DPReview has never taken sides in the row over whether image stabilization should be provided in-body or in the lens. But we'll now list the absence of in-body stabilization as a Con in our camera reviews. We'll explain what's happened to change our stance.

For at least the last decade, our internal policy has been to have no preference between in-body and in-lens stabilization. It's an approach that was sometimes seen as fence-sitting but was driven by two logical precepts:

Our primary reasoning was based on our guiding principle that it makes more sense to concentrate on the photographic impact, rather than which technology was used. For many years, a majority of users bought DSLRs that came with a stabilized lens and had essentially the same experience as those users who bought a camera body with stabilization. The in-body advocates talked a good game about 'every lens being stabilized' but we didn't hear reports from thousands of distraught DSLR buyers wondering why their shots were blurry.

There are genuine advantages and disadvantages to both approaches

Secondly, there are genuine advantages and disadvantages to both approaches. As a broad generalization, in-body stabilization does a better job of correcting the translational movements that affect close-up and wide-angle photography, while in-lens designs cope better with the large pitch and yaw corrections needed with long focal lengths. So why would we give credit for one and ignore the benefits of the other?

The result was close enough to a wash that we felt it was inappropriate to have a preference.

So what's changed?

That's now changed, again for two main reasons. The first is prevalence: for our review Pros and Cons, we tend to recognize a new feature as a Pro if it's rare or exceptional but then acknowledge its absence as a shortcoming, once it's become a common (and hence expected) feature. So, for example, we started off seeing USB charging as a 'Pro' (with an associated 'Con' if it was used as an excuse to leave an external charger out of the box), when it first appeared. But, now that it's become commonplace, it's the absence of USB charging that would be seen as a negative.

Now that Ricoh, Olympus, Panasonic, Sony, Fujifilm and Nikon all offer in-body IS on at least some of their interchangeable lens cameras, its omission starts to look like a factor we should highlight to would-be buyers.

The second factor is the growth in importance of video. There may have been little difference in performance between in-lens and in-body stabilization when it came to stills, but the same can't be said for video. Watch TV or movies and you'll see all sorts of camera movements, but what you won't tend to see is roll: left-and-right rotations that cause the horizon to tilt. In-lens stabilization can correct for pitch and yaw but it can't correct for roll: which gives in-body stabilization an immediate advantage.

There may have been little difference in performance between in-lens and in-body stabilization when it came to stills, but the same can't be said for video

We'll still test each camera to see how well its IS works: our approach of assessing effect, rather than technology means we'll check whether digital stabilization in concert with in-lens stabilization provides a good degree of correction. But don't be surprised if you see Cons highlighting the lack of in-body stabilization in our future reviews.