Camera technology has leaped forward in the last decade. Digital cameras have become affordable – that's easily the biggest change. Autofocus has become almost universal – the second biggest change. We would argue that optical image stabilization, or vibration reduction, is nearly as significant.

It has made new pictures possible for photographers at every skill level: pros shoot more available light, and shoot long telephoto lenses with less support. Point-and-shooters essentially do the same – stabilization allows automatic cameras to select longer shutter speeds in preference to higher ISOs, and to leave the flash off longer. And without stabilization, there simply wouldn't be a market for pocket cameras with 10x zooms.  

Optical image stabilization is a phrase that Panasonic uses for its stabilization system, but we're using it as a general term for any camera that moves an element in the optical system to compensate for camera movement. Most cameras move a lens element, but some move the image sensor instead. The idea is simple – if the camera moves up, the lens moves down, and the net result is no movement of the image. The idea has been around for many years, but advances in electronics have only recently made it cheap enough for mass-market cameras.

Five Snapshot Cameras
We compared five snapshot cameras with stabilization so we can outline the uses and limitations of stabilization and compare the various methods currently in use. We've chosen cameras with relatively short zooms – 3.6x is the longest range in the bunch. All the cameras are small, and they are simple to use. With highly automated exposure modes, they're designed to function as point and shoots.

The cameras are: the Panasonic Lumix FX-50 ($400 online), the Nikon Coolpix P3 ($352 online), The Pentax Optio A10 ($262 online), the Canon PowerShot SD700 IS ($358 online) and the Sony Cyber-shot T30 ($395 online). Of these, the Pentax A10 Shake Reduction system moves the image sensor, while the other systems move a lens element.

Test Procedures
In our standard testing procedures, we strive for consistency and repeatability. We use the same lights for every test, the same charts and as much as possible, the same procedures. Consistency is the most important trait we can maintain in our lab, to ensure that all of our results, from camera to camera and from year to year, really are comparable.

For all of our tests except the ones we've put together for this review, we eliminate every source of camera shake that we can identify – we mount cameras on a heavy tripod, we use the self-timer delay, and on DSLRs, we flip up the mirror when we can.

To put it mildly, things are different in this particular test. To evaluate image stabilization, we had to provide wobbles. We tried to replicate the standard user experience. Having photographed a couple hundred weddings over the past 15 years, I've had a pretty good look at the standard use of point and shoot cameras. Among wedding guests at most receptions, standard grip seems to have the right thumb on the back of the camera, the index finger on the shutter release, and the other three fingers of the right hand wrapped around the front of the camera. On small cameras, the right pinkie sometimes wraps to the underside of the camera. The left hand firmly grasps a glass of white wine, a beer, or a plate of broiled scallops wrapped in bacon. Our test procedure omitted the refreshments.

It's not ideal to shoot one-handed – no one on staff at digitalcamerainfo.com shoots with anything less than two hands in their personal work. Still, in an effort to see what stabilization offers the typical user, we shot our tests one-handed to see which, if any, could compensate. At shutter speeds from 1/30 to 1/60, holding a camera with two hands can improve sharpness about as much as optical stabilization can.

Close-up Test
We photographed a dollar bill with all five cameras. The full frame shots are horizontals, and show the entire George Washington portrait, plus some of the text above and below. They were shot under tungsten light. The bill was taped onto a desk, and the cameras were held in one hand. The cameras were set to “macro” or “close-up.” Exposure and white balance were set to automatic.

The crops below are all full-size – they're 100 percent crops of the cameras' images. At this resolution, the whole images would be in the range of 30 x 40 inches, so these are closer views than users will typically employ. Washington's size is not completely consistent from image to image – since the cameras were hand-held, it wasn't possible to match image magnification exactly. Also, the unstabilized images vary enormously in quality – shooting at speeds lower than 1/20 without stabilization yields inconsistent results.

Nikon Coolpix P3

Panasonic Lumix FX-50

Sony Cyber-shot T30

Canon PowerShot SD700 IS

Pentax Optio A10

The Sony T30 shows plenty of improvement between the two shots, but the stabilized image still shows a bit of motion blur. Both the Canon PowerShot SD700 IS and the Nikon Coolpix P3 tamed motion blur, but were a bit out of focus. The Panasonic Lumix DMC-FX50 also wiped out motion blur, but the image sharpness is severely limited by JPEG artifacting. The Pentax Optio A10 looks outstanding in this set, showing much better detail.

Long Exposure Test
We photographed a scene in low light to see how the cameras handled relatively long exposures. Because the cameras do not have straightforward manual controls, we ended up with a range of exposure times, from 1/4 of a second to a full second. Given that range, these results are not directly comparable. Still, they show that the cameras can yield usable images at very low speeds. Again, our sample images are shown at 100 percent: each pixel on the computer display corresponds to a single pixel in the image file.

The Nikon Coolpix P3 image pair shows the most significant improvement. The P3 recorded a 1/2-second exposure for each shot. Its unstabilized image is the shakiest of any camera tested. That's odd – it's a large camera with a comfortable grip.

Nikon Coolpix P3

The Sony DSC-T30 logged 8/10-second exposures, and the stabilization system couldn't handle such a long exposure. Of all of the stabilized images, the T30's is the only one in which the word “PAPRIKA” isn't legible. Blaming the long exposure is reasonable, but the Panasonic Lumix DMC-FX50, which clocked exposures of 1 full second, rendered the text. Again, the Panasonic's JPEG compression appears to limit sharpness.

Sony Cyber-shot T30

Panasonic Lumix FX-50

Both the Canon PowerShot SD700 IS and the Pentax Optio A10 have maximum apertures of f/2.8, and therefore managed exposures of ¼ second. The SD700 IS performed better, rendering, just barely, the word “HUNGARIAN” above PAPRIKA.

