GFX100 II - A7R V - Phase One XF Comparison
PHOTOGRAPHY GEAR
Fujifilm GFX100 II - Sony A7R V - Phase One XF Comparison
So here we are, 2024 has come and the choice for camera systems is better than ever. I recently had the opportunity to test the new Fujifilm GFX100 II and its new Tilt-Shift lenses. So how is the combo GFX100 II and T/S lens performing?
Well, let's get to the point: I think that the Fujifilm GFX100 II produces the best image quality of all the digital cameras I have used so far. Colors are great, ISO performance is solid, and the output looks natural and beautiful. Additionally, the sharpness and construction of the GF 30mm T/S lens is great. Unfortunately, due to time constraints I was only able to test the widest of the 2 lenses in the kit received which is the most relevant one for architectural and interior photography work. I should also mention that I received the kit at the worst possible time when our Alberta winter was at its peak. With temperatures between -40 °C and -30 °C for 2 weeks, my hopes to get out and shoot a nice building exterior vanished for the sake of both the camera equipment and my fingers. I still thought a real-world example would be better than shooting charts and stuff nobody really cares about, so I decided to photograph my recently renovated kitchen. I understand that reviewing sharpness and digital noise on a web browser can be challenging, which is why I have decided to share all the RAW files and comparison photos here for you to review. Image quality, sharpness and colors are all subjective to some extent depending on one’s eyes, expectations, and requirements.
Before jumping into it I should add that:
- I purchased my Phase One camera in 2016 and the IQ260 digital back was first released in 2013! The Phase One system used in these tests is consequently rather dated in comparison to the current GFX100 II and Sony A7R V. I didn’t have access to the latest IQ4150 digital back for these tests, so comparing ISO and dynamic range would be unfair. Sharpness on the other hand can be evaluated even though the Phase One XF doesn’t have T/S lenses or any kind of movement available. A better comparison would have been to use Rodenstock and Schneider lenses on a technical camera with a newer Phase One digital back.
- The IQ260's sensor was not cleaned before these tests. I apologize for all the dust spots! I know what I am doing this weekend…
- The Canon TS-E 24mm that I use on the Sony camera is stellar. Back in 2017, I had purchased a first copy of the lens that I returned after a few days because I was not satisfied with the output. Quality varies from lens to lens unfortunately, and I read over the years many reports of people complaining because they had a bad copy of that lens. I have no way to know at this point in time if discrepancies in quality exist between Fujifilm T/S lenses. I tested the lens provided as is with nothing else to compare it to. Your experience with your own lens may vary.
- I am not affiliated with any of the brands mentioned in this short review. I currently shoot both Sony and Phase One cameras for work.
- These results are based on my own expertise and opinion. The best way to evaluate a camera and/or lens for your specific needs is to rent them or borrow them from a friend.
- A great image is not, and will never be, the sum of its pixels reviewed at 100% scale on a computer screen. Many factors contribute to creating outstanding work with different cameras and lenses. I follow the work of hundreds of talented individuals that inspire me daily. They all shoot a variety of equipment, and all have their own style and vision. We get paid by clients for our creative process, consistency, and professionalism. None of us get hired for the size and quality of our pixels.
Sony A7R V - Fujifilm GFX100 II - Phase One XF
Method
The goal was to test the different camera-lens combinations, not to create a portfolio image or a hero-shot of the kitchen. Everything was shot with ambient light only. Each camera was shot on a tripod in the exact same position and tethered to a computer running the latest version of Capture One. The composition is slightly different due to variations in focal lengths and image formats (4:3 for both the GFX and Phase One XF, and 3:2 for the Sony). Each shot was repeated 5 times over 5 days at the same time of the day to make sure that results and focus accuracy were reproducible. I always tend to overexpose shots by about half a stop compared to a camera’s internal light meter reading because I recover highlights with bracketed exposures and luminosity masks. I consequently shot these images the way I would for a work assignment.
