D70 focus problems??

Discussion in 'Digital SLR' started by Erik Persson, Jan 7, 2005.

  1. Erik Persson

    Erik Persson Guest

    When I compare the pictures from my D70 and my PowerShopt G3 the G3
    pictures seems very much sharper. This is the case even when I increase
    the sharpness in the settings. Maybe, it's hard to say for sure, there
    is a little back focus. By the way, this is with the 19-70 kit lens.

    I'll do some research tomorrow to see if there is an actual bacfocus.

    Anyone else with the same problem?

    /erik
     
    Erik Persson, Jan 7, 2005
    #1
    1. Advertisements

  2. Erik Persson

    Colm Guest

    Are you shooting the same subject with both? Same aperture and shutter
    speed? Tripod or handheld?
    Any examples?
     
    Colm, Jan 7, 2005
    #2
    1. Advertisements

  3. Erik Persson

    Ed Ruf Guest

    I believe there was a back focus issue with early models.
    ________________________________________________________
    Ed Ruf Lifetime AMA# 344007 ()
    See images taken with my CP-990/5700 & D70 at
    http://EdwardGRuf.com
     
    Ed Ruf, Jan 7, 2005
    #3
  4. Erik Persson

    Erik Persson Guest

    To be honest I haven't investigated it to that degree. It is more a of a
    thing I've noticed when I compare pictures from the two cameras. I have
    a large number of very crisp G3 pictures, and none of the D70 pictures
    are as sharp. My feeling is that the use of a tripod does not make any
    difference.

    I'll put some examples on my webserver - will post the URL later.

    /erik
     
    Erik Persson, Jan 7, 2005
    #4
  5. Erik Persson

    Erik Persson Guest

    I got the camera in august 2004 and it has (and had from the beginning)
    the latest? (A 1.01 B 1.02) firmware. As I understand, it was a new
    camera (they had to order it for me) and not something that had been in
    the store for a long time.
    Wasn't it the very first cameras that had the focus problem?
    Did this problem persist until late summer 2004?

    It could be that there is more in camera sharpening in the powershot,
    but I can't see any halos on the photos.
    The AF systems is different in campact and SLR digitals as far as I
    know. The photos are all taken with AF so this _maybe_ could have some
    impact - guessing here though...

    I'll try to investigate this further and make my own focus test shart.

    /erik
     
    Erik Persson, Jan 7, 2005
    #5
  6. Erik Persson

    C J Campbell Guest

    This guy claims to have a test for it. You might want to give it a try.

    http://md.co.za/d70/chart.html#ActualChart
     
    C J Campbell, Jan 7, 2005
    #6
  7. Erik Persson

    Erik Persson Guest

    Thank you!
    /erik
     
    Erik Persson, Jan 7, 2005
    #7
  8. Erik Persson

    Owamanga Guest

    He only tests the center horizontal focusing sensor. Wouldn't it be
    prudent to test each one separately?, there are 5 others.
     
    Owamanga, Jan 7, 2005
    #8
  9. Erik Persson

    bmoag Guest

    There is lens sharpness and digital sharpness.
    You can compare manual to autofocusing on your own to see if the autofocus
    is really the problem.
    The D70 at default jpg settings seems to apply significantly less sharpening
    than P&S and prosumer cameras and ramps up the contrast much less, all of
    which affects apparent image sharpness. Also the physically very small
    apertures in lenses used for smaller sensor cameras will tend to increase
    depth of field and apparent sharpness regardless of the focal length of the
    lens.
    However, comparing raw images of the same subject from a Nikon D70 and a
    Sony 828 I find there is an undeniable softness to the raw image coming off
    the Nikon D70 compared to the Sony 828 that is not simply due to the
    difference in megapixel count between the sensors, using the Adobe raw
    plug-in for both cameras rather than the Nikon/Sony raw converters and
    trying to use equivalent focal lengths and apertures.
    I do not know if this is a sensor or software issue: it is possible that
    in-camera processing is applied to the raw image differently in both
    cameras.
     
    bmoag, Jan 7, 2005
    #9
  10. Erik Persson

    Erik Persson Guest

    Now I have done some tests comparing my G3 to my D70 with 18-70 kit
    lens. The G3 lens is 35-140mm eqiv with aperture 2.0-3.0. The D70 18-70
    is 3.5-4.5 with a 27-105 eqiv (1.5 crop factor). I compared the D70
    (with the 18-70) to the G3 at the same aperture and at each cameras/lens
    wide angle resp telephoto limit using a home made focusing chart. I used
    a tripod.

