1.5X Sensors VS. Full Frame and other questions...

Discussion in 'Digital Photography' started by Progressiveabsolution, Apr 27, 2006.

  1. Progressiveabsolution

    Mark² Guest

     
    Mark², May 4, 2006
    1. Advertisements

  2. Progressiveabsolution

    Mark² Guest

    Laik bilong yu, Roger.
    I fail to see how anything presented here offers evidence that a tripod
    offers no advantage over hand-held at the "rule" limits...which is, after
    all, what this is all about. This image MAY be sharper than it appears due
    to severe noise, blooming, etc. But if you're seeing fine detail that
    reveals sharpness...I am not. This shots seems to be reduced to a very
    basic shape outline photo. If its sharp, I'd expect to see more fine
    detail/texture revealed, but instead, anything at the "detail" level is
    merely a mush of color.
     
    Mark², May 4, 2006
    1. Advertisements

  3. Progressiveabsolution

    ASAAR Guest

    From the beginning I disagreed with Alfred that hand holding would
    produce images as sharp as if a tripod was used. But as I
    previously noted, his image appeared to be quite sharp, and if it
    wasn't sharp enough to satisfy some, I saw no evidence that it would
    have been due to camera movement. The importance of remembering
    that this was supposed to be an evaluation of Alfred's "hand
    holding" claim seems to have been forgotten. As usual, it was by a
    few of the cliquish regulars that so easily reverted to their
    tendency to discredit the ausländer that doesn't seem to share their
    DSLR passion, by finding something else to criticize. They bring
    back memories of Golding's LOTF.
     
    ASAAR, May 4, 2006
  4. Progressiveabsolution

    ASAAR Guest

    What you fail to see is any evidence that Alfred's shot would have
    been improved by using a tripod. If there's any blur due to camera
    movement it should be distinguishable from other causes of a lack of
    sharpness. It's possible to detect camera movement even in low
    quality images produced by PDAs and cell phones. Can you identify
    the direction of motion in Alfred's shot? Do you care?
     
    ASAAR, May 4, 2006
  5. Progressiveabsolution

    Rich Guest

    Something a bit easier on the eyes. Taken with a camera with a 2/3rds
    sensor. 100% crop.
    http://www.pbase.com/andersonrm/image/56575926
     
    Rich, May 4, 2006
  6. I hadn't actually looked at your crop of the spire... but it
    *is* "blown up". (When I looked at it on my monitor it had just
    about the same size it would have if the image were printed at
    40"x60" on paper. Your view might be smaller, or larger.)

    The enlargement happens when someone (anyone) displays it in a
    browser window that is different than the actual image size in
    pixels and has a different dots per inch resolution than would
    be used for a printed image. The window on the monitor used to
    view it may be larger or smaller, pixel wise, than the
    particular image being viewed. That means it will be enlarged
    or reduced differently for everyone who looks at it with a
    different monitor and browser window size!

    I would, for example, suspect that you are looking at it with
    perhaps a 1024x768 resolution monitor... or is it actually
    800x600 or 1280x960??? Viewed full screen the results will be
    different on each of those.
     
    Floyd L. Davidson, May 4, 2006
  7. Between the lense and the sensor, that is the entire problem.
    It *is* sharp, and there is no indication that it would be
    sharper if it had been bolted to the pavement with 1" steel
    screws.

    Look at that loop of wire or whatever it is sticking out from
    the right side of base of the center spire. Look at the edge of
    the roof with blue sky above it down in the right hand corner.

    It might be lense aberrations (the EXIF information says the
    lense is at maximum focal length and the aperture is at maximum
    too), or it might be the sensor. But the "unsharpness" that can
    be seen is *absolutely not* due to camera movement.
     
    Floyd L. Davidson, May 4, 2006
  8. Progressiveabsolution

    Mark² Guest

    I don't think that's correct. I'm fairly certain my crop is an accurate
    100%.
    Here's what I did:

    I viewed the image at 100% (which is, in this case, an image measuring
    2592x3888.
    That is the full-size image which he posted...for a file size of 3456.32 KB.
    This is in keeping with his claimed 3.5MB size when he posted the image.
    To the contrary, I suspect that others may be viewing the full size image in
    a state that is adjust DOWN to their browser window. When you select the
    option in IE, or Firefox, it ONLY enlarges up to the original size...not
    larger. Also...I saved the target of his link, and got the image I used to
    extract the crop. Viewed in ACDSee and Photoshop, the image dimensions are
    correct, and my crop is indeed 100%...no more.
    Also notice that Alfred hasn't protested. This is a 100% crop.
    Check with Alfred and he should confirm this.
    -Mark
     
