How is anyon using DSLRs for video?

Discussion in 'Amateur Video Production' started by Tony, Dec 29, 2013.

  1. Tony

    Tony Guest

    I brought my Canon T3i to my son's soccer game back in Oct and I
    recorded the game. What a friggin disaster. I had a 140 -250 zoom lens
    and was excited to get those great action shots you see of sporting
    events with the dept of field blur behind the subjects. Well, I
    focused in and if they move a foot or two closer or further, they are
    out of focus. How the hell does anyone shoot in-focus video when
    shooting manual focus? The focal distance changes every second, plus
    you have to frame the subject, who happens to be running 15 mph in
    either direction. Anything I am missing?
     
    Tony, Dec 29, 2013
    #1
    1. Advertisements

  2. *GOOD* autofocus...;-). For this, generally mirrorless cameras
    work better due to the way AF is implemented in those, although
    Canon (and likely other dSLR makers) are beginning to put AF
    pixels on the sensor - but Panasonics still do video the best
    (their G5, G6, GX7, GH3 have the highest quality video around
    at present for "still-cameras-that-can-also-shoot-video"). With
    some cameras now, you can even touch the rear screen or maneuver
    a spot in the frame (even up to the edge or corner) in the
    eyelevel finder to place the focus spot manually and accurately.
    --DR
     
    David Ruether, Dec 29, 2013
    #2
    1. Advertisements

  3. Tony

    Brian Guest

    There may be an option on that camera called Servo AF which will
    continuously focus on the subject.
    You could try zooming back from the moving subject when recording then crop
    the video in a video editor.

    This site will give you recommended setting for photographing a Soccer
    match.

    http://www.exposureguide.com/soccer-photography-tips.htm
     
    Brian, Dec 30, 2013
    #3
  4. Tony

    Brian Guest

    The Canon EOS 70D does all of the above.
    Camera manufactures are seeing a need to improve on the way the camera
    handles video recording now that people are using dSLR to record video.
     
    Brian, Dec 30, 2013
    #4
  5. This can help, but some cameras are better than others for not
    "hunting" as much when focus is momentarily lost...
    This would not work - the DOF would be the same with the
    original (zoomed in) as with a wider view cropped to the
    same framing as the first - and the overall sharpness
    would be reduced (everything else being equal).
    BTW, long lenses (narrow shooting-angle) are much easier
    to buy, carry, and handle for smaller-sensor formats (my
    MFT 300mm [600mm-equivalent in FF-35mm] is tiny and very
    light compared with 600mm lenses made for 35mm...).
    --DR
     
    David Ruether, Dec 30, 2013
    #5
  6. I forgot to add that with MFT, there is also a DOF advantage
    at a given aperture and angle of view compared with 35mm or
    crop-35mm formats, aiding focus...;-)
    --DR
     
    David Ruether, Dec 30, 2013
    #6
  7. Tony

    Brian Guest

    In addition to what I have posted. What you may be missing is experience in
    using a dSLR camera as a video camera. It takes practice to get the shots
    you want.
     
    Brian, Dec 30, 2013
    #7
  8. Tony

    Brian Guest

    A video camera does not have this problem as its designed for the purpose
    of recording moving objects but I understand that it can be a problem with
    some dSLR cameras. The Canon 70D has a special focusing system so the lens
    don't need to hunt for the correct focus.

    I was thinking more about his comment on trying to track a moving object.
    It reminds me of trying to video record a bird flying in the air that has
    been zoomed into.
    While its true that you loose some sharpness, if the video has been
    recorded in Full HD then cropping the video a bit does not have any
    noticeable effect on the video quality.
    Now I could be wrong but recently I was listening to a processional
    photographer who said that you can crop a video taken with a dSLR camera
    with less loss in video than a video camera due to the larger sensor.
     