BTS image capture and basic settings in Capture One
The following cameras were used:
Sony A7R V 61MP BSI-CMOS sensor 36x24 – Pixel pitch: 3.76 micron – Base ISO 100
GFX100 II 102MP BSI-CMOS sensor 44x33 – Pixel pitch: 3.76 micron – Base ISO 80
Phase One XF IQ260 60MP CCD sensor 54x40 – Pixel pitch: 6.0 micron – Base ISO 50
Once again, the Phase One was used for reference for sharpness only. It would be absurd to compare the performance of a newer BSI-CMOS vs a CCD sensor in terms of dynamic range, colors and low light performance. A better comparison would have been the IQ4150 with its BSI-CMOS sensor, but I didn’t have access to one during testing.
The following lenses were used:
Canon TS-E 24mm f/3.5L II (35mm format)
Fujifilm GF 30mm f/5.6 T/S (field of view on 44x33 sensor equivalent to 24mm in 35mm format)
Schneider Kreuznach 35mm LS f/3.5 (field of view on 54x40 sensor equivalent to 22mm in 35mm format)
Sharpness
First, let me say that all 3 lenses might have their own limitations and strengths, but they will all produce sharp images. Most pictures end up on clients’ websites or print publications that will never fully reveal their weaknesses, if any. Additionally, comparing different lenses operating in front of various sensors in a completely fair manner is impossible. That being said, I did my very best to evaluate each lens in a real-world example.
Sharpness tests have been performed at base ISO for each camera operating at full resolution. Medium format cameras have a shallower depth of field, so I couldn’t shoot all 3 cameras at the exact same aperture. I first had to determine for each camera/lens the aperture that produced sharp images in the center while maintaining the sharpest corners possible for the view photographed before hitting diffraction. I photographed the kitchen at various apertures between f/8 and f/22 at base ISO for each camera. I thought this range of apertures was the most relevant for architectural and interior photographers. In terms of sharpening settings, let’s agree that “no sharpening” at all would make it very difficult to assess because all lenses require a minimum sharpening to produce clean images. I consequently had to decide on a baseline and used the default sharpening in Capture One for the GF 30mm T/S lens with the following values for all 3 cameras. Sharpening amount: 185; radius: 1; Threshold: 1; Halo suppression: 0. Diffraction correction was left at 0 in the lens correction panel. This topic can be debated endlessly which is why I am sharing the RAW files here, so you can decide on your own sharpening baseline to review the images.
GFX100 II - ISO 80 f/11 - Basic settings
A7R V - ISO 100 f/8 - Basic settings
Phase One XF - ISO 50 f/14 - Basic settings
GFX100 II + 30mm T/S: the image shot at f/11 is the sharpest overall. The image shot at f/8 lacks depth of field and the ones taken at f/16 and f/22 are softer showing that lens diffraction is no myth after all. Note that I found it rather challenging to nail the focus with that lens. I had to compromise because when I was happy with the center, I lost a bit of sharpness in the corners, and vice versa, suggesting a rather pronounced field of curvature. Maybe my lack of experience with the lens and camera is to blame but I would be interested to hear what other people think about this.
Sony A7R V + Sigma MC-11 adapter + Canon TS-E 24mm: the images shot at f/8 and f/11 are extremely close. I believe f/8 is a tad sharper in the center though, and f/11 better on the edges. Like in the case of the Fuji camera things soften up at f/16 and beyond.
Phase One XF + Schneider Kreuznach 35mm LS: overall the image is the sharpest around f/14-f/16. Corners and center are similar at these apertures. I honestly don’t see a difference between f/14 and f/16. The same observations as the 2 other camera/lens combos apply regarding wider and smaller apertures.
All the sharpness comparison screenshots are available for download here in the "Sharpness" folder.
