Add a comment or send Thom feedback on this article.

  Fuji S2 or Nikon D100?

Two digital cameras built on essentially the same platform. Which one is better?

  Updated 21 Dec 2003

Yes, the relative sizes in the photos at left is correct. Physically, the S2 Pro is a slightly larger camera.

On the front there's not much to distinguish the two other than the names and the Fuji's larger size.

The backs are quite different, though. It's hard to tell from these small shots, but the autofocus sensor direction pad on the back of the D100 is shorter, stiffer, and stubbier than the S2 Pro's. And while both have a bewildering array of buttons on the back, I must say that I find the D100's buttons a bit more disorganized than Fuji's. The Fuji uses the same digital control interface as on the S1, which is easy to learn and fast to use. The D100 relies a bit too much on the color LCD for setting things, in my opinion. Some users will like that, though, as the color LCD certainly gives plenty of feedback.


Feature Differences

Since both cameras are based upon the same basic film body, there aren't a lot of differences between them. The differences really work out to these:

Fuji S2 Nikon D100
Sensor 3rd Generation SuperCCD (Fuji) 2nd Generation D sensor (Sony)
Real Photosites in Capture 3024 x 2016 3008 x 2000
ISO 100, 160, 200, 400, 800, 1600 200-1600 in half stops, 3200/6400 (boost)
Exposure 1/2 stop increments; compensation +/- 3 stops 1/3 stop increments; compensation +/- 5 stops
Depth 12-bit 12-bit
Flash 1/125 sync; TTL on any Nikon-compatible flash 1/180 sync; TTL only on DX flashes (D-TTL)
Storage SmartMedia and CompactFlash (I/II) CompactFlash (I/II)
Video Out NTSC or PAL (depends on region) NTSC and PAL (selectable)
Communication USB 1.1 and Firewire USB 1.1
Power 4 AA Nimh rechargeable batteries, 2 CR123a EN-EL3 lithium rechargeable
Weight 1.7 lbs (760g) 1.5 lbs (700g)
Other Voice annotation; PC Sync terminal; multiple exposure! In camera pixel interpolation to produce 12-megapixel images.

Anti-mirror vibration reduction; optional battery pack (six AA) with vertical release, 10-pin camera control connector, and voice annotation;"more green" sRGB mode! Camera frame is actually based on F100, though controls and viewfinder are clearly N80-based.

Price US$2395 street price (may be lower). Includes batteries, USB and Firewire cables, Fuji FinePix Viewer software. US$1695 street price. Includes battery and charger, Nikon View 5, trial version of Nikon Capture.
Availability Working prototypes were first seen by outsiders in March according to posts on the net. Camera appeared in July 2002.

Working prototypes were shown at PMA in February. First shipment was in late June 2002.

Commonalties Most Nikon N80 attributes, such as five-sensor CAM900 autofocus, matrix/center/spot metering, viewfinder on-demand grid lines, built-in internal Speedlight, 30 second to 1/4000 shutter speeds, numerous custom settings, etc.

Not a lot of functional differences, are there? The Nikon gets the nod with a slightly simplified battery situation (lithium also works better in cold) and lighter weight, the Fuji gets the nod with slightly more features, most notably the much faster Firewire communication support. Careful readers will note that the optional MB-D100 Battery Pack adds a number of interesting abilities to the Nikon, including a fully supported vertical release and a Nikon-standard 10-pin release.

But the bottom line in features boils down to this: which sensor does the better job? That's actually not as easy to answer as you might think (see the Image Quality section, below).


top | home

What about Sync Speed?

Many recent digital camera forum postings have been devoted to the issue of the faster sync speed of the Nikon D100 compared to the S2. After all, 1/180 is faster than 1/125. True, but the usual reason photographers want faster sync speeds isn't to stop motion (sports being an exception), but to allow more flexibility in use of flash for fill. The biggest request usually is to be able to use larger apertures to throw the background out of focus.

Sync Speed isn't the only factor that enters the picture (pardon the pun). ISO value also makes a difference.