    Brian, Dec 30, 2013
    #8
  9. The mirrorless cameras are better at this generally than dSLR
    cameras (with mirrors), and mirrorless cameras with stills,
    focus is generally quick and accurate (and without any need to
    calibrate the individual lens AF). Some dSLRs are beginning to
    include better methods of focusing to get around some of their
    inherent focusing problems, but...;-) Also, video cameras and
    mirrorless cameras (at least the ones I've seen, can still
    visibly "hunt" a bit if they lose focus while videoing, but it
    isn't as bad, or as often, as with most dSLRs, but it is still
    there...
    UM!!! I have SERIOUS doubts about this! ;-)
    "Less" rather than "none" is the relevant part - but it also
    depends on the inherent image quality of the cameras themselves
    as much as on the formats. With the Panasonic TM700 camcorder,
    you can crop its image *slightly* while digitally stabilizing
    its video image in post (with adding some careful sharpening
    afterward) with little (if any, unless compared frame-by-frame
    side-by-side) noticeable difference in sharpness, even with its
    small sensors (3), but the video images of most dSLRs are fairly
    soft to begin with in video-mode (and with noticeable aliasing
    also sometimes visible) without cropping, and those faults get
    worse with cropping (the "near" exceptions are with the Panasonic
    G5, G6, and GX7 with video, but even the GH3 has some aliasing
    problems with video...).
     
    David Ruether, Dec 30, 2013
    #9
  10. Tony

    Brian Guest

    Up to 3x digital zoom seems OK.

    Olympus cameras were known for good lens quality. My first camera was an
    Olympus pocket sized camera. It only took 1.3 megabyte photos but they were
    very sharp thanks to the lens.

    Out of interest which camera do you use the most or do you pick the camera
    your going to use by the type of subject your going to shoot?
     
    Brian, Dec 31, 2013
    #10
  11. Um, we may be looking at different things, but in my experience,
    3X enlargement of video (at least in post) would result in VERY
    poor-quality video; if you are referring to 3X magnification
    in-camera with a camera that has a high-resolution sensor, and
    if the camera is digitally redistributing the number of pixels
    used from the sensor downward from all of the originally-used
    number toward the video 1:1 resolution of 1920x1080 in order to
    increase the video image magnification, this *might* work very
    well. And, I think this is correct, since I just did a little
    figuring...;-) On Panasonic still cameras, there is in video a
    2.4X "extra tele" mode. The sensor pixel-width of G5 and G6
    cameras is 4608 pixels. Dividing that by 1920 pixels gives,
    Ta-DAH!, 1920 pixels! Hey, I must try this with video! With the
    Panasonic 45-175mm zoom, this would give (with *video*...), an
    effective 420mm lens (840mm-equivalent in FF-35mm terms!)! And,
    while we're at it, my Panasonic 100-300mm zoom would offer the
    35mm-equivalent of 1,440mm! Hmmm, I will try this to see what
    it looks like, and whether or not this isuseful to do...;-)
    Thanks!
    Olympus does make very good lenses, and often unusually
    fast and wide ones. BUT!!! That does not tell the whole
    story (this issue came up recently while talking with a
    local videographer who was thinking about MFT lenses for
    a possible Black Magic pocket video camera, which is one
    of three makers that offer MFT-mount cameras with compatible
    electrical contacts in the mount, with Panasonic and Olympus
    being the other two). And, here comes the "but": all three
    body-makers use the contacts in different ways, and Pan. and
    Oly. have different ideas about what should be done where,
    which makes lens choice NOT obvious! Panasonic designs lenses
    with some corrections NOT included in the lens itself (like
    for linear distortion, illumination-roll-off, and chromatic
    aberrations - but they put the stabilization, if any, in their
    lenses), and this often results in cheaper and smaller lenses,
    with those corrections being made in the camera body. Olympus
    makes their lenses with the corrections made in the lenses
    themselves, not in the bodies, but their lenses have no
    included stabilizers (that function is handled by the camera
    body), and this sometimes results in their lenses being
    considerably more expensive, and sometimes MUCH larger and
    heavier. Now come the problems with mixing brands and bodies:
    Olympus lenses on Panasonic, Olympus, and Black Magic bodies
    very often show (for me) too much residual red/green CA to
    be acceptable for use with shooting JPGs and video - but
    have stabilization available only on Olympus bodies; Panasonic
    lenses on Olympus and Black Magic bodies generally show too
    much linear distortion and CA, and all do have stabilization
    on Oly. bodies - but "all is (mostly...;-) well" when Pan.
    lenses are used on Pan. bodies (including having generally
    very low CA showing in images shot with JPGs or video).
    As for "3rd-party" MFT lenses, they have no built-in
    stabilization and must correct (best they can) the other
    lens problems within the lens itself, but do have AF - and
    do have stabilization when used on Oly. bodies. More: for
    video shooters, video quality and type varies considerably
    between the camera makers, with Olympus video being the
    lowest-quality, Panasonic offering the sharpest video, and
    Black Magic offering more pro-oriented "super-16" format
    and the ability to shoot in Pro-res 422 and DNG formats.
    Simple, huh? (Ain't marketing gran'...?;-) So, I stick with
    Panasonic for bodies and for most of my lenses, but there
    are some exceptionally good 3rd-party lenses available for
    MFT - and, BTW, I'm now working on a set of lens reviews
    of the many lenses I've tried for MFT on Panasonic bodies.
    Most of the writing is done, but I'm going to try to include
    photo samples...
    NOT "subject"!;-) The "deciders" for me are where I'm going,
    how light/heavy/small/big the all-smallish bags I will carry
    are going to be, and (most important!) "whim"...;-) I have a
    dozen or so lenses to choose from, three basic camera types
    (all Panasonic) and six different camera bags to hold different
    combinations in (all "smallish" - I don't like carrying much
    weight or bulk while "having fun"!).
    --DR
     
    David Ruether, Dec 31, 2013
    #11
  12. Tony

    Brian Guest

    That was what I was meaning. I'd be interested in the result.
    Have you considered using charts designed to test lens.
    I've had no luck in finding a lens test chart on the net that I can print
    and use to test lens but there should be something on the net. I have used
    small text when testing camera lens in the past. A problem with some lens
    is that they are sharp in the middle but are less sharp near the edges of
    the image.

    I feel the same when it comes to weight. Some powerful telephoto lens can
    be big and very heavy. Also a bigger lens means buying a bigger camera bag.
     
    Brian, Dec 31, 2013
    #12
  13. I do not use test charts since using them has many associated
    problems: lens performance can vary radically with focus distance,
    and the test-chart distances often fall within the distance setting
    where the lens may be nearly at its worst (I prefer using highly
    detailed and familiar subjects near infinity-focus); lenses are
    rarely truly flat-field, especially close-in, and using a flat
    target may then be inappropriate; EXACT focus is ESSENTIAL with a
    close-in test chart (or targets), and the results of inconsistent
    focus with using a 3D close-in target are easily seen in dpreview's
    results...; EXACT parallelism with a close-in target is also
    ESSENTIAL for meaningful results, and this is very difficult to
    achieve (and very easy with infinity-focus targets). If you want
    inconsistent results, try using test charts...;-)
    Ah, so you've noticed...!;-) So many people see a sharp center
    and call the lens sharp regardless of what the edges and corners
    look like. This IS NOT me! 8^) Some VERY famous lens-producers
    commonly make lenses that, while very sharp over most of the
    frame, are soft in the corners at ALL stops; these I do NOT
    consider to be adequately sharp lenses for my purposes! Acceptable
    (with digital) are lenses that can be sharpened toward the edges
    and corners to look evenly sharp everywhere in the frame, IF no
    other practical option is available.
    And lugging the &%$#@ thing around, and trying to unpack it at
    least somewhat gracefully - and then there is the extra gear to
    hold the "enormous-mess" up with...;-) I do this only with my
    500mm mirror on 2X converter (for 2,000mm-equivalent on MFT), but
    then I generally carry the works only a short way from my front
    door; the 100-300mm (out to 1,200mm-equivalent on MFT) has never
    yet been on a tripod, but I still have sharp photos shot with it
    (aren't good stabilizers gran' on small MFT gear...?;-).
    --DR
     
    David Ruether, Dec 31, 2013
    #13
  14. OK, HERE ARE THE RESULTS:
    I shot two clips - the first with a good zoom lens set at 14mm,
    plus 2.4X magnification (=34mm, done in-camera by cropping the
    sensor's 4608x3456 pixels down to 1920x1080, the resolution of
    HDTV); the second with the zoom lens set at close to 34mm, plus
    0X (=34mm, as video would normally be shot, with the camera
    down-sizing the whole image in real-time [60fps] to 1920x1080).
    Surprisingly, both looked about equally good(!) both in motion,
    and with the video viewed frame-by-frame, but in motion the
    second one showed very subtle aliasing on one small part of the
    image (an edge view of a single CD case with white edges and
    black label, on three shelves of CDs + more in image height).
    The second also showed subtly more edge compression-effects.
    I made frame-grab stills of three subjects common to both clips,
    to compare these with stills, and I will send two pairs of these
    to Brian (uncorrected for color, unsharpened, etc.). Overall,
    the photographer was right, so long as the above is done in
    the camera, and not during editing - and for some good, logical
    reasons, the crop looks *very* slightly better (with differences
    that would most likely be concealed with a camera less sharp with
    video than the Panasonic G6 I used is...).
    --DR
     
    David Ruether, Jan 1, 2014
    #14
  15. Tony

    Brian Guest

    So far I've used a bush in my garden to test for sharpness and correct
    colours. There are very small leaves on the bush which should be sharp on
    good quality lens. I was told recently that I should set the same aperture,
    shutter speed and ISO for each lens and camera that I'm testing.
    I noticed this when viewing so e test photos for different camera lens when
    downloading the full photo and enlarging it 100%. I would imagine any lens
    that does not have this problem would be expensive to buy and is likely to
    be a professional lens.
    I remember so years back that people who owned a certain camera said it was
    too heavy so the next version of the camera was lighter but people then
    said it was not heavy enough to keep the camera still when taken hand held
    shots.
    Some of the Sony dSLR camera bodies are getting to be very light. But in
    some causes manufactors of cameras seem to have added weight to the body of
    the camera such as the Nikon 7100 compared to the lighter Nikon 7000.
    If MFT are light cameras then that is one good thing going for them.
     
    Brian, Jan 1, 2014
    #15
  16. :

    [VASTLY abridged...]
    There are good and cheap lenses around. The best I've seen for
    even/great performance are the Zeiss $4,000 Otus 55mm f1.4 (for
    FF-35mm), the $250(!) Sigma 60mm f2.8 (for MFT and NEX), the
    $250 Rokinon 7.5mm for MFT, and a few others. Surprising to me
    is how poor some "famous-brand" lenses can be at *ALL* price
    levels(!). Weird! But it goes to show that salesmanship is
    often worth more to a manufacturer than inherent product
    quality is...;-)
    --DR
     
    David Ruether, Jan 1, 2014
    #16
  17. Tony

    Brian Guest

    I'm told that its to do with brand names. You could have two identical lens
    and if one of them has the manufacture of a well know camera brand such as
    Canon then it would sell at a higher cost. Some lens are at a ridiculous
    price but if the buyer is a professional then he can get his money back
    with his photography job. The same thing seems to happen with camera
    batteries so Canon had to offer a few extra features that another
    manufacture could not offer such as registering the battery in the camera
    so you can keep track of power levels of extra batteries. If the battery is
    very cheap then it might not give good performance
    A good photo is not always due to lens quality but depends more on the
    person behind the lens..
     
