Pentax K-x Digital Camera Review

Color

The Pentax K-x was less accurate in its color reproduction than most cameras we test. To determine color accuracy, we shoot the X-Rite ColorChecker chart under bright studio lighting, in each available color mode, and analyze the resulting images using Imatest software to measure the color deviation from the known chart values. The score is based on the result for the most accurate color mode.

The K-x offers six color mode choices, the most accurate of which is called Natural. This is the only available choice that's properly saturated, at 103.5%, The next closest to ideal, Bright mode, is oversaturated to 123%. While it's not unnaturally vivid, even the Natural mode exhibits substantial color shifts, most notably when reproducing orange and yellow shades. Blue, red and magenta, though, are reproduced quite accurately. Flesh tones are off a bit, but it will take a critical eye to notice this shift. More on how we test color.

NOTE: Because of the way computer monitors reproduce colors, the images above do not exactly match the originals found on the chart or in the captured images. The chart should be used to judge the relative color shift, not the absolute captured colors.

The default color accuracy for the K-x in its best mode is less precise than the other cameras in our line-up. Color shifts are most noticeable for orange and yellow, which fortunately aren't that prominent in most photos.

Color Modes

The K-x offers seven Custom Image settings, which control color along with saturation, hue, brightness, contrast and sharpness, all of which can be adjusted. Six of these settings are in color: Bright, Natural, Portrait, Landscape, Vibrant and Muted. The final choice is monochrome. For more information and image samples comparing Custom Image settings, see the Picture Effects section.

In the table below are same-size crops from our test images for five of the color modes. You'll find samples taken in Muted and Monochrome modes in the Picture Effects section of this review.

White Balance

The K-x is a mediocre performer in white balance adjustment accuracy, lagging the competition using the automatic white balance system and falling short of the color accuracy levels we expect after taking a manual white balance reading.

We check white balance performance by shooting the X-Rite ColorChecker chart under daylight, incandescent and compact fluorescent lighting produced using the X-Rite Judge II light box. The test photos are analyzed using Imatest software, with color error measurements compared against the known values of the ColorChecker chart.

Automatic White Balance ()

The automatic white balance system handles daylight illumination well, but it's flummoxed by compact fluorescent and incandescent lighting, like that produced by household tungsten bulbs.

Custom White Balance ()

We demand a very high level of color accuracy after going to the trouble to take a custom white balance reading with an SLR. The K-x didn't live up to these standards, Shots taken under incandescent lighting were much improved compared to the automatic WB results, but still no great shakes compared to other tested cameras.

The K-x couldn't match the exemplary results delivered by the K2000 in daylight shooting, but its WB accuracy holds up well compared to the other tested cameras.

Incandescent illumination is consistently the most challenging light source for automatic white balance systems, and the K-x is not alone in failing to cope well. The GF1 was the noteworthy exception to the orange-tinged rule here.

The Pentax K-x handled compact white fluorescent lighting poorly compared to the competition, including the exceptional K2000 results.

Overall, the Pentax K-x offers underwhelming white balance accuracy levels.

White Balance Options

The K-x provides a generous selection of white balance presets, including a variety of fluorescent bulb types.

The on-screen preview system works well when setting white balance. By pressing the exposure compensation button, you capture a preview image at current settings. Then, as you cursor through your white balance preset options, or fine-tune the setting along the green-magenta and/or blue-amber axes, you see the effects of your changes live on the preview image (which you can save as well, if the spirit moves you).

Taking a manual white balance reading is simple enough. You shoot a white or neutral gray surface under current lighting conditions. You can then make the selection area larger or smaller, and move it around the screen, to specify the precise area you want to use for taking a reading.

Long Exposure

Our long exposure testing combines color accuracy and image noise analysis at slow shutter speeds (between 1 and 30 seconds), at low light levels. The K-x didn't fare badly in the noise component, but its bright light color problems carried over when the lights were low, leading to a mediocre overall score. More on how we test long exposure.

The good news is that the color didn't shift when the shutter speed changed — consistency here is a virtue — and there was only slight oversaturation. However, the color hues weren't reproduced accurately, leading to a low score here. As expected, results were very close whether long exposure noise reduction was turned on or off.

Image noise results aren't great, but at under 1% across the shutter speed range the images are certainly useable. As we often find, long exposure noise reduction had no positive effect, and doubled the elapsed time for each shot. Long exposure noise reduction systems work by taking a second exposure with the shutter closed, determining the noise pattern in that shot and mathematically removing these flaws from the original. However, since most image noise is unpredictably random from shot to shot, the system rarely works and, in fact, often increases visible noise.

The Panasonic GF1, with its noisy Micro Four Thirds sensor, posted a slightly lower score than the Pentax K-x, but all of the other comparison cameras (including the non-video Pentax K2000 model) posted higher scores in our long exposure test.