Olympus EVOLT E-330

The second chart illustrates the same test, but plots the colors on a gamut map. Each small square shows where the ideal colors land, and the circles show where the E-330's colors are. The length of the lines between the squares and circles indicates the degree of error in the E-330's color. The closer a spot is to the center of the chart, the less saturated it is. If the line between the circle and the square runs tangent to the center – clockwise or counter-clockwise, rather than in or out – the hue is shifted.

The E-330 gets saturation just about perfect – 100.2 percent saturation is unusually good. Unfortunately, it shifts the colors significantly, earning a high 6.85 color error score. We shot our best image at ISO 100, with noise reduction on.

Still Life Scene
We photographed our colorful still life scene with the EVOLT E-330 and the 14-45mm kit lens under tungsten light. Click on the image below for a link to the full-resolution shot.

Resolution / Sharpness (3.47)
We test resolution by shooting an industry-standard test chart under controlled lighting, with the camera on a sturdy tripod, using the camera's self-timer to limit vibration as much as possible. We shoot the chart at a range of focal lengths and apertures, then analyze the images with Imatest software. Imatest delivers results in "line widths per picture height," a measure that we can use to compare cameras with varying sensor sizes. We report the best results we can achieve.

The E-330 delivered 1432 lw/ph horizontal, with 6.08% under-sharpening and 1413 lw/ph vertical with 6.43% under-sharpening. These results are not very impressive, however, users can eke out a bit more apparent sharpness digitally in post-processing, even with out-of-camera JPEGs.

Noise – Auto ISO (9.46)
The EVOLT E-330's auto ISO noise performance is among the best we have ever recorded, primarily because the E-330 set the ISO to 100, its minimum value.

"Noise" looks a bit like grain in a film photograph or static on a television image. It is basically the variations in tone that the camera's electronics add to the image.

Noise – Manual ISO (11.07)
Noise was a real problem for the EVOLT E-300, the E-330's predecessor. Fortunately, Olympus has made great strides in cleaning up the noise issue with this new camera. The E-330's score is certainly competitive in its category, and, more important, good enough that noise won't be a distraction in prints.  

Image noise in some cameras looks very much like film grain, and, in those cases, can actually be appealing to users who want a gritty feel to their pictures. It might be too much to call this an advantage, but it's a quirk. The EVOLT E-330 doesn't have it. Though its noise is reduced, compared to the E300, it remains blotchy, and looks more electronic than some other cameras' noise.

Low Light Performance (6.0)
We tested the EVOLT E-330 in low light settings ranging from 60 lux, which is enough light to read by comfortably, to 30, 15 and 5 lux. 5 lux is about the light level in a room lit by one of two candles, and is challenging for any digital camera. We set the E-330 to ISO 400 and shot at a range of shutter speeds. We test many DSLRs at ISO 1600, but chose not to do that with the E-330 because Olympus warns that its settings above 400 are an extended range and the manual has reservations about image quality at the highest ISOs.


As the below chart shows, the E-330's noise rises as exposure time increases from 1 to 30 seconds, but it doesn't rise steadily – it jumps between 1 and 5 seconds, remains near steady from 5 to 10 seconds, jumps at 15, and then again at 30. E330 users who plan to work in low light should buy wide-aperture lenses, and try to keep exposures under 1 second.

Dynamic Range (6.0)
The Olympus EVOLT E-330 uses a novel imaging chip: the N-MOS, or negatively-charged CMOS chip, is new to us, so we have been curious about its performance. Judging by the E-330, the technology does not offer a particular advantage as far as dynamic range goes.

When a photographer takes manual light readings, they usually optimize exposure for tones in the middle of the range – not the darkest or lightest. Most subjects fall mainly in the middle range, but scenes very often include bright highlights and dark shadows as well. Dynamic range is the measure of how well a camera can capture detail in highlights and shadows while still handling the middle range well. We test dynamic range by photographing a Stouffer chart: a strip of film showing a row of rectangles ranging from essentially clear to nearly opaque. When the chart is lit from behind, the clear rectangle is 13.3 EV brighter than the darkest one. We analyze the images of the chart with Imatest software, which yields ratings for high and low quality dynamic range. High quality indicates the range the camera can distinguish with only 1/10 of EV of image noise, and low quality indicates the range with up to 1 full EV of noise.

The E-330's score at ISO 100, 7.79 at high quality and 10.6 at low quality, puts it below the middle of the pack for entry-level DSLRs. Its dynamic range drops significantly from ISO 100 to 400, where its performance is significantly below other DSLRs. Interestingly, it doesn't drop much further from ISO 400 to 1600, so, at its highest ISO, it's about average.

Speed / Timing
Start-up to First Shot (7.31)
It took the EVOLT E-330 at least 2.63 seconds to start up and take a shot in our trials. We assume that a fair portion of this delay is due to the dust removal system, which runs every time the camera is switched on. We think there ought to be a way to skip the dust routine, because the long start-up time is unacceptable in a DSLR – speed is one of the DSLR design's most important advantages, and start-up time is a significant part of that. The EVOLT E-330 has a delay more suited to a compact camera than a DSLR.

Shot to Shot Time (9.51)
The EVOLT E-330's specs quote a 3-frames per-second burst rate. We got that rate, but only with the live preview shut off. At to Super High Quality, RAW or TIFF, the camera only manages a burst of 4 frames at a time. With the live preview turned on, the burst mode slowed to about 1 frame per second, but continued indefinitely. A 4-frame buffer really isn't good enough, particularly at 3 fps. Having the burst rate slow to 1 fps with live preview is also very limiting.

Shutter to Shot Time (8.2)
Shutter to shot measures the lag between the instant the shutter is pressed and the instant the picture is actually taken. Our best trial with the EVOLT E-330 was 0.3 seconds without prefocusing or live preview. In live preview "A" mode, the lag increased to 0.4 seconds, and in "B" mode it rose to 0.5 seconds. The increased delay in "B" mode makes sense – the shutter is open, and the main sensor runs the display, so, when the user hits the shutter release, the shutter has to close and then open again for the actual exposure. It's not clear why the delays are so long for "A" mode or normal shooting.

Even 0.3 seconds is far too long for a DSLR – that's too much of a delay for action shooting, and much longer than that of competing cameras.