P&S vs DSLR - Does this argument make sense?

Discussion in 'UK Photography' started by aniramca, Jul 26, 2007.

  1. aniramca

    Alfred Molon Guest

    Interesting. What was the exposure time of the P&S?

    In any case you tested one camera, so you can't make generic statements
    about all P&S cameras. You claimed that all P&S have this kind of delay.
    Less of an issue with compact P&S as DOF is huge.
     
    Alfred Molon, Aug 1, 2007
    #81
    1. Advertisements

  2. Thanks for looking! I doubt that even you would only lose 1% in the
    circumstances I described! ;-)

    The main problem is that there is no time to focus and auto focus - even
    fast - is distracted by the background.
    Sure. And I would add (from my years of lugging a box full of lenses
    around) a single good quality zoom increases the opportunities lost from
    changing lenses!!

    I'm not claiming that any one camera type is better than any other, btw.
    There's a place for all of them. (Yes, I know - the trash can for
    some!!)

    Mike

    [The reply-to address is valid for 30 days from this posting]
    --
    Michael J Davis
    <><
    Some newsgroup contributors appear to have confused
    the meaning of "discussion" with "digression".
    <><
     
    Michael J Davis, Aug 1, 2007
    #82
    1. Advertisements

  3. No, I've tested several Fuji, Nikon, Canons.
    1) Small sensor P&S cameras have a subset of the DOF of
    larger format cameras.
    http://www.clarkvision.com/photoinfo/dof_myth

    2) When your subject is large in the frame, depth of field
    is an issue, particularly when you need the fastest shutter speed
    to freeze the action. Of course if the subject is small
    in the frame it doesn't matter much. Like the tourist who
    takes a picture of a bear in Yellowstone, gets home and sees
    it is a tiny speck in the frame. Here is an example of a
    Yellowstone grizzly bear, full frame image at f/5.6:
    http://www.clarkvision.com/galleries/gallery.bear/web/grizzly.bear.c07.03.2003.IMG_5682.b-600.html
    Note the plane of focus is at the eyes and the back end of the bear
    is slightly out of focus. As the bear moved, occasionally ran,
    the focus continually shifted. If you were manually focusing,
    you would need to adjust focus every fraction of a second.
    Then as the speed goes up it becomes very difficult to manually track
    focus. That's when predictive autofocus comes in to get the shot.

    Roger
     
    Roger N. Clark (change username to rnclark), Aug 1, 2007
    #83
  4. I doubt it too. Note above I said 90% success rate, not 99%.
    Also note your image of the puffin is mainly a pan shot, the
    easiest action shot for a camera as the focus shift is small
    (to zero). As the subject direction turns closer to you, it becomes
    more difficult. I know puffins do fly very fast so you shot
    is difficult due to the speed.
    Again, this is where selecting a single focus point becomes critical,
    and using predictive autofocus. The challenge then is to keep
    the focus point on the subject's eyes while keeping good composition.
    Example: full frame, brown snake eagle, predictive autofocus
    tracking on the eye, headed at the camera:
    http://www.clarkvision.com/galleries/gallery.bird/web/eagle.c01.18.2007.JZ3F7355b-700.html
    Some may say just prefocus on the branch, but there were
    many branches, and I didn't know which branch the bird was
    flying to.

    Here is an example with a cluttered background:
    http://www.clarkvision.com/gallerie...ll.cranes.c12.01.2004.JZ3F7332.arl.d-700.html
    A single autofocus rectangle was kept on the bird's eye (closest
    bird). They were tracked all the way to take-off.
    I agree. My point is understand your tools and their limitations.
    Select the best tool you have for each job. I often travel with
    just a small P&S camera, but when I want to do high end photography
    I carry the bulky DSLR.

    Roger
     
    Roger N. Clark (change username to rnclark), Aug 1, 2007
    #84
  5. Sorry, yes, I misread that!
    Yes, it certainly makes your point!
    Love it!
    Yup! As an amateur, I don't think I'll go back to loads of lenses, but
    the temptation is always there... ;-)

    Thanks for your comments. I hadn't realised how advanced some of the
    camera are.

