I’ve always been somewhat fascinated by the idea behind the Foveon sensor. Making use of the different wavelengths of red, green and blue light, the Foveon sensor stacks three layers of photosites, each recording a different color for the final image. The advantage: no color moiré due to a lack of need for demosaicking, and thus no need for an anti-aliasing filter. The result: uncompromised sharpness, and theoretically high color fidelity. So when I got the chance of using a Sigma SD14 recently, I just had to take the opportunity to experience the Foveon sensor myself. Oh, and yes, I know, we’re late with this article …
The Sigma SD14 was introduced back in 2006, and even for that time, its specifications were pretty unspectacular. The slightly-smaller-than APS-C sensor has an effective resolution of 4.7 megapixels, ISO goes as high as 800, the screen has a 150k pixel resolution and it takes up to 3 frames per second. Compared to the competition, that was pretty low-end even back then. On the upside, the SD14 has a nice 0.9x magnification viewfinder with 98% coverage, which is pretty good, can be remote controlled, and has a very user-friendly interface with straight menus, well-placed buttons and a top LCD panel. Its real strenght, though, is said to be its Foveon sensor, which allegedly delivers outstanding resolution and color fidelity. Having heard so many enthusiastic reports about the Foveon technology, I wanted to see for myself.
DISCLAIMER: This is in no way meant to be a technical and precise review of the Sigma SD14. Rather, it’s a report on my time spent with the camera, and my assessment of the quality of its output. So don’t expect any in-depth discussion of the placement of the remote control connector or the layout of the menu. Also, don’t expect any studio measurements of signal-to-noise ratios, or MTF charts of the lenses I used. This is a highly subjective piece, with some scattered objective information here and there.
The SD14 is a pretty straightforward camera. Most buttons are where they are supposed to be, although it does take a little time to get familiar with how the camera works. The menus are almost minimalistic, offering no more functionality than you absolutely need. What I particularly liked was the quick panel with direct access to ISO, white balance and a couple other often-needed settings. I hardly ever needed to access the main menu at all, since the functionality of the camera is limited to the basics. Mode wise, it offers P, S, A and M. You have a couple drive options, and that’s it.
The shutter button is placed nicely on the slanted top of the grip, and is surrounded by a clicky-wheely-dial that changes aperture, shutter speed and a bulk of functions that you can tweak via the top LCD. As I said, no much need to dig into the menus. Everything’s readily accessible.
The body lies in your hand very nicely. I don’t have particularly large hands, and I found it to be a nice fit, if not a tad on the large-ish side. So larger hands than mine will probably still feel comfortable. What I found peculiar is that the camera feels so heavy when you hold it. Compared to an entry-level DSLR from one of the major manufacturers, this thing weighs almost a ton. But that can only be a good thing, meaning it’s probably very solidly built.
The built-in flash is nothing to get excited about. It’s a built-in flash: it does its job, and that’s it. If you want to properly light stuff, get proper lighting equipment. The autofocusing was quick and reliable with the two lenses I used. Nothing to complain about. The SD14 has five AF points that light up when correct focus is achieved — one in the center, and four to the top, bottom, left and right of it. Nothing spectacular, but absolutely sufficient. I have always been using the center focus point, anyway.
The viewfinder is moderately large and bright. For a camera with a 1.7x crop factor, I’d say it’s fairly large. Nothing like a full-size full-frame viewfinder or those absolutely gorgeous VFs in the old 70′s and 80′s SLRs like the Pentax ME or Olympus OM-series, but then again nothing like the tiny peephole you find in most entry-level DSLRs — you know, the ones that make your eyes hurt if you take more than a couple pictures. All in all, I found the SD14′s viewfinder to be okay. Again, nothing spectacular, but also nothing to worry about.
With the Sigma SD14, I was using two zoom lenses: the Sigma 18-200mm f3.5-6.3 DC OS (the link leads to the latest version of that lens) and the Sigma 18-50mm f2.8 EX DC. While the former comes with an integrated optical image stabilizer, the latter trumps with a constant f2.8 aperture. The 18-200 is a nice all-rounder, especially if you’re on vacation and want to go light, while the 18-50 worked nicely for portraiture and closeups, and in situations where the superzoom’s mediocre light-gathering capabilities were of no particular use.
I really liked the reach and convenience of the 18-200mm, but that’s about it. Overall, I found its image quality to be mediocre. Sure, it’s sharp when stopped down, but not much more. I admit, I’m a sucker for fast lenses with shallow depth-of-field wide open, and this lens provides nothing of that. But that’s not what it’s intended for. As I already mentioned, this is probably an awesome travel lens, and the OS (image stabilization) helps a lot when shooting at the longer focal lengths stopped down.
This lens was more to my liking, due to its constantly large aperture — especially at the long end. At 50mm and f2.8, the lens was able to provide some moderately shallow depth-of-field for separating your subject from the background. Translated to 35mm full-frame terminology, the 50mm end of the lens corresponds to an 85mm lens, which is the perfect focal length for portraits. While pretty mushy wide open at the wide end, the tele side of the lens produced acceptably sharp pictures with nice detail even at its largest aperture setting. As to be expected, stopping down only improved things.
If I were to buy a Sigma today, I probably wouldn’t get either of the lenses. I would probably get the Sigma 20mm f1.8, 30mm f1.4 and 50mm f1.4 lenses, which would make a perfect 35mm-50mm-85mm setup with fast apertures for lots of creative depth-of-field control. But I won’t be getting a Sigma anytime soon.
