Subject and Composition versus Equipment

Last week-end, I contributed a photo to a silent auction. It's one of my favorite images from the Road North series called Sunset on the White House. Given that the photo was going to be shown in the context of a silent auction, with people milling around comparing donations ranging from art to golf outings, I thought that I'd attach a rather detailed description of my artist's statement and a description of the image.

When I looked at the EXIF data for the image, I was shocked to discover that I'd taken it several years ago, using an original Canon DRebel and a crummy 70-200 consumer lens. This was the first of a series of DSLR cameras (10d, 20d, now 5d) that I've owned and certainly the lowest quality of any of them. The lens is not well-regarded by any of the equipment review sites that all of us frequent.

However, it remains one of my favorite images from the Road North project and I've exhibited the image upsized to 16x24 without any ill effects. It just goes to show you that the camera and the lens really don't matter a heck of a lot. It really is the subject and the composition that count.

Sunset on the White House

Full Frame - Marketing Gimmick or Not?

Now that I have a Canon 5d full-frame camera, I'm wondering whether I just got caught up in a marketing push by Canon (and now Nikon) or whether there is something really magical about a full-frame camera. Here's why: if you wanted to buy a camera system plus three lenses that covered everything from wide angle, the cost of full-frame is pretty steep. Let's compare (using prices from B&H):

• APS-C package
• Canon 40D with 28-135 IS lens $1,500
• Canon 10-22 EF-S lens $690
• Canon 70-300 IS lens $549
• TOTAL: $2,739
• FF package
• Canon 5D with 24-105 L lens $3,330
• Canon 17-40 L lens $679
• Canon 70-300 DO lens $1,100
• TOTAL: $5,109

Why did I choose somewhat lesser quality lenses for the APS-C package? The smaller sensor uses the centre of the lens, so sharpness out to the corners is less critical than with the full-frame sensor. My experience with a 10d and now a 20d is that you get almost identical performance with cheaper lenses and a smaller sensor.

As you can see, full-frame comes at quite a penalty in cost. In fact, it is nearly double the cost of an equivalent APS-C system. Is the difference worth it or is it marketing hype that has driven me to buy an expensive full-frame system?

Here's an article by Ken Rockwell that goes into a lot of detail on the full-frame advantage. Ken describes why he likes his Canon 5d and how it compares to his Nikon APS-C cameras. Here are the salient points:

• Sharpness: It isn't just the number of megapixels that determines how sharp an image is. When you look at a photo, the sharpness (assuming you've taken it in focus) depends on the resolving power of the lens as well as the resolving power of the camera. Modern APS-C cameras are capable of pretty good resolving power (measured by how many black and white alternating lines you can distinguish in a millimeter) and rival full-frame. However, lenses are not necessarily up to the task. Theoretically, if you had a perfect lens, capable of outresolving any camera, the APS-C 40d sensor would actually outresolve the 5d's sensor (the 5d has around 8 megapixels within the APS-C sensor coverage area, where the 40d has 10 megapixels). But, lenses aren't perfect and the sensor actually passes the limitation of the lens. The 5d produces sharper images because it spreads its pixels out over a wider area. Think about a photo of a picket fence. Let's say that a lens can resolve 40 fence posts per millimeter. In a full-sized sensor, this works out to 40x35 = 1,400 fence posts in the photo. For an APS-C-sized sensor, the lens is only capable of showing 40x22 = 880 fence posts in the photo. Regardless of how well the two sensors resolve the fence posts, the APS-C camera starts off with fewer to begin with. Assuming the sensors can out-resolve the lens, you'll have a much sharper photo from the full-frame image.
• Noise: Big pixels are reputed to produce less noise that smaller pixels. It makes sense that a larger pixel will catch more photons of light, requiring less amplification than smaller pixels. Noise is a by-product of amplification. Now, it is true that Camera manufacturers are working hard to produce less noise in their circuitry and are working hard to put faster processors in their cameras so that sophisticated noise reduction programming can be applied, but why not start with less noise to begin with? Now, I'm not a big expert on camera noise, but the folks who produce Noise Ninja are and they develop noise profiles on most DSLR cameras. I thought I'd run a couple of 1600 ISO photos through Noise Ninja, one from my 20d and another from my 5d to see what the default noise profile was for each one. The result was a noise reduction level of 32 for the 20d and 23 for the 5d. I'm not sure if these numbers are linear or not, but there is a significant difference in values between the two cameras. I'm not sure where the 40d would come in, but I suspect that it would be comparable to the 20d if you turn off the noise post-processing in the Digic III processor.
• Colour differentiation: As a corollary to the lower noise argument, if the pixels are producing less noise, it should be easier for the camera's processor to detect the true colour of a the light being received by a pixel as opposed to muddying it in response to random coloured noise.
• Ability to use wider angle lenses: There is a limit to how wide an angle you can build into a lens and I would speculate that Canon's 10-22 AF-S lens is probably pushing it. This lens delivers a similar angle of view to a 16-35 mm lens on a full-frame camera. Canon makes a 14 mm lens that delivers a wider angle of view than possible with an APS-C camera.
• Brighter, larger viewfinder: My 5d has a much larger, brighter viewfinder than the 20d and it is such a pleasure to use!
  

