How do you test a lens' accuracy

[donmash]

Can you test the accuracy and focus of a lens? What about the auto focus functions? Can these be tested or does it have to be sent back to the manufacutrer?

Input would be appreciated.



Thanks

[SmartWombat]

There's good advice on the Canon CPS Pro Support KB page:

http://www.cps.canon-europe.com/kb/d...jsp?faqId=1070


Carry out the following focusing test at a reasonable light level (office lighting or brighter) to ensure optimum results.

1. Place the camera on a sturdy tripod and use a remote switch when firing the shutter. For maximum quality, consider the use of mirror lock (e.g C.Fn 12 on the EOS-1D)

2. Select a focusing target with adequate detail from centre to edge. A newspaper page is a good choice.

3. Make sure that the target is totally flat and as parallel as possible to the camera.

4. Set the camera to One-Shot AF and manually select the centre focusing point.

5. Compose the picture so that the test target fills the frame from edge to edge.

6. For each of your lenses, make one exposure at the maximum aperture of the lens and another at f/8.

7. Open the images in your computer and make two copies of them: one 'as is' and another with a liberal amount of unsharp mask applied. Here are some recommended settings: amount, 300%; radius, 0.7 pixels; threshold 2 pixels.

8. Make A4 prints on a high-quality inkjet printer using photo paper and photo ink. Compare the results.

If the camera has a consistent focusing problem with several different lenses, then it's reasonable to assume the camera needs an adjustment. If you are getting sharp results from most lenses but not all, then it's reasonable to assume that the lens in question may need an adjustment.

[mwfanelli]

Can you test the accuracy and focus of a lens? What about the auto focus functions? Can these be tested or does it have to be sent back to the manufacutrer?

Input would be appreciated.

My techno-luddite tendancies are coming forth again. My method is very simple. Put the lens on a camera and take photos. If there is a defect, you will see it. If you see nothing, there is no defect or it is irrelevent.

[donmash]

Paul,

Thanks for the input, I will check it out.

[Trevor Ash]

Somehow I knew you were a Canon shooter.

And I don't mean that in a demeaning way. It's just one of those things

[another view]

I keep it easy - take your 70-200, shoot it wide open at 200mm and at close to (but not at) the minimum focusing distance, within a foot or two. This will give you a very shallow depth of field so you'll know what's going on. Use any lens for this, but a 200mm at f2.8 and a distance of about 6' will give you a very shallow DOF - moreso than most lenses that people usually own. In this case it will be obvious if it's right or wrong. And of course there's your preference in USM treatment to consider too.

Obviously you want to do this on a tripod with all the other stuff that Paul recommends (or just a self timer will be OK too). Trevor knows I'm not a Canon shooter though...