Time for another one of my unscientific photography tests. I've heard it said that using short focal lengths for portraits makes for distorted images and big noses. I'd never really thought about it before, but today I did. And it started me wondering whether this is true and if so, why. After all, it's the same image you're taking isn't it?

In order to in test the theory I enlisted the help of Tina's stuffed koala, Eukie. I took 5 shots of his cubby little face, trying to fill the frame with has head from ear to ear. I took one shot from really close up at 17mm, one from a bit further away at 40mm, one at 50mm, one at 70mm and a final one from half way across the garden at 200mm. I didn't worry about sharpness or correct exposure. Here's the results, 17mm first, 200mm last:

The effect is somewhat subtle as you look at most of the images, but what stands out is the way his nose looks so much bigger on the first image. On the 17mm shot his nose is 108 pixels high from top to bottom. On the 200mm shot his nose is 93 pixels high - that's 15%. The other values are 40mm - 101 pixels, 50mm - 96 pixels and 70mm - 93mm.

As I stress with these tests, they're not scientific. I didn't measure things too accurately or frame Eukie's face too carefully. But there's not much doubt that the effect exists and is quite obvious at focal lengths up to about 50mm. After that the images looks much the same, with little apparent difference between 70mm and 200mm. You can see it if you look at those last two images, but you need to look carefully. On the other hand, here are the 17mm and 200mm shots together:

Not much question about the big nose effect then...