The latest Contest!
Greetings, you’re listening to the Liam Photography Podcast, I’m your host Liam Douglas and this is Episode 183.
So as many of you know Fujifilm recently held their X Summit on-line presentation and announced their new cameras and lenses and although some of the news is exciting, especially for Fujifilm X shooters, most of it was “ho-hum” for me personally.
Now I know what you are thinking, “But you’re a Canon shooter, why would you care about what Fujifilm announced?” Well, remember I am also a Fujifilm shooter, I own the GFX 50R Medium Format mirrorless system, and although I do love the camera, it does have it’s short comings. No GPS and Contrast Detect AF are the two HUGE issues I have with the platform.
I understand that the camera is also slow in the frames per second speed, but I do not care about that, I know that Medium Format is not built for shooting sports and I am fine with that. But, come on, WHY in God’s name would you use Contrast AF when EVERYONE KNOWS Contrast AF sucks. It’s slow, it’s inaccurate, it’s just not the AF system that ANYONE should ever use in a camera in this day and age. I mean they put Phase Detection AF in the GFX 100 and 100S, why not the 50R, 50S or 50S Mark 2? I mean it isn’t like adding Phase Detect AF would drastically raise the cost of making these cameras. The have Phase Detect AF in all the X Mount bodies, so why not in ALL the GFX bodies? To me it’s just a way to screw the consumer, but maybe that’s just me, but you can buy a sub-$1,000 camera and get better AF than in a Medium Format system that costs anywhere from $4,000 to $6,000???
Then we have the X Series announcement, which was the X-T30II, which is basically an X-T30 with new firmware??? Ok, remember a few years back when the Fujifilm CEO at the time said that they were using the “Kaizen” model which was giving Fujifilm shooters basically a new camera every time they released a new firmware that would be packed with new features and enhancements? Well, apparently the NEW CEO at Fujifilm has decided that doing firmware updates for free that enhance the user’s camera and adds new features wasn’t a good model to make money as the days of “Kaizen” are over and now, instead you have to buy the next Generation of an X Series camera to get the stuff you should have gotten from a firmware update. The X-T30II has NO new hardware and is strictly as Fujifilm calls it, “firmware on steroids”. Don’t get me wrong I am not saying Fujifilm doesn’t make great cameras but now, instead of keeping their users excited about the next set of feature enhancements they can get for the camera they already own, they want you to keep buying the next revision year over year.
That may be fine for your hardcore Fujifilm X Loyalists, but it’s not going to generate excitement for other shooters to switch platforms, especially when Fujifilm has been sorely lacking in innovating camera technology. Fujifilm doesn’t have the deep R&D Pockets that Canon and Sony have, so they cannot compete there and they cannot mass produce cameras the way Canon and Sony can either so they are kind of stuck there, which to me is why they should be giving their customers the most bang for their buck.
Fujifilm might be able to keep things exciting by making “niche” cameras, kind of like a less expensive Leica brand. They have had some success with that with their Instax Brand of cameras that young people really love to use as their creative outlet. They also offer their Instax cameras in some really cool and unique colors, which kids love, making the Instax instant camera system hugely successful, couple that with the ability to print right out of the camera and you have a winning combination.
With Fujifilm’s new CEO stating earlier this year that they are not planning to sell off the camera division as it’s their “contribution to society”, doesn’t instill a lot of confidence in me. If you remember, Olympus’ CEO said the same thing and then they sold their camera division off to Japan Industrial Products, effectively killing off the Olympus camera brand.
Now I am hopeful that unlike Olympus’ CEO, Fujifilm’s new CEO is being honest about the camera division not going away, but they need to step things up in order to continue to thrive and survive. Fujifilm has very beautiful designs for their cameras, especially with the number of their bodies that are design to look like vintage cameras from the film days. Now that will NOT be enough to keep Fujifilm alive, especially when Nikon is now doing similar designs with their new Z fc system and Sony’s new A7c system is going after that “niche” market as well. Cameras that look retro and are just a lot of fun to shoot with, those are things that get people excited about photography and excited about brands. Fujifilm dominated here for a long time but now Nikon and Sony are coming for Fuji’s jugular with their new models in that same space.