    Mike

    [The reply-to address is valid for 30 days from this posting]
    --
    Michael J Davis
    <><
    Some newsgroup contributors appear to have confused
    the meaning of "discussion" with "digression".
    <><
     
    Michael J Davis, Aug 1, 2007
    #85
  6. aniramca

    Alfred Molon Guest

    F5.6 is nothing at 700mm (1050mm equiv.) on a DSLR, i.e. will give you a
    very shallow DOF. But F5.6 on a compact camera even at 105mm focal
    length (equivalent) will give you a huge DOF, so even if the bear moves
    back and forth it's not a problem. Obvioulsy with just 105mm you can't
    be too far away from the bear.
     
    Alfred Molon, Aug 1, 2007
    #86
  7. Actually, for this case, the small format camera has
    nearly the same as the depth of field of the larger
    camera over the animal!

    There are several misconceptions here. Depth of field is
    controlled by aperture, and defined circle of confusion in
    the final print, not f/ratio. Second, when considering
    distance to an animal and different distances with different
    focal lengths, the DOF implications are changed over the
    classical view.

    Let's do an example. Consider two points on a bear:
    The bear's eyes and a point 1 meter further from the
    bear's eyes to a point on its back.

    Consider the bear is 20 meters away and the point on the back
    21 meters away from the lens. Compute the circle of confusion (COC)
    if the focus is on the eyes for the point at 21 meters.
    With the 700 mm lens I get a blur of 30 pixels (on a DSLR
    with 7.4 micron pixels). Now change to a 100 mm lens and
    move closer to the bear (7x closer): the bear's eyes are at
    2.857 meters and the point on its back is at 3.857 meters.
    Mow you see the DOF effect: the subject is closer so the near
    to far point is a larger percentage of the range. I compute
    the COC to be 23 pixels, so a slight improvement.

    Now let's scale the camera format down 3x to 2.47 micron pixels.
    With the same total megapixels and field of view, the true
    focal length is now 33.3 mm for the 2.857 and 3.857 meter
    subject distance. The COC of the point on the back is now
    7.7 pixels. Sounds good, but the problem is the smaller
    pixels collect less light. The aperture of this camera is
    only 5.95 mm (33.3mm/5.6). The larger format camera collects
    9x the light per pixel at the same f/ratio, so the larger
    camera can have it's aperture reduced 3x in diameter:
    the 100 mm lens working at f/17 delivers the same
    total photons and same depth of field.

    Now let's go back to the 700 mm lens at 20 meters.
    At f/5.6 the lens diameter is 125 mm. Reducing the aperture
    to f/23 (22.76) gives equal depth of field on the animal at 20 meters
    as the small format camera (2.47 micron pixels) with the
    animal at 2.857 meters. But the aperture of the f/23 larger
    format camera is 30.76 mm, collecting 26.75 times the light.
    The lens magnifies more (7x), and the subject is 7x further
    away, but the pixels are larger, so the light is reduced by
    7*7*7*7/(9*9) = 29.6, so giving total light levels within
    10% of the subject in the small camera at 7x closer range
    (29.6/26.75 = 1.1).

    Of course photographing the bear at 2.857 meters away is
    not smart. The larger format camera does a great job and is
    safer!

    More on depth of field and digital cameras at:
    http://www.clarkvision.com/photoinfo/dof_myth

    Roger
     
    Roger N. Clark (change username to rnclark), Aug 2, 2007
    #87
  8. aniramca

    Paul Furman Guest

    Part of the problem here is the small format camera can get no faster
    than equivalent f/23 wide open. You don't notice that because:

    "Sounds good, but the problem is the smaller
    pixels collect less light."
    For smaller prints, this makes the small format camera look awesome.
    Thanks for your explanations Roger.
     
    Paul Furman, Aug 2, 2007
    #88
  9. My earlier comment on your 'talking sense' about dof, relates to
    something I did 50 years ago as a teenager, and demonstrated that for a
    given coc (in the final print) and a given image size (eg. your bear at
    24" high), the depth of field was only dependent on aperture and
    independent of focal length. (i.e one is close for a wide angle lens and
    distant for a long focus one),

    However, I suspect that my sums all related to 35mm cameras, but I can't
    think that the arithmetic would be affected by neg size.

