How do you test a zoom for focus accuracy?

This is what I am talking about. This image has had two levels of sharpening applied (USM and a high pass layer), so it doesn't look too bad, but that's not the point. Look at the guy coming in from the left, he is much sharper than the intended target (the runner). I had the camera set to center point only in AI servo on a 20D.

BryanP wrote in post #2019854
The point of the test is to check for front or back focusing. It's assumed that if the focus is spot-on then back focusing or front focusing at any distance should be close to or pretty much non-existent.

Well, that was my assumption too....till 2 minutes back when I read the post about back focussing on a 24-70 Canon. There are 2 who have similar issues and the back focus is only at 50mm and below.

edit: Found the thread - http://www.photography​-on-the.net/forum/showthre​ad.php?t=218612

So, all you new lens owners - prepare to test your lens at every focal length

One good reason to buy primes

kram wrote in post #2020033
So, all you new lens owners - prepare to test your lens at every focal length

I did that . Does that make me bad ?

I really started noticing what you see above at FLs beyond 180mm. Maybe it is a little more than a calibration issue . Or maybe there is just less DOF field at 300mm (I am sure that is what I was seeing (Ihope)).

Travis

Travis F wrote in post #2020092
I did that . Does that make me bad ?

Travis

No, just really tired

I plan to return some reading on this - if/why the shorter focus lengths seem to have a higher tendency to be back/front focus.

Definitely test at different focal lengths. I just bought the Sigma 100-300 f/4 EX and, while it's spot on at 300mm, it badly front-focusses at 100mm

If this helps, this is the most realible focus test chart I have used, because the target is in a perpendicular plane to the axis of the camera and big enough (no way to miss it with good light):

http://www.canon-dslr.com …/Canon_SLR_Focu​s_Test.htm

gcogger wrote in post #2020637
Definitely test at different focal lengths. I just bought the Sigma 100-300 f/4 EX and, while it's spot on at 300mm, it badly front-focusses at 100mm

Yes, if it´s a zoom, test it at different focal lenghts, some F/stops wide mainly and repeat the series, letting the distance ring in different positions, I means make AF with the objective comming from infinitum, repeat making AF comming from the closest distance, etc ---> Repeat several times for consistence.

I had a consistent back focus problem with my original EF 70-200 f/4 L, but at 200 mm f/4 was working pretty acceptable.