Fuji FinePix S20 Pro

Accurate color and saturation, with a good exposure as well. High resolution, but high image noise in the shadows.

The extreme tonal range of this image makes it a tough shot for many digicams, which is precisely why I set it up this way, and why I shoot it with no fill flash or reflector to open the shadows. The object is to hold both highlight and shadow detail without producing a "flat" picture with muddy colors, and the Finepix S20 Pro did a great job here.

The shot at right was taken with a +0.3 EV exposure compensation adjustment, less than this shot usually requires. The shot is a bit contrasty, but midtone detail is pretty good. The S20 Pro's Auto, Daylight, and Manual white balance settings all produced great color here, but I chose the Manual setting as the most accurate overall.

Skin tones are appealing, but a bit more pink than in real life. (In actuality, I've found that most people seem to like this look better than a more accurate, neutral rendering, feeling that it looks more "healthy." Each person will have to decide this for themself though. I personally would like just a hint less pink, but given the choice between this and a more technically accurate skin tone, would probably choose this.) The blue flowers in the bouquet look very good as well. (Many digicams have trouble with this blue, often producing strong purple tints, but the S20 Pro gets it just about right.) Color looks great elsewhere in the frame, with good saturation, although the red flowers are quite hot and oversaturated. Resolution is high, with strong detail even in the shadow areas, but image noise is moderately high.

Highlight Detail - The "SR" Advantage?
The S20Pro uses Fuji's "SR" sensor technology, combining normal- and low-sensitivity sensor elements to create a sensor capable of preserving more highlight detail than conventional approaches. The SR technology has been a bit controversial, with some reviewers reporting that they see no difference between it and ordinary CCDs. I go into the reasons for this in my main review, but this shot here is an example of the sort of subject that the SR sensor was specifically made to handle, with enough strong highlight detail to trigger the use of the low-sensitivity sensor data.

To show the difference between the SR technology and a conventional CCD's, I've cropped portions of this image shot with the S20Pro and the Olympus C-8080 and shown the results side by side below.

Note that these were shot within a few minutes of each other, so the lighting is essentially identical between the two of them. The folds of Marti's shirt are positioned slightly differently, so specific areas of the image may not correlate exactly between the two shots. You can definitely see areas with the same shading though, and I've highlighted an area in each shot that not only has the same lighting, but that shows an excellent example of white-on-white detail as well. Note too, that I selected shots that were exposed the same by both cameras, as evidenced by the brightness of the wall behind Marti. - Comparing the same area in both images, the average brightness matched to within two luminance units (197 for the S20Pro vs 195 for the C-8080). This tells us that any differences we might see in the highlight detail is not influenced by the overall exposure level.

In the comparison images below, I first show the crops exactly as they came from the camera, and then after an extreme "levels" adjustment in Adobe Photoshop(tm). (The shadow slider was pushed up to a level of 150.) The purpose of this adjustment was to bring the extreme highlight detail down into a range that's easy to see on a typical CRT. (This is an example of detail that is much more readily apparent in a print than onscreen, on a CRT.) - Click on either image to see a full-sized view (1:1 as it came from the camera).

To view the entire exposure series from zero to +1.0 EV, see files S20OUTMP0.HTM through S20OUTMP3.HTM on the thumbnail index page.

Good resolution with strong detail, but a little noisy. Good exposure and color.

This close-up shot was taken with the S20 Pro's default exposure setting, which resulted in a similar exposure to the wider shot above. Midtone detail is fairly strong, as is detail in the brightest highlights, further evidence of the benefit of Fuji's SR technology. The camera's 6x zoom lens prevents any geometric distortion of Marti's features, a concern in close-up portraits like this one. Overall resolution and detail are much higher here, especially in Marti's face and hair. However, image noise is moderately high, particularly in the shadow areas, which obscures definition of the finer details.

To view the entire exposure series from -0.3 to +0.7 EV, see files S20FACAM1.HTM through S20FACAP2.HTM on the thumbnail index page.

Normal Flash +0.3 EV

Slow-Sync Mode +0.3 EV

Good intensity and coverage with the built-in flash, with nearly accurate color.

The S20 Pro's built-in flash illuminated the subject pretty well at its default exposure setting, although I preferred a +0.3 EV exposure compensation adjustment for the main shot. Overall color is quite good, though the background incandescent lighting results in a slight orange cast on the background, which also spills onto Marti's features and shirt. Still, color is believable and bright, with pretty good accuracy even in the blue flowers. (They're slightly dark and purplish, but still quite good for this shot and the difficult lighting.) The camera's Slow-Sync flash mode also produced good results, though with a stronger orange cast. Even though the exposure is slightly longer here, I still preferred a +0.3 EV exposure compensation adjustment.

To view the entire flash exposure series from -0.7 to +0.7 EV, see files S20INFM2.HTM through S20INFP2.HTM on the thumbnail index page.

To view the same exposure series in Slow-Sync mode, see files S20INFSM2.HTM through S20INFSP2.HTM on the thumbnail index page.