    Mark², May 4, 2006
  9. Progressiveabsolution

    Mark² Guest

    Part of what's going on in Alfred's shot is that because he insists on
    hand-holding, he is forced to use large aperture (or noisy high iso
    settings). This is part of the whole issue. He insists that his
    compromises to allow hand-holding don't impact the image, and they clearly
    do. He suggested (on my f20 1/13th second shot) that I could have simply
    bumped up the ISO so I could hand-hold the shot. This is what I'm seeing as
    a disregard for the realities of how to achieve critically sharp images.
    For critically sharp images, you can't insist on hend-holding if it means
    you are forced to use a large f4.8. No wide lens is at its sharpest at that
    aperture, but since he doesn't believe there's a benefit to tripods that
    can't be matched by hand-holding at this extension, he's stuck with the
    image he's posted. It's a mess, and I won't be giving up the tripod any
    time soon.
    http://www.pbase.com/markuson/image/58828940/original
    The above was the shot in question.
    -Mark
     
    Mark², May 4, 2006
  10. It is enlarged. You just don't understand the effects of
    displaying images on a monitor.
    Wrong.

    It depends on the dot pitch, the number of pixels, and the size
    of the window it is displayed in. Which is to say that *you*
    only controlled one part (the number of pixels), and can have no
    control over whether any given viewer is seeing it enlarged,
    same size, or smaller.
    Exactly. That image is 2592x3888 pixels, and lets say you are
    using a 1024x768 display, where your browser or image editor is
    giving you a 900x600 image display. That might be about 10"
    high and 15" wide on a 19" monitor (I didn't work it out
    exactly, but that is close enough to demonstrate what happens).

    Your window is showing you only 900x600 pixels out of 2592x3888,
    which means that you are looking at a 10" high by 15" wide piece
    of an image that is actually 65 inches high! You take what you
    call an "unenlarged 100% crop" of that, and post it as a 900x600
    image (I'm using round numbers... but when I checked I see you
    posted an 800x589 pixel image... so this calculation is *very*
    close).

    You are looking at what amounts to hugely enlarged crop of
    the original, much as if it where printed at about the 40"x60"
    size that I mentioned previously.

    Note that anyone who views it on a 19" monitor that is using
    800x600 resolution is going to see it enlarged even more, while
    a 19" 1280x960 monitor would show it less enlarged.
     
    Floyd L. Davidson, May 4, 2006
  11. Then you need to re-examine what has been found in the image
    being analyzed. It is "quite sharp". It would not be sharper
    if a tripod had been used, because there is *no* indication of
    blur from movement.
    You are looking at a blow up that is equivalent to a 40" x 60"
    print but continue to think it is the same as if it were an
    8x10.

    I agree that the image would print very nicely at 16x20 and
    appear extremely sharp (Mr. Clark's appraisel above). I think
    it would print better than many at 40" x 60" too, given that we
    are viewing it at a distance of less than 2 feet and it is
    unlikely that a print that large would be displayed where the
    viewing distance would ever be that close.
    In a 40" x 60" print viewed at a distance of less than two
    feet, what do you expect?
     
    Floyd L. Davidson, May 4, 2006
  12. Only if you know the size of the original from which it was
    taken, and failing to account for that made your analysis
    incorrect.

    He *does* have a point!
     
    Floyd L. Davidson, May 4, 2006
  13. How can that be? It was shot at 1/50 with the aperture set to
    f/4.8, so it doesn't appear to have been a particularly bright
    day.
    He certainly was not using a "fast enough shutter" for a focal
    length of 72mm.

    Rather, it looks like Alfred has a remarkably stable hand!
     
    Floyd L. Davidson, May 4, 2006
  14. Progressiveabsolution

    Mark² Guest

    You're talking about the wrong image.
    This post didn't refer to Alfred's image, but rather Scott's image.
    Go back and look. Its a reference to this:
    http://www.sewcon.com/temp/16mp.jpg
     
    Mark², May 4, 2006
  15. Progressiveabsolution

    Mark² Guest

    One needn't analyse an entire frame to judge a 100% crop.
    For that matter, no monitors can even SEE a full res image from his camera
    at 100%, which means you can only EVER check a "crop"...since you can't see
    it all at once.
    A reference shot can be given for scope/relative size, but a 100% crop is
    most certainly sufficient for judging sharpness. If it wasn't, you couldn't
    ever judge it.
     
