Home / Culture and Society / Science and Technology / Review: The Fuji FinePix S9100 digital camera

Review: The Fuji FinePix S9100 digital camera

Please Share...Print this pageTweet about this on TwitterShare on Facebook0Share on Google+0Pin on Pinterest0Share on Tumblr0Share on StumbleUpon0Share on Reddit0Email this to someone

Bottom line: love the camera body and the user interface, good lens, good grip, but utterly hamstrung by cheap CCD sensor. Read on for the details. I won't bore you with the specs, which you can check out for yourselves. I'm going to focus on real-world use, which is what really counts. I purchased the FinePix S9100 because I wanted a good camera that would tide me over until I purchase a great DSLR outfit (I'm eyeing the EOS 5D). I had no DSLR expectations from the S9100. I just wanted a decent digital camera with a good grip. I don't like smallish cameras made for a woman's hand, because they're too light and don't feel right in my hand. The S9100 was pretty close in dimensions to medium-sized DSLRs, and that was a strong selling point for me. I also liked the FinePix S line's reputation. People kept saying these cameras take really good photos, and I wanted to see for myself. There were other selling points, such as the much-touted low light sensitivity, the 9 megapixel resolution, the 10x manual optical zoom, and the fact that it used AA batteries. I have a whole slew of rechargeable AA NiMH batteries at home, and I looked forward to the day when I could use them properly.

S9100 frontSo, I got the camera yesterday, and went out immediately to shoot with it. I've used so many cameras by now that unless I'm using a new DSLR, I don't need to read the instruction book. The menus of the S9100 were also arranged very well, and I was able to find and set all of the options I wanted within minutes. So, within 15 minutes of opening the box, I had the camera configured and the strap and lens cap attached. That made me pretty happy. I like cameras that are easy to use.

I started taking photos before getting out of the house, and that curbed my enthusiasm. The focus time was longer than I expected, comparable to and even longer than the focus on my Kodak EasyShare v610, which is a compact point and shoot. That didn't bode well. At any rate, I pushed forward, and made it outside. The plan was to get sushi with Ligia, then go out into one of the local parks and take photos as the dark set in. This would give me a chance to shoot across the whole ISO range.

At the sushi place, I took more shots, and found two things that were pretty annoying. First, there was some serious lag time between shots. I shoot in RAW format, and the S9100 apparently takes a really long time to write the photo to the card. S9100, back, angled, LCD outWhat's more, there's no burst mode in RAW mode. That's right, you can only take one picture at a time, then wait until it gets processed and written to the card before you can take another one. I had to sit there counting second after second while the write light was on, unable to do anything else. And no, it wasn't my card's fault. I've used that card (120x CFII) with competent DSLRs like the Canon 30D and Olympus E-500, and it works beautifully. Second, when shooting at ISO 1600 inside the restaurant, there was a whole lot of noise in the shots. That really put a damper on things, but I wanted to get out and take plenty of photos in the forest, then complain.

Once we got outside and I got used to the long write times, using the camera was kind of nice. The flip screen was great. It allowed me to use some really interesting angles. I'd have had to guesstimate some of the shots if I only had a viewfinder to look through, since there was no way I could have craned my neck into those positions. I also liked the zoom lens. I like to twist lens barrels, I can't help it. It gives me that tactile feel I need from my camera. The nice rubberized grip worked very well. Holding the camera in my hand, it was easy to forget that it wasn't a DSLR. It feels very good, it's balanced, and the buttons are just where they need to be. I had no problems using them. I loved their placement. I also loved the camera's two Macro modes, one for closeup shots and one for really close shots of insects or other such tiny things. That's a great feature!

As it got darker and darker, I switched to a higher ISO, and the camera worked decently up to 800 ISO in the twilight. Every time I'd switch to 1600 ISO, the noise was unbearable. S9100 backBut I figured, hey, I'm in the middle of a forest with no ambient light, and I'm also shooting handheld. Maybe this is to be expected. So I wrapped things up and we went back home. As we pulled into our garage, I looked at the lights in the parking lot and realized there was plenty of ambient light there to test out the 1600 ISO. I ran out, camera in hand, ready to test things, only to be disappointed once more. Every time I switched to 1600 ISO, the noise was too much, and there was serious pixel streaking going on. At the highest aperture (f2.8) and shutter speeds of 1/30 and above, there were no decent images to be gotten with the S9100, even if I stood right underneath a lamp post.