So let's go outdoors for a typical fill flash situation. We'll say that we're shooting under midday sun, so the exposure by definition is f/16 at 1/ISO. On the D100, the fastest ISO we can set is 200, while on the Fuji we can set ISO 100. Thus, we get:

D100 -- f/16 at 1/180
S2 Pro-- f/13 at 1/125

In the case of the S2, we're about a 1/3rd of a stop hot, while we're about a 1/6th of a stop hot with the D100 (due to the increments being used). So, you actually get about a half stop more aperture flexibility in fill out of the S2 Pro.

On the other hand, if you're shooting action, the extra half stop of shutter speed the D100 gives you might be useful.

Either way, we're talking about a difference of about half a stop. So which do you want? More aperture (choose the S2 Pro) or more shutter speed (choose the D100)?


SB-50DX Use Difference

The S2 Pro supports firing both the internal flash and a hot-shoe mounted SB-50DX simultaneously in TTL modes, and works out the correct exposure when doing so. The D100 will fire only the internal flash in D-TTL modes. You can turn the internal flash to manual flash mode and fire both the internal flash and SB-50DX in the hot shoe simultaneously (also in Manual flash mode), but this isn't very useful, as both fire at full power and have different GNs! The more I look at the nuances of D-TTL flash, the more I find little things that don't work as expected.


Look and Feel

A number of posts on the digital camera forums have noted one seeming difference between the two cameras: the Fuji, they claim, appears to be a little more "grafted on" to the N80 than the D100, which appears to be a modified N80. One went so far as to call the S2 (and its S1 predecessor) a Frankencamera.

I'm not sure I can agree with the contention that the Fuji S2 will be more awkward in use, however. The battery situation is about the only kludgy aspect of the Fuji design, in my opinion, and even that can be simplified if you don't use the internal flash. In some ways, the Fuji design has one clear advantage: the camera stuff stays with the basic Nikon N80 controls, the new digital aspects are controlled with the added black and white LCD and four buttons, and as S1 users know, Fuji has done a pretty good job of making this simple and direct. If you haven't used an S1, which features the same basic design, you shouldn't poo-poo this "separation." In practice, most of Fuji's digital UI is far simpler and faster to use than those on the original Nikon D1 (and even the subsequent D1h and D1x, and D100). It certainly is nice to have most of the digital tools organized into one spot. Moreover, Fuji has done a nice job of integrating the custom settings with the color LCD, essentially doing what Nikon should have done. The digital parts of the D100 work much like the D1's do, which is to say very well integrated but not perfect. There's no absolute winner here--either camera is simple to control and shoot with once you've used it for awhile. Still, I like the Fuji's controls a bit better than the D100's.

The D100 seems to once again have continued Nikon's arbitrary relocation of the autofocus sensor direction pad (the Fuji's has moved from the N80 position, too). The Fuji feels more comfortable in my hand while shooting, with a more natural right hand position. Both cameras have buttons and connectors where I would naturally be holding the camera with my left hand, though the Nikon's rubber cover extends the full length of the left side, meaning that if you plug any cable into the camera, the left side hand position is compromised. With Fuji's design, those with small hands might be able to still grasp the camera with their left hand with a cable plugged in.

The position of the color LCD on both cameras is going to provoke a lot of discussion, too. While you won't get nose prints on the Fuji S2, you might get chin prints. You also can't "peek" at the color LCD quickly due to its extreme position in the lower left corner of the back. Nikon's choice seems better. You won't get nose prints on it, and you don't have to move the camera far from your face to see the color LCD (to review a histogram, for example). The protective cover Nikon supplies is clear (but prone to breaking) while Fuji's is partially opaque and not very secure. Both cameras use the horrible N80 rubber eyecup, which you will lose very quickly. (Hint: get the N50 or N60 half cup, which fits on the N80-style viewfinder; you'll still have to watch carefully to keep from losing it, but it does a better job of isolating the stray light.)

Fuji's image review system is far poorer than the Nikon's. First, you have this annoying delay after taking a picture unless you turn on "Preview," which is a surefire way to lose images someday (you have to confirm that you want each image saved, or it will be thrown away before saving; e.g., you can't use recovery tools to find the lost image). Second, the camera has to be active to view pictures, which means that most of the time I'm pressing the shutter release partway before pressing the Play button. An awkward and poor design.

