Home    Search    Gallery    How-To    Books    Links    Workshops    About    Contact

Ritz Camera

adorama

I personally buy from Ritz, Adorama and Amazon. I can't vouch for any other ads.

 

Fuji Fujifilm S5
Performance

© 2007 KenRockwell.com

Please help KenRockwell..com

Fuji S5

Fujifilm S5 Pro

I got this one here ($1,899 and free shipping). I'd have gotten it here or here, too. It helps me publish this site when you get yours from those links, too.

back to top of Fuji S5 Review.

Pages with explicit comparisons:

Color

Dynamic Range

Film Simulation Settings

High ISO Performance, General

High ISO Color Rendition

High ISO Definition

Be sure to read the Introduction and Specifications and Observations sections for more performance information.

Features Unique to the Fuji S5

The S5 offers a live on-LCD preview from the CCD for focusing, which no Nikon does. I didn't try it. Fuji cautions if you use it for too long you can heat up the CCD and have some hot pixels.

Fuji has some simulated film modes and wide dynamic range modes, which may or may not help you get the look you want. If they do, then get the S5. Photography is an art, only you can see if you like the results for your work.

I prefer my D200's automatic zoom-to-active-AF-area playback to the Fuji's cute "face detection" feature. Face detection didn't do much for me and rarely worked.

Fuji went so far as to make their own BM-6 LCD protector. It says says "FUJIFILM," not Nikon or BM-6.

The S5 can tell you haw many shots it's gotten off. Go to Menu > Set Up > Maintenance Counter.

The S5 has the ability to turn off some of the noise reduction for astronomy. This is in Shooting Menu > NR > ORG. (OK, most Nikons have that, too.)

Body Mechanics and Optics

Same as my Nikon D200.

Viewfinder

Same as D200.

LCD Monitor

Same as D200. The Fuji S5 seemed a little cooler, but not enough to worry about.

Feel and Usability

The Fuji S5 is worse than the D200. Just like its frame rate, the Fuji S5's controls and menus are clumsier to manipulate than the D200.

Getting the S5 to do something, like play and zoom an image, takes more pushes than the D200.

For instance, you have to push the S5's play button twice to see the image if the camera has gone to sleep after several seconds, but only once if it's still awake. This makes it difficult to fly the camera by feel, because I always seem to hit the play button the wrong number of times. This means most of the time I hit PLAY once or twice, guess wrong, and see nothing.

After you make a shot, you have to get to the play mode to see other images. The D200 gives me full playback freedom after every shot.

The Fuji S5 locks up its menus while it's writing to the card (it ignores the MENU button), while the D200 never stops anything while writing to the card.

The Fuji S5's menus are ugly. Come on Fuji, photography is art. The menu graphics look like they came from hardware limitations of the 1980s. They are legible, but ugly. Fuji does a great job on the graphics on the outside of the box, but didn't do anything for the menus.

There is no Cloudy White Balance setting. Use the Flash setting instead.

There is no Basic JPG setting.

There are no Shooting Menu Memory Banks and no Custom Setting Memory Banks as on the D200.

There are no easy White Balance trims via the dials and the top LCD. You can tweak them around in the menus, but you only get one master setting which applies to everything and doesn't show on the top LCD. This menu item is misspelt in the Fuji S5 as "AMB/PRECET WB FINE TUNE," which means that the fine tuning value applies to Auto WB and the Sun, tungsten, shade, and etc. fixed settings.

There is no way to preface file names with anything different than DSCF. The D200 allows one to prefix files as you see fit; my files start as KEN.

The Fuji S5 comes in a nice box with pretty pictures of the camera, unlike the simple gold Nikon boxes.

Auto Focus

AF is the same as the D200. AF speed is less relevant on the Fuji S5 because the S5 has such a slow frame rate. If you can shoot precisely enough to deal with the slow frame rate, AF is more than up to it.

AF Accuracy, Consistency and Offset

All camera and lens combinations vary, and they vary from sample-to-sample. That's why the Canon 1D Mk III provides a unique menu item to fine-tune this for each Canon lens you own. With my Nikon gear I check each piece I get, and send it back if it isn't compatible with the rest of my system.

Like my Nikons, AF results are consistent shot-to-shot. It's much better than my Canons, which have some percentage of shots that are simply way out of focus. My S5 shots have the same focus from shot to shot.

My S5 had some offset with some of my lenses. It tended to focus a little behind the subject with some lenses, which means unsharp results with those lenses wide open.

It was perfect with my Nikon 80-200mm f/2.8 AF-s and Nikon 18-200mm VR.

It was off a little with my Nikon 12-24mm.

It was too far off to use my Nikon 28mm f/1.4 reliably at f/1.4.

Again, every sample of camera and every sample of lens varies. If you worry about this, you must pick and choose every piece of gear as do, which will make you go insane.

Exposure

Exposure is excellent, as it is in my D200, although I usually prefer results with the Fuji S5 at -0.7 in daylight versus 0.0 with my D200.

Just like my D200, the Fuji S5 is consistent and doesn't require me fiddling from shot to shot as I have to do with some other cameras.

Flash Exposure

Flash control seems the same as the excellent system of my D200.

Speed: FPS and Buffer Depth

The Fuji S5 is the slowest DSLR I've ever used.

If you're shooting sports, the Fuji S5 is way too slow, unless you're the sort of guy skilled enough to shoot a 4x5" press camera.

The S5 only runs at 1.5 FPS with only a shallow 7-shot buffer. I easily can overflow the S5's buffer, something that has never happened to me with any other DSLR.

The S5 can hit 3 FPS, but you need to shoot above ISO 1,000 or go into the menus and turn off the default automatic dynamic range adaptation and fix it at 100%. In these modes the S5 only uses half its pixels.

At short shutter speeds, the S5 fills its 7-shot buffer in 3.9s. After that, mine took 3.4 seconds per frame (0.3 FPS) for each shot as the buffer wrote to my super-fast SanDisk Extreme IV card. This was at the default Medium (native) JPG Normal setting.

(Note: seven shots took 3.9s from the first click to the seventh. That means only six complete shots were made, not seven, for calculating FPS.)

If you go into the menus and turn off the default automatic dynamic range adaptation and set it to 100%, the buffer expands to 16 shots. Otherwise it's only 7 shots regardless of file format, resolution, ISO or image size settings.

How slow is this? My Canon SD700 point-and-shoot runs at over 2FPS with a buffer so deep I've never found its bottom, even in continuous 2FPS shooting for dozens of uninterrupted sequence shots. Yes, the Fuji can refocus and change exposure between shots, but this also slows down the Fuji S5 even further if it has to make corrections between frames.

Definition

Definition is how sharp something looks. It means much more than resolution, but there's no direct way to measure definition short of a concatenated modulation transfer function. You simply have to look at an image to see its definition.

Here are crops from 100% images from each of four cameras at their highest resolution settings. The Nikon D40 is only a 6MP camera, so I upsampled it's 3,008 pixel wide image in Photoshop (Image > Size > 4300 pixels wide to match the others > bicubic sharper) to the same width as the Fuji S5 and Canon 5D. The Fuji S5, Nikon D200 and Canon 5D images are directly from the original JPGs, no extra sharpening.

Guide Image

Complete Guide Image

Crops from 4,000 x 2,700 pixel (approx.) images at 100%.

The Canon 5D clearly wins, which as the highest resolution camera and full-frame, it ought to.

The Nikon D200 is in second place.

The 6 Megapixel Nikon D40 and Fuji S5 are tied. The D40 looks sharper, but that's from more sharpening, while the S5 uses less to optimize it for portraiture.

Sadly, this shows us that the large image size of the Fuji S5 is simply upsampled (interpolated) from a 6 MP image, and worse, that it's not a very good interpolation. Notice the very slight jagginess of some of the near-vertical lines from the Fuji. Photoshop did a better job interpolating the results from the D40 than the Fuji did in-camera.

