Fujifilm FinePix F50fd

Color (7.97)
Every camera reproduces colors differently, from the shades of a blue sky to skin tones. Sometimes cameras oversaturate colors, making them more vibrant but less natural, or undersaturate colors, making them appear muted or dull. We test cameras’ color accuracy by photographing an industry standard GretagMacbeth ColorChecker test chart and comparing the colors the camera reproduces with the colors of the test chart. The ColorChecker is made up of 24 color tiles, representing colors that vary in both hue and saturation. The image below shows how well the Fuji F50fd’s colors match up with the ideal colors. The outside squares show the colors the camera reproduces, the inside squares show the ideal color of the chart corrected for luminance, and the inner rectangles show the ideal chart colors under a perfectly even exposure.

As you can see in the image, a number of the F50fd’s colors stray from the ideal colors in both hue and saturation. In particular, notice how different the inside square is from the outside square in the yellow and blue tiles in the third row. The camera’s colors are quite undersaturated. The graph below shows this same information in a different way. The background of the graph shows the entire color spectrum. The ColorChecker’s colors are represented by squares and the camera’s colors by circles. The length of the lines connecting the squares and circles shows the amount of color error.

The white circle in the center of the chart indicates an almost perfect manual white balance, but the accuracy of the colors surrounding it is less than perfect. Blues and yellows are particularly inaccurate. Overall, there is a trend toward undersaturation, making colors appear muted and dull, as we saw in the image above. This will have a noticeable impact on your photos, such as a shot of a sunset. The sunset’s yellows will be considerably dulled, and blue sky will have a purple tinge. We’ve seen much worse color accuracy than this, but we’ve also seen much better.

Resolution (14.93)
With a whopping 12.1 megapixels, the Fuji F50fd is one of the least expensive 12-megapixel cameras released to date. We test camera resolution by photographing an industry standard resolution test chart at varied focal lengths, apertures, and shutter speeds. We then run the images through Imatest to determine the camera’s best possible resolution. Imatest measures resolution in units of line widths per picture height (lw/ph), which refer to the number of alternating black and white lines that can fit in the picture frame before becoming blurred.

Click to view the high resolution image

Other high-resolution cameras we tested this year have been impressive, but the resolution of the Fuji F50fd is truly astounding. The camera resolves 2776 lw/ph horizontally with 6.6 percent oversharpening, and 2420 lw/ph vertically with 5.9 percent undersharpening. Not only are these the highest lw/ph values we’ve ever seen, but the sharpening levels are just right, so there isn’t much image artifacting. Clicking on the resolution chart photo above brings up the full resolution version, where you will notice that every line on the chart is sharply resolved, even the long, thin trumpets. The one problem with the F50fd’s resolution is that it isn’t maintained over the entire frame. As you can see in the image above, the right edge of the photo is blurred and shows signs of chromatic aberration. Photos taken with this camera will stand up to extensive cropping and enlarging, but beware of blurry edges. In terms of resolution, the F50fd is leagues ahead of any other point-and-shoot we have tested this year, including much pricier models such as the Canon PowerShot G9.

Noise – Manual ISO (5.42)
A drawback of increasing the amount of megapixels is more image noise. Image noise appears as sandy grains or splotchy patches, scattered randomly around a digital image. Noise becomes much more apparent in low light or at high ISO sensitivities. More megapixels packed on a sensor of the same size means the pixels are smaller, and smaller pixels lead to more noise. We test noise levels by photographing our test chart under bright studio lights at all ISO speeds a camera offers. We then run the photos through Imatest, which determines noise levels by the percentage of image detail the noise drowns out. The graph below shows the F50fd’s noise levels throughout its ISO range.

Noise levels are manageable at ISO 100, but images are very noisy above that. At ISO 800 and 1600, the noise is overwhelming. Close inspection reveals the noise is extremely splotchy and jagged, and contains random patches of purple, yellow, and blue. It is apparent in smooth tones at ISO speeds as low as 200 (see the still life images further down the page). Overall, the F50fd is slightly noisier than average, but significantly noisier than its predecessor, the F40fd. This shows the tradeoffs Fuji had to make when creating a high-resolution camera.

Noise – Auto ISO (1.29)
We also evaluate noise levels with cameras set to Auto ISO under the same bright studio lights. Under our studio lights, the F50fd chooses ISO 400, which results in a lot of noise and a poor Auto ISO score. This is a camera you will want to keep at ISO 100 as often as possible.

White Balance (4.96)
Every type of light source, from fluorescent lights to broad daylight, has a different color cast. The human eye automatically adjusts for this, and cameras must too. For cameras, this is called white balancing, and the Fuji F50fd has three methods for it: Auto white balance, white balance presets, and Manual white balance. Manual white balance is usually the most accurate, but requires the use of a white or gray card. We test cameras’ Auto white balance and presets by photographing the ColorChecker test chart under four types of light: flash, fluorescent, outdoor shade, and tungsten.

Auto (3.84)
Using the Auto setting, the camera does a mediocre job white balancing under fluorescent light, and a poor job under flash, outdoor shade, and tungsten. These results send a clear message: avoid using the F50fd’s Auto white balance.