The D100 has more viewfinder information and more accurate battery level display. The Nikon implementation of histogram and highlights is better than what Fuji provides, as well.



top | home

Rear Sync Use Difference

The S2 Pro again follows the 35mm flash specifics: when you select Rear Sync, the pre-flash is cancelled (it's actually cancelled all the time). With the D100, the D-TTL system never cancels the pre-flash, so with longer shutter speeds you'll see two flashes (the pre-flash and the flash). Fortunately you'd rarely use Rear Sync with long shutter speeds for portraits, but if you do with the D100, be aware that your subject is going to react to the first flash (the pre-flash) and think the photo has already been taken (it has not).


CompactFlash Differences

Here's a strange one: shooting with Microdrives on a Fuji results in faster save times than with a Lexar 24x card. On the Nikon, the Lexar 24x card saves faster than a Microdrive. In both cases, the difference is significant and meaningful (on the order of 200kbps, or about a 10% difference).


Image Quality

I've now looked at quite a few images off a number of D100 and S2 Pro bodies. Plenty of S2 Pro/D100/D60/D1x comparison photos also exist on the Web, so I'll leave it to you to browse and find them on your own. My remarks are primarily based upon observation of images I've taken at workshops with the two cameras side by side.

Even casual perusal of images from these cameras shows us that they have very different personalities. Perhaps this is an over-generalization, but here goes:

  • Out of camera sharpness is different. At default settings and in JPEG modes, the S2 Pro pictures appear quite a bit sharper than the D100. Nikon's JPEG engine seems to produce very smooth, artifact-free images, but edge sharpness suffers. The Fuji produces sharper edges, but you do have to be on the alert for moire. Either way, there's a compromise to be had. Still, if you want to minimize post-production and don't mind shooting in sRGB color mode, the Fuji wins the "out-of-camera" race with JPEGs. With raw formats, sharpness isn't an issue. [The D100 does have one advantage in some situations: anti-mirror vibration reduction helps keep telephoto images sharp when using long shutter speeds (1/15 to 1/2 second, typically).]
  • Color rendering is different in sRGB color modes. I see it most in the greens and skin tones, but the S2 Pro appears to have a slightly saturated color spectrum, pushing face tones towards warmth and greens towards that Velvia pop that photo editors seem to like (extra yellow in the green). The D100's images look more neutral, though sometimes they have a slight coolness to them. But the D100 also allows you to shoot in AdobeRGB color space, and there the D100 gets a clear advantage in color neutrality.
  • White balance handling is different. Fuji's automatic white balance appears to work in a wider range of color temperatures than Nikon's, though not by a lot (say 3800-6400K versus 4000-6000K). Both cameras have problems getting decent automatic white balance when incandescents are present. The D100 has more white balance selections available, which is nice (assuming you know how to interpret color temperature). Both cameras can set custom white balance settings, but neither is a snap to set. Overall, I'd say that Fuji has to work on improving the white balance options.
  • Contrast is different. The D100 seems to have slightly less contrast than the S2 Pro, which seems slightly over-contrasty (here I'm talking about JPEGs, and generally with contrast settings on "normal"). While I'm always pulling up shadow detail on D100 images for "taste," I find that I have to do it with many Fuji images with wide dynamic ranges. At the other end, I've had a more difficult time holding highlight detail on the Fuji (but that is partly due to the histogram implementation). Personal taste enters into the picture here. Neither camera produces the contrast levels I'm looking for, but that may not be true for you.
  • Dynamic range is nearly identical. If you measure your dynamic range simply by whether there is detail available at each stop, you'll come up with about 7.5 stops of dynamic range for both cameras. I consider "noise" when determining usable dynamic range, and here there are a few visible differences. Curiously, the S2 seems to do slightly better with JPEG, the D100 a bit better with raw formats. When I try to pull up shadow detail on a D100 JPEG, some of the detail is false (i.e., noise). This happens less often with the S2. With raw formats, this pattern is reversed. One side note: since you can see individual channel histograms on the Fuji, you can sometimes "play with" controls to help things out a bit. Noise is generally the biggest problem in the blue channel, so if I see a poor histogram in the blue channel, I'll do something about my exposure or color balance while still in the field. In particular, I can tolerate a bit of blow-out in the green channel if it boosts the blue channel levels away from the left edge of the histogram.
  • When you significantly boost shadows using Curves and see the Fuji noise, it will be objectionable. The S2 Pro has a tendency to produce a diamond-patterned noise when you try to expand it's dynamic range by boosting shadows using Curves. The D100 just delivers random noise, which is slightly less objectionable.
  • Either way, both cameras are state-of-the-art. The question that I and many others had was whether the 6-megapixel consumer cameras would equal the output of a D1x, the former resolution champ in the SLR world. The answer I've come to over the past few months is yes, they are. Amazingly, though, the D1x still holds its own against both these cameras. If I were told that I could only use one of the three, I'd probably pick the D1x or D100, but I would not be unhappy being told that I could only use the S2 (remember, I use raw formats almost exclusively). If I were shooting only JPEG, my assessment would flip, favoring the S2, though I'd still be happy with either the D100 or D1x.