So, there are advantages to full-frame, but do any of them really matter when all you want to do is take beautiful photographs? In reality, the one that smacks you in the head is the brighter, larger viewfinder. It makes it easier to compose an image because all the elements in it are larger. Pretty elementary really. The ability to buy a 14mm lens doesn't turn my crank -- 17 mm is fine. I really haven't seen the colour improvement yet, but I've only had the camera for a couple of days. Sharpness and lower noise will be important when I create larger prints, but right now my printer limits me to 13 inches wide anyway.

So is it hype or reality? I'm still waffling. Maybe the differences will become more apparent as I use the 5d more, but right now I think the premium is probably too high. I'm sure it is tough to build large sensors in the quality needed. Probably a high percentage of sensors get turfed at the assembly line due to flaws. But, does that justify spending twice the amount of money on a full-frame system. I'm not sure it does.

Canon 5d First Impressions Part II

It's day two of my Canon 5d experience and I'm still down for the count with this chest cold. I finally got tired of sitting around on my butt and ventured forth for a walk just after lunch. One of the projects that I'm currently working on is a portfolio of construction images for my two year-old grandson. He's absolutely in love with construction equipment and Bob the Builder, so I've been taking photos of bulldozers and diggers for his bedroom wall.

There is a construction site not far from our house where they have knocked down an old shopping mall and are building a new-style shopping "village" with open-air stores and street parking. There are bulldozers and diggers galore on the site. Where better to test the Canon 5d?

The main challenges were the lighting and the distance between spectators and the action. The lighting at noon was pretty harsh, with the sun nearly directly overhead. Stout chain link fencing kept the spectators well away from the machinery and the fencing was fairly high.

At first, I took only my Canon 70-300 mm DO telephoto zoom lens to try to get close enough to the action. This proved to be a good choice -- the machines were right in range and the business end of the lens is only 58 mm wide so it fit between the fence links. After I got home, I decided to drive back later in the day and take a couple of wide angle shots with the Canon 17-40 mm L lens. The 77 mm diameter lens wouldn't fit inside the chain links, but I was able to find a higher perch to take the panoramic view that was needed.

And how did the Canon 5d perform? Let me first give you a caveat. I'm agin pixel peeping in general unless the performance of the sensor is getting in the way of capturing the image you want. The purchase of the 5d was done because of its full-size sensor (and how Canon lenses behave with it) and its compact size. It wouldn't have bothered me necessarily if the sensor had been 8 megapixels or even 6 megapixels. It's interesting to note that the review of the Canon 17-40 lens referred to above, done by Michael Reichmann, was written in 2003 when a top-of-the-line Canon digital camera was an 11 megapixel model. No one complained about having not enough pixels.

My comments will be constrained to how the camera felt in the field, whether it intruded into the picture taking and whether it produced usable images. Here are my comments:

• The camera handles extremely well. Both lenses seemed totally in balance with the weight of the camera. I can't imagine using the 70-300 DO lens with, for example, a Canon Rebel DSLR because the lens weighs a lot more than the camera does and would dwarf it physically. On the other hand, I've seen the pros wrestling with Canon 1Ds models of various vintage and couldn't imagine toting one of those beasts on a nice walk on a sunny day. It would be enought to give you a back ache. Again, the camera would outweigh the lens and be out of balance.
• As with yesterday's shoot, I loved the sound of the shutter. It sounds like my film camera.
  