    As a result of this thinking I applied for a patent because I then
    realised that a rangefinder camera could thus show dof in the finder (a
    given spacing was equivalent to the dof) - I was rapidly disabused as a
    company called Leica had made a similar application 18 months earlier!!
    :-(

    I need to do all the sums again to be sure there wasn't another
    assumption inherent in my approach.

    Mike

    [The reply-to address is valid for 30 days from this posting]
    --
    Michael J Davis
    <><
    Some newsgroup contributors appear to have confused
    the meaning of "discussion" with "digression".
    <><
     
    Michael J Davis, Aug 2, 2007
    #89
  10. You must have a later model of Sony R1 than mine. If I wave my hand in
    front of the lens the delay is very obvious in the display. However, I
    don't think it's seriously longer than the mirror flap delay of an
    SLR, because I find it just as easy (or difficult :) to catch
    pre-focussed action shots as on my old film SLR. In fact all my
    initial problems with action shots on my R1 were due to my believing
    all the rude things said about the action delay problems here and in
    other places. It was only when I realised shutter delay wasn't a
    problem that I started getting the action shots I wanted.
     
    Chris Malcolm, Aug 2, 2007
    #90
  11. aniramca

    Alfred Molon Guest

    You have to compare same with same.

    A "crop 5" P&S has the DOF of a full frame camera at the same F-stop and
    *same equivalent focal length*, i.e. F5.6 on a crop 5 P&S gives you the
    same DOF as a full frame camera at F28.

    A crop 1.5 DLSR instead has at F5.6 the same DOF as a full frame camera
    at F8.4.

    At F8.4 there is way less DOF than at F28.
     
    Alfred Molon, Aug 2, 2007
    #91
  12. aniramca

    Alfred Molon Guest

    You are mixing together two different things, LCD display delay and
    shutter lag. We were only talking about live preview, not about shutter
    lag.

    It's true that the R1 is not a camera for action shots, because it will
    refuse to take the picture when you press the shutter. The engineer at
    Sony who designed this behaviour into the camera should get fired.
     
    Alfred Molon, Aug 2, 2007
    #92
  13. aniramca

    Henry Hank Guest

    So far, the only drawbacks to an SLDC (single-lens digital camera) vs D-SLR
    (both designs are P&S cameras, get it right people) seem to be: focus-delay
    time, some extra noise at higher ISOs which is about the same amount of grain
    that appears in film of the same ISOs, and a deeper DOF.

    Focus-delay time can be zero if you know how to use hyperfocal distances.

    Image noise can be an issue if you are frequently taking all your photos in
    conditions that require the use of higher ISOs. There are many ways to
    compensate for noise in small-sensor cameras.

    Deeper DOF can be and is an asset for many people. Especially for
    macro-photography where no D-SLR in the world can take equivalent photos with as
    much of the subject in focus. I was thrilled when I found out my newer SLDC had
    a deeper DOF than my previous ones, macro photography being one of my main
    interests. In the past I had to use tedious image-stacking methods to achieve
    the same thing that I can now obtain all in one shot. With moving subjects like
    flowers in a breeze or a fast crawling insect it was totally impossible to
    achieve this in the past. No D-SLR in the world can capture equivalent photos.

    That leaves me with only two drawbacks. Slower auto-focusing time, and more
    noise in higher ISOs. Both I can easily deal with by knowing their limitations
    and improving my own photography skills. By properly hand-holding a camera my
    auto-focusing time becomes less than a 1/4th of a second even in very low light
    conditions. That's far faster than anything I can do manually. I guess I can
    scratch focus-delay off the list of drawbacks too. Noise at higher ISOs being
    the only one left. I find that my SLDC camera has almost no noise at ISO-200 (I
    now leave it set on 200 as my default), and tolerable noise at 400. Even at 800
    I can produce very nice images using 3rd-party noise filtering software. Noise
    in small-sensors is not a problem if you know what you are doing. I will admit
    that my latest SLDC was a luck of the draw as far as sensor noise is concerned.
    I happened to get one with some of the lowest noise in this model, or any that
    I've used. This only goes to prove that low-noise small sensors are possible to
    create. It's just a matter of quality-control.