    At wide angle one of the objects on my test chart had a size of 65 mm on
    the G3 and 70 mm on the D70 (roughly the same distance between the
    sensors and the object). The D70 image was however very much larger in
    the objects direction so the D70 picture thus corresponds to a smaller
    focal distance. It is however of importance below when I compare the
    resolution, since even though the objects on the G3 were smaller in
    terms of number of pixels they occupied, the resolution and sharpness
    seemed to be the same as with the D70!
    I found the percieved sharpness about equal as well as the resolution
    (that is, if a could tell 2 points apart in one of the pictures I could
    do the same in the other). There was some sharpening artefacts in both
    images and none were better than the other. The G3 had a larger DOF (but
    you could take the G3 down to aperture 2.0 if you want a small DOF). The
    D70 DOF was quite a bit back, but the object I focused on where in focus
    (maybe there were a very slight back focus).
    The G3 thus performed equally well as the D70 even though the objects on
    the G3 was a little smaller (in of number of pixels).

    At the telephoto limit one object where 233mm in the G3 and 191 mm in
    D70. The G3 picture was virtually free from sharpening artefacts whereas
    the D70 had some halos (I set the D70 sharpening to medium high). In
    this test the resolution of the G3 was superior to the D70, possibly
    because of the larger tele. I found the sharpness to be better in the G3
    picture (but not by much) as well. In this test the D70 focused
    accuratly with about equal DOF at both sides. Again, the G3 had a larger
    DOF (but you could be set the aperture to 3.5 instead of the tested 4.5).

    When it comes down to a pixel by pixel comparision, that is if a object
    takes up the same number of pixels, the G3 seems superior. The G3 had no
    problems with setting the correct focus (besides being much slower), but
    D70 18-70 had some minor problems.
    I have some problems explaining the larger DOF of the G3, especially
    since it were more in the telephoto range, but I guess this is because
    of the smaller optics that can be used secondarily to the smaller CCD.

    There was some focusingproblems when using the D70 with the kit lens,
    and the G3 seemed overall a little bit sharper (pixel by pixel), but I
    can't say that this was because of bad focusing.

    This test does not explain the differences I've seen before!

    /erik
     
    Erik Persson, Jan 7, 2005
    #10
  11. Erik Persson

    Erik Persson Guest

     
    Erik Persson, Jan 7, 2005
    #11
  12. Erik Persson

    Erik Persson Guest

    Interesting! If the only difference between the both raw formats are the
    softness then Nikon is using an inferior CCD or bad processing.

    I found the D70 to be less sharp (not much) than the G3 _even_ though
    there were a little bit more halos in the D70 image. That's compatible
    with your findings. The sharpening algorithms etc have to work harder.
    The relative lack of sharpness could be the result of bad sharpening
    algorithms etc, but in the light of your findings it is more likely to
    be the quality of the input to the sharpening algorithms that is the cause.

    To my knowledge Canon is leading the pro digital SLR league at the cost
    of high prices. The D70 should however compete well against the EOS
    300D, and generally be a better by even though it's a little bit more
    expensive. I haven't seen any nice and well done comparison between the
    two when it comes to image quality (with kit lenses) though. 300D will
    not have as much noise, but apart from that what should one expect??

    I don't know if I could do better then the AF system. When I look
    through the viewfinder in AF mode my impression is that of good
    focusing. On the other hand, my test showed that there really were no
    (or at the very most very slight) back focusing at the very wide angle
    of the lens and no problems at all at the tele end of the lens. It seems
    that the focusing isn't the problem (if there are any problem :)).

    My tests could not explain the differences I see in my other photos.
    Maybe I'm just wrong and the cause is that I have so much more G3 photos
    and among all these, there are some that are perticulary well.

    /erik
     
    Erik Persson, Jan 7, 2005
    #12
  13. Erik Persson

    Ed Ruf Guest

    What do you mean by this? Sounds like you are are near or at the close
    focusing limit of the kit lens, ie 1.2 ft.
    ________________________________________________________
    Ed Ruf Lifetime AMA# 344007 ()
    See images taken with my CP-990/5700 & D70 at
    http://EdwardGRuf.com
     
    Ed Ruf, Jan 7, 2005
    #13
  14. Erik Persson

    Ed Ruf Guest

    How exactly are you comparing these different sized (pixel wise) images?
    ________________________________________________________
    Ed Ruf Lifetime AMA# 344007 ()
    See images taken with my CP-990/5700 & D70 at
    http://EdwardGRuf.com
     
    Ed Ruf, Jan 7, 2005
    #14
  15. Erik Persson

    Erik Persson Guest

    The first comparisions where more of "this one was sharp at 100%
    pixelwise, lets see what the D70 pictures look like. Hmm, that one
    wasn't as sharp, neither that or that or that... I have heard of a D70
    back focusing problem. Could it be so, or is it something else.".
    The latest comparision was more systematic. I took a number of pictures
    at the telephoto and wide angle ends of both the G3 and the D70 18-70
    and compared them. At wide angle there were little difference in the
    sizes (pixel wise) of objects in the picture. (se another post)
    At the telephoto end however there were since the G3 has a longer 35mm eqiv.

    I guess you want to tell me that the D70 is sharper than the G3 since a
    more "correct" test would be to compare the pictures at equal sizes
    (disregarding the different width to height ratios). This is of course
    one way to compare the pictures. One important factor is however how
    much one can crop and increase the size of the pictures. A test
    comparing pictures pixel wise is thus important, as it tells me that
    some of my extra pixels won't give me an increase in picture size but
    are "wasted".

