Overall Image Quality
Lets start with colour. Kodak digital cameras have always had a good reputation for colour reproduction, indeed the DC260, 265, 280 and 290 probably had the best colour of any digital camera at their time. And the DC4800 carries this tradition well, in "Saturated" mode colours are vibrant yet accurate and not overblown, colours which normally suffer (greens mostly) are faithfully reproduced and the DC4800's colours remind me quite a bit of the Fujifilm S1 Pro upon which I gave so much praise.
Moving on to detail at a pixel level and I'm less than happy with the DC4800. The imaging system itself has some noticeable problems (which I'll cover in detail below), dollar for dollar it didn't provide the same detailed impact I'd expect from a camera of such "calibre" coming from Kodak. In print it may be true that some of these artifacts may not be visible, and certainly for web use sizing the images down will produce very pleasing images. However most people's reasoning for going to 3 megapixels was the ability to print at larger sizes either now or in the future, sizing the DC4800 images upward didn't work well and made the artifacts far more visible.
Purple Fringing (Chromatic Aberrations)
Sorry Kodak but the lens on the DC4800 isn't up to scratch, chromatics are visible at almost any zoom and are obviously worse at full wide (which is pushing such a small lens system to its extremes). I noticed the chromatics from the outset and it wasn't hard to pick out a couple of real life examples with chromatics that would be visible in print.
So far in our testing of digital cameras we've surmised that chromatics are enhanced by blooming which is an effect of the overflow of charge from one pixel to the next (on the boundary of very bright to dark). Thus what you're seeing below will be chromatic aberrations "amplified" by the blooming effect.
Jaggies / Moiré
Another problem I noticed in quite a few DC4800 shots was the presence of unnatural jaggies on diagonal lines and visible moiré patterns on repeat detail. As far as the jaggies are concerned this could simply be a problem with either the CFA interpolation or sharpening algorithm, either way it produces noticeable "steps" on diagonals which shouldn't be there. The problem of moiré patterns was more visible in finer details such as hair (as shown in the example below) or repeat areas of an image (such as railings / fence posts into the distance).
Barrel and Pincushion Distortion
With such a "wide" wide angle (28 mm equiv.) we expected the DC4800 to have some barrel distortion, and indeed it measured at 1.1%, however that's a pretty good performance compared to what we normally see on 38mm equiv. digital camera lenses. At full tele, pincushion distortion was so small as to be unmeasurable.
|Barrel Distortion, 1.1% @ Wide Angle||Pincushion Distortion, Unmeasurable|
The DC4800 has several methods for setting White Balance, obviously there's Auto white balance where the camera will attempt to measure the light colour and adjust white balance accordingly (this only tends to work under certain light conditions), otherwise you can choose from the preset white balances, enter the Kelvin temperature of the light used or manually set a white balance using the 4-way controller pushing white balance towards cyan, magenta, green or yellow (this works quite well once you get the hang of it).
|Outdoors, Auto WB||Outdoors, Daylight WB|
|Incandescent, Auto WB||Incandescent, Tungsten WB|
|Fluorescent, Auto WB||Fluorescent, Fluorescent WB|
The Auto White balance seems to work fairly well, obviously daylight (or in our case overcast daylight) produced the best overall balance (thrown slightly blue by the overcast nature of the day).
Dynamic range simply defines the range of light the camera is able to capture before it either loses detail in darkness (shadows for example) or blows out a highlight (edges of chromed metals are good examples of this). Most consumer digital cameras only have a 8-bit analog to digital converters, plus their CCD's are not built to have a particularly large dynamic range.
Using our new dynamic
range measurement method we measured the DC4800's dynamic range as
(higher numbers are better except for noise):
Pretty average results for a 3 megapixel digital camera, fairing about as well as the Coolpix 990 (though the 990 performs far better than the DC4800 for the "best case" results; ISO 100, sharpening disabled) and a little better than Canon's G1, noise appears to be a factor in the DC4800's images, and thus takes a big bite out of dynamic range at higher ISO's.