Olympus 14-42 EZ test after repair

Discussion in 'Native Lenses' started by Klorenzo, Mar 9, 2016.

  1. Klorenzo

    Klorenzo Mu-43 All-Pro

    Mar 10, 2014
    I just got the 14-42 after the second warranty repair for the same problem: the lens focused a little too far often giving a clear soft look that could be confused for shutter shock. Manually focused the lens works fine.
    First time they did an "AF review" and replaced one lens element, this time they replaced the AF motor but the problem is still there.

    I did a few tests this morning. The lens, at 14mm, focuses a little too far 2 times out of 3. At 42mm the behavior is better, still focuses too far but less and is more consistent in the overshot.

    This is a best/worst comparison at 14mm:

    (sorry for the ugly target, I needed something flat).

    So I took the time to extract the focus distances from the EXIF files and make a chart out of it. I know that many consider this information useless but I found the variation consistent with my eye inspection of the pictures. I noticed that this value depends on the focal length so it is not a simple "absolute" number. Focusing the same object reports 2.5 meters at 14mm and 3.5 meters at 42mm (real distance was about there).

    This is the chart for the 14mm shot series shown in the above comparison:

    The scale on the left is meters, blue is AF and red is MF. My focus is much more consistent and is almost 4 meters shorter. Here the "spread" (std dev) is about 3 meters, is seems almost random.

    At 42mm the AF "spread" is about 10 cm but still 0.6 meters above my manual focusing point. Here the AF shots and the MF ones are almost indistinguishable by eye. Notice that the scale changed.

    I also took the O60 for a comparison:

    Now is my focus that seems like I'm drunk but the scale is again different: there is a 15cm scattering in my shots (BTW the same target as before now reports 5 meters). The O60 is admirably consistent. We disagree on the focus distance by about 20cm but the result is just the same.

    I know that AF precision is limited, but here the variation is big and the baseline is off too.

    There are a few things that I get from this:
    - a good AF motor matters as much as lens sharpness, shutter shock and all
    - this may explain some of the different opinions of this lens
    - maybe building/calibrating a pancake zoom is really hard (maybe there is a reason for the P12-32 not having the MF ring other then size?).
    - all tests done in the reviews are with manually focused lenses (of course)
    - some other lenses with mixed reviews (12-50?) may have a similar problem?
    - honestly I do not know if to pay for another shipment for repair or just sell it for the P12-32 or P14-45. I hate to sell "broken" things but I really have little faith in the next repair...
    • Informative Informative x 1
  2. Lawrence A.

    Lawrence A. Mu-43 All-Pro

    Mar 14, 2012
    New Mexico
    Real Name:
    Hmmm. Mine arrived yesterday. I had one before and liked it but nonetheless sold it to concentrate on primes. I'm now concentrating on small. I though I exhibited similar behavior once last night, in a dim room and focusing rather close, but after reading this I'll put it through its paces tomorrow, when I'll have a chance to get out with it. I like my last copy a lot and never noticed a similar problem. Thanks for the heads up.
  3. Lawrence A.

    Lawrence A. Mu-43 All-Pro

    Mar 14, 2012
    New Mexico
    Real Name:
    Got a chance to check my copy out today, and everything seems to be fine. If I use the small focus point on the E-M5 that is exactly where it focuses, time and time again. Don't know what happened last night.
  4. SojiOkita

    SojiOkita Mu-43 Top Veteran

    Feb 23, 2014
    I had the same problem with Olympus repair service and the 14-42EZ (which had exactly the same AF problems as yours.)
    I sent it twice through the store I bought the lens. No change.
    I sent it once directly to Olympus to see if was better (I paid the shipment this time). No change.

    So I emailed Olympus to say that their repair was a shame, and they paid for another return, and they finally changed the lens.
    The lens I got seems a little lower optically but the autofocus is fine.
    I don't use the lens anyway, but I don't want to have a defective lens (I'm probably going to sell it in the end, and it's easier when your lens works fine.)
    • Like Like x 1
  5. Klorenzo

    Klorenzo Mu-43 All-Pro

    Mar 10, 2014
    UPDATE: I got the lens back, a couple of weeks ago. Exactly like before...

    So I wrote to the UK support service where I was told the lens was sent (I later discovered that it went to Portugal for repair). They asked me to update the camera firmware (it was 1.2, I installed the 1.3 update that AFAIK is only relevant for the O300) and provide a few samples. Fine.

    I took 6 series of 20 test shots each (tripod, self timer, etc.) 10 with AF and 10 MF (120 total). Different focal lengths, focus distances, etc. I even took "backstage" shots of the tests, with the SCP, AF box and Magnify box locations.

    The difference IMO is huge, constant and clearly visible also downscaled to fit the screen size (17% downscale).
    This is an example:


    On flicker you can find the two full size images. Here you can find the whole 120 shots I sent him if you are curious (375MB):

    Dropbox - test-May-5-2016-JPG.zip

    Well, first he told me that he cannot see any difference. And that it is normal for AF to be less precise then MF (with CDAF on a flat subject with good contrast!?). I tried to tell him that at 14mm f/3.5 the depth of field is huge so small AF errors should not be visible anyway.

    Today this thing:

    "Please stick the largest newspaper you can to the wall,
    Set the camera on a tripod, turn off the IS, adjust the camera so that the newspaper fills the picture and the camera is parallel to the wall.
    Using a shutter speed of 1/250 or more take an image using S-AF without any face detection and using a self timer (I would advise 10-12 seconds depending on how steady the tripod and floor are).
    Now without any magnification or peaking take another picture using MF.

    Repeat the test for 14 and 42mm at apertures of 5.6 and 22 (8 pictures in total) making sure the paper is the only thing on the screen."

    It took me 10 emails, 14 days, a few hours to do the test, upload, answer the email, to get to this.

    This morning I wrote to the customer service kindly asking to transfer the case to someone else for a "deep gap in communication between us".
  6. Klorenzo

    Klorenzo Mu-43 All-Pro

    Mar 10, 2014
    Just a quick update. The customer service passed my request to that very same person that kindly told me to go to hell.

    So I went back to the Italian support, they finally seemed to care about the problem, they acknowledged the problem from those very same samples and, free of charge, they sent a courier to pick up the camera. After another month I got the camera back with another AF motor replacement and, this is new, the focus helicoid replacement. From a quick test now the problem is solved. The lens does not seem as sharp as I hoped after all these reviews but it is reasonable and much more usable than before (EDIT: I used the lens for a few real shots and it is plenty sharp, no complaints at all). So quite an happy ending for this story and a big thanks to Polyphoto for solving this (eventually).
    Last edited: Oct 23, 2016 at 3:32 AM
    • Like Like x 1
  7. SojiOkita

    SojiOkita Mu-43 Top Veteran

    Feb 23, 2014
    Quick update also for me ;)
    I tested the exchanged lens when I got it back and it was fine.

    I had planned to sell it as I never use it (I prefer my 12-32).
    So before selling it, I made new tests and in fact the AF is not fine.
    It's not as worse as the first one, but it's not usable from my point of view...