• The controls of the camera fall readily to hand. Of course, it helps if you've had a number of Canon cameras, but that's the beauty of sticking to one brand. All three cameras in my bag (Elan film, 20d and 5d) have identical controls, so there's never a false moment when you're screwing around looking for a dial or messing with a menu.
• I was shooting raw files and developed them in Adobe Lightroom. The results were pretty decent. Given the lighting conditions, I expected that the focus would be spot on (lots of light to work with) and it was. I expected the photos to be too harsh and contrasty, but the camera seemed to have enough dynamic range to produce usable images. The histograms were spread out nicely and only a touch of exposure/blacks adjustment was needed. There was not a single image with highlights blown out. That's pretty incredible given the strong light and the highly reflective metal surfaces out there in a construction site.
• I did cheat a little and zoomed in to 3:1 to make sure that my sensor wasn't a dud. Noise was very, very low as expected. At ISO 100, I didn't even bother to take the photos through a Photoshop and Noise Ninja round trip.
• The overall impression was very positive. The photos required virtually no adjustment in Lightroom other than very minor exposure tweaking, a touch of clarity and just a touch of saturation (the camera is set at a the default saturation and is a tad low for my liking, especially for construction pics for a toddler).
• If the 5d wasn't new and in need of a good work-out, I would have chosen the 20d for this project. Nearly all the shots needed just a touch of cropping to get the subject into a perfect position. In other words, the subjects were just a bit beyond the range of the 300 mm end of the zoom with the full frame camera. The 20d would have filled its frame with the subject and would have brought its full 8 megapixels to bear. By cropping the 5d, I estimate I was down to around 7 megapixels or more. Oh no! Not a megapixel short! In reality, either camera would have done really well.
• However, the 5d was perfect for the wide angle shot at 17mm. I got nearly the whole construction site into the viewfinder and that was pretty cool. The full-frame sensor pushed the Canon 17-40 mm lens to its limit at the wide angle end and there was some vignetting (darkening of the image at the corners), but Lightroom has a tool that makes it a snap to fix this problem.
  
• Yes, I could have bought the Canon 10-22 mm EF-S lens and fitted it to the 20D to get the same result as the 5d with the 17-40 zoom for much less money, but my main walk-about lens is a Canon 24-105 L lens and it is designed perfectly for a full-frame sensor like the 5d.
  

In summary, the Canon 5d is just a pleasure to use. It blends into the background as it should and frees the photographer to get into that lovely zone where the concern is purely for the subject and the light.

Here are a couple of the photos from the shoot. The first was shot with the 70-300 DO lens and the second was shot with the 17-40. Both were given a very light dusting in Lightroom:

Big Digger, little Digger

The WHOLE Site!

Canon 5D First Impressions

I've been lusting after a Canon 5D for a while, but the price tag is pretty extreme considering that you can buy a perfectly decent 20D or 30D (and now 40D) for half the price. I've been getting along quite nicely with a 20D for most of my work, supplemented by a Canon Elan film camera. Several things came together nicely to make it possible to buy the 5D at last: the Canadian dollar rose to match the US greenback, the 40D was announced, negatively impacting the 5D used market and I did a consulting gig that game me a little extra dough. As a result, I was able to find a nice 5D on Craigslist.org and I've had the camera for a couple of days.

Unfortunately, there has been a nasty cold going around the office, so I wasn't able to get out and really put the camera to work. I've been relegated to using the camera in my wife's lovely garden, taking photos for my "fading beauty" project. In this project, I'm looking for the beauty in all things that age, whether it be flowers or people. I find that flowers in particular are far more interesting when they are past their prime and the flaws that were hidden when they were in perfect bloom are now starting to give character to each one.

So here are my first impressions of the 5D, from the point of view of someone who has owned a 10D and a 20D:

• Physically, the camera is lovely. If you are used to the Canon high-end consumer cameras, you'll find that all the controls are exactly where you'd expect them. The camera is slightly heftier than a 20D, so if feels very solid and reliable in your hands, like holding a piece of solid metal.
• The shutter is much quieter. It sounds more like my Canon Elan film camera than my 20D. The shutter noise of the 20D is one of my only dislikes about that camera. It sounds like a chiken clucking every time the shutter is pressed. I'm going to enjoy the sound of the 5D.
• The viewfinder is a joy. After peering through the small viewfinders of the APS-C line, I'm delighted with the large, bright viewfinder of the 5D.
  
• The display screen on the back is a little bit disappointing. Yes, it is large and bright, but when it is displaying the histogram along with the image, the image is teeny tiny. There is a lot of unused real estate for some reason.
• The image quality is excellent, but it is early in the game to comment on this right now.