top | home

New Prices Change Values

With Nikon dropping the D100 down to US$1499 it now clearly is the value leader here. Yes, the S2 prices have been slowly dropping, too, but there's now a substantive difference in prices that I don't think is justified by differences in body quality or image quality. Unless there's a feature of the S2 Pro you absolutely need (wireless TTL, for example), I'd simply get the D100 body and use the extra money to buy another lens.



Reader Comments and Queries

Robert Joplin asked how the D100 compares to the D1x. In particular, he wanted to know about the resolution differences and what the ISO 200 base speed might mean:

Thom's Response: The D1x and D100 are both interpolated (both use Bayer pattern filters) as is the Fuji S2. The only real difference is the regularity and alignment of the photosites, which effects their size and may introduce some minor artificating if not corrected for. But at these photosite specifications, you're not likely to see a lot of difference. The Fuji S2 does an in-camera doubling of photosite data to 12 megapixels for JPEG images, a D1x NEF image can be nearly doubled to 10.4 megapixels by running it through Bibble, Capture, or QImage Pro, which invent intermediary vertical pixels from the sensor data. Both of these higher outputs can be quite good on most subjects, but show slight artifacting on others.

As for ISO, high ISO values make it difficult to use fill flash in bright light, especially when you consider the limited 1/180 sync speed of the D100. The S2 with a sync speed of 1/125 and an ISO of 100 actually is a slightly better alternative in this regard (well, by a third of a stop). For the internal flash on an N80 you can't set Standard TTL without sacrificing other features, such as wireless flash use.

Daniel Eppelsheimer asks: Why would Nikon (D100) not have the FireWire and now the Fuji S2 (compared with S1) have Firewire as an option? It seems if they are moving in opposite directions in the same market.

Thom's Response: FireWire is not ubiquitous in consumer computers (of my three computers, only two of them have it, and the two that do are from Sony and Apple, not exactly mainstream). True, you could add it to most machines, but I think Nikon simply looked at the least common denominator for their consumer model. Fuji aspires to have the S2 used in studio settings (note the addition of a PC sync socket), and there FireWire would almost certainly be needed for quick preview of results. Note, too, that Fuji's camera comes from their Professional division, which caters mostly to the studio crowd, though it obviously has spillover into the general market. So, you are correct, they are moving in opposite directions, though not necessarily after the same market.

David Dyer-Bennet writes: Based on the facts you give, I see a clear advantage to the Fuji on the battery front. I *hate* specialized proprietary batteries, so the fact that it uses ordinary common AA Nimh cells is a *big* win for me. It means I can have a second (third, fourth...) set of charged batteries ready without spending however many hundred dollars Nikon would charge me for a second battery pack. And use my existing charger at home, in my car, wherever, rather than hoping the Nikon charger supports car power.

