Results 1 to 10 of 10

Thread: Canon focus mystery

Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. #1

    Re: Canon focus mystery



    I think that Trowski hit the nail on the head (after the lensrentals.com article author). Misalignment between imaging sensor and AF sensor is the culprit.

  2. #2

    Re: Canon focus mystery



    Consider also that there is a third path: from the main mirror to the focussing screen. If the screen is not in the right position then the viewfinder can appear out too. I believe Canon use thin shims to align that. I had an original EOS1 that had VF disagree with AF.


    Paul

  3. #3

    Re: Canon focus mystery



    Hey Trowski, thanks! That article definitively unveils many mysteries about phase-detection autofocus, it was the explanation I was looking for!


    I think some bodies could just "tell" the lens to move its focusing group a certain distance, but don

  4. #4
    Senior Member neuroanatomist's Avatar
    Join Date
    Jan 2009
    Location
    Boston, MA
    Posts
    3,922

    Re: Canon focus mystery



    Well, the article does explain why AFMA is necessary, but doesn't really address Bruce's fundamental mystery as I interpret it. I've seen something similar myself, as has scalesusa it seems.


    For example, in my case:
    • The 16-35mm f/2.8L II requires an adjustment on the 5DII that is 4 units negative relative to the adjustment on the 7D.
    • The 70-200mm f/2.8L IS II requiresan adjustment on the 5DII that is 4 units negative relative to the adjustment on the 7D.
    • The 85mm f/1.2L II requiresan adjustment on the 5DII that is 2 units positive relative to the adjustment on the 7D.
    • The 100mm f/2.8L Macro IS requiresan adjustment on the 5DII that is 6 units positive relative to the adjustment on the 7D.
    • The 24-105mm f/4L IS requires the same adjustment value on both my 7D and my 5DII.




    <div>I had thought there would likely be systematic differences - a particular camera was x units off in one of 'true', and a particular lens was y units off of 'true', in theory if you knew those values for a given camera and lens you could predict what the AMFA should be for the combination of them. But in practice, that's not the case - each combination (at least with the bodies and lenses at my disposal) seems unique.</div>
    <div></div>
    <div>So, from a theoretical standpoint is remains a mystery. From a pragmatic standpoint, it's irrelevant - I will just continue to do an AFMA for each body+lens combo I have...</div>

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •