| We've begun including links in our reviews to a Thumber-generated
index page for our test shots. Normally, this page (and our links below)
contain a host of information on various exposure settings, extracted from
the JPEG "EXIF" file headers. It appears that the DSC-F505 doesn't
write this information to its files, so we aren't able to bring you any
of that data. Still, the index page is a good visual guide to ALL the shots
we've posted. |
 |
|
Outdoor Portrait
(520.9 K) Wow! - Let's hear it for Zeiss optics! This shot is generally
the first we look at closely in our analysis, as it combines a wide range
of factors into a single image. (Skin tones, highlights, shadows, strong
primary colors, resolution.) When we looked at this shot, the sharpness
and detail absolutely leapt out at us: Check out the exquisite resolution
in the model's hair. (It really puts the other 2 megapixel cameras to shame,
at least on this particular shot.) Color is very good as well. Automatic
white balance (520.9 K) won out here, producing
a slightly warm tone that we felt was more representative of the true scene
colors than the cooler tones produced by the daylight
white balance setting (519.7 K). Color saturation
is just right, and colors are quite accurate, with the exception of the
blues, which have just a shade of the purplish cast that so many cameras
produce on this shot. Exposure accuracy was about typical for this shot,
with a compensation of +0.5 EV being required for our main
shot (520.9 K). At this level, the strong highlights
in the shirt are just beginning to blow out, but there's exceptional detail
visible in the shadows if you play with the images a bit in Photoshop(tm)
or another image editor. Overall, an excellent performance! The tables below
samples of a range of exposure compensation settings, shot with both Automatic
and Daylight white balance settings.
Exposure compensation series, Daylight White Balance
|
|
||||||||||||
|
Closer Portrait
(514.7 K) Again, excellent color and tone, with accurate, pleasing
skin tones. Absoutely superb detail again, although the variable lighting
of this outdoor shot makes it a little harder to use the model's hair as
a point of comparison. (The F505 shot shows a lot of highlights in the hair
due to the low (late fall) sun angle, that aren't evident in pictures taken
with other cameras earlier in the year.) Still, the overall impression is
of really exceptional sharpness. For our main shot
(514.7 K), we again used +0.5 EV of exposure compensation.
The table below shows the results of other exposure settings, ranging from
0.0 to +1.5 EV.
|
||||||||||||||
|
Indoor Portrait, Flash
(400.2 K) This shot often causes problems for digicams, due to the
rather strong incandescent room lighting it's taken under. The DSC-F505's
flash tube is apparently color-balanced to match household incandescent
unusually well, as evidenced by the complete lack of bluish highlights here.
The F505 offers variable flash intensity, and we chose a shot taken with
the flash in its "low" setting as our main
shot (400.2 K) for this test. Varying the flash
intensity has a fairly subtle effect though, as seen in a version shot with
the "normal" flash setting (410.4
K). The exposure compensation adjustment apparently has no effect
when the flash is enabled, since this shot (408.8
K) looks virtually identical to the prior one, shot with no EV adjustment
dialed-in. |
||||||||||||||
|
|
Indoor Portrait, No
Flash (431.6 K) This scene is a very tough
test of a camera's white balance capabilities, given the strong yellow cast
of the household incandescent lighting it's shot under. The DSC-F505's white
balance system worked exceptionally well in its incandescent
setting (431.6 K) though, producing an unusually
well-balanced picture. Other white balance modes performed adequately, at
least when compared to the rest of the current (December, 1999) digicam
field. Here's a slightly reddish sample (434.5
K) shot with automatic white balance, and a slightly
yellowish version (431.2 K), captured with the
One-Push white balance setting. As usual, this shot required a moderate
amount of exposure compensation, and we chose the version with +1.0EV of
adjustment as our main shot for the category. The table below contains samples
shot with incandescent white balance and other exposure adjustment settings
ranging from 0.0 to +1.5 EV.
|
||||||||||||||
|
House Poster - House
1 (643.7 K), House
2 (840.6 K) Our test of the DSC-F505 marks
the beginning of our transition to a new "House" poster, as high-end
prosumer cameras are beginning to approach the limits of the previous target.
(The new poster is printed photographically at a continuous-tone resolution
of 304.8 dpi, from a ~400 megabyte file scanned from a 4x5 film transparency
- More than enough resolution to last through several more years of consumer
camera evolution!) for some time to come, we'll be shooting test images
of the old poster at high resolution, to provide a basis of comparison with
previously-tested cameras. We'll reserve our full resolution/quality series
for the new poster though. For purposes of comparison with previously-tested
cameras, here's a shot of the original House poster
(643.7 K).
