December 11, 2017, 06:34:01 AM

Author Topic: My favorite method of focus calibration. Cheap, fast, accurate and foolproof.  (Read 2946 times)

Luds34

  • EOS 5DS R
  • ******
  • Posts: 907
Since i have a DPAF camera (70D & 6Dmk ii) i adjust focus by comparing the difference of focus  movement of the lens in live view vs. OVF movement. In live view i magnify ten times to check the focus on the target. When the lens doesn't move after switching and refocussing between live view and OVF i think the value stored is the right one.
Useually i use approx. 50 times focal distance.
Anybody with same sort of experiences?

Yes, I do something similar. I just shoot a test chart and compare between liveview and PDAF. I usually try to defocus near and far, just to make sure there is no discrepancy depending on where the focus started from.

BTW, while I did do this with my old 70D and new 6D2, this method worked fine with my old 6D. DPAF is just so much faster, the old contrast detection, while very slow, is just as accurate. In fact I'm quite sure all the mirrorless cameras fine tune to "accurate focus" with contract detection after letting the phase detection get it most of the way there.

canon rumors FORUM


picturefan

  • EOS Rebel SL2
  • ***
  • Posts: 97
I'm used to some EOS Cameras, which the 7D MKII is the latest of them.
4 (!) L-Lenses and some no-canon-lenses do not focus correctly, means I need to do some "time consuming" afma with them.
The only high aperture lens delivering tack sharp images with the 7d MkII-af is the nifty-fifty, stm-version.
Not really satisfying for such a pricey equipment...
DPAF is working well, but not fast enough for some wildlife.

Sporgon

  • Canon EF 400mm f/2.8L IS II
  • *********
  • Posts: 3463
  • 5% of gear used 95% of the time
    • www.buildingpanoramics.com
This thread caught my eye because I have found that checking focus on a sloping target is unreliable.

I have my own quick method: usual things, good light, solid tripod etc. I have a well defined vertical target and ensure the camera is lined up horizontally on the centre point. Using BBF I take a shot using TTL phase AF. I then switch to live view and x10. Gently moving the focusing ring I see if I can improve on the sharpness, noting which way I'm turning the focusing ring, rear or far. I do this four times. If there is a consistent difference I then adjust the AFMA by the amount I guess it's out, if any, normally starting at +\ - 5. Then repeat four times. Assess.  Hone in from there.

Maybe I'm lucky, but all my lenses are naf all out.

This method isn't as easy to do on a STM lens or an original 5D.

Talys

  • EOS 7D Mark II
  • *****
  • Posts: 587
  • Canon 6DII
Neat stuff, thanks for sharing :D

I think you'd have some problems with this method using telephoto lens, just getting enough distance to the target.

A long time ago, I used a cheap cardboard foldout from Amazon that cost something like $5.  It worked fine; I still have it mounted to tripod-friendly plate.  Then, I bought a fancier LensCal, which worked a little better.  Dot Tune is totally free, but it has not worked as consistently for me.

The problem with all of these methods is if you have more than a couple of lenses.  I think I have something like 8 zoom and 5 prime lenses (at least, that I care about AFMA with)... it takes FOREVER to AFMA them all manually.  I can blow a whole day, and still second guess myself.  It also sucks if I buy a new zoom lens, because with 4 bodies, it means that I have to do 8 calibrations.  Or, at least 4 calibrations, for my 2 favorite bodies.

If your camera supports Reikan FoCal in full auto mode, I highly recommend it.  Sometimes I may disagree with it slightly, but overall, it saves me so much time.  If I'm suspicious of the result, I will use a LensCal and double-check it, but pretty much always, I'm the one that is wrong :) 

canon rumors FORUM