I have a much better test than this one now. Here it is.
I’m leaving this post up for historical reasons, but I strongly recommend that you use the new test.
From the mailbag:
I just found your blog and have been really enthralled by all the data you’ve got! I had a question I was hoping you may be able to help with. I’ve got a Sony 16-35 lens, and am worried it’s decentered. Trying most of the standard decentering tests @ 16mm seem like they are super sensitive to test setup issues. For example, even though I got a test chart printed at 20×30, that only allows me to put the camera ~10 inches away which leads to a lot of false-positives if the lens isn’t perfectly parallel with the chart. Do you have any other good methods to check on such a wide lens? Thanks for the help, and please keep up the blog!!
I have published some semi-complicated decentering test methods that don’t require precise alignment:
And some quite complicated ones that don’t require precise alignment:
But there’s an even easier way, for which I am indebted to Professor Hank on DPR. All credit to him. I’m going to give you my version, so, if there’s a problem, it’s most likely my fault.
You’ll need a dark room — thankfully, not a darkroom (those are getting harder and harder to find), just a room that’s dark. Ideally, you want to get at least 50 times the focal length of the lens you’re testing away from the target — more on that in a sec — and, within reason, the farther away you can get, the better.
You’ll probably want a tripod, although it’s not strictly necessary.
And you’ll need a small LED flashlight. Within reason, the smaller the better. The ones that operate from a single AAA cell are good. The ones that use just one AAAA cell are harder to find. If the flashlight has power settings, set it to the dimmest setting. It’s easy to get one that’s too bright, and I’ve never found one that’s too dim. The flashlight should have only a single LED, not a group of them.
- Put the flashlight at the far end of the room facing the camera.
- Turn the flashlight on.
- Go back to the camera
- Turn it on, too.
- Douse the lights
- Set the camera’s ISO knob to base ISO.
- Open the lens all the way.
- Focus it to as close as it will focus. That’s right, set the camera to the minimum focusing distance.
- Center the out of focus image of the flashlight.
- If you’ve got a mirrorless camera, adjust your shutter speed until the zebras go away and/or the right side of the histogram comes off the end about a stop.
- If you don’t have a mirrorless camera, take pictures, look at the histogram, and adjust exposure until the right side of the histogram comes off the end about a stop.
- For other cameras, take pictures and adjust the shutter speed until that happens.
- Is the blurry disk all within the frame? If so, you’re done. If not, adjust the focus until there’s a fair amount of black around it.
With a mirrorless camera, you don’t even have to take a picture. Just look at the disk.
- Is it evenly illuminated?
- Is it round?
- Are the edges well defined and the same all the way around the disk?
If yes, congratulations; you have a well-centered lens. If no, you’ll have to do additional testing, like the tests I linked to above, to find out if the decentering is going to damage your photographs.
I’ve been waiting to post this until I found an example of a decentered lens. Unfortunately, or fortunately, depending on how you look at things, I haven’t found such a lens. I suspect that decentering is not as common as it is portrayed on the ‘net, and that many perfectly good lenses get returned because their owners don’t set up their tests precisely.
Here are a couple of sample images. One has been cropped, because the OOF flashlight didn’t com anywhere near filling the frame.
You will notice, if you look carefully, that the Sony image is not quite as circular as the Otus one. This is a departure from perfection that I would not call decentering, or at least, the kind of decentering that would motivate one to send the lens back. In other testing the Sony shows that it’s not decentered in a way that might significantly affect normal photography. The non-circular opening is a result of the diaphragm not fully opening when the lens is set to wide open; you can see the image of the edges of the blades.
You’ll also notice some patterning in the disk. Don’t worry about that. Some of it maybe from fine details in aspherical elements in the lens (you’re actually looking at magnified marks left by the machines that milled the lenses or the molds). Some of it may have to do with the light source, if it’s not small enough or you’re too close. And, of course, this is an excellent way to discover dust.
Also, don’t worry if there are flattish edges equally distributed around the periphery of the disk. The diaphragm may not be completely out of the way, even though you’ve opened up the lens as far as it will go. You’re looking for bulges, or an ovoid shape.
Occasionally, you may see a strip across the image, like this:
Of that happens, don’t worry. It’s just the flashlight pulsing the LED at high frequencies. Ignore it.