Even though I've been using the Canon Elan for quite a while and have done a lot of travelling with it, I've used it mostly for wide angle shots, taking advantage of the wider angle of view of the full frame camera. My every day camera has been the 20D, so there were very few times when the Elan was used for macro photography. As you can imagine, the fading beauty series demands a lot of medium distance macro work, where you're trying to fill the frame with a flower and its immediate environs. The 20D does this very well with my Canon 24-105mm lens. The closest focal point of the lens, combined with the effect of the small sensor manages to fill the frame with the image I'm looking for.

Imagine my surprise when I found one of the negative effects of converting to a full-size sensor. The 24-105 lens has a limited macro capability and I was already as close as I could go with the 20D. At the same distance, the 5D has a much wider angle of view, so the flower no longer filled the frame and the image included too much extraneous stuff. Clearly, I was going to have to find a better lens for this type of work.

Last Christmas, I asked my wife to buy me a set of Kenko extension tubes for macro work and I'd not really made much use of them. The combination of a 50mm lens with a 12mm extension tube was okay, but the 20D's crop factor just seemed to fill the image with too much of the subject. Enter the 5D and the combination now seems to work really well for photographing images of flowers. Happily, as one door closes, another opens up!

Here are some samples of my fading beauties taken with the 5D an the combination of a Canon 50mm lens with Kenko 12mm extension tube:

It's the subject stupid

I enjoy Alec Soth's blog. Every day, he has something interesting to say about the art of photography. Lately, he's been focusing on the teaching of photography, largely due to an experience that he had doing a freelance teaching stint. He seems undecided on the question of whether photography can be taught or not. Certainly, the technical side of photography can be taught, but can you teach someone to see? Can you teach them how to pick a subject, how to time the shot and how to compose the image? Alec isn't sure and neither am I.

However, there are lots of articles about the artistic side of photography and I've been reading them quite avidly to try to improve my art. For example, Alain Briot writes quite often on the art of photography. He's written the featured article on Luminous Landscape called "Developing your Vision".

My favourite article is a discussion between two accomplished photographers, David Hurn and Bill Jay, a chapter in their book On Being a Photographer. You can read the chapter on picking a suitable subject here. It made a point that I hadn't really given a lot of thought to, but it is so simple that it hits you right on the forehead -- doh!

You should photograph subjects that you are passionate about! Of course. How could I be such an idiot? Don't bother with crap that is just pretty or stuff that other people are passionate about, focus on the subjects that turn your crank.

There are a couple of important corollaries to this point. First, you have to research your subject to find the right images. Second, you should pick subjects where your passion intersects some other group of people or you'll be taking photos purely for your own enjoyment.

This reminds me of Ed Burtynsky. No wonder he is so popular and successful. I heard him speak once and he is absolutely passionate about the industrial landscape. He researches each photograph for days or weeks -- stalking his subject like a hunter in the forest. He may visit a target subject many times looking for the right vantage point, the right time and the right lighting conditions.

Similarly, Alain Briot specializes in taking photos of the grand canyon and environs. He hikes for miles and miles looking for the right vista, the right time of day and the right weather conditions.

Yes, that's it! It's the subject stupid. Now, what am I passionate about? That's the question.

Framing your Fine Art Prints

If you're anything like me, you love nearly everything about photography. There is the joy of quietly stalking a subject, whether it be animal, mineral or vegetable, and the lovely zen moments that you feel when you are clicking away.

Then there is the magic of making images come alive using tools like Photoshop and Lightroom. Images that sometimes come out of the camera flat and two dimensional can sparkle and take on a fantastic depth with some judicious application of curves, levels and local contrast.

Finally, there is the process of printing out your images. First, you do a draft or two on matte or semi-gloss to see if you have a photo that wants to be an oil or a watercolour. Then you pick from your treasure trove of lovely, clean sheets of paper. You might want to use Epson Luster or something textured like Moab Entrada. You load it in the printer and press the print button and, a few minutes later, you have a beautiful print.

I think the best moments happen when the print comes off the printer and you have this lovely image in your hands. You move around the room trying it out in different kinds of light, looking at all the things that made you fall in love with this particular image. It may be the composition or the colour or the subject itself.

And then what?

This is the cruelest part of the whole process. You realize with a sinking feeling that you are not Ed Byrtynsky. There are no loyal fans lining up to buy your prints. Your walls are already full of all the prints that your spouse will let you hang and your easel is stacked full of previous endeavours. Maybe you've sold a few prints and, perhaps, you have hopes of a showing in the not-too-distant future, but what do you do with that lovely image that you just printed off?