Finally, I switched it back to 100 ISO to try out some long exposure shots. I set it to a shutter speed of 4 seconds, and snuck it between the branches of a tree to stabilize it. The sky was filled with beautiful shades of blue that begged to be captured. The preview screen, which is supposed to compensate for the shutter speed and show me what the photo will look like given my settings, showed the sky exactly as I wanted it to look. I took a few shots, trying different angles, and according to the camera's display, each photo looked fantastic. I couldn't wait to get back inside and have a look at the photos on my computer.

After the shots were all loaded into Lightroom, Ligia and I sat at my laptop to have a look. What we noticed made us very unhappy. A lot of the shots were out of focus, even though they had seemed to be in focus on the camera's screen. When we used the Loupe to view good shots at 100%, all of them were trashed. I have no better way of putting it. It looks like the sensor isn't really meant for 9 megapixels, and FujiFilm is forcing it to produce that resolution through interpolation. But that results in some really cheap-looking shots at full-size. All of the detail is lost, and a whole lot of white pixels are seen instead. Really, the photos are that bad! To put things in perspective, the photos from my Kodak v610, which is a 6.1 megapixel camera released last summer, and my Panasonic Lumix FZ20K, which is a 5 megapixel camera that's about three years old, are better than the photos from the S9100! Then I remembered that a cheap camera from Fuji, the FinePix A700, also uses the same 1.6-inch Super CCD HR sensor. So what we've got here is a sensor from a $157 camera, being used in a camera that originally retailed for over $600. (Now it goes for about $420). That hardly seems appropriate to me, and as the say, the photos tell the truth. Have a look at a crop from one of the photos taken with the S9100 below. It's cropped directly in Photoshop, at 100%, with no other editing whatsoever.

Sample photo from the S9100

Do you see what I mean? That photo's no good, and every single one of the photos looks like this at 100%. All of the detail is gone because of that cheap, overworked sensor. Fuji might as well not have released the S9100. The inadequate sensor ruins it.

Oh, and remember those gorgeous long-exposure shots of the sky? They were all completely dark when I viewed them on my laptop. I mean pitch dark! And yet they appeared beautifully exposed on the camera's screen. What happened? I'll tell you: the camera can't adjust the live preview accurately when composing the shot, and what's worse, instead of reading the real image from the card and displaying it on the screen after taking it, it re-displays the stored live preview image instead. So I had no real way of knowing what those photos looked like when I took them. I suppose I could have switched to playback mode, but who'd have thought that the camera's display would be this inaccurate?

After we saw all this, Ligia and I looked at each other, and we knew what had to be done. Even though I prefer to test out cameras for a month so I can get a really good feel for their usability, given the S9100's shortcomings, there was nothing else to do but to wrap it back up and sit it nicely in its box. It's going back. I was so disappointed. I loved the body, loved the grip, loved the zoom and ease of use, but when it came to its most important feature, the sensor, I just couldn't live with it.

Here's hoping Fuji sticks a good sensor in this camera at some point in the near future. Until then, my advice to you is to stay away from it.

Powered by

About Raoul

  • James

    The Author’s comments says it all:

    “I’ve used so many cameras by now that unless I’m using a new DSLR, I don’t need to read the instruction book”

    The Pride comes before the Fall. If the Author had bothered to open the Instruction Book, he would have found several ways to get from Point A to Point B and get fantastic photos along the way. But then again, he was more intent impressing his lady friend at a Sushi Bar. This camera review is a joke of the first order that doesn’t do this camera justice. How many 16″ x 20″ prints have you printed that were taken with this camera, bub? I have printed over 50 and they rival photographs I have taken with my Toyo 45AII using a 135mm Fujion Lens…but then I’ve already lost you because you don’t have a clue.

  • Kye Wood

    You whole problem came from shooting raw – and not reading the instructions.

    If you wish to appear credible, ask for someone to convert the raw files you had originally taken, using the Fuji Raw converter – then go hard and ugly if you still feel like a robbed consumer.

    To quote you:
    “I doubt Adobe would screw up RAW conversion so badly that their software would be to blame for the high noise and poor photo quality encountered with the S9100.”

    To quote TrustedReview – re Fujis Raw Converter VS the ones you used.
    “Batch processing a few dozen RAW files can take the best part of an hour. Unfortunately using the Camera RAW plug-in for Adobe Photoshop doesn’t produce anything like the same quality, and can result in very noisy high-ISO images, so you’ll just have to be patient.”