Canon PowerShot SD700 IS

Pentax Optio A10

Telephoto
The five cameras tested have moderate telephoto lenses, from a 140mm equivalent on the Canon PowerShot SD700 IS to a 102mm equivalent on the Panasonic Lumix DMC-FX50. These lenses are suitable for portraiture and candid shooting, but they aren't the powerful zooms one might use for sports or nature photography. Their relatively short focal lengths are also less taxing on image stabilization systems.

We tested the cameras at their maximum zoom settings, shooting a copy of our standard resolution test chart outdoors in open shade, producing exposures from 1/60 to 1/80 of a second. The chart is a high-contrast ISO chart, made up of black shapes on a white background. Any gray, or blur is an artifact of the camera, not part of the chart. The cropped images shown below are reproduced at 100 percent, so there has been no re-sampling of the images. Again, our crops are very small portions of the actual images.

Panasonic Lumix FX-50

Canon PowerShot SD700 IS

Nikon Coolpix P3

Pentax Optio A10

Sony Cyber-shot T30

The Panasonic Lumix DMC-FX50 has two stabilization options: continuous, which stabilizes the image whenever the image sensor is active, and image-only, which stabilizes only while an exposure is being made. In the illustration below, the left image was taken with no stabilization, the center with continuous, and the right with image-only. Image-only produced the superior result, but both stabilization options significantly sharpened the image.

The Canon PowerShot SD700 IS offers the same stabilization options as the FX50, plus a panning mode. Panning mode is meant for stabilizing images shot while tracking a moving subject, and was not appropriate for this test. Canon's continuous and shot-only modes are both sharper than the unstabilized shot. The two stabilized shots are so similar that we recommend setting the SD700 IS to continuous stabilization, which makes the viewfinder a little more convenient.

The Nikon Coolpix P3 offers “Vibration Reduction” and “Vibration Reduction Active.” Nikon's manual says that the Active setting is for more severe shake than the regular setting. In this test, the two settings performed equally. At 1/75 of a second, there wasn't much improvement over the unstabilized shot at either setting. By cranking down the ISO, we got the shutter speed down to 1/15 of a second. At that speed, the Vibration Reduction system is very effective – normal is in the center, and Active is to the right – but the images still aren't very sharp.

The Pentax Optio A10 offers a single stabilization mode, which works well. The A10's image looks sharper than some others, perhaps because it is more contrasty. The Sony DSC-T30 also has a single mode, and in this test, it produced much better results than any of the other cameras.

Video
All five cameras record video at standard quality – 640x480 pixels and 30 frames per second. As we examined stills from video clips, we didn't notice a worthwhile difference in resolution between the stabilized and unstabilized shots. On the other hand, stabilization smoothed out our unsteady hold on the camera.

Click on the thumbnails below to view a compressed video clip of the stabilized footage from each model. To view the unstabilized clips (image stabilization off) or uncompressed video footage, click on the text links below.  

Panasonic Lumix FX-50

Compressed Video - Panasonic Lumix FX-50 (without stabilization)
Full Resolution Video - Panasonic Lumix FX-50 (with stabilization)
Full Resolution Video - Panasonic Lumix FX-50 (without stabilization)

Canon PowerShot SD700 IS

Compressed Video - Canon PowerShot SD700 IS (without stabilization)
Full Resolution Video - Canon PowerShot SD700 IS (with stabilization)
Full Resolution Video - Canon PowerShot SD700 IS (without stabilization)

Nikon Coolpix P3  

Compressed Video - Nikon Coolpix P3 (without stabilization)
Full Resolution Video - Nikon Coolpix P3 (with stabilization)
Full Resolution Video - Nikon Coolpix P3 (without stabilization)

Pentax Optio A10

Compressed Video - Pentax Optio A10 (without stabilization)
Full Resolution Video - Pentax Optio A10 (with stabilization)
Full Resolution Video - Pentax Optio A10 (without stabilization)

Sony Cyber-shot T30  

Compressed Video - Sony Cyber-shot T30 (without stabilization)
Full Resolution Video - Sony Cyber-shot T30 (with stabilization)
Full Resolution Video - Sony Cyber-shot T30 (without stabilization)

The big performance difference between the cameras was how they handle panning. Two cameras, the Panasonic Lumix DMC-FX50 and the Nikon Coolpix P3, yielded relatively smooth pans, while the other cameras introduced a jerking motion as the camera moved. The Pentax Optio A10 had a more subtle effect than the Sony DSC-T30 or the Canon PowerShot SD700 IS. Apparently, the Pentax, Sony and Canon attempted to treat the pan motion as if it were unsteadiness, holding the frame steady as long as they could, and then jumping forward to the new position.

For static shots, all of the cameras shoot better video with stabilization turned on. For pans, only the Panasonic and Nikon shots are more pleasing with stabilization.

Conclusion
Consistently, stabilization improved each camera's performance in still shooting, so our confident recommendation is to look for stabilization when shopping for a compact camera. Casual snapshooters should get it, and leave it turned on.

Stabilization contributes to image quality, but it's only one element in a complex system that includes the lens, sensor and image processor. Good stabilization isn't enough to make a good camera. For example, the Nikon Coolpix P3 and the Panasonic Lumix DMC-FX50 have effective stabilization, but their images still aren't as sharp as the shots from the Sony DSC-T30 or the Pentax Optio A10.

The Pentax Optio A10 is the unusual camera in the bunch – it moves the CCD imaging device, instead of a lens element. We have assumed that moving a lens element is the more logical way to go, because the lens element is lighter than a CCD, and doesn't need any electronic contacts to the rest of the camera. Based on the A10's performance, we'd say that the moving CCD system can equal the best moving-lens systems.