Samsung GX 1S

Color (9.97)
We test color accuracy in a standardized process: shooting an industry-standard GretagMacbeth color chart under tungsten light. We shoot at a variety of settings to discover each camera's potential, and run the resulting images through Imatest software's color testing package to evaluate accuracy. Imatest creates two graphics: the first is a composite based on the camera's image of the GretagMacbeth chart, while the second is based on a color-space diagram.

On the first chart, the large background squares are the colors as the GX-1S recorded them. The vertical rectangles inside those squares are the colors as they should be reproduced, and the small squares are the ideal tones, corrected only for luminance.

GX-1S - Bright Color Mode

GX-1S - Natural Color Mode

The second graphic shows a circle and square for each color. The square shows the ideal color, the circle shows the GX-1S's rendition, and the distance between the two shows how much error there is.

GX-1S - Bright Color Mode

GX-1S - Natural Color Mode

The GX-1S's Natural mode is much more accurate than its Bright mode, but it still over-saturates color. In Natural mode, the GX-1S oversaturated by 109.1 percent, and showed a 5.74 mean color error. The color error score is comparatively good, but many DSLRs have saturation scores within a few percentage points of the perfect 100. In Bright mode, the GX-1S over-saturated 127.5 percent and had a mean color error of 6.33. The Bright mode scores are less accurate than many point-and-shoots, and create images that are highly unrealistic and impossible to edit. In both modes, the GX-1S oversaturated warm tones in particular, which ought to yield very rosy complexions and bright flowers.

Still Life Scene
We have provided two full-resolution shots of our much-loved still-life scene, as captured by the Samsung GX-1S in its two respective color modes.

GX-1S - Bright Color Mode

GX-1S - Natural Color Mode

Click on the images above to view the full resolution files.

Resolution / Sharpness (3.21)
A digital camera's resolution isn't simply a matter of how many pixels it records to a file. Images made of pixels can be, and to some extent always are, blurry. No camera delivers the maximum resolution theoretically possible for its pixel count.

To test how close cameras come to the ideal, we photograph an industry-standard resolution chart. We use a sturdy tripod and bright lights, testing cameras at a variety of focal lengths and apertures to get the best performance each camera can deliver. After analyzing the images with Imatest software, , a leading package for testing camera performance, we report our results in line widths per picture height, (lw/ph). This measure is independent of the sensor’s physical dimensions , allowing the direct comparison of any two digital cameras, regardless of sensor size.

Click on the image above to view the full res version.

The Samsung GX-1S achieved its best results at ISO 200, with the lens set at 55mm and f/11. The GX-1S's image delivered 1253 lw/ph horizontally, with 0.403 percent oversharpening, and 1458 lw/ph vertically, with 7.32 percent oversharpening.

The GX-1S's results are considerably worse than other 6-megapixel DSLRs, delivering about half the resolution such a file size could theoretically record. In practical terms, the GX-1S's images shouldn't look sharp as 8x10-inch prints, while shots from better-performing 6-megapixel DSLRs will look excellent.

Noise – Auto ISO (8.23)
"Noise" is texture in an image which isn't part of the subject matter. In film photos, it's called grain; in bad television reception it's called static; and, in all media, it detracts from a detailed rendering of the subject The GX-1S performed very well in our noise test for the Auto ISO setting, offering almost exactly the noise performance the camera would deliver if set manually to ISO 200.The GX-1S's option for setting an upper limit on the ISO in auto is responsible.

Noise – Manual ISO (12.32)
We tested the GX-1S for noise at each available ISO setting.  The results are plotted in the graph below; the horizontal axis displays the ISO settings, while the vertical axis indicates the corresponding noise levels.

The GX-1S delivered very good noise performance in manual mode as well. When the image mode is set to "Natural," rather than "Bright," the GX-1S surpasses the performance of the similar Pentax *ist DL. The two cameras' noise curves are very similar. Notably, the GX-1S's noise level increases gradually at higher ISOs – there isn't a single point at which the GX-1S suddenly deteriorates precipitously.

Low Light Performance (6.5)
Our low-light testing involves taking long exposures at high ISOs. We run the tests at 60, 30, 15 and 5 lux. 60 lux corresponds to subdued indoor lighting. 30 lux is dim, like a living room lit with a single 40-watt bulb. 5 lux is barely light enough to see. We look at both noise and color performance in low light and test cameras' noise-reduction routines in the process.

The GX-1S has an effective noise reduction system. With it, the GX-1S performed well – far better than without. Noise reduction didn't make a difference with 2-second exposures, but it kicked in at exposures of 4 to 6 seconds. At 10 seconds, it cut noise in half. At 30 seconds, an exposure without noise reduction was 2.5 times noisier than and had almost twice the color error of an image taken with noise reduction. Any noise reduction system will decrease resolution, however, and the GX-1S has relatively poor resolution to start with. Still, the improvement in color and noise means that shooting the GX-1S without noise reduction is not an option.

Dynamic Range (7.25)
Many photographs include both bright and dark subjects. Even with film cameras, it was a challenge to record the full range of light and dark that many scenes present. Unfortunately, digital cameras have not solved the problem. In fact, the failure to capture a full range of tones is one of the most common problems with digital photos. The span from light to dark is called dynamic range, and we use Imatest software to compare how well cameras capture it. We photograph a standard Stouffer test target with a very large dynamic range. Imatest analysis shows us how wide a range of tones the camera catches at various quality levels. We track High Quality, which measures the range with noise levels no higher than 1/10 of an EV, and Low Quality, which measures the range with noise up to 1 whole EV.

The GX-1S did a good job on our dynamic range test. At ISO 200, its Low-Quality result, 11.4 EV, is the best we have recorded. The more significant ISO 200 High-Quality result, 6.92 EV, was more typical, but still very good for an entry-level camera. At high ISOs, the GX-1S's range doesn't match the best high-end DSLRs, but it’s better than its price competition.

GX-1S Dynamic Range - ISO 200

GX-1S Dynamic Range - ISO 3200

Speed / Timing
Start-up to First Shot (9.02)
Our best speed, from turning on the GX-1S to getting an image, was 0.9 seconds. That's a bit slow for a DSLR. Most turn in results of around half a second.

Shot to Shot Time (9.24)
The GX-1S shot 8 JPEG frames at 2.65 frames per second. That's slow. Most sub-$1000 DSLRs shoot at about 3 frames per second, so the GX-1S doesn't compare well. Worse, two and a half frames per second won't keep up with sports action or the more active kinds of snap shooting. This speed is a real limitation.

Shutter to Shot Time (8.08)
The GX-1S was also slow to get off shots as the shutter was pressed. Our typical delay was 0.4 seconds when the camera wasn't prefocused. When we prefocused, the GX-1S still disappointed: our typical delay remained 0.17 seconds, much slower than other DSLRs.