Now as I mentioned before, Fujifilm’s new CEO realizes they cannot keep giving major updates away for free, especially when you have the cost of staffing a software development team to stay on top of giving away major new features and enhancements with firmware updates. But Fujifilm does need to come up with other ways to stay relevant and in the game when facing the fierce competition from all the other companies. According to the most recent reports, Canon and Sony own basically 90% of the global camera market, which leaves Fujifilm and others to fight over the remaining 10%, which is not much of a market and makes for little opportunities for profit.
If Fujifilm can continue to come up with exciting “niche” cameras they have a good shot at staying in the game. Fujifilm makes great products, but another area where I think they strayed was going with their X-Trans sensor in the X Series bodies. Some people claim it gives better image quality and it might, I have not shot with one so I would not know there, but a close friend of mine, Brent from the Latitude Photography Podcast gave Fujifilm X a try to see if he could switch from Canon and he was not happy with the way the X-Trans sensor performed. I spoke with Brent about it a bit on Facebook Messenger and he said that for landscapes and travel photography he ran into weird issues with some odd looking artifacts in his images that were caused by the X-Trans sensor.
I am still confused as to why Fujifilm went with the X-Trans sensor in their APS-C cameras but then use a Bayer sensor in their GFX bodies. It would seem that it would be more cost effective to use the same type of sensor in both camera platforms. And since the GFX platform is an image quality beast of a system I am not really sure where the thinking that the X-Trans sensor is better for image quality comes from.
Fujifilm introduced the first X-Trans sensor in 2012 in the Fuji X-Pro1 and the idea behind this sensor was to make a new way for the sensors color filters to create the image. At their core, image sensors are like black and white film: They only see a range of brighter or darker tones, with no color information attached. To let them detect color, an array of tiny red, green and blue filters are essentially painted onto the chip, so each sensor pixel will only respond to one of those primary colors. It’s as if you had three cameras, each capturing monochrome red, green and blue images that were then combined together to produce a full-color result.
There are two problems with this approach. First, the camera has to do some number-crunching to turn each of the separate red green and blue pixels into full-color RGB ones. This isn’t terribly difficult with conventional sensors, but a bigger issue is that the regularly repeating patterns of colored pixels can result in moiré patterns when the subject contains finely-detailed, repeating patterns. (Think of texture in cloth or subjects like venetian blinds.) If you’ve ever held two pieces of window screen at an angle to each other, you’ll have seen the broad swirls of light and dark the conflicting patterns create. The same thing can happen in your camera when a pattern on the subject happens to align in just the wrong way with the regular array of colored pixels on the sensor.
The classic way of dealing with this is to put what’s called an optical low-pass filter in front of the sensor. This basically produces a very controlled blurring of the image, so sharp edges and abrupt color and tonal transitions in the subject won’t cause problems by interacting with the pixel pattern. Rounding off those sharp edges in the image makes the moiré problem go away, but at the expense of a much softer-looking image.
There’s really no way around this; it’s a mathematical fact of life, despite what non-mathematicians might tell you. No amount of processing or fancy algorithms can guarantee to eliminate moiré in all cases, if the subject detail is too fine relative to the pitch of the pixels.
Most cameras these days have dispensed with low-pass filters, getting away without them thanks to sensor resolution outstripping the resolving power of many lenses. In many cases, the lens can’t resolve detail fine enough to cause a problem with moiré, so it’s basically acting as the low-pass filter in the system relative to the fine pitch of the pixels. With the lens blurring the finest subject detail, the camera doesn’t need its own low-pass filter to blur it further.
Of course, you can see the problem with this: What happens if your lens can resolve such fine detail?
