AFMA/FoCal advice requested

[Jura]

I recently bought the canon 16-35mm F4L IS USM on the back of a wave of good reviews. Upon getting the lens in hand I've followed the lens testing advice here http://www.canonrumors.com/tech-articles/how-to-test-a-lens/ which was very useful.

I've been finding the focus on the lens a little off and not owning focal I used the dot tune method to see what was going on at 35mm, 24 mm and 16mm and I'm throwing up a required +1AFMA at 35 mm, a -1/-2 AFMA at 24 mm and a +6 at 16mm.

I've never had an issue of a zoom lens showing different requirements at different focal lengths. What would members of the forum suggest I do?

1) Dial in +4 and see if that works, I know wide angles are less sensitive than a fast teliphoto prime..
2) Buy FoCal to get more data (does foCal alow you to tune the lense at different areas of the zoom range or is that only the sigma dock?
3)Return the lens and ask for a replacement (Amazon purchase). I want to avoid this if possible as this friday I'm going to India then Australia for a few months and really wanted to use the lens on this trip.

Any and all advice welcome.

 

[neuroanatomist]

I've found FoCal to be reliable, DotTune not very reliable. While you can test at multiple FLs, the 5DII allows only one AFMA value per lens (newer bodies allow a wide and a tele value for zooms, and interpolate for intermediate FLs).

If your values from DotTune testing are accurate, you'll need to replace or service the lens, or pick something like +2 (as you say, DoF is deep with wide FLs unless the subject is very close).

 

[AlanF]

It is quite usual for zooms to have different AFMA for different focal lengths. FoCal is, as Neuro writes, very reliable and DotTune doesn't work for some of us.

 

[neuroanatomist]

It is quite usual for zooms to have different AFMA for different focal lengths.

True. IMO, it's important to test at intermediate FLs. Personally, I test at 1-3 of them (more with greater zoom range). My 24-70/2.8L II is 0 at one end and +5 at the other, and the intervening points (35 and 50mm) fall right on the regression line. If they did not, I'd have sent the lens for service.

I did have one lens similar to the OP, a 28-300L that I bought used and had the same values at both ends (+1), but the three intermediate FLs gave a sort of 'V' shape (–8 at 200mm). I decided to sell the lens (better IQ with the 24-70 II + 70-300L, ended up with a small profit on the sale anyway), but had I kept it it would have gone to Canon.

 

[Jura]

Thank you for the recieved advice.
I guess biting the bullet and purchasing Focal is the best thing I can do. If after the trip I'm finding the focus with the 16-35 a problem it can go for a tune up.
I'll be interested to see how my other dot tuned lenses compare with the focal results and will post my findings here.

I wont be upgrading bodies for a while yet so will have to wait for the functionality to tune different values at different focal ranges...

 

[Mt Spokane Photography]

Canon considers AFMA that is within +/-5 to be within spec. Try setting your AFMA to +3. that would put it within spec over its range, and if your AFMA values are correct, images should be sharp.

If you use the lens mostly at one focal length, I'd optimize for that focal length.

I'm one who does not trust dot tune, but it should get you in the ballpark and likely close enough.

If you compare images with the Dot Tune AFMA with the exact same using contrast dect in live view, that should tell you if the AFMA is right or close enough. If the image is sharp with live autofocus, but not with the AFMA you have set, try a different value about 5 to 10 points closer and further and then start adjusting towards the right direction.

Remember, AF systems do vary, so take 10 careful shots on a tripod with eyepiece covered at each AFMA setting. Reset the lens to infinity or minimum focus before each shot. Its a pain to do all this manually when Focal Pro automates the process, but you will get there.

If you use live autofocus and images are not pinpoint sharp, the lens is likely defective.