Fuji S2 Pro
You
can find excellent complete reviews of the S2 on the web sites referenced on my
links page. This article just
covers some of the fine points that come from extensive use of the camera.
The
official name of this camera is “Fuji FinePix S2 Pro,” but I hate that
“FinePix” moniker. It always
makes me think of Radio Shack’s “Realistic” brand name – just a little
bit overstated and ironically unimpressive.
Nevertheless, the camera is widely regarded as having the best image
quality of the various 6 megapixel digital SLRs.
This originally meant the Nikon D100 and Canon D60, but undoubtedly still
applies to the much expanded group: Canon
10D and 300D, Nikon D70, Pentax D*ist, Sigma S9 and S10, etc.
Not that the image quality is remarkably superior – it isn’t – just
a little better. It is better in
terms of apparent sharpness due to the SuperCCD “honeycomb” structure, and
better than most in terms of noise level. The
exception on noise is that Canon’s CMOS sensors are somewhat better at the
lowest ISO settings, but quickly get much worse at higher settings.
Also, some people complain that the S2’s noise pattern (in dark areas)
is more visible or bothersome than others although it is mathematically a lower
level. Personally, I have never had
much trouble with such noise.
The S2's apparent advantage in image sharpness is an interesting story: The difference between Fuji's SuperCCD and other image sensors is that the photo sites are arranged along a 45-degree diagonal rather than the normal horizontal and vertical rows and columns. Obviously, this does not do anything to increase the theoretical resolution since you still have the same number of samples in the same area. However, in the real world (either the natural world or man-made) lines tend to fall more on the horizontal and vertical axes than on the diagonal. A Fourier transform of an image will generally show that there is more information in the relatively horizontal and vertical. It happens that sensing an image with rows and columns of photo sites applies the greatest part of resolution on the 45-degree diagonal, while Fuji's "rotated" sensor provides more resolution where it is needed, and less on the diagonal.
Here
are some of the issues that people sometimes complain about in the S2:
Batteries:
The
S2 takes two different kinds of batteries; the usual 4 AA batteries, plus 2
CR123A lithium batteries. The
CR123A’s operate the “camera” portion (auto focus motor, exposure
metering, built-in flash, etc.) while the AAs power the image capture and
processing functions. This is
annoying because you always have to keep spares of both types of batteries
around. Actually, you can operate
the S2 without the CR123As, except for the built-in flash, but performance may
suffer somewhat. In my experience,
the CR123As last a long time and everything works better when both battery sets
are used. In fact, I am more
bothered by the fact that the camera does not behave gracefully when either
battery set gets weak.
Auto Focus:
This
is a complex issue, if indeed there is an issue.
Many people have complained that the focus is not good, but this can be
the result of one or more of several factors.
The worst of these is that, apparently, some S2s were built with poor
positioning of the auto focus sensors and/or the CCD image sensor itself.
The result is that focus is not set to the distance indicated by the
sensor. I have tested my S2 and
believe the focus accuracy is good, but others have reported poor accuracy and
improved performance after having the camera serviced.
A
similar focus problem is that the auto focus sensors (at least the center one)
is not positioned quite in the center of the area indicated in the viewfinder.
In my camera (and I think this is true of most S2s), you should place the
point that you want to be sharpest at the bottom of the center focus area
indicator (assuming you have elected to use this sensor).
The camera also has some sophisticated dynamic auto focus modes that can
produce unexpected results, and that often leads people to think that the camera
is not working correctly.
Lastly,
many people just don’t know how to analyze sharpness.
Specifically, the S2 can produce a 12 megapixel image file, although it
really has only 6 million photo sites. If
you compare such a file at 100% (each pixel in the file is displayed as one
pixel on the screen) and compare it with a similar file from a 6 megapixel
camera (without interpolation), the larger image will appear to be less sharp.
Exposure control:
High-end
35mm cameras typically provide shutter, aperture, and sensitivity controls in
1/3rd-stop increments. The
S2 provides ˝-stop increments. Personally,
I think this is a really silly complaint. The
difference between the “best” setting in one case versus the other is only
1/6th of a stop. That
much error is barely noticeable and can be easily adjusted afterwards without
noticeable degradation of the image. Commercial
photographers may insist that they don’t have time to “tweak” images.
They use JPEG mode (which limits the dynamic range captured) and want to
print directly from the camera with no fuss.
So I understand their concern, but I have no sympathy for it.
Nor do I believe that a person in that much of a hurry can often get the
exposure correct to 1/3rd stop while shooting an event.
I shoot raw mode almost exclusively, and spend a great deal of time
processing each image. I would not
dream of selling a print without carefully studying it to make sure it is the
best it can be.
Flash sync:
The
S2 will synchronize flash with shutter speeds up to 1/125th second
using the built-in flash or up to 1/180th second using external
flash. Some people want flash sync
at higher speeds because this allows you to use flash to fill in shadows when
the primary exposure requires a high shutter speed.
I guess this is a valid concern to some people, but I also suspect that
it is primarily a gear-headed ego problem because “better” cameras generally
support higher flash sync speeds. I
just don’t use fill flash. I use
studio flash, but it doesn’t care about flash sync speed because there is no
significant light from other sources to contribute to the exposure.
In fact, none of the film cameras I own will sync at over 1/60th
second.
Focal-length multiplier:
Because the S2’s image sensor is smaller than a full 35mm frame, there is an effective 1.5X “multiplier” on the focal length. Aside from causing some confusion, this makes it difficult to get a very wide angle lens at a reasonable price. I bought a 17mm lens to get an effective 28mm perspective, but it is not a very good lens because it is quite difficult to make good lenses at such extremes. On the other hand, the multiplier is a boon to telephoto users, and I tend to use long lenses more than wide lenses.
Here
are the things that bother ME about the S2:
Viewfinder:
Because
the S2 is built on a 35mm film camera body (the Nikon N80) and its image sensor
is smaller than a full 35mm frame, the viewfinder is simply masked to show the
area actually captured. This makes
for a rather small image, but more importantly, it leaves a big gap between the
bottom of the image and the information display below it.
Since I wear glasses I don’t mind too much that the image is small.
On my 35mm film cameras it is often difficult to see the whole image at
once. With the S2 I can easily see
the whole image, but always have to struggle to then see the information
display.
The
S2 is the first “real” camera I have used that is auto focus.
I am specifically excluding the Olympus C-3000 digital camera, which is
more of a “point and shoot” consumer camera.
I’ve learned to use auto focus fairly well, but some of my lenses are
manual focus and sometimes I prefer to use manual focus even with auto-focus
lenses. The problem is that the
viewfinder is really made for auto-focus operation: Not only is there no focus-assist feature such as micro prism
or split field, it is also not very “focus sensitive.” That is, the screen is relatively transparent, which makes
for a bright display, but makes it hard to tell where the image is sharp.
The only way to accurately tell when it is in focus is to use the
indicator from the auto-focus circuit. That
means that you can only adjust focus at the currently selected focus sensor and,
as noted above, the indicator is hard to see!
So the end result is that you might as well use auto focus anyway.
Exposure metering with manual lenses:
A
great thing about Nikon 35mm cameras is that, unlike Canon, the lens mount is
compatible with almost all of the older, pre-auto-focus lenses. This is great because it means that you have a much wider
selection of lenses, including some excellent models that can be purchased quite
inexpensively. Unfortunately, along
with automatic focus, you also lose ALL exposure metering with these older
lenses. The reason is that the
mechanical linkage that these older lenses used to “index” the lens aperture
has been replaced by an electronic interface.
It seems to me that this is really a choice that Nikon made to decrease
the usefulness of older lenses. That
is, it would be fairly easy to make the system work: All that the camera needs to know to calculate proper
exposure is the difference between the “wide open” aperture of the lens
(where metering is done) and the aperture selected for exposure.
Since the aperture control on the camera body does not work with such
lenses, it could instead be used to convey this information to the camera.
For example, if I were using my 180/2.8 lens at f/8, I would set the
difference to 3 stops. The camera
could then be used in manual or aperture priority mode with full metering
capability. Changing the aperture
is clumsy, but this is still far better than just giving up all metering.
Mode memory:
The S2, like most digital cameras, has a lot of preference settings. Fuji recognized that it would be totally unacceptable to require that the user go through all these selections every time the camera is turned on, so they provide a memory for these settings. But they could easily have provided multiple memory sets, as some other models do. That is, you could have a quick selection that set all the preferences to one of several pre-determined modes. For example, one set for outdoor shooting, one for indoor available light, and one for studio work. This would cost almost nothing to implement and would make the camera much easier to use.
Flash Memory:
There are two problems with Compact Flash image storage on the S2: Speed and capacity. Speed is always an issue with flash memory, but it can be alleviated somewhat with good firmware. I use a 1GB microdrive, which should be somewhat faster (at least in writing) than most flash cards, but shows little if any advantage in the S2. In fact, there is a significant disadvantage in that the camera's power-up delay is increased by the time needed to "spin up" the microdrive. I use a 256MB flash card instead of the microdrive when shooting events because it takes too long for the camera to be ready to shoot otherwise. But there is no need for the camera to wait for the drive to spin up, especially from "sleep" mode (where the camera is on, but has turned off things like the drive to conserve battery power): It could have been designed to store images in its local memory until the drive was ready.
The capacity issue is simply the result of supporting only FAT16 (the "File Allocation Table"), which limits capacity to 2GB. This has not been a problem for me, but it does seem like a rather silly design choice.
Copyright (C) 2004 Greg Marshall
No portion of this web site may be copied or reproduced in any manner without express written consent.