If you're as data-stingy as I am, the D40 JPG was only 1.6 MB while the Fuji's crummier JPG sucked up 3MB of my disk space. The D200 took 3.5MB (normal JPG; I usually use BASIC) and the Canon took 4.5 MB (normal, the smallest of two settings). I usually shoot my D40 at BASIC, which makes a file just as sharp in half the space. An 800 kB D40 file would have been sharper than the 3 MB Fuji file, although the D40 BASIC file might have a tiny bit of noise along the edges if you're looking at this magnification.

Caveat: If you printed an entire image this size, it would be almost four feet (1.1m) wide!

Let's get real and compare all the cameras set down to the Fuji's native 6 MP resolution.

Here are crops from 100% images at the Large JPG setting of the D40 and the Medium settings of the other cameras. In these settings each camera made images about 3,000 x 2,000 (3k x 2k) pixels.

Crops from 3,000 x 2,000 pixel (approx.) images at 100%.

Again the Canon 5D completely smokes all the other cameras. The D40 and D200 are pretty close, and the Fuji S5 is last. The Fuji S5 uses less in-camera sharpening than the others; it could be improved somewhat in Photoshop.

It's extremely hard for anyone, even engineers, to devise tests to test what's intended without being degraded by other effects. Lest one be concerned (I was) that these are defocus or diffraction issues, the aliasing of the star targets coming next show me that I'm putting more optical detail on the sensors than they can recover. People are always sending me images they think are unsharp for me to fix, and 99 times out of 100 it's equipment operating normally or photographer error. That's not the case here: the Fuji, if you're sticking your nose too far into it as we are here, is only about as sharp as a 6 MP D40.

Resolution

Resolution is easy to test, but unimportant unless you're printing several feet wide. This is easy to see on star targets, but not important unless you're shooting the Wall of Shame.

Even if you are printing that big, rarely does one need more than the 6MP provided by the S5. Plenty of full-time wedding and portrait photographers are still earning livings with 6 MP cameras and making any size prints wanted by clients.

Resolution is a sore spot for the S5, since it's only as good as my 6MP Nikon D40, but without the D40's aliasing (artifacts).

The Fuji S5 is much poorer than my Nikon D200 or Canon 5D, but as I've been cautioning for years, megapixels don't matter. It doesn't matter how many pixels you have if they are the wrong color!

The S5 has the same real resolution as my 6 MP cameras, regardless of how I set the S5's dynamic range. I tried. The S5 only uses 6.1 MP in the standard dynamic range mode or above ISO 1,000, which lets it get to 3 FPS. In the other Dynamic Range and film simulation mode the S5 uses the extra pixels to expand highlight range (the z or intensity axis), not to increase spatial resolution (the x/y axes, or up/down/left/right).

Fuji S5 Nikon D200
Nikon D40

Canon 5D

Star Target Resolution, maximum resolution settings.

The star target lets us see exactly the resolution and aliasing issues of each camera. In case you're concerned that there were defocus or diffraction issues which limited the results, the aliases (wavy lines or colors) at the unresolved resolutions show us that the optics are putting details there and that the sensor simply isn't resolving.

Since each camera creates a different image size at its largest setting, I scaled all of them up to the same size as the largest and then cropped out the chart. If you printed these full images at the same magnification as above, you'd have 40" (1m) wide prints!

The D40 has the same resolution as the S5, however the D40 has a lot of aliasing. Aliasing is the false-colored bands. If you shoot fabrics or screens you'll see occasional weird color effects if the patterns get fine enough and you're in perfect focus.

The Fuji S5 is nice because it has no aliasing. Use it on and fabric (and screen doors) and there is never a problem. This is a big disadvantage of the Nikon D40, D50 and D70: they sometimes put falsely-colored bands over textured objects! For this reason some wedding and fashion photographers won't touch them. The Fuji S5 has no problem with textures and fabric, and the Canon 5D and Nikon D200 have no problems in real shooting either.