Inside the Sigma SD14 works a nominal 14.7 megapixel Foveon X3 sensor. The nominal pixel count arises from the fact that the sensor uses three layers of light sensitive silicon cells, each sensitive to one of three light wavelength spectra that translate into the colors red, green and blue as our human eyes perceive them. Since for each single pixel in the final image the information of three stacked red-, green- and blue-capturing pixels is added together, the factual resolution of the SD14′s RAW and JPEG output is 4.7 megapixels. However, since there is no need for interpolation of color information, the Foveon sensor’s output shows much more fine detail per pixel than that of a conventional Bayer pattern sensor. (For more in-depth info on the Foveon sensor technology, I recommend you indulge in the lecture of the corresponding Wikipedia article.)
100% crop of the above: under the right conditions, and with a little tweaking in pp, the SD14′s output shows very crisp and fine detail, and good separation of subtle tonality shifts:
Or at least that’s what they say. While I must say that the color rendition of the Foveon sensor is pretty unique and has an almost film-like quality in the way subtle tonal differences as well as overall contrast differences are rendered, it did not blow me away like I expected. This might be due to the fact, though, that I’ve been spoiled by the output of the Leica M8 for two years now.
As I said, the color rendering of the Foveon sensor is pretty special and almost film-like, and I liked it quite a lot. I do see a huge difference to what conventional Bayer sensors produce, but I have a hard time finding the correct words to describe it. It just looks different. Sharpness-wise, I can’t really make out any difference to what I see in the pictures from the Leica M8. If anything, the Leica’s DNG output appears to be even a bit sharper and have a crisper detail than that of the SD14. The color rendering of the M8 also appears more natural to my eyes. More like slide film, whereas the output of the SD14 is more like that of color negative film. If that makes any sense at all.
Was this shot taken on Kodak Portra 160?
The Foveon sensor as it is built into the SD14 has a couple of downsides, though. One is its tendency to produce an overall greenish-yellowish tint in its pictures. This can be corrected when shooting RAW, but I needed to dial in lots and lots of extra magenta in order to get the color straight in some pictures.
Next, it’s the way the sensor handles bright reds. Bright reds are always a problem and many digital capturing devices have trouble separating the subtle hues, but this one does an especially horrible job. In many pictures containing bright red elements with subtle shifts in hue in the original scene, the object as it appears in the pictures seems to be of one continuous bright hue that eventually clips at some point. Here’s an example of what I mean:
The way the Foveon sensor reproduced the reds in my son’s sweater jacket is simply horrible.
Thirdly, the high ISO performance of this particular Foveon sensor is ground level. Heck, I think it’s even below that. Use any ISO setting other than the base 100, and you will get big fat colored blotches all over the place. Even when pushing ISO 100 shots these blotches can appear. I say can, because they don’t seem to appear consistently. Some pictures show them to a larger, some to a lesser degree. Some don’t show them at all. It’s as if the sensor was kind of moody. I had to dial in tons of chroma noise reduction to get rid of the SD14′s color noise, with the effect of also getting rid of lots of fine color detail.
Where the sensor absolutely shone was when there were spades of light. Like outside on a bright, sunny day. In such a case, the color rendition was often amazing, there was no noise visible at all (neither luminance nor chroma), and with the lenses stopped down there was lots of fine detail. Under ideal circumstances, the pictures that come from this sensor can be absolutely amazing, with an almost organic look. You wouldn’t tell these are digital captures. More like super-slow super-low-grain slide film projected to a perfectly clean and white wall from a perfectly calibrated, super-bright projector.
While the Sigma SD14 isn’t a bad camera overall, it’s also not a spectacular camera. If you’re purely a photographer, you will appreciate the straightforwardness of its ergonomic layout. If the light(ing) is right, its sensor procuces spectacular results, but the files almost always need a lot of tweaking in post-processing to get the most out of them. By todays standards, the output is no longer as amazing as it might have been when the camera was first introduced. The latest-generation Foveon sensors are far far far better — by a really huge margin.
Lots of light is what the SD14′s Foveon sensor wants. Then, and only then, it can produce stunning results.
Still, the output of the SD14 has something going for it, which is its unique look and the sharpness and color fidelity you can get under the right circumstances. But that alone isn’t sufficient, not to me at least. All in all, the body is too heavy, the sensor is too moody and the output too inconsistent to really recommend the camera. If you absolutely must have a Foveon-equipped camera, I suggest you get one of the older DP models which feature the same sensor, but in a much smaller package. Or, you save a little more and get the SD1 right away.
If you’re mainly a studio shooter or shoot under highly controllable conditions, the SD14 (or its successor the SD15) might still be an interesting choice. The files have so much quality that large prints should be no problem at all, and as I mentioned, with the proper lighting and some fine-tuning in post-processing, the results can be amazing. So, if you’re on a budget, why not get an older-model Sigma SD.
For me, personally, this is not a camera I would buy, for the reasons stated. I did enjoy my time with it, though, and captured some really nice pictures. However, I’m much happier with my Micro Four Thirds gear, which is much smaller and can output some pretty nice quality pictures that are good enough for my taste.
Please Support The Phoblographer
We love to bring you guys the latest and greatest news and gear related stuff. However, we can’t keep doing that unless we have your continued support. If you would like to purchase any of the items mentioned, please do so by clicking our links first and then purchasing the items as we then get a small portion of the sale to help run the website.