Ultimately, the Sony was shot at f/8 (f/11 is very close), the Fuji at f/11 and the Phase One XF at f/14 (f/16 is very close). Now when comparing all 3 I would say that we are splitting hairs. The Phase One lens is the sharpest in my opinion except in the top left corner. The Fuji lens is significantly better than the Canon on the left edge. Honestly, I will let you decide for yourself. What seems obvious to me though is how relaxed and natural the rendition from the GFX is when viewed at 100%. The pixels don't look as “stressed”, in other words as "digital", like in the Phase One’s and Sony’s images that are almost too sharp. It is especially obvious after final sharpening with the following settings - Sharpening 300; radius: 1.2; Threshold: 0.8; Halo suppression: 0. Diffraction correction was left at 0 in the lens correction panel.
Full image - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Center 100% - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Top Right Corner 100% - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Top Left Corner 200% - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Right Edge 100% - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Left Edge 100% - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Bottom Left Corner 100% - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Bottom Right Corner 100% - GFX100 II-A7R V-Phase One XF - Advanced sharpening
Color
I set the white balance for all the images by clicking in the middle section of the wall by the arch on the right. It should be obvious to you that the GFX colors are the best. The whites look “more white” off the bat and everything looks very natural. Sony made some improvements on the A7R V using AI-based technology to tweak auto-WB. I have been using the camera extensively for over a year now and I can attest to this - the whites look better than they did on the A7R IV and previous cameras. What is also true is that the A7R V saturates the blues like no other A7R camera before it! It is particularly visible when shooting exteriors - skies are significantly saturated. In our example, look at the Kitchen Aid Mixer on the countertop and the cabinet doors below. Because I know the space, I can tell you that the most accurate colors are the ones from the GFX image. Period.
Color Comparison: blues - GFX100 II-A7R V-Phase One XF
The IQ260 is close but images from the CCD sensor have typically a red/magenta cast particularly visible in blue skies as well. See a closeup of the ceiling below. Let’s not throw Phase One cameras under the bus though because newer digital backs such as the Trichromatic ones and the IQ4150 have undoubtedly better and more accurate colors than the IQ260 (remember it came out 11 years ago. Pretty amazing!)
Color Comparison: whites - GFX100 II-A7R V-Phase One XF
All the color comparison screenshots are available for download here in the "Color" folder.
Dynamic range
I did a simple test to evaluate the dynamic range. It is by no mean perfect and scientific but gives a rough idea of highlight/shadow recovery. I shot 4 images for each camera: -2 EV; -1EV; +1EV; +2EV – the standard being the base exposure shot previously. I applied the same basics adjustments to these images and corrected the exposure: +2 for the -2EV image; +1 for the -1EV image; -1 for the +1EV image and -2 for the +2EV image.
Dynamic Range Test - GFX100 II -2EV/-1EV/+1EV/+2EV Recovered
Dynamic Range Test - A7R V -2EV/-1EV/+1EV/+2EV Recovered
Dynamic Range Test - Phase One XF -2EV/-1EV/+1EV/+2EV Recovered
The most interesting images are the -2EV and +1EV because at +2EV with an already overexposed window, the highlights are gone for good (cold but sunny outside!) It seems to me that the GFX is significantly better at recovering shadows in the -2EV exposures. There is barely any noise under the counter on the left. Unsurprisingly, the old CCD sensor on the Phase One is the worst in that regard. In the +1EV shots, the Sony seems to recover more details in the highlights. Additional testing would be required but it looks like the GFX does better with underexposed images (clean shadow recovery) than it does with overexposed ones. Exactly the opposite to the Sony in my experience.
Dynamic Range Test - GFX100 II-A7R V-Phase One XF -2EV Recovered
Dynamic Range Test - GFX100 II-A7R V-Phase One XF +1EV Recovered
All the dynamic range comparison screenshots are available for download here in the "Dynamic Range" folder.
Shift (rise)
I just had time to test the rise between 0 and max for each lens. I didn’t have a chance to check the tilt capabilities. The Phase One lens doesn’t shift, so it is out of the race for this category.