    D-SLR owners just *LOVE* focusing on the negative aspects of life, don't they.
    (an intended pun) So instead lets focus on, and add in, the many ADVANTAGES of
    an SLDC, this is the part in the movie where the D-SLR owners run screaming:

    Quiet operation -- does not scare wildlife, does not alert others so you can do
    candid photography, or in places where the self-appointed camera-cops and
    control-freaks are running around telling everyone what they can or can't do.

    No dust on sensor -- you can take photos in harsher and dustier environments
    without fear of ruining your camera or images. You won't return from your
    vacation only to find out each and every photo you took was ruined by some stray
    dust because you had to change your $2000 lenses to get the needed zoom ranges
    on your camera. Which reminds me, why is it that D-SLR owners never add in
    lens-changing time in their lists of delays? How convenient that they always
    leave out that agonizingly long time-lag in their photography routine when
    telling others how things truly are. Let's add in one more related SLDC
    advantage: no fear of changing lenses in inclement weather conditions. If you
    change a lens in more humid and warmer air then the air temperature inside your
    camera then you risk getting condensation on your D-SLR's sensor, ruining all
    your photos without you realizing it, or making you wait a long time before you
    can bring the camera body up to ambient air temperatures. I don't even think
    twice about taking my SLDC out into an ice-storm or other harsh conditions to
    get those photos that no other photographer will try to get. You should see some
    of my flower and other nature photos taken during pouring rains and ice-storms.
    I'm quite proud of some ultra-rare orchid photos that I took in a remote,
    rugged, and steamy swamp, when dew was dripping off of everything from the
    super-saturated air. I wouldn't dare take a D-SLR into a place like that.
    Investing my time, money, and life-experience into using SLDCs has allowed me to
    obtain photos that no D-SLR owner on earth will *ever* get.

    Sorry, I got distracted. Back to more SLDC advantages ...

    High-speed flash sync -- you can use flash on many SLDC cameras up to their
    highest shutter-speed. For example, your image of someone taking a jump-shot
    won't have blurred halos around them due to the ambient lights and the long time
    it takes for the D-SLR's focal-plane shutter to traverse the width of your
    sensor.

    No distortion of fast subjects -- focal-plane shutters in D-SLRS will distort
    the propositions and shapes of anything that moves during the use of high
    shutter speeds. It's just the nature of the focal-plane beast.

    100% accurate viewfinder -- Slow shutter motion-blur effects and DOF aperture
    effects are correctly relayed to the EVF. A true WYSIWYG preview. EVF/LCD
    equipped cameras are the only ones that give you a totally accurate presentation
    of what your final image will look like before you press the shutter.

    Live histograms and other important information available -- relayed to you at
    all times during your photography. In some of the cameras they even include DOF
    calculations, your zoom's focal-length, and much much more.

    Longer life-time -- the slow and noisy mechanical shutters and mirrors in D-SLRs
    have a significantly shorter life-span than SLDCs that don't depend on these
    ancient contraptions.

    Video & sound recording capabilities -- some SLDCs rivaling and beating
    video-only equipment.

    True *continuous* high-speed frame rates -- some SLDCs allow you to take up to
    2.5 frames per second up to the storage capacity of your memory card, resulting
    in many thousands of photos if you so need. You are not limited to the memory
    size in an internal memory-buffer before your high-speed slows down to a crawl,
    as what happens in *all* D-SLRs.

    Convenience -- weight and volume, plus single-lens vs. three or more bulky
    lenses costing many thousands of $. For equivalent focal-length range and
    capabilities you can fit an SLDC in a large pocket compared to requiring a
    sturdy backpack for the D-SLR system.

    Cost -- a few hundred compared to a few thousand. Generally an SLDC will be at
    least 1/10th the cost of the equivalent capability in a D-SLR, with many added
    features that a D-SLR can't have and will never have (see above).