    Brian, Jan 1, 2014
    #17
  18. Um, see the second half of my comment just above your last one...;-)
    To get specific, a friend bought a Leica in the old film days, plus
    several lenses, incliding a Leitz 21mm. It was expensive, and TERRIBLE!
    Lenses vary by sample, and a second one was much better. Another friend
    (also during the film era) bought into the Canon system for pro work,
    and i don't think he ever could find a decent sample of the Canon 20mm.
    Another friend got a "deal" on a Canon 5D Mk. II plus 24mm f1.4 lens,
    and hascomplained about how poor it is ever since. He sent me a sample
    photo, and it wasn't very good - but I suspect that it was a poor sample
    of a lens that can be very good. He also bought a Canon 50mm f2.5 macro
    lens to get a sharp lens, *BUT EVEN AT F8*, a lens that should have been
    sharp to the corners wide-open was still uncorrectably soft in the
    corners! U-N-A-C-C-P-T-A-B-L-E ! So, I suggested a 50mm f1.8, figuring
    that surely by f4, that lens would be good everywhere in the frame,
    but no - it was also soft in the corners at f8(!!!). BLEAH! Not that
    Nikon escapes this completely, either. See my "SLE[MN])" at --
    http://www.donferrario.com/ruether/slemn.html and check the "RATINGS"
    column for variation in image sharpness numbers (and number tried).
    Some lenses varied widely, and a few lenses were jut plain not very
    good.
    With Panasonic lenses, I've returned some for exchanges after checking
    them also. -->*LENSES VARY NOT ONLY BY DESIGN QUALITY, BUT BY SAMPLE
    VARIATION*<-- And, this is regardless of price! ALWAYS check out
    lenses at purchase for performance while you can still return them
    for exchange or refund - and buy from shops that permit you to do this!
    (That is, if you care at all about the quality of your gear - brand
    and price guarantee NOTHING.)

    As for the second part, yeah, yeah, BUT! Why be limited in the range
    of what YOU can do by using inferior-quality lenses, and by being
    forced by the limitations of your gear when that is not necessary?!
    Some cheap lenses can be top-class. My $250 Rokinon fisheye is better
    than Panasonic's $600 fisheye (but it took six samples to get a
    good one...). The $250 Sigma 60mm f2.8 beats "by a mile" all other
    lenses I've seen for MFT at ANY price, and with attachments, it's
    also an excellent macro lens. Anyway, time to stop ranting.....;-)
    --DR
     
    David Ruether, Jan 1, 2014
    #18
  19. While playing around with this again this morning, I discovered
    something new (and quite useful for video). I again tried various
    ways to set up items and modes to shoot video while zooming with
    the on-body (Panasonic G5 and G6) zoom lever with my two "power
    zoom" lenses that work with it. I had been complaining about
    exposure "bobbles" while zooming, but I may have found a solution
    that (almost) works well (well, much better than before, anyway;-).
    As one of the many changes I tried, I switched from center-patch
    focusing to 23-patch auto-selection, with "continuous-focus" turned
    on in the video menus, and, Ta-DAH...!, almost completely smooth
    exposures resulted while zooming (although it still worked better
    zooming wide-to-narrow-angle, rather than the reverse, but even
    that was much better). I keep learning new things about these nifty
    cameras...;-)
    --DR
     
    David Ruether, Jan 1, 2014
    #19
  20. Tony

    Brian Guest

    I look at a great photo in a camera magazine which has details on what
    camera used, the lens used and the exposure used and think to myself if I
    had the same camera and lens then maybe I can get a photo similar to the
    one printed in the magazine. But what we don't get to see is all the bad
    photos the photographer took before getting a great photo of a subject such
    as landscape. So is it the photographers skill and could he had taken a
    great photo with a lower quality camera and lens or is it then lens and
    camera quality that makes it possible? Some soft focus photos look great
    using the soft focus effect.

    In some cases its the photographers fault and not the camera or the lens if
    the photographer gets bad photos. Some like to fiddle around with the
    camera and alter settings that is better left to experienced photographer
    to change. For example if you set the camera to a very high ISO then the
    quality of the photo is going to suffer. But I agree with you that some
    branded lens should give better quality but not always.

    One thing I don't like is testing the camera or lens in a shop. There is
    nothing in the shop to test it on unless you focus on a row of cameras for
    sale and you can't test it on distance subjects. I feel camera shops should
    have at least a vase of flowers to test cameras/lens with. The last time I
    tested a camera the only thing I could do is to focus on the shop across
    the road by aiming at the entrance of the photography shop. Its good to
    view test photos on the internet for that reason.
     
    Brian, Jan 2, 2014
    #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.