GFX100 II: the lens performs very well up to about 10mm. After that the corners get slightly softer, but it is still very good especially for exteriors where the corners will be sky 90% of the time. Remember, only standard sharpening is applied, so there is room for improvement in the computer. When shooting interiors, I rarely shift more than 5mm, so there is nothing to worry about with this lens. Note that the vignetting is minimal as well which is quite remarkable considering the size of the sensor.
Ful image - Shift (rise) 0/+5mm/+10mm/+15mm - GFX100 II
Top Right Corner 100% - Shift (rise) 0/+5mm/+10mm/+15mm - GFX100 II
Top Left Corner 100% - Shift (rise) 0/+5mm/+10mm/+15mm - GFX100 II
A7R V: for an adapted Canon lens to a Sony body, I would say it is great. It also softens up around 8-10mm of rise. I suspect it would be even better at f/11 but the center will start degrading a tiny bit. I wish I had shot these at f/11 to be honest, but I forgot. I have enough experience with that lens on the Sony to attest it is a solid performer though. If you don’t trust me, check my portfolio – more than 50% of the images were shot with that lens on different Sony bodies over the years. It is also half the price of the Fuji GF 30mm T/S that is excellent, but clearly doesn’t perform twice as good as the Canon lens in my opinion.
Ful image - Shift (rise) 0/+5mm/+10mm/+12mm - A7R V
Top Right Corner 100% - Shift (rise) 0/+5mm/+10mm/+12mm - A7R V
Top Left Corner 100% - Shift (rise) 0/+5mm/+10mm/+12mm - A7R V
All the shift comparison screenshots are available for download here in the "Rise" folder.
ISO
GFX100 II: solid performance from the Fuji here. I will let you decide, but it is very clean. The colors hold well too, and the base ISO is truly fantastic.
A7R V: the smaller sensor does well, but not as good as the Fuji. The base ISO is noisier, and the colors degrade at ISO 3200 and beyond. When was the last time you shot architecture at ISO 3200 though? Note that ISO 80 on that camera is just an extension of the base ISO 100. It is not native like on the Fuji.
IQ260: I included it for reference, but CCD sensors are not known for their high ISO performance. CMOS sensors are just better at this game, and it shows.
ISO Test - GFX100 II
ISO Test - A7R V
ISO Test - Phase One XF
All the ISO comparison screenshots are available for download here in the "ISO" folder.
Tethering speed and file size
I recorded the following tethering speeds with the same IQWire 15ft USB-C to USB-C cable on my 2020 13-inch MacBook Pro (intel-based): GFX100 II 4.0s ; A7R V 2.6s. The Phase One IQ260 uses a Firewire connector, so I thought it was irrelevant in this category. The average RAW file on the 102MP Fuji is about 209MB. The average on the 61MP Sony is 128MB. It is 70MB on the 60MP IQ260. I edited all 3 final images with the exact same layers in Photoshop and saved the corresponding .PSB file for each camera. It took almost twice as long to save the multi-layered Fuji file which was almost double the size (2018 Mac Mini 6-core Intel Core i7, 64GB DDR4, 2GB SSD). There is a reason why I still don’t wish to work with 100MP files...