    I could keep going with a hundred more thing things like some SLDCs that have a
    viewfinder that you can see from virtually any direction and long working
    distances, or even composing yourself in the image for documentaries, etc., but
    I think you get the drift. The plusses of owning an SLDC ****FAR**** ****FAR****
    ****FAR**** ****FAR**** ****FAR**** outweigh any drawbacks compared to owning
    and paying for a D-SLR.
     
    Henry Hank, Aug 2, 2007
    #93
  14. aniramca

    dj_nme Guest

    This would make sense if you'd written "...has the same DoF of a full
    frame camera at the same aperture size..."
    So for a "crop 5" digicam, it uses a 10mm lens compared to a full-frame
    camera with a 50mm as a "normal lens"***.
    For an aperture set at f1:2.8 on a "crop 5" digicam (aperture size of
    3.57mm) equals the DoF of a full-frame camera set at f1:14.

    If you've only got a table of DoF for 35mm full frame, then the DoF for
    a different film or sensor size can be calculated by by multiplying the
    aperture (f stop) of the full frame by the crop factor and looking at
    the entry for the equivalent focal length at the resultant aperture setting.
    EG: for the "crop 5" digicam, look at the entry which matches the "35mm
    equivalent" of the lens you're using at 5 times the aperture setting of
    your digicam.



    *** (to be technically correct, a "normal lens" is of a focal length
    equal to the diagonal of the sensor/film frame. For a "crop 5" digicam
    this should be 8.6mm and for a 135 full-frame camera, 43mm)
     
    dj_nme, Aug 3, 2007
    #94
  15. aniramca

    dj_nme Guest

    I imagine the only way for you to take a picture exactly (or as close to
    as is humanly possible) to when the shutter button is pressed is to use
    manual focus, otherwise the camera will try and get an AF lock first.
    This is assuming several things: the camera isn't processing a previous
    shot, the memory card isn't full (some won't trip the shutter if it's
    full) and that the delay between what's happening in front of the lens
    and display on the LCD and/or EVF doesn't make synching the two events
    (action and shutter button pressing) unlikely.
     
    dj_nme, Aug 3, 2007
    #95
  16. aniramca

    Alfred Molon Guest

    Compare same with same. You have to use equivalent focal lengths, or if
    you prefer the angle of view. The 10mm crop 5 lens has the same angle of
    view of a 50mm crop 1 (full frame) camera, therefore it's a 50mm equiv.
    focal length.
     
    Alfred Molon, Aug 3, 2007
    #96
  17. aniramca

    dj_nme Guest

    Congratulations: you have just agreed with what I wrote.
    That is exactly why I chose 10mm for the "crop 5" and 50mm for the 135
    full frame.
     
    dj_nme, Aug 3, 2007
    #97
  18. aniramca

    Paul Furman Guest

    I'm still unclear about this. If f/stop is not comparable at equivalent
    focal length, I think ISO and dynamic range must be the bridge
    accounting for any difference.
    Correct me if I'm wrong but the difference in DOF can be compensated by
    increasing the ISO on a dSLR. SLDC's (P&S) typically have a good deal of
    noise reduction built into the jpeg creation as well, I believe. Another
    factor is dynamic range.

    I think the slower AF time is largely due to less light to work with.
    Yes, it's possible to take great photos with all sorts of limitations.
    The limitations of a DSLR included.

    No, I'm aware of the tradeoffs. If I moved to large format, I'd only be
    able to take photos with a tripod. You could underexpose the large
    format image & push the development at high ISO but that wouldn't make
    any sense... if that's what you want, just use a smaller camera. I
    noticed the differences when moving from small digital to dSLR: it takes
    some getting used to and there are definately trade-offs.

    You can get an 18-200 lens & never change it if you want, just as you
    could set ISO to 800. But yeah, I never heard of a 24-400 lens.

    Yes, though some DSLRs have hybrid electronic shutters. I don't
    understand your blurred halo example.

    That's a pretty obscure one... I don't know.
    Yes, I agree here, I miss that. Big bright optical viewfinders have
    advantages too though. Of course you can get a pro model with a 100%
    viewfinder.

    Oh yes, certainly.

    It's a matter of priorities for each individual.
     
    Paul Furman, Aug 4, 2007
    #98
    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.