Camera IBIS and Lens OIS in Canon EOS R5 – How Good is It?

I’m afraid that this piece is going to be a little bit technical, so for those who aren’t inclined that way, but who own a Canon EOS R5 camera, let me just say that Canon has really brought it home here: a Canon EOS R5, equipped with a native Canon RF mount lens that includes lens-based image stabilization, can indeed get up to 8 stops of improved low light performance. That simply means that you can shoot images with shutter speeds up to 8 stops slower than you normally could and still get sharp, in-focus images in low light conditions. Up to 8 stops. Ok, that’s the non-techie part. And you already knew that anyway. Non-techies can leave now. But stay and continue reading. It seems that “up to” has proven very confusing. Read on to find out why.

There seems to be a big debate about whether Canon’s claims are true. This debate seems to stem from a misunderstanding around how lens-based stabilization interacts with camera-based stabilization in the R5. I stumbled across this debate and thought it was worth clarifying, if only for my own use. I don’t like confusion.

Canon themselves have put out a video explaining the matter. Unfortunately, it seems more marketing-oriented than user-oriented, but still fills in part of the gap.

But really, the first step in enlightenment is understanding why you need stabilization at all. Hopefully most of us are past that, but here goes. As you know, cameras record light. In low light conditions, it takes longer to record the correct amount of light to properly expose an image. The longer it takes, the more likely it is that something can interfere with the camera during capture. In this case, any vibration, unintentional small movement or deliberate small movement of the camera can ruin an image.

This is especially true when hand-holding your camera. The human body is a fountain of vibration. We are constantly in motion, from large positional changes, to repeated actions like breathing and heartbeats, to micro-movements as electrical impulses flow along our nerves, even when we try to be still. Our bodies rock and our hands shake no matter how old or young we are.

We also operate in three-dimensional space, meaning that movements can occur along three axes or three planes, typically labelled x, y, z. X is universally recognized as the horizontal axis, y the vertical axis and z the front to back axis. In each of these axes/planes, there are two types of movements: linear and rotational. Linear is simply up/down, left/right or forward/back. But there are also rotational movements in these same axes, typically labelled pitch (x-axis), yaw (y-axis) and roll (z-axis).

Are you still with me? We can move unintentionally in any of an infinite number of combinations of micro-movements in any or all of these three axes/planes. So the technology to compensate for them has to be not only very sensitive to detect the movement in the first place, but very fast to program, transmit and activate the solution, all before the next micromovement happens. It’s astonishing that the latest cameras can do this so well.

Let’s spend a bit of time on the history of this technology. It’s long been recognized that some form of mechanical stabilization independent of the photographer was both necessary and helpful in specific shooting situations. Initially this was as simple as putting the camera on a tripod and using a shutter release cable or a timer to dampen the vibrations of our touches. Early cameras were always operated on tripods. It was only when cameras became smaller and more portable that we discovered the limitations of our stability and had to find other solutions.

Built-in image stabilization first appeared on the scene in consumer-grade cameras as early as 1994, introduced by Nikon as Vibration Reduction. It was a two-axis solution. Canon followed the next year with their version. Initially, this only appeared in lenses and for the longest time, was only in lenses. I’m speculating, but since lens optics and lens components were already moving elements for zoom and autofocus purposes, it may have been easier to build stabilization into this existing system of chips and motors. Digital in-body image stabilization appeared around 2008, and relies not on mechanical components, but on the way the data from the sensor is processed in-camera and recorded to the card. Sensor-based in-body image stabilization (where the image sensor physically moves in response to vibration) first appeared around the same time (2008) but multi-axis versions really started to take hold around 2013.

Imagine how difficult this is to deliver: for those of us who wear glasses, when we update our prescriptions, we place our heads in a fixed headrest and stare at a target fixed to the wall. Even so, the doctor has to try several optical elements to get the prescription exactly right. Imagine how difficult it would be if not only our heads were moving freely but the target was as well. No wonder it took so long to develop this technology.

Canon and all major brands have improved this technology with every generation of lens and camera. And therein lies perhaps the biggest issue in this controversy. Canon has promised that we can use older EF glass with the new R-series cameras. And indeed, that it true. But they conveniently omitted some details. One of the major claims when the R5 was released was that this camera could allow for up to 8 stops of image stabilization. That was huge and still is.

But, the actual number of stops of benefit you will experience depends on the lens/camera combination you choose. Here are some things to be aware of:

• not all Canon EF glass has optical image stabilization
• not all Canon RF glass has optical image stabilization
• even if it has IS, not all Canon EF glass has image stabilization that can fully interact with in-body stabilization of the R5; the older the lens, the less likely that it can fully interact, even if it says “Stabilizer” on the barrel
• even if it has IS, not all Canon RF glass has image stabilization that can fully interact with in-body stabilization of the R5 in all axes/planes; the less expensive the lens, the less likely that it can fully interact
• the only way to know if the lens and body are fully interacting is through a moving hand indicator on the viewscreen or viewfinder; if fully, then the hand will have a plus sign beside it; when no plus sign appears, you have no idea just how much stabilization is active
• if there is a stabilization switch on the lens, turning it off turns off both lens and body stabilization; there is no way to leave one on without the other
• there are no menu options in-camera for managing stabilization, except when using a lens with no native stabilization
• if using a lens with no native stabilization, a menu choice suddenly appears in the Shoot 7 menu to allow you to turn on or turn off body stabilization

So what’s the bottom line: as with all things R5-related, the camera provides a somewhat complicated set of working parameters for image stabilization. Sometimes you get 8 stops, sometimes 5, sometimes 2 or 3, and in very specific cases only, you can turn it off. Sometimes it’s 2 axes of stabilization, sometimes all 5. And we haven’t even discussed video. So the debate can be put to rest: everyone is right.

To help you out, here is a chart that summarizes all of the actions and interactions, surprisingly issued by Canon themselves.

Many of us (including me) wanted a smart camera. Many of us (including me) don’t really appreciate how smart it is. It’s like having a child that quickly exceeds your talents and skillset. You still need to have control, but you want to give full rein to the smart one to showcase their abilities. Now at least for R5 image stabilization, you can.