Hidden inside the retro-designed body, the Fujifilm X100F has Fuji’s latest X-Trans III sensor. The first generation X-Trans sensor was introduced on the X-Pro1 camera in 2012. But what exactly is this unique sensor and what does it mean for your photographs?
Sony is who manufactures Fujifilm’s APS-C X-Trans sensors. They are actually ordinary sensors found in a number of different cameras by Sony, Nikon and Pentax. What’s different is the color filter array placed on top. Instead of using the traditional Bayer color filter array found on almost every digital camera since its invention, X-Trans uses a completely different pattern.
The Bayer mosaic uses a 2×2 pattern with 50% green, 25% blue and 25% red light sensitive sensor elements. It was developed in 1976 by Bryce Bayer who was working for Kodak. The advantage of Bayer is its simplicity, while the disadvantage is that it is subject to aliasing.
The X-Trans mosaic uses a 6×6 pattern (made up of four different 3×3 patterns) with 55% green, 22.5% blue and 22.5% red light sensitive sensor elements. The advantage of X-Trans is that it’s not subject to aliasing, while the disadvantage is that it’s much more complex to process.
The original reason Fuji developed X-Trans was to eliminate the need for a low-pass filter, which blurs the image slightly to prevent aliasing. However, it was discovered that as resolution increased the need for a low-pass filter decreased, so many modern Bayer cameras don’t have one.
Digital cameras use the green sensor elements for luminosity information, while blue and red are for color information. Because the X-Trans has more green sensor elements (55% vs. 50%), it has a larger dynamic range (in the shadows) and better high-ISO performance than the same sensor with a Bayer color filter array.
Because X-Trans is much more complex and has to be handled differently, Fuji went back to the drawing board and rethought camera processing. A few things came out of this.
First, they adjusted how digital noise is handled, making it less saturated and more random, which produces a film-grain-like result. Comparing really high ISOs, there is a distinct difference between X-Trans and Bayer–not necessarily the amount of digital noise, but the look of the noise.
Second, to maximize dynamic range and prevent clipped highlights, they programmed the software to underexpose while boosting midtones and shadows to the correct exposure. This is why ISO 200 is the base ISO. The camera is actually at a lower ISO, but increasing everything except the highlights to an equivalent of ISO 200. This is also why the extended ISO 100 setting has a smaller dynamic range. Some have called this cheating, but I call it smart programming.
Third, they took camera-made JPEGs seriously, and decided that they shouldn’t look like crap. Fuji studied their films, and they also studied what photographers typically want their pictures to look like, and made JPEG settings that could actually produce out-of-camera results that photographers would actually appreciate. The X-Trans III cameras can even add fake (yet convincing) film grain.
The disadvantage of X-Trans is that it’s significantly more complex. It requires more processing power. It’s not quite as fast. A big obstacle for Fuji has been heat. There’s a little more guesswork within the algorithms, and it doesn’t always get things right. Adobe has struggled with how to appropriately handle the RAW files.
For me, the advantages outweigh the disadvantages. Fujifilm has smartly designed their cameras to deliver results more in line with bigger and more expensive gear. That’s why there is a passionate cult-like following from their users, including myself.