Updated 6/25/2010 to offer greater detail on our testing methodology and add sample video clips from each phone to the video comparison section.
If a picture really is worth a thousand words, this summer's camera-equipped smartphone pack should save you several minutes on your monthly voice plan.
The megapixel wars have officially migrated to the phone world: Sprint's HTC Evo 4G and Verizon's Motorola Droid X both sport 8-megapixel cameras, while AT&T's Apple iPhone 4 and T-Mobile's Samsung Galaxy S offer shooters going up to 5 megapixels. All four phones also boast 720p high-definition video capture at 30 frames per second.
But as any camera buff will tell you, megapixel counts and boasts of "HD video capture" rarely mean a thing in terms of performance--especially when it comes to small-sensored camera phones and point-and-shoot cameras. We wanted to see how each of these four superphones performed, as cameras, in the real world.
We ran them through the gauntlet of the PCWorld Labs' subjective testing for still image and video performance. Here's how the Droid X, Evo 4G, iPhone 4, and Samsung Galaxy S stacked up to one another in our formal tests for color accuracy, exposure quality, sharpness, distortion, video quality in bright and low lighting, and audio capture.
How We Tested
With each phone, we used a truncated version of our regular testing methodology for point-and-shoot cameras. We affixed each phone to a tripod and shot two images with the flash turned off. All images were taken using full auto mode with no post-processing, using the maximum resolution for each device:
1. One still-life scene with a color chart and delightful random objects to rate exposure quality and color accuracy. Daylight-balanced 6500k lights were used to light the set. (see example at left).
2. A target chart and printed text to evaluate sharpness and distortion levels (see example below).
We use print-outs of each image to rate each component of image quality. All test images were printed using a Fujifilm Pictrography 3500 Silver Halide Printer, and the printer was recalibrated after printing each set of test shots. A panel of five judges examined each photo and video, then rated each of them independently for color, exposure, sharpness, and distortion.
For video testing, we shot a moving scene twice from a tripod with each phone. While shooting each test clip, we played the same audio clip through speakers to evaluate how well each phone picked up sound. In the first test video clip, we shot in bright indoor lighting. In the second test video clip, we shot with the overhead lights turned off and a floor lamp turned on behind the camera to evaluate low-light footage.
To rate the quality of each clip, the videos were viewed on a 30-inch-diagonal LCD monitor that was calibrated to a color temperature of 6500k using a Spyder calibrator. Each clip was evaluated by the same panel of five judges.