    Mark², May 4, 2006
  16. Progressiveabsolution

    Mark² Guest

    Ha!
    Yes I do, chum.
    Its irrelevent, because I didn't use a browser to view the file!
    Nope.
    Not if you're viewing at the actual pixel level in Photoshop.
    See above, oh master of the assumption!
    :)
    Clearly you know nothing of how I viewed the image.
    I have complete control over screen rendition in Photoshop and in ACDSee.
    No.
    I viewed the full size image--which meant I could only see a portion of it
    in Photoshop and ACDSee.
    You are simply wrong, bub.
    When you view an image pixel for pixel, there is no enlargement unless I
    tell it to do so.
    You are making completely wrong assumptions about how I viewed it...the
    program I used to view it...and my awareness of how this works.
    I didn't rely on a browser AT ALL. Save'???

    That might be about 10"
    Only if the person viewing the image is an idiot who uses a web browser to
    make an analysis of an image.
    I NEVER do this.
    I copy the target file (as referenced by the link)...and view it in either
    photoshop or ACDSee.
    From there, I have full control, and am quite aware of what a
    pixel-for-pixel view (or "actual pixel" view) is.

    -Mark
     
    Mark², May 4, 2006
  17. It was shot at ISO 160. You have another bogus claim.
    Mr. Clark has demonstrated the sharp edges in the image, hence
    it is a bit unreasonable for you to continue claiming there is
    any blur at all form camera movement. You can't demonstrate
    that here is, and almost everything you have claimed has been
    shown to be incorrect.
    It depends, of course, on the camera. If you were shooting that
    at ISO 50... I suspect he is correct. At ISO 400 you could
    shoot at approximately 1/100 second. That of course assumes you
    have a camera with decent noise characteristics at ISO 400,
    which (at least in Alfred's case) may not be true, but in yours
    may well be true!
    His image was in fact fairly sharp. But yes I believe it is
    possible, though not guaranteed, that it would have been better
    to shoot at less that maximum aperture, less than maximum focal
    length, and that would have required a shutter speed
    significantly longer than 1/50s, and he would have had to use a
    tripod.

    Now, given that the image appears to be quite good enough for a
    16x20 enlargement, we have little choice but to bow to his
    judgment that what he did do was indeed "good enough". The
    proof is in the pudding, and he has pretty good pudding.
     
    Floyd L. Davidson, May 4, 2006
  18. Progressiveabsolution

    Mark² Guest

    Apparently you haven't been in the thread from the beginning.
    My ISO comment is in reference to his comments on MY image...not Alfred's.
    Go back and read a bit.
    Really?
    No noise?
    No blooming?
    All is sharp?
    Detail is present?

    Wow.
    Here we go again...
    400 ISO was not acceptable to me, nor is it necessary to make that
    compromise the shot referred to (which again...was my shot here:
    http://www.pbase.com/markuson/image/58828940/original ).
    That was the point of contention.
    Go read this thread, Floyd ..
    You've clearly missed the crux here.
    Good man!
    I strongly disagree with Roger on that one...

    , we have little choice but to bow to his
    I've NEVER disputed this last part.
    I have (many times) mentioned that if its good enough for HIM...great!
     
    Mark², May 4, 2006
  19. Go for it, Mark. Bought one last month, took it to Chile for 2.5 weeks,
    magnificent! As is the 24-105 I bought with it. (Unfortunately I had to
    leave the lens in SA with my daughter, whose 28-135 broke down while we
    were with her - what we do for our children!).

    David
     
    David Littlewood, May 4, 2006
  20. This is wholly incorrect. There is no such relationship between object
    detail size and sensor resolution. Your assertion completely overlooks
    the magnification provided by the lens. Photomicrographs can show
    details down to around 250 nanometres - about 1/32 of the typical pixel
    pitch of a DSLR - and this limit only exists because of the diffractive
    nature of light, not because of sensor limitations. (With appropriate
    optics, the 250 nm resolution image could even be projected on, and
    resolved by, your extreme hypothetical sensor.)

    David
     
    David Littlewood, May 4, 2006
    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.