• Members 4 posts
    April 15, 2023, 10:25 p.m.

    I'm a little confused about image stabilization - when it will be of help and when it won't.

    As best as I understand, image stabilization is a process in which either a sensor and / or lens elements will move in order to attempt to cancel out camera movement that would otherwise blur an image.

    I can see this as certainly being of value in reducing camera shake (either slow shutter speed, or shooting with a long telephoto lens), and perhaps driving along a bumpy road.

    My issue is that I often take photos while in movement - not on the run, but perhaps on the walk - which results in camera movement from footsteps, body movement or perhaps unsecured camera movement. I'm mostly on the move, rarely stopping to compose or take a shot. To me, these seem like fairly gross movements that are unlikely to be negated or substantially minimized by sensor or lens element movements.

    Is image stabilization a function that will do much good in those kinds of situations, or are they problems that are too much for it to handle effectively?

    Thanks in advance.

  • Members 1737 posts
    April 15, 2023, 10:36 p.m.

    Yours is certainly not the use case for which IS systems are designed. CIPA has specified a series of camera movements for IS testing, and most cameras are optimized to that test procedure (see lies, damn lies, and benchmarks).

    www.cipa.jp/image-stabilization/index_e.html

    www.cipa.jp/image-stabilization/documents_e/VMVA_for_DC-X011_E.pdf

    But why don't you do some testing and see how well it works for you?

  • Members 142 posts
    April 15, 2023, 11:08 p.m.

    If your camera takes video, compare results while walking etc with it on and off. For me (Panasonic FZ1000) the difference is night and day.

    Sherm

    www.flickr.com/photos/32989985@N07/sets/
    P950 album: www.flickr.com/photos/32989985@N07/albums/72157713228488992
    P900 album: www.flickr.com/photos/32989985@N07/albums/72157671637486978
    RX10iv album: www.flickr.com/photos/32989985@N07/albums/72157719453957168

  • Members 4 posts
    April 15, 2023, 11:54 p.m.

    Thank you for confirming my suspicions.
    Testing would be the most reasonable way to proceed except that it would involve purchasing two cameras and then returning one or the other - neither inexpensive, nor fair to the dealer(s). Sadly, none of my existing cameras is in sufficient shape to perform a reliable test (hence my hurried search for a new one).
    Right now, I'm leaning Sony a6400, but it does not have image stabilization, and I don't think that the kit lens does either.

    Artie

  • Members 1737 posts
    April 15, 2023, 11:56 p.m.

    Rent.

  • Members 4 posts
    April 16, 2023, 12:01 a.m.

    Interesting. I shoot very little video, other than by accident.
    This would obviously require a camera with IS, while my count of useful cameras is down to 0.
    Are the impacts of IS on video comparable to those on stills?

    Thanks.

    Artie

  • Members 142 posts
    April 16, 2023, 12:19 a.m.

    I'd think the mechanism within the camera would be comparable at comparable shutter speeds, but haven't directly tested that.
    Sherm

  • Members 360 posts
    April 16, 2023, 6:37 a.m.

    IS certainly helps, but walking results in such wide range of motions, motion speeds and accelerations, that it will obviously not solve smearing on its own. It's too much for this system. You need more tools. Best walking/holding/handling technique, maybe gimball stabilisation, wide aperture lens and good sensor for fast shutter speeds if taking photographs.
    That's how you move towards better results.

  • Members 4254 posts
    April 16, 2023, 6:55 a.m.

    IS is designed to help reduce/eliminate what is commonly referred to as 'camera shake'.

    When the camera is on a tripod it is usually recommended to turn off IS since the camera should not move during the shutter actuation.

    Having IS on while the camera is on a tripod will mean the camera will go looking for camera shake that doesn’t exist and that could cause some blur in the image.

  • Members 17 posts
    April 16, 2023, 7:05 a.m.

    It is just magic, isn't it? One of the few really excellent developments over the past few years. It seems to "just work" for me. I don't do video (much) but I've seen some really startlingly impressive stuff other people have taken and my hand held shots just got a bit crisper. Good article here.

  • Members 84 posts
    April 16, 2023, 8:51 a.m.
  • April 16, 2023, 9:06 a.m.

    Yes - just use the link button and put in the YouTube URL

    A Youtube video

  • Members 1737 posts
    April 16, 2023, 3:18 p.m.

    My biggest problem with IS when the camera is on the tripod is that the IS system can recenter the sensor during the exposure.

  • Members 140 posts
    April 18, 2023, 11:09 a.m.

    Well… you can turn the stabilization off, so testing is easier by just turning stabilization on and off.

  • Members 56 posts
    April 26, 2023, 1:36 a.m.

    When we look at the mechanism of how IBIS will do (of course IBIS of every brand might be different, but at least true for certain brand at least), your question should be addressed easily.

    The following was an youtube link about an interview with Olympus concerning their IBIS system. It uses a few gyro sensors located around the sensor housing unit to detect motion and would apply the appropriate stabilization. On a stable surface, e.g. tripod, if no movement will be detected surely no stabilization action would be applied. The Panasonic 5-axis IBIS system is also using similar design and hence my cameras will never add unnecessary stabilization which will eventually blur an image. It is actually not how smart the system, just the design can avoid unwanted stabilization only. Under a modern date stabilization system, it is very precise and accurate.

    Please fast forward to 3:29 for the general explanation, and from 4:30 showed the IBIS component.
    www.youtube.com/watch?v=zrLGZSQ0ulU&t=1s

  • Members 1737 posts
    April 26, 2023, 1:39 a.m.

    I know of no cameras that use gyros for accelerometers.