    Dude, we all make mistakes.

  • otto uberswengen

    auf wieder sehen!
    My farewell remarks will include something from a well respected pro and camera reviewer.
    See if you can guess who( to make some fun out of this dreary and endless tit-for-tat)
    Quote 1: “If you can’t make decent and clear pictures from your existing camera, it’s not the camera.”
    Quote 2: “Magapixels have nothing to do with a great photograph”
    Quote 3: ” If you can’t sink a putt from any distance, it’s not the putter’s fault”
    Quote 4: “a poor carpenter blames his tools…”

    Q3&4 are from Otto who is about to attend to a gas leak before reviewing reviewers of super-market trollys…
    Have a great day everybody and don’t sweat the small stuff…
    contemplate at your leisure:
    GOD: Just because you are one of my sons does not mean I won’t let you die.
    Buddha: Cease Mind, Cease Suffering.
    Lao Tzu: This that can be named is not THAT.
    George Bush: It’ll take some time before we can restore chaos.
    with the warmest regards to anyone who might read this,

  • otto uberswengen

    If by chance other readers might have an interest in this thread, do look up Steves Review, Trusted Reviews and of course DP Review where the results from the S9000/S9100 looked extremely convincing. Trusted Review gave the cam 10/10 and images from Steve(protraits) and DPR were nothing short of extremely good.
    Debating about Megapixels and file-size can be a world upon itself. Something like giving a bunch of men a measuring tape.
    It’s obvious that DSLR are superior (otherwise pros would all carry just a compact in their pockets) when it comes to IQ, but for a 6X4 or a 5X7, (and considering the PRICE)
    cameras such as the S6500fd can be just spectacular.
    I would tend to argue that for most people who smoke, lighting a cigarette with a blow-torch is a bit of an over-kill. Similarly an S6500fd or even the “dreaded” S9000/S9100….is all any sane person might require for a decent photograph provided you know the limitations and use the tool appropriately.
    For extreme IQ, I would recommend going to medium format.

  • bivith

    This site mentions the resolution doubling in RAW.


  • bivith

    One final point. Your review mentions using Lightroom for the capture. Every other review that I’ve read that mentioned the RAW function, also mentioned that all the 3rd party convertors did a poor job when compared to Fujifilm’s own Hyper Utility software. They also mentioned that Hyper Utility was pig slow and unintuitive, but the final results were far superior.

    There’s also a freeware raw convertor out there specifically for Fujifilm cameras called S7RAW which is what I use, and it’s pretty decent too.

  • bivith

    After checking, you’re correct it is just the one diode per photosite on that camera. I’ve also checked and the RAW files are 18MP, so you’re conversion software must be pulling it back to the rated 9MP. I’ve no idea why this is!

    As for noise at 1600 iso, point me in the direction of a non DSLR that doesn’t have, and compare it to the fuji. I think Fujifilm’s real strength is in the in camera processing. I’ve used a Panasonic Lumix super zoom, and the noise at 100 iso shocked me, but since the F10 Fuji has always produced great jpeg’s at high ISO’s. The praise for lower noise is always when compared to other compacts and small sensor cameras. No-one is claiming that it can do better than a DSLR.

    perhaps the s6100 is the camera your after, which is a 6-megapixel in a very similar body. It’s missing a hot-shoe though which is a shame, and the sensor is slightly smaller at 1/1.7-inch.

    These are some my shots on the s5200/s5600 which has a 5MP 1/2.5-inch Super CCD sensor.

    The first 2 pages approx, are using that camera, and the most recent ones are when I started using it in RAW mode. I didn’t have to do a lot of processing, not really much more than I would do with a JPEG anyway.

    Workflow was, using S7RAW software for conversion, pull back exposure to rescue highlights if necessary/possible, small amount of sharpness, and maybe a curve/gamma adjustment to pull back some brightness lost to exposure adjustment and generate tiff file. Open in Paint Shop Pro XI, and use Neat Image plug-in for noise reduction, and then use Paint Shop Pro to punch up/adjust contrast, and sharpen. Nothing spectacular there really.

    Maybe you got a bad model 🙂

    Anyway, I’ll be on the lookout for an s9100 when the price drops. I think the s6100 looks like a better camera, but the lack of a flash hotshoe is a sticking point.

  • bivith

    Forgive the abuse of the apostrophe in my previous post!

  • bivith

    I do use Fuji cameras and there a couple of points to make.

