lens provides a usable and uniquely affordable super-wide capability. At its maximum aperture of f/4, the sharpness and contrast are disappointing. It also has the distortion that’s typical of many super-wide lenses— correcting that distortion is what makes the best of these lenses so expensive (and bulky). But if you keep it between f/8 and f/16, and avoid photographing office buildings or graph paper, you can get very pleasing results. I’ve even sold 11x14 prints of pictures I took with it. If you want an affordable super-wide lens for your Canon EOS film SLR, it is certainly worth looking through the remainder bins for one of these photographic cubic zirconias.

Digital Cameras and Lenses

I “converted to the Digital Faith” in April 2005 with the purchase of a Canon Digital Rebel XT (called the 350D in Europe and, so help me, the Kiss Digital n in Japan). The (relatively) low price finally convinced me that it was time to go fully digital after six “hybrid” years of shooting negative film for scanning. I had carefully considered the heftier (and costlier) 20D, but I decided that even irrespective of the lower price, the smaller and lighter Rebel XT would suit me better.

I initially chose the (discontinued) Sigma 18-125 f/3.5-5.6 zoom as my “standard” lens based on its range and a number of good reviews of its performance. That would be supplemented with a Tokina 12-24 f/4 for true wide-angle coverage. B&H and Adorama had the Sigma lens readily available, but the Tokina was scarcer than hens’ teeth. When I called the camera stores in my area, I found that one of them had the Rebel XT and the Tokina in stock— at the same price as B&H and Adorama— but not the Sigma. I’m pretty sure I called every camera store in metropolitan Los Angeles, but not one of them had the Sigma.

So I went to the local store and bought the camera body, the Tokina lens, and the associated paraphernalia of memory cards and filters. I figured I could use my perfectly good 28-105 until I decided what to do about a more convenient option. I have since read numerous complaints about quality control problems with the Sigma lens and the difficulty people have had finding a good copy. This convinced me that buying it mail-order would not be a good idea.

I can’t provide a “scientific” review of the Tokina 12-24 (several publications have already done that), but here are some notes about it. Physically the lens is rather large and bulky, but any good lens this wide inherently needs quite a bit of glass to correct distortion. Advertising copywriters would probably call it “substantial.” It’s nicely finished with focus and zoom controls that operate very smoothly, in contrast to the jerkiness of every Canon “consumer” lens I’ve owned. It has the same 77mm filter diameter as all the other lenses of this type. Despite the bulk, it’s not horribly heavy and it balances well on the Rebel XT. It has a constant f/4 maximum aperture, and internal focusing that greatly simplifies use with a polarizer. There’s no need to get a special “thin” polarizer. Contrary to what some reviewers say, I don’t notice any vignetting or fall-off with a standard-thickness Hoya HMC polarizer, even at 12mm.

Unlike other Canon-mount lenses, you switch this one between auto and manual focusing by moving the entire focus ring instead of the usual small switch on the lens barrel. To set manual focus, you pull the ring toward the camera; to set auto focus you push it toward the front of the lens. I find this a bit awkward, since it takes a good bit of force to move the ring. But I suppose that’s better than a looser clutch that could slip accidentally. The process of pushing in the ring is also likely to displace the focus setting. That can be a problem if you want to stop down in A-DEP mode; you’ll need to make a mental note of the focus setting and restore it after you’ve pushed the ring to manual focus. Another small quibble: The red dot to align with a corresponding red dot on the body when mounting a lens is on the on the back of the metal mount rather than on the barrel where it’s found on most lenses. That makes mounting the lens unnecessarily difficult. Nothing that a little dab of nail polish can’t solve.

Optically, The Tokina is quite sharp from corner to corner, although the corners start to get a bit soft when the lens is wide open at 12mm. At high magnifications small details break down into pixels rather than blurring, suggesting that the lens has more resolution than an 8-megapixel sensor.

The dreaded purple fringing of chromatic aberration common to ultra-wide lenses is very noticeable whenever there’s a high-contrast edge, such as tree branches against a bright sky. Some copies may have more of a problem with this than others. I had to return my first copy because of noticeable front-focusing of distant subjects. The second copy focuses correctly and has noticeably less purple fringing. Photoshop versions since CS5 include a profile for this lens; the “Remove Chromatic Aberration” check-box in Adobe Camera Raw’s lens correction tab disposes of the fringing quite effectively.

My camera seems to have trouble auto-focusing accurately at the widest settings, between 12 and 15mm. I suspect that autofocus systems inherently have problems with very wide lenses; my Elan II has similar difficulty with a 22mm lens. Regardless, it’s easily (if not always conveniently) dealt with by zooming in, focusing, and zooming back out.

I’m generally pleased with the optical performance (and with the impressive sweep of a 12mm lens), although I’m less pleased with the “user interface” of that slip-clutch focusing ring. Since the competition is Canon’s very expensive 10-22, Tamron’s 11-18 (with less range and a smaller maximum aperture), and Sigma’s 10-20, I think the Tokina is a good choice.

After seven months of using the 28-105, I chose as its successor Canon’s venerable 28-135 f/3.5-5.6IS. I put together a spreadsheet with the focal lengths I used for my Sedona and San Juan Capistrano trips in 2005. I found that I used the 90-105mm range more often than I suspected. In addition to being overpriced at the time, Canon’s 17-85 isn’t long enough. The 28-135 seemed the best available choice almost by default, as the combination of it and the Tokina provides the focal lengths I use without the need to carry more than two lenses.

