Nothing particularly unusual here; the S10's fairly light touch with noise reduction at higher ISO values produces results that look quite grainy, but you do at least retain some detail, although we did find the noise reduction at ISO 50 and 100 to be a little aggressive. Aside from a few exceptions (such as Panasonic) there's very little difference in the appearance of any of these 1/2.5 inch CCD cameras at anything over ISO 200, though it's good to see a fairly linear relationship between noise and ISO here, meaning Nikon isn't turning up the NR too much at the higher settings.

  ISO 50 ISO 100 ISO 200 ISO 400 ISO 800

Noise graph

Indicated ISO sensitivity is on the horizontal axis of this graph, standard deviation of luminosity is on the vertical axis.

Low contrast detail

What the crops and graph don't show is the effect of noise reduction on low contrast fine detail such as hair, fur or foliage. An inevitable side effect of noise removal is that this kind of detail is also blurred or smeared, resulting in a loss of 'texture'. In this test the crops below show the effect of the noise reduction on such texture (hair) as you move up the ISO range. Note that this is a new, standardized version of this test - all future tests will be directly comparable.

100% Crops
ISO 80 ISO 100 ISO 200
ISO 400 ISO 800  

As the crops show, at ISO 50 and 100 there is not a great deal of smearing of low contrast detail, and ISO 200 is usable. We did find that in 'real world' shots there was some loss of very fine texture to noise processing even at the lowest ISO settings - if viewed at 100% - but nothing you'd really miss in a normal print. At ISO 400 you're losing a lot of texture (though there's many cameras that are worse!) and at ISO 800 the noise itself is masking most of the finest detail.

VR (image stabilization)

Unlike the last Nikon Coolpix we tested with a VR function (the Coolpix P3), the S10 uses a CCD-shift system rather than a moving lens element. And unlike most of the compact camera CCD-shift image stabilization systems we've tested this one seems to work very well. It's not quite as reliably effective as the lens IS systems used by Panasonic, Sony and Canon, but it does do a very good job when you are shooting a stop or two below the recommended minimum for the focal length chosen (i.e. down to about 1/100th sec at full zoom).

If you push it harder (such as the first example below) and take a couple of 'safety' shots, you'll usually get at least one 'keeper'. This is where Nikon's excellent 'Best Shot Selector' (which takes a series of shots and only saves the sharpest) proves its worth.

We can't run our usual IS test suite on the S10 as there is no control over shutter speeds (or any display of shutter speeds), but the examples below show just how much difference VR makes when shooting at the longer end of the zoom using a shutter speed up to three stops longer than would be needed for hand-held shooting without VR.

1/16th second, hand-held, 230mm (equiv.) 100% crops
VR on VR off
1/35th second, hand-held, 63mm (equiv.) 100% crops
VR on VR off