Auto White Balance

Incandescent White Balance

Manual White Balance

Great color with the Manual white balance option (surprisingly good with Auto too), good exposure as well.

This shot is always a very tough test of a camera's white balance capability, given the strong, yellowish color cast of the household incandescent bulbs used for the lighting. The S20 Pro's Auto and Incandescent white balance settings both produced slightly warm-looking images, but were within the realm of what I'd consider acceptable, but the Manual setting produced really excellent results. Overall color is pretty good, if slightly cool, with good skin tones. The blue flowers are dark and purplish, but that's to be expected with this difficult light source. The shot at right was taken with a +1.0 EV exposure compensation adjustment, which is about average for this shot.

To view the exposure series from zero to +1.7 EV, see files S20INTP0.HTM through S20INTP5.HTM on the thumbnail index page.



ISO Series:

The S20 Pro's lowest noise levels are generally higher than on competing cameras, thanks in large part to a minimum ISO value of 200. That said, noise levels aren't that bad at the 400 and 800 settings. At ISO 1,600, the camera averages the data from adjacent pixels to reduce noise in exchange for a smaller image size. The results aren't bad, but image detail is compromised even beyond what you'd expect from the reduced pixel count.

Auto White Balance

Daylight White Balance

Manual White Balance

Accurate color with the Manual white balance setting, and good resolution, though details are a little blocky-looking in the fine foliage.

The S20 Pro's Auto white balance setting produced just slightly warm results here, but the Manual option produced the most accurate white value and overall color. The Daylight setting was warm. Resolution is pretty high for what's essentially a 3 megapixel sensor (there are 6 million active sensors, but they're processed in groups of two, so the net resolution is that of a sensor with 3 million pixel locations), though details in the tree limbs and front shrubbery appear somewhat pixelated and ridden with artifacts. These are the result of the interpolation used with SuperCCD sensors. Oddly, as evident as this is on-screen, prints from SuperCCD cameras tend to look sharper than images from conventional cameras with the same pixel resolution. The image is fairly sharp from corner to corner, with only slight softness visible along the left side.

Decent resolution, but with low definition, and only moderate dynamic range.

This image is shot at infinity to test far-field lens performance. NOTE that this image cannot be directly compared to the other "house" shot, which is a poster, shot in the studio. The rendering of detail in the poster will be very different than in this shot, and color values (and even the presence or absence of leaves on the trees!) will vary in this subject as the seasons progress. In general though, you can evaluate detail in the bricks, shingles and window detail, and in the tree branches against the sky. Compression artifacts are most likely to show in the trim along the edge of the roof, in the bricks, or in the relatively "flat" areas in the windows.

This is my ultimate "resolution shot," given the infinite range of detail in a natural scene like this, and the S20 Pro captures pretty good detail, for a camera having what's basically a 3-megapixel array. However, details in the fine foliage, as well as in the tree limbs over the roof, appear blurry and less defined than the more linear details of the house front. In-camera sharpening is consistent throughout the frame, and there's virtually no falloff in sharpness in the corners, indicating a good-quality lens. The camera picks up moderate detail in the bright white paint surrounding the bay window, which is a difficult area for many digicams, but the shadow area above the front door shows less detail and higher noise. (Fuji's SR technology helps a little in this shot, but the low percentage of strong highlights in the scene as a whole doubtless results in relatively little of the low-sensitivity sensor data being used.) Overall color is very good, with very appropriate saturation levels and accurate hue. The table below shows a standard resolution and quality series, followed by ISO, sharpness, color, and white balance series.

Resolution Series:

ISO Series:

Sharpness Series:

Color Series:

Color Series
Standard Color	Black & White	Chrome

White Balance Series:

Color Series
Auto	Daylight	Manual

Lens Zoom Range

Excellent 6x zoom range.

I routinely shoot this series of images to show the field of view for each camera, with the lens at full wide angle, at maximum telephoto (6x, in this case), and at full telephoto with the digital zoom enabled. The S20 Pro's lens is equivalent to a 35-210mm zoom on a 35mm camera. That corresponds to a moderate wide angle to a pretty substantial telephoto. Following are the results at each zoom setting.

Auto White Balance

Daylight White Balance

Manual White Balance

Near accurate color with the Manual white balance, good resolution and detail as well.

This shot is often a tough test for digicams, as the abundance of blue in the composition frequently tricks white balance systems into producing a warm color balance. The S20 Pro's Manual white balance setting produced the best, most accurate color here, despite a slight magenta tint. The Auto and Daylight settings were tricked by the large amount of blue in the image, and produced warm color casts, but not to any extreme extent. The magenta tint gives the blue background a purplish tint, also noticeable in the deep shadows of the blue robe. Resolution is high, with good detail in the embroidery of the blue robe. Detail is also good in the beaded necklaces and instrument strings, though details are again somewhat fuzzy.

Standard Macro Shot

A VERY tiny macro area, with great detail in the paper bill.

