Sony A200

The A200 stood up well to our extensive battery of laboratory tests. Granted, there was no performance breakthrough here that set our hearts aflutter, but more to the point for an inexpensive camera like this, we didn't find any deal-breaking flaws. We were pleased to see that even penny-pinchers can take photos with good image quality with the A200.

Color (8.34)
While there's a widespread fixation with how many megapixels a camera can boast, color reproduction is actually the quality in a photo that draws the most attention as people paw through your prints. To see how well an individual camera stacks up against the competition in this regard, we run lab tests shooting an industry-standard Gretag Macbeth color chart under precisely controlled lighting conditions. The resulting images are then processed using Imatest image analysis software, which can determine even the most subtle differences between the known values of the color squares on the chart, and the color information recorded by the camera.

The ideal color in this test result represents the original color,
the outer rectangle the camera's rendition.

The A200 performed well when it comes to color accuracy, especially for a camera at such a low price. In the chart above, the original color from the Gregtag Macbeth chart is shown by the small rectangle at the right of each square patch.  The color captured by the camera is shown by the large outer rectangle, while the captured color with luminance correction applied is shown by the inner square. As you can see, no glaring disparities appear anywhere in the color spectrum. An additional Imatest chart represents the reported color shift graphically:

The squares indicate the original color chart,
the circles what the A200 captured.

Resolution (9.33)
The ability to capture fine details is clearly another major factor in selecting the right camera for your needs. This is especially true with today's high-megapixel cameras, which enable you to blow up a small section of the recorded image into a large print. However, the number of megapixels doesn't necessarily correlate with useful resolution in the photos you take. To find out the true resolution performance of a camera we take it into the lab and take hundreds of test shots of a standard resolution chart, trying out the full gamut of camera settings. Then we fire up Imatest software to determine how many alternating horizontal and vertical lines the camera actually captured (measured in line widths per picture height, or lw/ph).

The industry-standard resolution chart that we use for testing

The A200 turned in decent results in this test, not quite on a par with the Nikon D60 and Canon XSi, but within shooting distance of these models. The maximum horizontal resolution measurement was 1803 lines, with slight undersharpening, with the lens set to a wide angle.  Performance was reasonably consistent as we zoomed in, though it did fall off noticeably at the longer telephoto range.

Noise – Manual ISO (10.60)
Visual noise is kind of like the noisy static you hear when the radio's tuned between channels – random irritating speckles caused by electrical glitches. In your photos this translates into blotches in areas of flat color and bright spots in dark portions of an image. Noise is particularly prevalent in photos taken in low-light conditions, and gets progressively worse as ISO settings increase.

To test a camera's noise performance, we photograph the Gregtag Macbeth color chart at the full range of supported ISO settings, then analyze the results using Imatest software looking both for image noise at particular settings and for trends that arise as the ISO gets higher.

In our manual settings test, noise levels started out at barely over half a percent and rose at a modest rate. Interestingly, the choice to turn noise reduction on has essentially no effect at all until you hit ISOs above 800, when the imaging processing software in the camera kicks in and makes a major difference.

The A200 results for noise were roughly on par with expectations. Nikons are known for their low-noise prowess, and the Sony clearly couldn't measure up to that standard, but the Sony did outscore the Canon XSi in the manual settings test.

Auto Noise (4.34)

We run a separate test to gauge noise levels with the automatic ISO setting engaged. The A200 did relatively poorly here. Even  under our bright studio lights, the camera chose an ISO setting of 400, where most cameras we test shoot at 200 with the same lighting. Higher ISO settings, more noise.

White Balance (7.87)
The human mind does a great job adjusting for the different colors of light in our environment, basically filtering what the eye sees so the illumination looks white except under extreme circumstances. In fact, though, the color of light from the sun is vastly different from that cast by a household lightbulb or a fluorescent tube. A digital camera has to adjust to these differences to produce photos that look the way we remember the scenes we shoot – hence, white balance controls. We test this by using a variety of illumination sources to shoot a standard color chart and using Imatest to evaluate the results. We run two separate batteries of tests for white balance, one with the camera's automatic setting, the other using each of the system's preset values under the designated light source.

Auto White Balance (4.82)
In the Imatest chart below we see the same color pattern layout as the color test results above: the color the camera captured is in the outer rectangle, the color corrected for luminance is shown in the center square, and the ideal color appears in the rectangular inset to the right. Keep in mind that this chart presents exaggerated differences to make comparison easier: you wouldn't see this level of white balance shift in an actual photo you snapped.

The overall white balance performance for the A200 is a decidedly mixed bag. The automatic settings had trouble with a variety of light sources, particularly shaded daylight and fluorescent illumination. Surprisingly, it handled tungsten illumination very well, which is a stumbling point for most of the cameras we test.

Preset (10.93)
The preset controls, on the other hand, produced very good results across the board, with flash and fluorescent lighting two notable standouts. In fact, based on these results, we'd advise taking the time to set the white balance manually whenever you're taking flash photos.
 

Exaggerated White Balance Errors (Presets)

Flash Preset WB - Flash Illumination

Florescent Preset WB - Fluorescent Illumination

Cloudy Daylight preset WB - Daylight illumination

Tungsten preset WB - Tungsten illumination

 
Still Life
To provide you with a useful comparison both of a single camera across the full range of ISO settings and of different cameras shooting the same scene from review to review, we include the following still life photos in our testing regimen. To view the full-resolution original, click on an image. These are large files, though (some more than 4MB), so they may take some time to download.

Low Light (8.32)
Shooting indoors without flash has become much more practical recently, as the sensitivity of sensors in even inexpensive cameras has increased and noise reduction technologies have improved. This is still an area, though, where cameras vary pretty widely, and differences are readily apparent in the final image produced. That's why we subject our cameras to two separate tests for low light performance. The key criteria: color accuracy as light levels drop, and the increase in noise as exposure times lengthen.

The A200 performed well in low-light environments, though two peculiarities did arise. First, we found that when shooting at 1 second under low-light conditions, the metering system produced grossly inaccurate readings (off by three or four stops), which threw off our initial results. We worked around the problem by setting the values manually, and it only occurred at that particular setting (though repeatedly), which probably indicates a software inconsistency of some kind. The second oddity: turning on noise reduction had virtually no impact on measured levels of noise. This is not a critical problem, though, since overall the A200 exhibited low-light noise levels that were perfectly acceptable across the board.

For the first test we maintain a constant light sensitivity setting of 1600 ISO but vary the light level, from 60 lux (roughly what you'd find in a normal room indoors) down to 5 lux (about the light level a single candle throws off in a darkened room).

As shown above, colors didn't drift out of whack badly when shooting in low light, even down to 5 lux, and the exposures are acceptably even. The chart below indicates how noise performance varies as exposure time increases. The levels here are pretty good overall, with only modest variation from a short exposure out to 30 seconds. As for the fact that turning the noise reduction circuitry on has nearly imperceptible effects on performance, it could be worse - we've seen cameras where noise reduction produced increased noise at certain exposure settings.

When compared to other cameras we've tested, low light performance for the A200 is very competitive, with only modest differences between this camera and most of the others in the group. Again, the Canon XSi fared particularly well in our recent lab testing, and outperformed the others handily.

Dynamic range performance for the A200 was slightly better than the Nikon D60 or Pentax K10D. It lagged the Canon XSi by a wide margin, but the Canon proved to be an outstanding performer in this area.