    Fuji use their own CCD’s mostly, and rather than being cheap, are regarded as being very innovative what with having two sensor’s per photosite, one for the lowlights, and one for the highlights, extending the dynamic range captured. Fuji RAW files are 14-bits per pixel, whereas a lot of DSLR RAW files are 12-bits per pixel. Because of these 2 sensors per photosite, Fuji RAW files tend to have twice the number of stated megapixels, which in the s9100’s case should load into the RAW convertor as an 18MP image.

    Yes, there is interpolation, but almost all digicams, including DSLRs, interpolate due to the bayer filter used when the image is captured. Also Fuji photosites are octagonally shaped, and are not arranged in straight rows, making interpolation unavoidable, but it’s nothing to do with extending the megapixel rating.

    As you end up with an 18MP image, which is double the number of photosites on the camera, the unprocessed images will look soft at 100%. Due to the anti-aliasing filters used on digital sensors, the image is also artificially softened to prevent moiring.

    Check the res of the unprocessed as RAW. If I’m right it will be much higher than the 3,488×2,616 stated JPEG maximum ( i would guess 4,982×3,737 aprrox. ). Try resizing it to 3,488×2,616 and then determine what it looks like at 100%. Some RAW converters take this into account though, and will resize it accordingly automatically.

    As for the noise, it’s a fact of life with digital sensors, but the Super CCD is generally regarded as one of the best performers in this area, but it’s still a small sensor at the end of the day. The in camera processing goes a long way to supressing noise, and Fuji do a good job of it, at the cost of some detail at high ISO. With RAW, 3rd party software like Neat Image should be used, and this gives much better results than the in camera processing.

    And finally you’ll need to tweak the levels, and curves, and apply the relevant sharpening to get the best out of the image, just as the camera does in JPG mode, but you have a lot more control.

    The preview you see on the LCD is a JPG embedded in the RAW file, with all the camera processing applied as per the settings of the camera when you took the picture. This is why it doesn’t look the same as the RAW file unprocessed. It’s been processed by the camera.

    And finally, people don’t look at 100% crops, they look at either pictures resized to fit their screens, or they look at prints, and from my experience Fuji cameras make great prints.

    The first time I used RAW on the Finepix S5200, I was a bit disappointed, for similar reasons, but after a bit of effort started getting fantastic results, and far far better pictures that I ever got with JPG mode, and in camera processing. I would expect similar with the 9100. That write speed is a killer though.

  • otto uberswengen

    Dear Pop,
    I have not bought the camera. I don’t have a soft spot for Fuji cams or any other cam.
    I am sorry you missed the “far-fetched” analogy.
    Never mind…
    My point is that the cam which you so much disliked at 100% crop, isn’t the way it’s intended to be used by 99.99% of users. Most people just want “fotos”.
    If you don’t mind my saying so,I just think your criticisms are a bit harsh, though it’s clear that you intended it that way, to get even for having bought such a “dud”…
    I am sorry for your disappointment and I don’t feel particularly “inadequate” for pointing this out to you.
    Why don’t you review things that you actually “like” and perhaps concede that most people who use non DSLR’s like to keep their photography simple, inexpensive and free of
    powerful likes and dislikes?
    My point if there is any, might be that perhaps, you did not review the subject the way it was intended for use. If you drive a people-mover like you would a Ferrari, you are bound to be disappointed…oh sorry, you are not into this kind of thing…no offence intended. Have a nice day, with which ever camera you end up buying.

  • otto uberswengen

    nothing personal and certainly not an “attack”…taking RAW shots in a sushi-bar seems to point to a very curious approach to “testing” a camera.
    I ask why it is, that other testers do not apply this strategy. They usually try a variety of scenarios to test a camera’s overall capability.
    According to Pop, the writers at “Trusted Camera Reviews” and “Steve” for example, have all got it WRONG! They have not the faintest idea that the CCD sensor in this camera is crap, yet they were able to get more than reasonable results with it. Moreover, they even have a certain amount of praise for it.
    A Stradivarius in the hands of a fool will sound very ordinary.

  • otto uberswengen

    You should have given it a bit more time.
    How come your findings are very much opposed to the existing ones?
    Perhaps you were expecting too much.
    Most people who buy cameras other than DSLR do not bother with RAW. It’s just too much of a waste of time.
    You ought to have found out A WAY of making the photos acceptable, telling us how you did it…
    Perhaps starting with the manual is not a bad idea.