The S20 Pro did exceptionally well in the macro category, capturing a minimum area of only 1.12 x 0.84 inches (28 x 21 millimeters). Resolution is excellent in the printed details of the dollar bill. (The brooch is soft due to the short shooting distance.) There's a lot of softness in all four corners of the frame, but strongest along the left side of the frame. This is unfortunately very common in digicam macro shots, the result of curvature of field at very short focal distances. The very close shooting range also renders the flash ineffective, and the exposure is somewhat uneven from the lens shadow as well. (A ring-light would be in order here, for the best lighting, but plan in any case on using external lighting for macro shots with the S20 Pro.)

Auto White Balance

Daylight White Balance

Manual White Balance

Accurate color, but a slight overexposure and higher than average image noise.

The S20 Pro's Auto and Daylight white balance settings both produced warm images, while the Manual setting produced a more accurate white value and good-looking color. The S20 overexposed the image slightly, but it still managed to distinguish the subtle tonal variations of the pastel swatches in the Q60 target. The bright exposure also left the white color block with a slight halo. (Lens flare?) The large color blocks are just about perfect, but the minor overexposure contributes to a similarly slight reduction in saturation. The shadow area of the charcoal briquettes has moderate detail, but with higher noise than I'd expect from a camera of this caliber. (Why, oh why won't Fuji give us lower ISO settings on their cameras? Many photographers would like the option to trade off ISO speed in order to get lower image noise.)

ISO Series:

Color Series:

Color Series
Standard Color	Black & White	Chrome

Low-Light Tests

Good low-light performance, with good color and clarity even at the darkest light levels.

The S20 Pro produced clear, bright, usable images down to the 1/16 foot-candle (0.67 lux) limit of my test at all four ISO settings. Color was good in most cases, though the dimmer shots had a warm color balance. The S20 Pro handles noise fairly well here, with reasonable levels at the 200 and 400 ISO settings. At ISOs 800 and 1,600, noise is much higher, as you might expect, and at ISO 1,600, the camera forces a smaller file size (1,280 x 960), as it averages data from adjacent pixels to help reduce noise. The table below shows the best exposure I was able to obtain for each of a range of illumination levels. Images in this table (like all sample photos) are untouched, exactly as they came from the camera.

	1fc 11lux	1/2fc 5.5lux	1/4fc 2.7lux	1/8fc 1.3lux	1/16fc 0.67lx
ISO 200	1.3 sec f2.8	2.5 sec f2.8	8 sec f2.8	15 sec f2.8	25 sec f2.8
ISO 400	1/1 sec f2.8	1.3 sec f2.8	4 sec f2.8	8 sec f2.8	13 sec f2.8
ISO 800	1/3 sec f2.8	1/1 sec f2.8	1.5 sec f2.8	4 sec f2.8	6.5 sec f2.8
ISO 1,600 (Small file)	1/6 sec f2.8	1/3 sec f2.8	1/1 sec f2.8	2 sec f2.8	3 sec f2.8

Flash Range Test

A very powerful flash, with good intensity and practically no falloff at the 14 foot limit of our test.

In my testing, the flash illuminated the test target very well, all the way out to 14 feet without any significant decrease in intensity. The flash is very strong, and exposures were almost too bright. Below is the flash range series, with distances from eight to 14 feet from the target.

8 ft	9 ft	10 ft	11 ft	12 ft	13 ft	14 ft
1/125 sec f3.1	1/125 sec f3.1	1/125 sec f3.1	1/125 sec f3.1	1/125 sec f3.1	1/125 sec f3.1	1/125 sec f3.1

ISO-12233 (WG-18) Resolution Test

High resolution (for a 3 million element array), 1,300 lines of "strong detail." Low barrel distortion, but higher than average pincushion.

The S20 Pro performed well on the "laboratory" resolution test chart, considering that it really has only 3 million "resolution elements" in its array. (It has 6 million sensors, but they're paired, so there are actually only 3 million discrete imaging locations on the array surface.) It started showing artifacts in the test patterns at resolutions as low as 750 lines per picture height, in both horizontal and vertical directions. I found "strong detail" out to at least 1,300 lines. "Extinction" of the target patterns didn't occur until about 1,600 lines.

Optical distortion on the S20 Pro is lower than average at the wide-angle end, where I measured approximately 0.5 percent barrel distortion. The telephoto end fared much worse, as I measured 0.7 percent pincushion distortion. Chromatic aberration is quite high, showing nine or more pixels of fairly strong coloration on either side of the target lines. (This distortion is visible as a very slight colored fringe around the objects at the edges of the field of view on the resolution target.) I also noticed some softness in the corners of a few shots, but it was for the most part pretty minimal (with the exception of the macro shot, which had much stronger distortion, not unexpected for an ultra-macro image).

Resolution Series, Wide Angle

Wide Angle	"Fine" JPEG	"Normal" JPEG
2,832 x 2,128	S20PRORESW2832F	S20PRORESW2832N
2,048 x 1,536	S20PRORESW2048F	-
1,600 x 1,200	S20PRORESW1600F	-
1,280 x 960	S20PRORESW1280F	-

Resolution Test, Telephoto

2,832 x 2,128 (Fine, Tele)

S20PROREST2832F

Sharpness Series

Sharpness Series
Soft	Normal	Sharp