Looking at either the old (643.7 K) or new (840.6 K) House tests, the impact of the Zeiss optics used in the F505 is again apparent. Detail is exceptional, clearly among the best we've seen. (Look at the detail of the branches against the sky, above the central gable of the house.) Color is very good as well, although we do see some noise in the relatively flat tints of the shingles on the roof. Some higher-resolution digicams don't do so well when shooting at lower
resolutions, but we found low resolution shots
taken with the F505 to be exceptionally clean and detailed. (They could
do with some in-camera sharpening though, as they're a little on the soft
side. When you sharpen them after the fact in Photoshop(tm) or other imaging
program though, the detail is incredible, given their modest resolution.)
The table below contains links to a standard resolution/quality series,
showing samples of all resolutions and quality settings the camera produces.
(With the exception of the "email" setting, which we neglected
to test.)
To show the effect of different white balance settings, we also shot this target in low resolution, with each of three "daylight-appropriate" white balance options: Automatic, Daylight, and One-Push. The results are shown in the table below. We felt that the Automatic setting produced the most accurate results, with the Daylight option coming out ever so slightly more yellowish, and the One-Push setting slightly cooler in hue.
|
||||||||||||||
|
Far Field Shot
(787.6K) 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.
Excellent color and tone here, but again the big story is the exceptional sharpness of the DSC-F505's lens. Right as it comes from the camera, this shot (787.6K) is at the very top of the field of current (December, 1999) 2 megapixel digicams. We found that we could go even further though (albeit by a small increment) by judiciously sharpening further in Photoshop(tm). (FWIW, we felt that unsharp masking settings of 0.4 pixels and 100% produced about the best on-screen results on a large/fine image. - Try it on a downloaded image, and see for yourself.) No question, this is the sharpest 2 megapixel digicam on the market! (IOHO, anyway.)
|
||||||||||||||
|
Lens Zoom Range In response to reader requests, we
now routinely shoot images showing the range of each camera's zoom lens.
The table below links to Standard-resolution images (to save download time),
shot at wide angle, telephoto, and telephoto with the digital tele option
engaged. Again, the F505's lens is the big news, with a long 5x zoom ratio,
ranging from 35mm focal-length equivalents of 38 to 190mm, stretching to
380mm in digital zoom mode. (Note that the shots below taken at the 640x480
resolution make the digital zoom look like it's working the same as a true
optical zoom would: At higher resolutions, it would stretch the same image
data shown here across the increased pixel area of the larger file size,
reducing sharpness...)
|
||||||||||||||
|
Musicians Poster
(689.3 K) For what seems to be a fairly straightforward
image, this shot causes surprising problems for many digicams, producing
color casts in cameras that have no trouble whatsoever with our other tests.
The F505 somewhat fell prey to this, in that the auto
white balance setting (689.7 K) produced images
with a very bluish cast. The daylight setting
(689.3 K) gave very good results, but is slightly
on the warm side. We initiallyselected the One-Push
white balance setting (675.5 K) for our main shot,
but ultimately decided we preferred the warmer skin tones of the daylight
sample shown above. As before, resolution is excellent, perhaps being somewhat
limited by the poster itself here. (Finest detail is evident in the Oriental
model's robe, in the bird's wings and the nearby fine silver threads.) Once
beyond the curve-ball thrown by this poster to the camera's white balance
system, the results were excellent. The table below contains the usual array
of resolution/quality samples, all shot using the One-Push white balance
setting.
|
||||||||||||||
 |
Macro Shot
(455.9 K) Like a number of digicams we've tested, the DSC-F505 achieves
its best macro performance with the lens set to its widest-angle focal length.
This results in less magnification than the 5x zoom and 3.14 inch (8 cm)
minimum working distance would suggest. We also found fairly evident barrel
distortion in macro mode, which you can see in the macro
test image (455.9 K). Still, macro performance
was about in the middle of the pack, with a minimum capture area of 2.7
x 3.5 inches (67.5 x 90.0 mm). In flash mode, the short working distance
and location of the flash head produced a significant shadowing of the subject
area, as shown here (522.9 K).
This problem was somewhat reduced by using a piece of white paper as a diffuser
between the flash head and subject, as seen here
(572.1 K). Both macro flash shots are somewhat dark,
as the subject and flash positioning apparently fooled the flash exposure
sensor into shutting down the flash a bit early. |
||||||||||||||
 |
"Davebox"
Test Target (402.4 K) If there's anything
amiss with a camera's color rendition, this test will smoke it out! This
test further demonstrates the excellent color-handling of the DSC-F505.