The D200's superior resolution is obvious, even though the image and files size of the Fuji S5 are much larger. This is bad: the Fuji wastes disk, card and file space. The Large JPG image of the Fuji S5 is interpolated from its native 3,024 x 2,016 to the 4,256 x 2,848 size shown above.

The Canon 5D is clearly the winner. It does have a little aliasing, but I had to work hard to excite it. Aliasing on my Canon 5D requires such a high resolution at the sensor that I've never seen it in real photography. To get these images I had to focus extremely carefully with several tries. I goofed slightly with the 5D: I used my fixed Nikon 28mm f/2.8 AIs manual focus lens on the Nikons and Fuji because I could focus it precisely and it's one of the sharpest lenses I own. I don't own a lens quite long enough for my full-frame Canon 5D to give the same field of view, so what you see is shot at at the 40mm end of my Canon 17-40mm L.

Since the resolution of the Fuji at its largest setting is only kidding (it's upsampled), let's set the Fuji back to its default resolution and shoot each camera at about the same setting.

For these next clips, the Fuji was set to its native 3,024 x 2,016 resolution. The D40 is also at its 3,008 x 2,000 pixel resolution. I set my D200 down to its medium (2,896 x 1,944 pixels). I forgot to set my Canon 5D to medium, so I changed it in Photoshop to 3k wide.

Digital camera files get sharper, as seen at 100% magnification, as you turn down their resolution. This is because even at their native settings, Bayer interpolation is used and every pixel isn't as sharp as it could be from a full RGB (Foveon) sensor or scanned film image. A true 12 MP camera set to 6MP gives a much sharper image, seen at 100%, than a native 6MP camera.

Fuji S5 Nikon D200
Nikon D40 Canon 5D

Star Target Resolution, 3,000 pixel-wide resolution settings.

The results are the same as above, just harder to see. Setting camera resolution to slightly lower values usually doesn't lower visible resolution. These images are cropped from 100% images without any of the slight resizing done at the first set of examples.

Power

The S5 has an excellent battery meter and can read battery charge to the nearest percent and the age of the battery, just like the D200.

The batteries look identical but are not interchangeable with Nikon. They fit in either camera, but data communication differs. Either battery reads DEAD in the other brand camera, even if full. I've tried; it doesn't work.

The chargers are interchangeable.

You're SOL unless your battery brand matches the camera, even though they fit in each camera and charge in each other's chargers.

Battery life is similar to my D200. Since I get between 200 and 3,000 shots per charge depending on how I use either, I'd have to use each camera exactly the same way to make any more precise comparison.

Playback

Playback is completely different from the D200. I prefer my D200.

I set one push of the big central nav button on my D200 to zoom way into the area of the photo that is supposed to be in focus. The Fuji S5 can't do anything like that. Push it on the Fuji and the LCD Monitor washes out to attempt to make it look brighter. It looks awful, but will help you see shadows on playback in direct sunlight. Thanks to Fuji for placing a flashing BRIGHT warning on the monitor when you do that, so you don't confuse the washed-out monitor with an overexposed image.

If I'm trying to get to a file in the middle of hundreds of other images, the Fuji S5 bumps me back to the last shot I made if I let it turn off the LCD before I get there. My Nikons wake back up where ever I was playing when they turned off, unless I make a new photo.

Face Recognition

There is a face recognition feature which sometimes lets you press the face button (lowest left rear) to zoom into a face. It rarely worked. You'll see a small icon matching that button on the LCD if the Fuji S5 has identified a face into which to zoom.

Zoom

Maximum zoom is much less than other Nikons.

Playback zoom happens with up/down pushes on the rear nav button.

I haven't figured out how to select among different playback images while magnified. Nikons do this easily, making it trivial to select the sharpest image.

Histograms

There are four full YRGB histograms, but sadly they are small and only one comes up at a time! On Nikons and Canons you can get three or four of them up at once. Seeing only one at a time on the Fuji makes them much less useful.

Shooting Data

Data is displayed along the top and bottom of the image, and two more screens show more intimate details like the lens used (an improvement over the D200) and the full IMAGE COMMENT.