It’s as much art as science to design a lens that not only has good optical characteristics, but that can also be manufactured reliably and repeatably. Any manufacturing process is a game of managing tolerances, balancing what you’d like to do in an ideal world and what you can actually achieve on the factory floor. This is even more true of lens manufacturing than other types. One way lens designers hedge their bets is to design lenses such that there isn’t a single point of absolute best focus, but rather a range over which the focus is more or less the same.
An ideal depth of field curve for a textbook-perfect lens and a sensor of infinite resolution would be a V-shape. In the real world, most lenses are designed to have a flattened bottom on the V, which helps accommodate minor variations in part tolerances during manufacturing. Fujifilm adopted a different philosophy, aiming for maximum sharpness no matter what, making the point of the “V” as sharp as possible. This means crisper images, but what about moiré?
To some extent, the better your lens is, the more likely you are to see moiré in images of subjects with repeating patterns in them. Part of the reason Fuji felt free to pursue this approach was that they already had a solution in mind, in the form of their X-Trans sensor technology.
X-Trans changes things up by using a more complex array of color filters. Rather than the two-by-two RGBG Bayer pattern used by most cameras, X-Trans uses a much larger six-by-six array that mixes up the pixel spacing. Depending on where and in what direction you’re looking, the spacing between pixels of a given color can vary quite a bit. This gives the camera’s image processor more spatial information to help it sort out what’s moiré and what’s subject detail.
Of course, X-Trans does still have a repeating pattern of colored pixels, so there’s not a 100% guarantee that you’ll never see a moiré pattern. The bottom line, though, is that X-Trans sensors give the camera a much broader range of spatial frequency information, so its X-processor has more data to use in striking the best balance between moiré and maximized sharpness. It takes a lot of number-crunching (see below), but the different spatial sampling frequencies give the processor more information to work with.
Like Super CCD before it, Fujifilm’s X-Trans sensor technology has continued to evolve over the decade or so since its launch. Since that first generation arrived with the X-Pro1 in 2012, three subsequent generations have refined and extended the technology.
The first X-Trans II chip debuted in the Fuji X100S in 2013. Harkening back to the Super CCD EXR II design that preceded it, Fuji’s X-Trans II sensors added on-chip phase-detection autofocus. Improved pixel circuitry also brought better dark-noise suppression, and the resolution jumped from 12.3 to 16.3 megapixels, as well.
In 2016, the third generation X-Trans III arrived with Fuji’s X-Pro2. It brought still higher 24.3-megapixel resolution, even more phase-detection AF points, and switched from aluminum to copper wiring, simultaneously lowering noise levels and significantly improving readout speeds and thus camera performance. Most recently of all, the Fujifilm X-Trans CMOS 4 sensor arrived in 2018 with the X-T3. This latest generation switched to a backside-illuminated design for even better light gathering capability, and once again boosted the on-chip phase detection pixel count significantly for more capable autofocus. It also increased the resolution a bit, to 26.1 megapixels.
X-Trans technology is fundamental to Fujifilm’s strategy. They view image quality as a key differentiator for their cameras and have invested a huge amount in developing their own sensor technologies in pursuit of it. X-Trans is just the latest iteration in that process, one that was planned years before the processor technology would arrive to make it feasible.
But from what I have read and researched on this subject the benefits of the X-Trans at reducing color noise in not necessary in the GFX line as the sensor and lenses are so much larger, the issues with color noise aren’t as prevalent. Still, it would seem to me to be a huge, possibly unnecessary cost to have two different lines of sensors for their camera systems, but with Fujifilm image quality is top priority and I know many people that have shot with both Fuji and non Fuji APS-C cameras say that the image quality on the X Series system are superior to any other APS-C platform.
Of course no system is completely perfect and I guess the way Fujifilm has arranged the color filter elements in the X-Trans sensor is what causes the issues in certain types of shots as Brent discovered when he tested the Fujifilm X-Series to see if he could use a smaller, lighter system for his photography.
Fujifilm did recently “retire” their X-Trans IV series of sensors earlier this year and many people wondered if they were going to move away from X-Trans in their X Series of cameras but now it has been announced that the next X Series cameras will contain a new generation of X-Trans sensor so it looks like their X-Trans is here to stay with the release of X-Trans V.
For my work, right now I am staying with Fujifilm’s GFX system for my Medium Format work but with the fact that they don’t offer Phase Detect AF only n their much more expensive GFX 100 and 100S models I am disappointed in that decision. I would LOVE to have the 100S for my work but at the moment it is out of my price range. The other big thing that Fujifilm has disappointed me on with all of their GFX systems is the fact that they left out GPS. I know I have complained about it before, but it really sticks in my craw that they won’t at least give us a new communications stack in their firmware to allow us to use external GPS units like we can with most any other camera system.
Fujifilm’s answer has always been use their Camera Remote app on your smartphone for GPS and Geo-Tagging. The problem is their app is buggy, I just tried to test use it again just now and the app won’t even link fully with my camera. I deleted the phone paring registration in the app and paired my iPhone 12 Pro Max again via bluetooth and then the phone connects to the Fujifilm GFX 50R’s WiFi network but the app just hangs when trying to activate Live View shooting mode. Hence the other problem with Fujifilm’s “solution”, the app doesn’t allow you to just pair the two via Bluetooth for GPS info sharing, the app wants you to shoot using it as well and I don’t want to shoot that way.
In addition, the app doesn’t make it easy to Geo-Tag your images as you have to power on the camera, connect to the camera’s WiFi network and then in the app go to Geo-Tagging and tell the phone to send the location to the camera. This is NOT an efficient way to Geo-Tag, which is something that pisses a LOT of Fujifilm shooters off. I mean come on, in this day and age it’s not that expensive to put a GPS chip in your camera or at least put the communications stack in your firmware so the customer can use an external GPS unit to Geo-Tag their images.
I have gotten so frustrated with this that when I shoot with my GFX system and need to have the images Geo-Tagged, I use my Canon with it’s GPS hotshoe unit or my iPhone to take a second shot and then copy the GPS data from one file to another in Lightroom. I was trying to use Capture One Pro 21 only to edit my photos, but Capture One doesn’t give you the ability to copy the GPS data. This is something I have complained to Capture One about and they said they are taking it as a feature request, which is fine, but how about you actually do it.
Ok, so wrapping up, I am done ranting about Fujifilm I do want to share that I had contemplated switching to the Hasselblad X1D 50CII camera system, which does have GPS built in for Geo-Tagging but they use the same sub-par Contrast Detect AF system that Fujifilm uses in the camera I already have so again my only option to get better AF is to upgrade to the GFX100S, since Fujifilm decided to keep that same crappy AF system in the new GFX 50SII instead of upgrading the second generation with Phase Detect AF.
Again, I am NOT saying that Fujifilm is going the way of the dinosaur at the moment, but they do need to come up with new ways to be innovative and keep customers using their systems and possibly switching to their systems. I had reached out to Brent to get his wording on the exact image artifacts he experienced when testing an X Series system but he was unavailable for comment. I may at some point rent an X Series system to test for myself and maybe I can replicate the issues he ran into.
In the meantime, let me know in the comments on the Facebook Group your thoughts. Are you a Fujifilm X Series shooter? Why did you decide to go with that system over other APS-C systems? Have you run into any of the weird image artifacts?
Also be sure to join the Liam Photography Podcast Facebook Group https://www.facebook.com/groups/liamphotographypodcast/ You can reach the show by call or text @ 470-294-8191 to leave a comment or request a topic or guest for the show. Additionally you can email the show @ firstname.lastname@example.org and find the show notes at http://www.liamphotographypodcast.com.
You can find my work @ https://www.liamphotography.net on and follow me on Instagram, Facebook and Twitter @liamphotoatl. If you like abandoned buildings and history, you can find my project @ http://www.forgottenpiecesofgeorgia.com. and http://www.forgottenpiecesofpennsylvania.com.
Please also stop by my Youtube channels Liam Photography