Thom's Response: I like the AA option, too, though the lithium batteries of the D100 perform better in cold weather than Nimh AA rechargeable in the Fuji (and last longer). The real issue, though, as you point out, is likely to be cost, since you can get four excellent Nimh AAs for less than $10. As for charging in a car, either way I use a power inverter in my vehicle, which doesn't require buying Nikon's specific car charger (though I end up with cords everywhere--good thing I have an outlet in the back of the vehicle where I can just let everything stay hooked up).

hcalahan writes: I think its great that the prices of good digital cameras are coming into a more realistic price range for most. Of the differences I have read about other than the obvious resolution superiority of the Fuji and the addition of the PC socket. The most noteworthy to me anyway is the ability of the Fuji to shoot several frames in a burst I believe 2.5 fps for up to 7 frames. I have not read in any of the Nikon rhetoric about similar capabilities of the Nikon. I am a big fan of Nikon but from what I am reading I think Fuji will get the nod from me. Looking forward to your further reviews.

Thom's Response: I'd expect burst speed and buffers to be very similar, and I doubt we'll know the exact numbers until just before the cameras ship. [In practice, I've found that the D100 shoots slightly faster than the Fuji in continuous drive. But the difference isn't enough to be important.]

R Patterson writes: As an S1 owner and user (portraits, nature & "light" photojournalism [no Afghanistan!]), I am quite impressed with Fuji's execution. At first, I was a little jealous of the D1 & D1x owners (missed the metal body), until I realized the following:

  1. I can clean the CCD myself.
  2. The S1 supports the IBM Microdrive right out of the box. In the early days, some D1s worked with it, others didn't.
  3. The S1 uses all the standard (SB-23, SB-25, SB-27, SB-28, etc.) Nikon flashes and the Metz SCA 3000 system for matrix-balanced TTL! In addition, the S1 supports standard Nikon multiple TTL flash. The S1 with an SB-23 is a great combo for tight areas! In order to achieve TTL flash with the Nikon digital SLRs you have to have a "DX" flash mounted on the body. It isn't clear from the Nikon literature, whether you can do multiple TTL flash (SC-17, SC-18 & 19 to other flashes) with the digital SLRs.

Moving forward to the S2, I think the ISO 100 is a great feature on the S2. I think in ISO 100 because that is the bulk of slide film I shoot (E100S). The minimum ISO 320 on the S1 has been a pain at times. In moderate bright light, I have to use ND filters to use wide apertures for depth of field control. I think the S2 gets it right having ISO 100 and Firewire. Firewire is the best for moving lots of large images (10MB+ files).

As for color, Fuji did a great job on the S1. The S1 images are very usable without a lot of tweaking before giving them to my clients. In addition, when shooting at the higher ISO 800-1600 the images aren't bad!

The proof will be in the output from these two cameras. If Nikon continues to improve their color and usability of the images, the D100 could be the one. I really wish Nikon didn't require "DX" flashes for TTL. However, after working with the Fuji product and seeing their execution, the S2 should be an awesome camera. We'll see if Fuji lowers their price after seeing the D100 and EOS D60 prices...

Thom's Response: Well, I'd say reasons 1 and 2 aren't particularly valid any more. D1 users clean their CCD (and Nikon now only says "do it at your own risk"), and Microdrives are supported by Nikon (the early 340MB drives were problems, though). As for #3, the N60 body on which the S1 is based doesn't perform pre-flash, so Fuji must have either gotten lucky on the reflectance of the CCD or figured out a compensation value to put in place. Also, the limitation on flash is ISO 400 with that body. So, while it works, it does so in a narrow range, that, as you point out, is problematic to use in bright light. As to the S2, I'm curious to see how Fuji will get around the pre-flash issue inherent in the N80 body design.

But your final thought is much the same as mine: the proof is in the quality of image versus price. If Fuji can retain their excellent color rendering and keep close to the Nikon price, this will be a hotly competitive duo. [And, again, after using both cameras side by side for several months, the differences in image quality aren't much different than choosing Ektachrome or Provia for a film camera. Some will prefer one, some the other.]

  top | home | Nikon | Gadgets | Writing | imho | Travel | Privacy statement | 2003 Thom Hogan. All rights reserved.