This is a series of images taken using the 300D's 18-55mm kit lens. I chose a variety of objects for the scene, consisting of some boxes for nice straight lines, some CDs for radial symmetry, a small remote controlled car, and then a nice brick background which should clearly show "deepening" or "flattening" of a scene as a result of perspective.

I set up the scene, placed the camera on a tripod, and proceeded to take five images from the same location. The images were taken at 18mm, 24mm, the "dot" on the zoom ring, 35mm, and 55mm.

I then imported each image into Photoshop Elements, and resized each to match the scale in the 18mm shot. Then I proceeded to overlay each resized image onto the 18mm shot to compare differences. I used Photoshop's "difference" layer mode so that any discrepencies would be strongly apparent.

Since I was overlaying images, I then proceeded to overlay the "dot", 35mm, and 55mm onto the 24mm image. Then I overlayed the 35mm and 55mm shots onto the "dot" image. Finally, I overlayed the 55mm shot onto the 35mm image.

Feel free to measure the differences in each image. Note that the differences are small--usually no more than about 0.5 percent of the total image width--even at the edge of the frame where lens distortions are most likely.

What this test shows is that a significant change in focal length (18mm to 55mm, a factor of about three) does not affect the perspective of the image as long as the camera does not move or change its pointing direction.