Image Stabilization – Your Camera Can Do What You Can’t

On your camera lens, you may find an inconspicuous button or buttons labelled “IS”, “OIS”, “VR” or something similar.  On the newest bodies, there is no button and you don’t need to turn anything on at all – image stabilization or IS is built into the body and is just there for you all the time.  What is it?  It’s a wonderful technology that stabilizes the image on your sensor when you can’t stabilize yourself or your camera.  We “need” stabilization to avoid a blurry image when the camera is moving.  And despite our best efforts, the camera moves ALL the time.

As humans, we are biological machines.  These machines are in constant motion, even if you do your best to stay still.  Our hearts pump, our nerves fire, our muscles twitch.  Sadly, the only way we can be perfectly still is if we are not alive.

There are also times when you have to move to get the shot – you’re tracking something, you’ve just found that perfect angle, but it’s on a branch up high in the air in the wind (not that you would climb up there), or you’re just plain in the wind and it keeps knocking you and your camera.  Or you are walking and vlogging, which seems to be popular now, and we naturally shake when we walk.  So what can you do?

The first solution, of course, is to select a shutter speed that might compensate for that motion.  We’re all taught the reciprocal rule:  that your minimum shutter speed should never be lower than 1/focal length of your lens.  Great plan, except in low light, deep shadow or other weird atmospheric or lighting conditions.  So what is Plan B?  Have our ever smarter and smarter cameras compensate for the motion.

Canon and Nikon both introduced their versions of IS in the mid-1990’s, so it’s been around for more than 20 years.

What is it exactly?  When first implemented, it was a technology that compensated for lens movements in two of the three axes of rotation:  yaw, which is rotational movement  horizontally, and pitch, which is rotational movement vertically.  As with all things, technology advances now make IS possible in 5 axes (which aren’t all technically axes):  vertical shifts, horizontal shifts, yaw, pitch and roll rotations.  Both lenses and camera bodies now have it.  And the most important thing to note:  the technology is relevant only when the camera equipment is moving; it has no impact whatsoever on a moving subject when the camera equipment is still.  If you don’t catch focus of that moving subject as it passes through your frame, your image will be blurry and IS won’t help you.

The technology senses the motion of the camera equipment using a series of gyroscopes of various configurations that notify the central processor of the camera that motion is occurring.  The processor computes an offsetting motion in either movable lens elements or in the image sensor itself (which is mounted on a movable platform), and implements the offsetting motion in a tiny fraction of a second.  This is repeated over and over during shooting.  The motions are minuscule, but are enough to allow for (in some cases) 6 stops or more of additional shutter speed flexibility.  It’s an amazing invention.

There’s been a lot of debate lately about in-body IS (also called IBIS) vs. lens-based IS.  The technologies are different.  The former moves the image sensor in response to equipment movement, the latter moves special lens elements inside the lens barrel.  So which is better?

The argument for lens-based IS is that it is built for the lens and every lens is different.  The unique combination of elements and moving parts on each lens means that one solution likely doesn’t fit all.  The argument against lens-based IS is one of convenience and economy.  Build it into the lens and the lens is heavier, bulkier.  And all lenses that include stabilization are substantially more expensive than the equivalent lens without it.

The argument for IBIS is that one size does fit all.  A movement is a movement is a movement and if the sensor platform can respond to that movement, all the better.  And of course, lenses can be lighter and more cheaply made.  The argument against IBIS is weight, since all possible scenarios have to be taken into account for every lens, meaning that a single response to a movement may not be appropriate.  IBIS may also have an impact on other activities that you may need to perform, such as sensor cleaning.  I’ve heard that stabilized sensors move when cleaned, making the challenge of one swipe across the surface a problem.

I mentioned in my last post that I had tested some cameras in preparation for purchasing a full-frame mirrorless.  One had 5 axis IBIS which could be combined with lens-based IS for even more stability.  Others had “only” lens-based IS.  In all honesty, for the shooting I tested, both performed equally well.  But I didn’t need 6 stops of stabilization since I don’t shoot in almost complete darkness and I’m not shooting 4K video while hiking up or down a mountain.  And the camera with IBIS weighed an incredible amount more than any of the cameras with a similar focal length length lens with only lens-based IS.   In all fairness though, that heavier camera also had other features that contributed to the weight.

I realized after testing several that amazingly clear, detailed, iconic images have been taken through the history of photography without either IBIS or lens-based IS.  Lenses and bodies continue to be sold without it.  Just because a camera/lens combination has it today doesn’t mean that everyone needs it.  I’ve used and still use bodies without it and I have and will continue to have lenses that don’t have it.  So at the end of my adventure, I realized that IBIS and IS were luxuries, not necessities, and made my most recent purchase decisions with that in mind.  I did buy some lenses that included it.

And like my last post, that’s the message after all.  Buy what you need, maybe even what you want, but not what others tell you that you must have.  That said, I am still fascinated by what engineers can accomplish.