Testing A Canon Image Stabilization Lens
For a while, I’ve been wondering exactly what I actually got for the extra money I paid for a Canon Image Stabilization lens. So I decided to do a little testing and find out, does it really work, and is it worth the money?
I bought the EF 28-135 USM IS lens to go with my Rebel XTi camera a little over 6 months ago. Since that time, I’ve shot with it exclusively in IS mode and I’ve been happy with the results. But I got to thinking, how much difference would it make if I turned off that IS and shot in some different lighting conditions. I found a couple different lighting situations around my house and took some shots both with and without Image Stabilization activated.
Most everyone knows the “rule” that the inverse of your lens focal length is the lowest speed you should shoot at to avoid blur. This means that if I’m shooting a 50mm prime lens, in theory, I shouldn’t be able to shoot a shutter speed any lower than 1/50 second and get a crisp image. This is no problem in outdoor or studio conditions, since most any outdoor shooting is going to be at a faster speed than that, and lighting is controllable in a studio setting. However, for indoor, nighttime or other conditions, 1/50 second is not a practical shutter speed. Before the advent of IS lenses, the only options were to either find more light, or use a flash. There are many situations where neither of these options are possible and in the past, the pictures just turned out a blur.
First lets look at a couple of shots that follow the focal length inverse “rule”
This is a shot taken at a focal length of 60 mm and a shutter speed of 1/60 second at f/5 with NO Image Stabilization:

Now here’s the same shot settings with Image Stabilization turned ON:

The first thing you notice is that there is little to no difference in the two images. This is a significant crop of the original image, used to show the flower detail. It looks as though the rule has been upheld so far. I took a couple of other test shots at 30mm and at 125mm and the results were the same, so I’ll skip posting these.
Now we’ll look at some lower light conditions where it’s not possible to shoot at an inverse shutter speed to the focal length.
The first example is of a lit lamp and the area around it. The focal length on this shot was 70mm and I shot it in Program mode, so the camera chose the shutter speed for the proper exposure. In this case, the shutter speed was 1/4 second at f/5.
First is the full frame shot with no crop and NO Image Stabilization:

Note the softness of the shot. The exposure is actually a little dark, but it was metering for the lamp, not the surroundings.
Now we’ll look at the full frame version WITH Image Stabilization:

The difference is striking. The picture, while still pretty dark, is much more clear.
Lets look at a crop from both photos to show detail:
NO Image Stabilization:

WITH Image Stabilization:

At these crop levels, the difference is much more clear. Without IS, there is significant halo/ghosting around the lamp. The IS shot shows no ghosting. While the focus isn’t perfect on either shot, its easy to see the difference in the two modes. The IS image is clearly better and provides a usable image even at the ridiculous shutter speed of 1/4 second.
The second low light shot I did was of some candles on a table, lit by sunlight through a window. The scene has a little more light than the lamp scene and the metering is more accurate. This time there is a little less range in between the focal length and the shutter speed. I was all the way open at 28mm and once again using a shutter speed of 1/4 second, but at f/3.5 this time.
Full frame NO Image Stabilization:

And WITH Image Stabilization:

There is a lot more going on in this photo than the lamp example, but the first thing I notice is that the candles look much more soft in the NO IS version of this scene. A look at the crops of the two images confirms this and more:
Crop with NO IS:

Crop WITH IS:

The differences here are striking once again. The crops of the photos show the real advantages of IS in these conditions. There is ghosting without the IS and the photo with IS, while not tack-sharp, is still very usable and free of ghosting.
After these tests, I am glad I went with the IS lens. While I may not notice the advantages it gives on a daily basis, it’s nice to know that they are there. If you think you will ever need to shoot in low light, without a flash, the Image Stabilization feature on this particular lens is definitely worth the extra money it costs.






Nice article. I too feel that buying a lens or camera body with stabilization is a must.
A more comprehensive test would be nice. All you did was compare 1 shutter speed, why not show the limits of the IS mode, although I probably wouldnt even use it at that speed 1/4 as the pictures are still too soft. Id like to see some at 1/8 and 1/16 and 1/30th. And compare them to a non – IS 1/60th shot
Nathan,
Thanks for sharing your tests. I was thinking about getting a cheap zoom to start with and now I’ve decided to wait and get something better. As for Kevin who wants a more comprehensive test… Let us know when you do your follow up article Kevin.
Shannon
Are there any different with tripod? L lens with IS lens is almost double the price, I am still thinking about Canon EF 70-200mm f/2.8L USM Zoom Lens ( Non-IS), but I have a 28-135mm IS lens to use walk around. Decision decision!