• Oct. 30, 2023, 6:22 p.m.

    In Body Image Stabilisation has matured over the last years.
    Starting with MFT now every Brand seems to have a solid level.

    But there are some differences in implementation features:

    • Image Stabilization linked to AF point
    • combined OIS / IBIS
    • Pixel shift
    • Hand held Pixel shift
    • Software ND Filter
    • Star Tracking
    • On Screen help for long time exposure

    What do you think will be most useful if any?

  • Members 1182 posts
    Oct. 30, 2023, 7:36 p.m.

    The most useful to me personally is 'combined OIS / IBIS' as in my Lumix DC-G9 + a 'Power OIS' lens.

    With my shaky hands, the more the better ...

  • Members 1070 posts
    Oct. 30, 2023, 8:45 p.m.

    This is an interesting question; but could you briefly characterise and explain these different systems for those of us who do not know about them?



  • Members 1756 posts
    Oct. 31, 2023, 7:51 a.m.

    it matured 7 years ago with the em5mk2 5 stops and thats it, anyone that claims 8 is dreaming unless you use a tripod and then you have 20 stops 😁 this will be the claim next year. my sony a7r2 was claimed at 5.5 and it would be lucky at 2 stops. my em52 is the industry standard measurement 😎

  • Members 796 posts
    Oct. 31, 2023, 5:21 p.m.

    Combined OIS and IBIS seems to have and edge when I use my Z7 + 24-200 , but I have done no testing.

    As to whether a camera has 3, 5 or even 8 stops, depends very much on the steadiness of one's hand holding technique. I do not know how they arrive at these numbers, or what they are measuring against. Hand holding is also dependent on focal length. I guess they measure against the old adage of the shutter speed matching focal length.

    I find I can get infinitely long sharp exposures with my three footed monopod technique, using IBIS, whilst my non IBIS D850, suffers from the micro vibrations that are present when I use this method.

    DSC_6217 3.jpg

    IBIS and OIS are probably dependent on the magnitude of the vibrations rather than time, are they not?

    DSC_6217 3.jpg

    JPG, 541.8 KB, uploaded by NCV on Oct. 31, 2023.

  • Members 1182 posts
    Oct. 31, 2023, 5:51 p.m.
  • Oct. 31, 2023, 6:34 p.m.

    Here is some Background:

    Modern Ibis uses 5 Axes stabilisation, but the CIPA Test is still from 2012 and not updated to this.
    So the claimed values of the manufactures could not be reached in real World.
    So the values are not really interchangeable between brand or even models if the Ibis mechanism was a new one.

    Long Time MFT was king of the Hill but with the last iterations, FF and even Medium Format caught up.

    I think the difference today comes with the extremes:
    How many secs can you take with wide angel lens
    How long can you go with 300mm or even more.

    So lets take a look a the points I mention:

    Nikon fc is the first one having Image Stabilization linked to AF point
    I could not believe this brings up much but have not seen test results. Also I doubt it is a help for very long exposure or long tele shots.

    I think every brand has combined OIS / IBIS in their latest models. This should help on tele shots because OIS can handle them better.

    Nikon fc is also the first Nikon having Pixel shift, Sony, Panasonic, Olympus and Fuji where early on this.
    The Ibis moves the Sensor a bit between multiple shots so you can get a higher resolution image or an image with more accurate colours.
    This needs a Tripod.

    Panasonic and Olympus got Hand held Pixel shift witch is not so accurate as the "normal" Pixel shift but do not need a Tripod to work.

    Pentax has a Star Tracking Mode witch can be used prevent star tails via Ibis.

    The Olympus Om-1 as many handys has a On Screen help for long time exposure this should help to pin the camera on the same object on multi second shots. This might help to bring the max secs for wideangel shots even further.

    Ibis might help on the following two Items:

    Olympus has a Software ND Filter to combine multiple shots into one with lower EV.
    Most Brand today have some kind of Focus Stacking combining multiple shots to one with a bigger Area in focus. This is mostly used for macro shots.

  • Members 1182 posts
    Oct. 31, 2023, 7:09 p.m.

    The official name of the standard is CIPA Standard DC-011-2015 Measurement and Description Method for Image Stabilization Performance of Digital Cameras (Optical System)


    Looking at that later standard I see that it is only for optical systems, not cameras ... hence the missing axes ... and hence it is only a partial response to @43review.

    Is there even a Standard for cameras i.e. sensor-based IBIS, I wonder ...

  • Members 272 posts
    Oct. 31, 2023, 9:45 p.m.

    That is a good question. The history of image stabilization dates back to long to how to mitigate vibration in airborne imagery platforms for military purposes. For the most part, the issue is to mitigate slight changes in the Euler coordinates, row, pitch and yaw to keep the camera axis stable for the exposure time. It may have been developed in the 1980's by Canon for consumer cameras but it was being implemented at the end of WWII for military applications. Then as reconnaissance went to space, the requirements moved to space and became more important. Of course the longer the focal length of the lens the more critical the issue.

    Today we have the traditional IS which corrects for in the Euler coordinates roll, pitch and yaw which in consumer camera normally takes place in the lens. With IBIS one can add the dx and dy axis to address movement in a plane. Lens IS works better for long lenses more or less because the sensor can't move far enough to compensate for angular movement of a long lens. For short to normal focal length lenses - IBIS seems to perform better probably because to roll, pitch and yaw corrections are so small they are corrupted by noise.

    It really should not matter and why does there need to be a standard for IBIS only. Vibration is vibration. The spec simply specifies the vibration profile and the IS subsystem of the camera is treated as a black box. Slap it on the vibration table and let the chips fall where they may. Does it work and how well. I don't for a nanosecond believe that a person can hand hold a lens 2X the length of the sensor diagonal for a two second exposure and get a rock solid clean shot 95% of the time. The standard is what it is. It is not absolute but a relative measure. If Nikon can reach 5.5, OM Systems 7, etc. then a relative comparison can be made. However, no matter what the marketing number says - don't count on IS saving your bacon with sloppy technique.

  • Members 1756 posts
    Nov. 1, 2023, 6:07 a.m.

    they havnt caught up at all. i have the latest a sony a7iv and the only image stabilization thats caught up is in video mode with active stabilization. lets be clear about ibis when comparing to MFT i can deliberately shake my mft cameras and clearly see how awesome ibis is. my a7iv when i shake the camera i see nothing change. when i hold the a7iv rigid and ever so slightly move the camera the effect is so tiny its virtually imposable to see but you can see ibis working.

  • Members 92 posts
    Nov. 20, 2023, 7:40 p.m.

    An interesting question is whether the method of rapid skewing of lens elements sacrifices optical quality more than faults introduced by to moving gthe sensor. One obvious effect of in-lens-corrections is that both the moving parts that have to be lightweight and the rest that needs to allow for one element danoing does sacrifice some potential quality- Not to speak of the cost of providing each lens with the necessary equipment rather than sticking to paying for sensor contortions