Tethered Shooting Test
Camera body and lens construction:
Coming from the Sony, the GFX100 II feels beefy and heavy in comparison. The camera fits comfortably in the hand though and the texture on the grip is amazing. It is more manageable than the Phase One XF that feels like a brick to haul around despite great ergonomics. The GFX mount is huge and so are the T/S lenses next to the Canon TS-E lenses that I use on the Sony. Something to keep in mind if you like lighter cameras. The star of the show for me on the GFX is the electronic viewfinder. It is amazing and a new benchmark for the competition: 100% coverage with 1x magnification and 9.44M dots of resolution. So clear and beautiful that the excellent viewfinder in the A7R V with the same resolution suddenly doesn’t look as good as I thought. It is removable too and can accommodate a separate tilt adapter that permits to adjust the angle at which the EVF is attached to the camera. In one word: awesome! Should I remind everyone that as I speak Phase One still doesn’t offer a single camera body with an electronic viewfinder? I have been begging for one since 2018… Regarding the back LCD I still think that the fully articulated bezel-less screen on the A7R V wins the battle. I won’t comment too much about camera menus because we all tend to have personal preferences. Sony’s menus have been heavily criticized for years, but I never had any issues with them. The GFX has clear and accessible menus as does the Phase One. I personally find the top LCD on the Fuji pointless in comparison to how useful and brilliant it is on the Phase One XF. I also noticed a few clunky things on the Fuji: by default, the front dial can be pushed in to switch between aperture and ISO control. I tried multiple times, and it was always hit-or-miss. It is probably preferable to set up a dedicated custom button for changing the ISO. I also very much disliked the small selector near the viewfinder to switch between focus modes that felt a bit cheap and dated for a camera of this price. My biggest gripe in ergonomics is the absence of a proper D-pad in the back of the camera to navigate menus and add custom buttons. The tiny joystick was a bit frustrating to use to select menu options. Overall, the camera is of course well-built and the viewfinder is a jewel. The T/S lenses are big and heavy as well. The 30mm T/S lens is as heavy as the Schneider Kreuznach lens (1.3 Kg. Ouch!) Surprisingly my copy didn’t come with a tripod collar mount which would have been good to better balance the weight on the tripod head and release some tension on the mount. Shift and tilt mechanisms are super smooth and seeing the amount of shift in mm displayed on the back LCD of the camera is kind of nice and addicting!
Investment
Here is a point that I wanted to address for a long time. A Phase One camera might be a lot more money, but it is the only system of the 3 that also comes with 24/7 support from a dealer. Over the years, I have needed assistance a couple of times with my Phase One XF and digital back – focus stacking is hard on the equipment. It always happened in the middle of shooting a new watch collection (sometimes late at night). Each time I grabbed my phone and called B3K Digital in Toronto, I received immediate assistance. Within 24-48 hours, I had a replacement camera body or digital back in hand to pursue my work with limited interruption. Would the same thing have happened with a Sony or Fuji camera? I would have had to wait for a month for the camera to be repaired with no replacement or locate a rental. How much is this service worth to you is the question? For time critical work, it is basically priceless to me.
In Canada, the GFX100 II body retails for $10,124.00. The GF 30mm T/S lens is $5,400.00. For comparison the A7R V body is $5,300.00 and a Canon TS-E 24mm $2500.00. The lens adapter is roughly $300.00. Basically, including a lens the Fuji is double the price of the Sony and about a third of the price of a Phase One.
Final thoughts
The Phase One XF is undoubtedly a great camera. What is true too is that it is now completely outdated and overdue for a replacement. A lighter body with an electronic viewfinder would be highly desirable. The beauty of the system resides in its modularity with a detachable digital back that can easily be used on technical cameras for additional control with outstanding Rodenstock and Schneider optics. The system now feels very overpriced against the competition though, despite the obvious quality of the latest IQ4 digital backs.
The GFX100 II in comparison delivers top-notch image quality, colors and tones at a more affordable price. The weight of the camera is manageable, and the electronic viewfinder is outstanding. It also costs twice as much as the Sony which overall is still a great performer.
After testing the Fuji camera, I have decided to stick to the A7R V for now for my architectural and interior work. Switching system always ends up being a costly exercise and I am satisfied with the quality produced from the Sony. I found the GFX100 II's overall image quality to be better, but only by a small margin. My position might be a bit unique because I also photograph watches and need the Phase One XF and its focus stacking tool that I would trade for nothing else. I guess I can't have them all! At this point in time, I personally chose to wait and see what the next generation of sensors will bring to the table in terms of image quality and innovation. For most people seeking the best colors and image quality, The GFX100 II is currently the best choice.
Retouched image - GFX100 II
Retouched image - A7R V
Retouched image - Phase One XF / IQ260