September 30, 2007

SUBJECT: Canon Mark III vs. Canon 40D

This is a rather long journal entry, but I felt it was necessary to help me sort out my own photography needs and also to share with others my opinions on the latest cameras available from Canon that are commonly used for bird photography.

Back in December of 2006 I sold my Sigma 300-800mm lens (affectionately known as the Sigmonster) so I would have cash-on-hand for the widely anticipated replacement for the Canon 1D Mark II professional camera. I was secretly hoping that this new camera would be a 12 megapixel, 1.3x crop camera. (The 1.3x crop means that the sensor is not full-sized compared to 35mm film, but cropped by 30%. This feature works to "magnify" the image by 30% when compared to a standard 35mm film camera.)

In March of this year, Canon finally announced the much anticipated Mark III as a 10 megapixel, 1.3x crop camera able to take 10 frames per second. This was a sport and wildlife photographer's dream especially with its superior autofocus that the professional series cameras are known for.

After waiting nearly three months I took ownership of a 1D Mark III camera. I went to take my first image and poof— ERROR 99! ERROR 99 is a generic error message that really didn't tell me what was wrong with the camera, but regardless the camera was dead in the water.

I returned the camera and within one week I had my second 1D Mark III. This one worked out of the box and within a few weeks, I had most of the options figured out. While other photographers were complaining about autofocus problems, I felt the camera worked as stated.

During the months of July and August, I used the camera extensively photographing birds-in-flight whenever I could. My goal was to hone my skills in time for the late-fall and winter arrival of bald eagles that frequently fish below the dams along the Mississippi River.

Then in September I started getting intermittent ERROR 99's with the replacement camera. The only way to clear this error was to remove the battery pack and reinstall it. Sometimes the error would go away with one try while other times it took many tries. I found this quite unacceptable and contacted my dealer who then contacted Canon and asked Canon about replacing the Mark III with another new one. Reluctantly, Canon agreed to do so under the terms that this be the last time that the camera be replaced with a new one. If I had any further problems with the Mark III, I would have to send it in for service.

Now I have owned many Canon digital SLR's and this was the first time that I have ever had a problem with one of their cameras. Considering that the Mark III is a professional camera and that I paid $4,500 for it, I expected the very best out of this camera.

I decided to send the camera back for a replacement. While waiting for my third Mark III, I happened to run across the Canon 40D at a local store. I could only test the camera in the store, but the autofocus appeared to work better than previous versions of Canon's mid-range DSLR's. About 30 minutes later, I was the proud owner of a 40D.

The next morning I headed out the door with my Canon 400mm f/5.6 and 500mm f/4 lenses to test this camera out on photographing birds-in-flight. I was amazed to find that the 40D's autofocus had greatly improved from its predecessors. As long as I kept the central focus point on the bird, the camera stayed focused! Returning home, I was stunned at how many keepers I had. Most of the photographs that I had taken were nice and sharp.

With this new revelation, I began to wonder if I really needed the Mark III. I decided to compare the differences between these two cameras from a bird photographer's perspective. So let's compare the two:

Resolving power is the ability to resolve detail in the subject that is being photographed. Getting as much feather detail as possible is one of the many goals of a bird photographer and the resolving power of the lens and camera play important roles in obtaining this goal.

In most instances, the photographer is focal-length-limited while photographing a bird. In other words, you just can't walk up to the bird and photograph it. You must photograph the bird from a distance. Moving closer to the bird will likely scare the bird away or you just don't have the time to move closer before the bird moves on.

So in order to get the most out of a focal-length-limited bird you need a long lens and preferably some form of a cropped sensor camera. Canon has three different kinds of DSLR's that feature different sensors: 1x, 1.3x and 1.6x.

A 1x sensor (commonly known as a full-frame sensor) will produce an image that mimics a 35mm film camera. Typically, the sensor size is close to 35mm in width. A 1.3x sensor crops the sensor by 30 percent and a 1.6x sensor crops the sensor by 60 percent. Here is an illustration of how each of these sensors would capture a photograph with the same lens from the same distance:

Using the above illustration, if all three sensors had the same number of pixels, the 1.6x cropped sensor would (in theory) provide the greatest resolving power.

Of course, not all cameras have the same number of megapixel's and when you meld that with the sensor crop, things can get confusing real quick. In order to compare the camera's resolving power, I find it useful to convert the cropped sensor's number of pixels to a full-frame sensor. Then we can compare apples to apples instead of apples to oranges. To accomplish this, here's a handy formula to use:


full frame size = pixels x crop factor x crop factor

Using this formula we can compute how many pixels a Canon 40D would have if it were a full frame DSLR:


full frame size = 10.1 x 1.6 x 1.6

full frame size = 25.856 megapixels

What this tells us is that if the 40D's sensor was made to be full sized, it would contain over 25 megapixels. This gives us a good indicator as to how well the camera can resolve a bird from a set distance. Using this same formula we can compute how other Canon DSLR's stack up to the 40D:

Camera Crop Factor Megapixel's. Full-Frame Equivalent in Megapixel's.
40D 1.6 10.1 25.856
Xti 1.6 10.1 25.856
1Ds Mark III 1 21.1 21.1
30D 1.6 8.2 20.992
1D Mark III 1.3 10.1 17.069
5D 1 12.7 12.7

As you can see in the above table, the 40D and the Xti have the greatest pixel density. But does this mean they have the highest resolving power for photographing a bird from a set distance? Not necessarily. There are other factors that come into play, one of which is noise. The amount of noise increases as the density of the pixels increase. The only way to really know how well a camera is going to resolve feather detail is to try them out.

I had done this a few months ago with the Mark III and an Xti. I borrowed a preserved northern flicker from the local nature center to test the resolving power of the two cameras. The flicker was set at 30 feet from the camera using the Canon 500mm lens with the camera set to ISO 200. Here is an example of what I found:

1D Mark III

The Xti outresolved the Mark III. This was to be expected however. The greater pixel density cameras usually win this test as long as noise levels are kept to a minimum. I would also expect similar results with the new 40D camera, but I don't have a northern flicker handy to accurately test this at this time.

Conclusion? The 1.6x vs. 1.3x crop at 10 megapixels does make a difference when photographing a subject that is at a set distance. The Xti and 40D will out-resolve the Mark III in a focal-length-limited scenario.

Onwards to other factors to consider...

Web design © 2007, Alan Stankevitz
All photographs © 2002-2007, Alan Stankevitz

This site requires Flash Player version 8 or above. Click Here to download Flash Player