Preset (6.09)
The camera is quite accurate under fluorescent light (using the "Fluorescent 3" setting under our white fluorescent lights), and mediocre under outdoor shade and tungsten. The presets’ performance isn’t stellar, but it’s a lot better than using the Auto setting.

Fluorescent WB - Fluorescent Illumination

Shade WB - Shade Illumination

Tungsten WB - Tungsten Illumiation

Still Life Sequences
Click to view the high resolution images.

Still Life Scene
ISO 100	ISO 100
ISO 200	ISO 200
ISO 400	ISO 400
ISO 800	ISO 800
ISO 1600	ISO 1600

Low Light (5.61)
Consumers are often confronted with less-than-ideal lighting situations, which is why we test low light performance. We photograph the ColorChecker test chart at light levels of 60, 30, 15, and 5 lux, and analyze color accuracy and noise levels. Sixty lux represents about the amount of light in a living room lit with two soft lamps, 30 lux is about the amount of light in a room lit by a single 40-watt bulb, 15 lux is how bright a room gets with a large television on, and 5 lux is as dim as a closet illuminated by the light of an MP3 player’s LCD screen. All shots are taken at ISO 1600 to find the limits of the sensor.

Low Light Tests
60 Lux	30 Lux
15 Lux	5 Lux

The F50fd can’t properly expose below 30 lux in Manual mode. Only when the shutter speed is manually adjusted in Shutter Priority mode can the camera obtain an accurate exposure. Also, color accuracy suffers in low light, becoming significantly oversaturated. However, this is better than the undersaturation we saw in bright light in the Color section above. Noise levels are extremely high, suggesting users should avoid ISO 800 or 1600 on this camera whenever possible.

We test long exposure performance in low light with the camera set to ISO 400 to standardize our testing. The F50fd has a long shutter speed option in the Setup menu, and turning this on unlocks shutter speeds as long as 8 seconds, but only in Night mode. The problem with this is Night mode can only be used at ISO 100. The longest shutter speed you can get at ISO 400 is 1 second, where the camera has significant color error and fairly high noise levels. Overall, low light performance is below average.

Dynamic Range (4.36)
Dynamic range, which describes the amount of gray shades a camera can discern, is an important image quality factor. A camera with good dynamic range can detect many shades of gray, preserving details in both bright highlights and black shadows. Good dynamic range helps prevent highlights from blowing out and shadows from becoming entirely black. We test dynamic range by photographing a backlit Stouffer test chart, which is made up of a long row of rectangles, each a slightly darker shade of gray, ranging from brightest white to darkest black.

The Fuji F50fd has very disappointing dynamic range, even at its lowest ISO speed setting, ISO 100. At this setting, the Fuji F50fd’s dynamic range is comparable to what most point-and-shoots produce at ISO 200 or above. At higher ISO speeds, the F50fd’s dynamic range drops even lower. This camera blows out highlights and loses detail in shadows, which hinders shooting in high contrast scenes. The poor dynamic range is due in large part to the camera’s high noise levels, and shows yet another tradeoff for its amazing resolution.

Speed/Timing – All speed tests were conducted using a Kingston Ultimate 120X 2GB SD Card, with the camera set to highest resolution and best quality unless otherwise noted.

Startup to First Shot (7.4)
The F50fd takes 2.4 seconds to turn on and snap its first shot.

Shot-to-Shot (9.5)
The F50fd has a plethora of Burst modes: Final 12, Top 12, Long Period, Final 3, and Top 3. Final 12 and Top 12 both fire only at a resolution of 3 megapixels. Final 12 takes shots every 0.2 seconds for 8 seconds and saves the last 12 images. Top 12 takes 12 shots every 0.2 seconds, lasting 2.1 seconds total. Long Period shoots full resolution photos every 2.8 seconds until the card fills. Final 3 takes full resolution shots every 0.5 seconds for 20.5 seconds, only storing the last three photos. And finally, Top 3 takes a quick burst of three shots, each 0.5 seconds apart.

Shutter-shot (9.0)
The camera has no measurable lag time when prefocused, but a 0.4-second lag when not prefocused.

Processing (3.0)
The camera takes a very leisurely 3.5 seconds to process one 4.5 MB full resolution fine quality photo shot at ISO 100.

Video Performance (2.10)
Bright Indoor Light – 3000 lux
We evaluate Movie modes’ color accuracy and noise levels by recording footage of our color charts under bright studio lights set to 3000 lux. In tungsten lights, the F50fd’s video has tremendous color error, and white balance can’t be adjusted. Noise levels, however, are nice and low.

Low Light – 30 lux
The camera has extreme color error in low light, as well, in large part because it can’t properly expose at 30 lux. Noise levels are incredibly high. This isn’t a camera you’ll be able to use to take videos at night or in a dark nightclub.

Resolution
We recorded footage of our resolution test chart at 1700 lux to test the Movie mode’s sharpness. The camera resolves 260 lw/ph horizontally with 21.9 percent undersharpening, and 343 lw/ph vertically with 15.8 percent undersharpening. These are mediocre numbers for a digital camera’s Movie mode. The crops below show how underexposed the footage is, as well as the ugly color cast.


Motion
We record footage of moving cars and pedestrians on the street to see how cameras render motion. The video of the F50fd has very jerky motion, streaky highlights, and terrible moiré problems. The Movie mode on this camera may be fun to fool around with, but does not have many practical uses.