You could hide your images away in portfolios or albums, but somehow that isn't the right thing to do. Images want to be seen.

Here's one solution that you might want to consider. It's a way of framing your prints in a fairly easy, cheap way and then rotating them on the walls of your apartment, house or studio so that each image gets some face time and you don't get tired of your gallery.

First, you make a little sandwich. Go to your art store and buy 16x20 matte board (museum quality). It should cost about $3.00 a sheet. While you're there, buy some foam core board and a can of spray glue (e.g. 3M). If you don't have one already, buy a plastic roller as well. Trim the foam core to around 16.5x12.5 inches. Foam core is tricky to trim because the foam catches on your Xacto knife. A guillotine actually works best if you have access to one at the office. Print out your image on either matte or gloss paper using 13x19 inch paper and crop your image to 13x17.

I know, I know, the Michael Reichmann's of the world would protest that you should crop your photo naturally to fit the composition, but sometimes compromises are necessary and we're only talking about small adjustments. We want the image to look nice in a commercial 16x20 frame with an inch and a half border around the image.

By now, you should have a 13x17 image, a slightly smaller piece of foam core and the 16x20 matte board. Place the image onto the matte board so that the boarder is even all the way around and mark the top two corners with a light pencil mark. Then, spray the glue onto both sides of the foam core. Do this outside so that your spouse doesn't yell at you for smelling out the place. Quickly, position the foam core onto the back of the photo and press lightly into place so that there is about a quarter inch of space around the foam core. Finally, flip the image and foam core over, line up the corners of the image with the two dots on the matte board you made earlier and press the foam core onto the matte board. Take the plastic roller and roll over your image, removing all the bubbles and making sure that the glue is well seated.

You now have an image mounted on foam core and matte board in such a way that it appears to be floating an eighth of an inch above the matte. Images can be easily stored in a drawer for future framing when their time in the rotation comes.

Now for the framing part. Again, go to your local art store or frame store and purchase the widely available Nexxt Suspense frame in 16x20 size. This is designed to be a floating frame, but it can be turned into a nice shadow-box style frame very easily.

Unpack the frame and throw away one of the pieces of glass. You mount your matte board to the back of the frame so that the sequence of pieces (viewed from front to back) looks like this: one sheet of glass, wooden spacer, matte board. Flip the little gizmos around to hold in the matte paper and you have a nice shadow-box frame to show off your image. The foam core gives the matte paper enough stiffness that it doesn't need anything behind it and it also makes the photo float up close to the glass.

The cool thing about this is that it's easy to rotate your images through the frames. Just turn the little gizmos on the back of the frame, remove one image/foam/matte sandwich and insert another and you have another image hanging in your living room.

This is a sure cure for the "I've got nowhere to hang this lovely image" blues!

Serendipity

I'm always amazed when I look at a photograph and find something unexpected that takes the shot from one level to the other. I guess I don't look at the world with enough focus to actually see these things ahead of time. Like Mr Magoo, the old cartoon character, I seem to stumble on them. The law of averages dictate that the more photos you take, the better the chances of finding something interesting and serendipitous. Here's what I mean.

I was taking photos of Edmonton, a very interesting city that goes from low-rise prairie town to high-rise city in a matter of one city block. I saw a woman exit my hotel and light up a cigarette. Terrific, I thought, here's an opportunity to add to my smoker project. I like taking pictures of people smoking outside buildings -- they have an lovely contemplative aura to them. It was a nice picture of a woman smoking, leaning up against the wall. When I developed the photo, I was surprised to see a passing bus in the frame that had a fragment of the words "peace of mind" in it. What a lucky piece of luck!


In a similar vein, I was bobbing around in a kayak on Sunny Lake in northern Ontario taking pictures of flowers in a small wetland adjacent to the lake. Later at home, I was developing the photos when I happened on a photo of a water flower that looked quite nice. The composition was fine and the flower was nicely in focus. When I looked more closely, I found an ant on the flower - posed like Yurtle the Turtle at the top of the plant, king of everything it surveyed. The photo transcended the original "post card" look and became a jumping off point for the viewer to create a story around the ant.

It just goes to show that there are terrific moments in time to be captured, when circumstances create a juxtaposition of elements that suggest a story or a mood. To capture these moments, you either have to be very good (think Henri Cartier-Bresson) or take a lot of pictures and be lucky!