A recent article on Panasonic filing a patent for sensor-based (in-camera) image stabilization (IS) got me thinking about the different image stabilization technologies; specifically which forms of IS are better overall. For those of you that don’t know, there are two primary types of image stabilization technologies currently used on DSLR cameras: sensor-based image stabilization, which uses a mechanism to move the sensor to compensate for camera shake, and lens-based stabilization, which moves a floating element within the lens to compensate for camera shake. Canon and Nikon both use lens-based stabilization, while companies such as Sony, Pentax and Olympus feature sensor-based image stabilization. Both types of stabilizing systems have their pros and their cons, but which is better? I can’t really give you an answer for which system is technically better, but I’ll the advantages and disadvantages of both systems, and let you know which I personally think is the better of the two.
Lens-Based Image Stabilization
There are people that argue that lens-based IS technology has a large advantage over sensor-based IS because each lens has its own mechanism that is built specifically for that lens, making it very accurate. Another advantage is that with lens-based IS, the image that you see in the viewfinder of the camera is also stabilized, something that you do not get with sensor-based technologies.
The disadvantages to lens-based IS is that image-stabilized lenses are typically more expensive than lenses without it. The added glass and electronics that are needed to produce lenses with IS add to the the overall cost of production. The difference of cost between lenses with and without IS vary from one lens to another, but some have differences that range in the upper hundreds of dollars. The difference between the IS and non-IS version of the Canon EF 70-200 f/4, for example, is more than $600.
Sensor-Based Image Stabilization
The biggest advantage that sensor-based IS has over its lens-based cousin is that any lens can be used. This can potentially save people a lot of money, especially when you want too have many different lenses in your kit. Another reason you save a lot of money with sensor-based stabilization is that you will always get the newest IS technology when you buy a new camera body. Because the life cycle of camera bodies are much shorter than those of lenses, the bodies will continually be updated with the latest and greatest image stabilization. Rather than spending money on upgrading every lens you have when new IS technology comes out, you only have to spend a fraction of the cost on a new body.
The disadvantage of sensor-based image stabilization is that you do not see a stabilized image in the viewfinder. This may not bother the user at all, but it is something to note. Another disadvantage is that lens-based IS is said to be more precise than sensor-based because it is built to work for all lenses. This gives more room for inaccuracy. In addition to this, the output image circle of every lens has to be larger because the sensor will be moving around. This means that although lenses are not designed with IS built into them, they still need to be designed to the specifications required to compensate for movement of the sensor.
I cannot give you a technical answer as to which image stabilization technology is better, but I really think that sensor-based is the way to go. It is by far the more economical route, and from what the Sony Alpha A900 (Full-frame DSLR) showed us, it works very well. Canon and Nikon may have chosen to go down the lens-based path for compatibility reasons, but it would be a dream come true if they ever choose to switch to sensor-based.
What do you guys think? Which Image stabilization technology do you prefer and why?