Optical image stabilization

As far as I can tell this is the first slimline Casio to feature a CCD shift stabilization system - presumably because this is the first to feature a long zoom range. The moving sensor works in conjunction with Casio's Anti-shake DSP (which is a slightly more sophisticated than normal ISO boost system) to - in theory - counter blur from both camera movement and subject movement. I say 'in theory' because, not to put too fine a point on it, it doesn't really work.

To be fair the anti-shake DSP side of things does what it's supposed to, forcing a higher shutter speed in order to avoid blur from moving subjects (the new motion analysis function apparently anticipates when it's needed), though there is a fairly significant quality hit if the ISO has to be boosted too far. But the other side of the anti-blur equation, the CCD-shift, has only a marginal effect on blurring due to camera shake. It's nothing like the optical IS systems used by Panasonic, Sony and Canon, for example; at best you can expect around a 1 stop advantage over an unstabilized camera (and to be honest that's being generous). It's not helped by the physical design of the camera, which is too slim and light to allow shake free shooting at the long end of the zoom.

We suspect that part of the problem lies with the folded optics design of the zoom, which places the sensor (the bit that moves to counter the camera shake) on its back, near the base of the camera. Since the CCD can only move in one plane, and that plane is parallel to the base of the camera, it can't correct 'up and down' shake at all.

There are four options in the anti-shake menu; Demo (stabilizes the preview image, which seems to work, but you can't take a picture), Image Blur (aims to increase shutter speed using ISO boosting), Camera Shake (CCD shift stabilization only) and Auto (uses both methods as the camera sees fit).

The stabilization test

In this simplified version of our SLR IS test, eight hand-held shots were taken of a static scene with the stabilization off and on (Camera shake mode). The shutter speed was decreased by a stop and repeated (from 1/800 sec to 1/10 sec). The zoom was set to its maximum position (266mm equiv.), the test chart was approx 3.5 m away from the camera.

The resulting images were then inspected and given a blur score - 'Sharp' (no visible blurring at 100%), 'Mild Blur' (the kind of camera shake that is tolerable at small print sizes), Heavy Blur (blur visible even at small print sizes) and 'Heavy Blur' (totally unusable due to camera shake).

As the charts below show we were able to get - if we're being generous - a one stop advantage. Unlike many competitor systems we've tried we were unable to get any completely sharp shots at below 1/50th second, no matter how many attempts were made.

Hand-held (one hand), no stabilization (266mm equiv.)

As you can see from the chart below once we dropped to below 1/100th second we had little or no change of getting a usable shot. Interestingly at 1/400th sec (the recommended setting using the good old focal length reciprocal rule) we still saw light blurring in 25% of shots because the design of the camera makes holding it steady in one hand very difficult (when I repeated the test holding the camera with both hands the entire graph shifted a stop to the right).

Hand-held (one hand), CCD stabilization on (266mm equiv.)

As the results below show there is only the smallest advantage to shooting with the IS on - at best you're getting a one stop advantage. It's still impossible to get a fully sharp image at 1/50 sec or less, and at 1/100 sec you're lucky to get one in four totally sharp.

Let's put this into perspective: at 1 stop below the recommended minimum the IS system won't guarantee sharp results, but it will avoid images that are totally useless because they are so blurred. At 2 or more stops below you might see a slight increase in usable shots, but it is very difficult to get a sharp result no matter how many 'safety shots' you take. Optical/mechanical stabilization is a key selling point of cameras in this class but if it doesn't work - and it's hard to realistically argue that it does here - it's pointless.

Image Parameters and in-camera effects

As mentioned earlier the EX-V7 has, like most modern Casios, a fairly comprehensive range of in-camera picture adjustments and, unusually, quite a few that can be applied to images after they have been saved.

The options on offer when shooting include the fairly normal stuff; contrast, sharpness, saturation (5 levels of each) and color filters (B/W, Sepia, Red, Green, Blue, Yellow, Pink, Purple) and some new options (thanks, we presume to the new processing engine): Dynamic Range and Portrait Refiner.

The Dynamic Range option allows you to lift the shadows in bright contrasty scenes (there are two settings). As usual there is a price to pay; a little extra shadow noise, but the output is certainly more pleasing and is quite subtle compared to some we've tried (Nikon's D-Lighting for example). The Dynamic range 'effect' can be applied either before or after you've taken the picture.

The Portrait refiner claims to reduce noise in skin tones, but to be absolutely honest we really couldn't see any difference at up to ISO 400.

Dynamic range settings compared

Dynamic range settings compared (post-shot application)
Original Dynamic range +1 Dynamic Range +2

In playback mode you get an unusually sophisticated set of options for changing saved images, including brightness, white balance and - as shown below - removing keystone distortion.

Keystone correction

The EX-V7's keystone correction feature attempts to automatically correct distortion, though it only really works if you're shooting square things (i.e. copying documents or prints). When it does work, it's surprisingly effective.

Keystone Correction
Original corrected