The spreadsheet confirmed my preference for focal lengths in the 18-40mm range, which means that the Canon’s 28mm short end makes me change lenses more frequently than I’d like. I can mitigate the problem of lens changes by first shooting a sequence with the Tokina and then switching to the Canon (or vice versa), but it’s still a hassle I’ll just have to live with. I was originally afraid that all those lens changes would rapidly accumulate dust on the sensor, but I’ve been pleasantly surprised to find that it’s not much of a problem. Some people insist that an electrical charge on the sensor when the camera is turned on makes dust more likely to accumulate when changing lenses. I don’t know whether this is true, but I am careful to turn the camera off before changing lenses.

The 28-135 is at least as sharp as the 28-105. And I know from years of staring close-up at film scans from the latter that it’s quite a fine performer. Canon first introduced the 28-135 in 1998, when it represented a breakthrough of sorts as the first “affordable” general-purpose consumer lens with Image Stabilization. Although that makes it old enough to qualify as a “living fossil” in a market where cameras become obsolete in a year, Canon still sells it; they even offered it as a $200 “kit lens” option for the 40D and 50D. The first-generation Image Stabilization system is less effective and less flexible than the version in current models, but it definitely works as advertised. That means it’s very helpful in many situations, but definitely not a magical replacement for a tripod. At 135mm I can get reliably sharp results at 1/50 instead of 1/250, the lower limit of reliable sharpness without image stabilization. (With an APS-C sensor you have to multiply the “one over the focal length” hand-holding rule by the crop factor of 1.6.) The effectiveness of the stabilization seems to decrease with the focal length setting, but it’s still helpful at 28mm.

The main disadvantage of this lens is that it’s much larger and heavier than the 28-105 (it uses 72mm filters instead of 58mm); it’s nearly the same size and weight as the Tokina. Also, it has noticeably more distortion at 28mm, but Adobe Camera Raw 6 (for Photoshop CS5) includes a lens profile for it that can automatically correct the distortion. I really would have preferred an 18-135IS lens, but that didn’t yet exist (Canon finally released this in 2009). As a full-frame lens, the 28-135 should be ready to serve as my standard lens once the price of full-frame DSLRs reaches reasonable levels.


For image file storage while traveling, I bought a SmartDisk FlashTrax portable hard drive (now discontinued) soon after I got my camera. It was an adequate but not truly satisfactory solution. Memory cards sufficient to accommodate raw files from a week of travel would have cost $900 in 2005. By 2008, the price of memory cards had declined dramatically, so I travel somewhat lighter. The FlashTrax (and the power supply that’s nearly as large as the device itself) now enjoys a happy retirement as short-term backup device for my computer.

When a point-and-shoot camera was more appropriate than an SLR in the film era, I used a Minolta Freedom Zoom Explorer. It’s very light and compact, and has a respectably sharp 28-70 zoom lens (a later version had a 28-75 lens). When I used it alongside my old Elan II SLR, I had difficulty telling which projected slides came from the Minolta and which came from the SLR. It’s quite a lot of fun to use. Like so many things today, it’s made in China and has a cheap plasticky feel to it. But what do you really expect for $149 (in 1997)? I most recently used it on a cruise.

More recently, I supplemented my Rebel XT with a Canon Powershot S100. It fits in a shirt pocket and weighs less than 200 grams. But it provides the full range of exposure controls and metering available on an SLR, as well as raw files. The image-stabilized lens has a very useful range, equivalent to 24-120mm on a 35mm camera, and is nearly as sharp as my SLR lenses. It doesn’t replace the Rebel XT, which offers lenses with wider and longer focal lengths, the ability to use a polarizer, better ergonomics, and a larger sensor with lower noise and better tolerance for bright highlights (among other advantages). But a tiny camera with excellent image quality that I can easily take anywhere opens up many interesting opportunities.

For a while I regularly used a lightweight Slik tripod. Although it was inconvenient, the tripod did help me get sharper pictures with slow film. Switching to 400-speed negative film mostly freed me from Tripod Tyranny. I still bring along the tripod when I travel by car, but I leave it at home if I’m flying.

Scanning and Processing

From January 1999 through September 2003, I scanned all the images for the Web site with a Hewlett-Packard PhotoSmart scanner (the original SCSI version). I replaced it in September 2003 with a Canon FS4000US. When that scanner broke in July 2012, I replaced it with a Plustek OpticFilm 7600i.

I use Ed Hamrick’s VueScan to scan film. For the PhotoSmart, it was an absolutely essential replacement for the Kindergarten toy software with which HP hobbled it. For the Canon FS4000US, VueScan offers numerous advantages over Canon’s FilmGet (which I discuss in my review of the scanner). I continue to use it with the Plustek scanner, again for the reasons I discuss in my review.

From early in 1999 until October 2005, I edited the images using Corel’s Paint Shop Pro. I switched to Adobe Photoshop CS2 in October 2005, and later upgraded that to CS3, and then CS5.

You can see five examples of how my scanning and image editing techniques have changed and improved in Then and Now, originally a special feature for the eighth anniversary of this Web site.

Commentary Contents

tedsimages.com Virtual Light Table Home Page