Photographer and YouTuber Michael the Maven has put together a video about one of the worst-kept “secrets” in optics: sample variation. Or: why you should immediately test every lens you buy to make sure you have a good copy.
If you’ve been a reader of this site for a while, or you’ve ever read anything by Roger Cicala over at LensRentals, you’ll be well aware that no two lenses are exactly the same. This is particularly pronounced with zoom lenses, where the sample variation can be huge—performance at any one focal length in the zoom range can vary drastically from lens to lens, even if you’re using the exact same model.
But since we don’t all have a test bench, the skills, or the patience to properly test every lens we buy and graph the MTF curve or create a field curvature plot, Michael shares his personal technique for testing each new lens he buys to ensure he has a good copy.
Essentially, his technique is to set his camera up on a tripod in front of a flat, textured surface like a brick wall and snap photos at various apertures: wide open, f/2.8, f/4 and f/8. Feel free to add in f/5.6 if you’re feeling comprehensive. If you’re testing a zoom lens, we recommend repeating this process at various focal lengths as well.
Try to get the sensor as parallel to the wall as possible, and inspect each photo from the center out to the edges—it should be immediately obvious if you have a really bad copy at any particular focal length.
Then, as a bonus test, shoot some power lines against a blue sky and see if the lens is producing any dramatic chromatic aberration, which will show up as color fringing at the high-contrast edges between the black wires and the blue sky.
This might seem like nit-picking or pixel-peeping, but it’s just plain good practice. If you’re using a 24-70mm f/2.8 and you shoot most of your work at the wide end, you want to make sure you get a copy that performs really well at 24mm; it won’t do you much good to find out a year down the line that your lens is soft at 24mm but great at 70mm if you hardly ever use the tele end.
Check out the full video to hear Michael’s full explanation of his testing process, alongside some examples shot with one of his lenses.
Source: petapixel.com