Data

Firmware

I used firmware versions 1.01 / 1.00 A / 1.01 B. There is an update, but the instructions are so poorly written and since it didn't work, I didn't bother to update my S5.

Formats

I only used JPG. My life has too few days remaining to mess with raw, especially since experience shows that most software can't read raw files from brand-new cameras as they come out.

File Sizes

The S5 creates bloated files. I hate this.

The Fuji S5 lacks the BASIC JPG mode I use all the time. You're stuck with at least NORMAL.

The Fuji S5 also lacks smart JPG encoding. All its JPGs are big.

The S5 isn't smart enough to vary the file size with image complexity. My D200 can be set to do this, and my Canon 5D does it even better by default.

Smart cameras vary file sizes to fit the complexity, detail level and contrast of the subject. The Fuji S5 doesn't - all the files are big.

The Fujifilm S5 is a data hog. It wastes bits on images that don't need them. This slows data transfer, clogs hard drives and CDs and slows workflow.

Every native-resolution (Medium, 3,024 x 2,016) JPG is 1.9 MB, whether it's a backlit forest or a blank sky. Every interpolated (Large, 4,256 x 2,848) JPG is 3 MB, regardless of the subject.

Better firmware would allocate more file size to the images that need it, and less to images that don't.

In Fuji's defense, these JPGs have no artifacts so you can fold, spindle and mutilate them later without worrying about image degradation.

Data Transfer

My Mac didn't recognize the Fuji S5 as a drive, as it does with all my Nikons. I had to use Apple's Image Transfer program, as I do with my Canons.

I would much prefer the Fuji to appear as a drive and save me the extra steps to transfer images. It should have worked, but didn't. In today's world, I didn't spend any more time trying to fix this.

I see no mass storage option in the Fuji S5's MSETUP > MAINTENANCE > USBMODE menu.

EXIF

ISO reads correctly in iView, even at ISO 3,200. This is better than my Nikons, which read "1" at ISO 3,200.

Flash WB erroneously reads as "Auto" in iView.

I can't see the lens used in iView, even though you can read it with in-camera playback.

User's Manual

244 pages in reasonably good English. Unfortunately, just like Nikon's, Canon's and other manuals, it doesn't have anything to say.

This is particularly sad with the Fuji S5, because there is so much brilliant image engineering inside that is never explained in the manuals.

If you've read this review and know how to use the D200, I've already pointed out what's new in the S5. Here are my humor notes from reading it:

First 11 pages: Stupid warnings and table of contents.

Page 20: First photo of a Japanese girl in front of flowers.

Page 48: First images of a porcelain cat waving, still no new information compared to the D200.

Page 51: Aha! Playback zoom works differently than the D200!

Page 141: First typo: "BILT-IN AF-ASSIST." (repeats on page 154.) This typo is also in the camera's menus!

Pages 230 - 237: Name, phone, street and email addresses for every Fuji support and repair center on earth. Nice!

Pages 238 - 241: More important warnings, like "Do not use in the shower," no kidding!

No index!

More Pages with Explicit Comparisons:

Color

Dynamic Range

Film Simulation Settings

High ISO Performance, General

High ISO Color Rendition

High ISO Definition

PLUG

If you find this as helpful as a book you might have had to buy or a workshop you may have had to take, feel free to help me continue helping everyone.

It also helps me keep adding to this site when you get your goodies through these links to Ritz, Amazon and Adorama. I use them and recommend them personally.

I got this S5 here ($1,899 and free shipping). I'd have gotten it here, too. It helps me publish this site when you get yours from those links, too.

Thanks!

Ken

 

back to top of Fuji S5 Review.

 

Caveat: The ads below come from a third party and I don't see or approve them. They are sent to your screen directly from a third party. They don't come from me or my site. See more at my Buying Advice page. Personally I get my goodies at Ritz, Amazon and Adorama.

Home    Search    Gallery    How-To    Books    Links    Workshops    About    Contact