    Then, as a bonus, I think it is interesting to try to find out why an
    image that occupies roughly the same number of photodetectors in the two
    cameras differ in terms of percieved sharpness (well, if it had done so
    more than it did :) )

    /erik
     
    Erik Persson, Jan 8, 2005
    #15
  16. Erik Persson

    Erik Persson Guest

    If that's the case, it would make me happy! I'll try to do the test
    again tomorrow at a larger distance. However, I think I was farther from
    the focus point than 40 cm but I can't remember for sure. I got a "focus
    lock" on objects much nearer the focus point (i didn't take any
    photos!!!), and there was no problem at 70 mm even though the distance
    from the focus point to the sensor was the same.
    I could be badly wrong here! It would be nice if you could give me an
    explanation in that case, ie why wasn't there any "problems" at 70 mm
    when it was at 35 mm.

    Anyhow, as I explained, the AF could not (if we extrapolate the test
    results to larger distances) be the reason that the pictures from the
    D70 looks less sharp (pixel wise) since the the focus point always were
    in focus (although the DOF was more to the back at 18 mm).

    /erik
     
    Erik Persson, Jan 8, 2005
    #16
  17. Erik Persson

    Ed Ruf Guest

    No, I'm trying to understand exactly what you are comparing, 1:1 pixel
    comparisons taken how? raw, jpeg, what settings?or something else. I own a
    Nikon 990, 570 and a D70. I still use all three.

    Yes, but you need to compare consistent processing. Try this test with all
    in camera processing turned off. Not just sharpening, but any contrast,
    brightening or saturation, etc. processing.
    I've done this with my 990/5700 and now my D70. You need to turn all in
    camera processing off to have any hope of a true unbiased comparison.
    ________________________________________________________
    Ed Ruf Lifetime AMA# 344007 ()
    See images taken with my CP-990/5700 & D70 at
    http://EdwardGRuf.com
     
    Ed Ruf, Jan 8, 2005
    #17
  18. Erik Persson

    Erik Persson Guest

    Ok, at first it was just pictures, mostly jpegs, taken with the two
    cameras. Settings varied. Not systematic in any way. Then I tried to
    understand if the AF was to blame - and it is probably not. As an
    "extra" I tried to see if there was any differences in picture quality
    with the settings I had applied to both cameras or if it just happened
    to be that among the more numerous G3 pictures I was lucky to find a
    couple that looked very sharp. To me it seemed like the G3 pictures
    where sharper even in the more standardized setting and in one instance
    had lesser sharpening artefacts - wich I find interesting. No clue to
    why though.
    Agrees mostly. But I think it is interesting to see how the camera
    processing works as well. Is one camera introducing more halos without
    incresing the "sharpness" compared to another camera etc. I'm only at
    the very beginning of the investigation, and later I'll try to see if I
    can locate the part, if any, to blame. For the moment I'm just having a
    number of pictures from the D70 and the G3 where the G3 pictures seems
    sharper pixel wise. My firts hypothesis was that the AF didn't work as
    expected since there's been a lot of talk about the D70 AF. But the test
    I've done indicates that the AF seems to work. But was there really a
    difference in sharpness or was it only a random occurence that hit me?
    But even using this more standardized procedure I found that the G3
    pictures looked sharper (and I could do this test at the same time as
    the AF test). Thus, it isn't likely that it was just some random effect
    that i noticed earlier. Accordingly i find there is reason to
    investigate this further.
    But so far I'm just guessing the reason of this sharpness difference -
    maybe it's the postprocessing, maybe the CCD, maybe the lens. I haven't
    come that far yet. The next step will for certain involve comparing the
    cameras with no postprocessing in raw mode! Thank you for the advice!

    In the end I want to take as nice pictures as possible with as little
    postprocessing as possible for most of them, leaving me more time to
    postprocess only the nicest pictures.
     
    Erik Persson, Jan 8, 2005
    #18
  19. Erik Persson

    Ed Ruf Guest

    If you are shooting at anything les than the 40cm min focal distance I have
    no idea. This is one area where a dslr with most lenses is not as capable
    as a P&S. I still use my 990 and 570 when I need real macro shots. Jus used
    my 900 the last few days fro shots within 3 in. My D70 won't do that.
    ________________________________________________________
    Ed Ruf Lifetime AMA# 344007 ()
    See images taken with my CP-990/5700 & D70 at
    http://EdwardGRuf.com
     
    Ed Ruf, Jan 8, 2005
    #19
  20. Erik Persson

    Erik Persson Guest

    I'll do the test from a greater distance tomorrow. It could be that I
    was to near.
    My g3 is superior to the 18-70 in macro as well. My brother has a nikon
    mikro 60 mm that i'm going to borrow the next time i meet him. Could be
    fun to test some makro photography with the d70.

    /erik
     
    Erik Persson, Jan 8, 2005
    #20
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.