All colors are unusually accurate, the only criticism we can find anywhere
being that the bright yellow is just slightly weak. Tonal range is also
excellent, with good detail in both the highlights and shadows. In this
shot, it was a very close call as to the most accurate white balance setting,
Automatic 402.4 K) winning
out by a nose over One-Push (92.7
K), which had a very slight cool cast to it. The Daylight
setting (94.8 K) produced a somewhat warmer tone.
As before, we've shot a full range of resolution/quality combinations and
put them into the table below, all shot with the Automatic white balance
option selected.
Resolution/Quality series
|
||||||||||||||
|
Low Light Tests With its exceptional capability in other areas, we were surprised that the DSC-F505 didn't fare better in low-light conditions. While the Night Mode Plus exposure setting does apparently increase CCD sensitivity somewhat, the camera was only barely able to get down to the 2 foot-candle (22 lux) light level, a full f-stop (EV value) brighter than what we consider necessary for good outdoor city-scene night photography. Color was quite accurate anywhere the camera could capture a usable picture, but the F505 wouldn't be the first choice for photographers interested in extensive night shooting. The table below shows the results we obtained with the DSC-F505 at light levels ranging from 8 foot-candles down to 0. 5 foot–candles (10EV to 6EV). |
||||||||||||||
|
Flash Range Test (New) (This test was added in August 1999, so cameras tested before that time won't have comparison pictures available. As we go forward though, all the new models will have similar tests available.) Sony specifies the maximum range of the onboard flash at 8.3 feet (2.5 meters). This agreed well with the results of our own tests, which showed the flash being quite bright out to 9 feet, falling off slowly after that. The table below shows the results we obtained at various distances, with the lens set to the telephoto end of its range.
|
||||||||||||||
 |
ISO-12233 (WG-18) Resolution
Test (398.7k) We've commented extensively
on the apparent sharpness of the DSC-F505's lens in our other tests, and
the resolution target results support our earlier observations. As is often
the case in this test, we can easily see detail reaching far beyond the
theoretical limits of the camera. The reason for this is that the camera
can in fact display detail at higher frequencies, but begins to introduce
"aliasing" once you get beyond the point at which theory says
it should run out of steam. In the case of the 'F505, we could easily see
detail out to 800 lines per picture height, in both horizontal and vertical
directions, but will "call" it at 700 lines per picture height,
as that's the point that roughly corresponds to the level of aliasing we've
been accepting in other cameras we've tested in the past. While (a few)
2 megapixel digicams have reached this level previously, the DSC-F505's
images look the most crisp of any we've seen to date (December, 1999) at
this resolution level. (Interestingly, the aliasing is much more evident
in the diagonal test elements than in the vertical and horizontal ones,
the opposite of what we usually find. This may help account for the very
high perceived resolution in our other tests, which tend to have detail
oriented along the vertical and horizontal axes.) We do observe though,
that the exceptional sharpness comes at the cost of some color aliasing
in the test patterns at very high spatial frequencies. (This effect could
conceivably also appear in shots of things like herringbone tweed fabrics,
etc.) Overall, a very impressive performance! As usual, the tables below
contain links to test images shot with the full range of resolution and
quality settings, at both wide angle and telephoto focal lengths.
|
||||||||||||||
 |
Viewfinder Accuracy/Flash Uniformity As
noted in the main review, the DSC-F505 lacks an optical viewfinder. Consequently,
the measurements here are for the LCD viewfinder only. The LCD is a bit
more accurate than most, ranging from a coverage of 89.5% of the final image
area at the telephoto end (184k)
of the lens' range to 93.5% at the wide angle end
(172k). In actuality, the coverage is likely identical,
the variation caused by the change in geometric distortion as we moved from
wide to telephoto settings on the zoom lens. (See below.) Flash uniformity
was very good at all focal lengths, unusually so at the wide-angle end. We now routinely measure lens distortion as part of our camera testing. The DSC-F505 showed some distortion at the ends of the lens' range, perhaps due in part to the difficulty of optimizing a zoom lens across a wide focal length ratio. At the widest angle setting, barrel distortion is a fairly noticeable 0.9% (far from the worst we've seen, but definitely noticeable if you have straight lines near the frame's edge). At maximum telephoto, the distortion switches to pincushion, albeit with a somewhat less noticeable magnitude of 0.6%. Chromatic aberration (observed as colored fringes around the edges of elements in our resolution test target near the extreme edges of the frame) is visible, but fairly minor, estimated at about a half-pixel, or 0.03%. There is some "coma" though (blurring of high-contrast edges) in the extreme corners, perhaps as much as 0.25%. (The coma is only evident at the extreme edges, vanishing completely by the time you're as little as 12% in from the edges of the frame.) |
Back to Sony DSC-F505 Review
Jump to Comparometer(tm) to compare with other cameras
Or, up to Imaging Resource Cameras Page.
Follow Imaging Resource: