Director/cinematographer Paul Leeming of Leeming LUT One and Visceral Psyche Films recently put Fujifilm’s X-T2 to the test alongside a range of other hybrid stills/video cameras, and believes that the X-T2 has the potential to be one of the best small, affordable Super 35 video production cameras on the market.
Mr Leeming comes from a classical feature film background and for some years owned and hired out several RED Super 35 digital movie cameras. In recent years he has adopted the Panasonic Lumix DMC-GH4 Super 16 camera as his mainstay, due to its advanced, near-complete moviemaking feature set.
The GH4’s successor, the GH5, due to appear sometime early 2017, looks set to acquire even more advanced moviemaking features. Meanwhile, there is a clear gap in the market for production-ready Super 35 cameras with Samsung discontinuing its very promising Samsung NX1 4K hybrid in 2015.
Unlike Samsung, Fujifilm has a long history of producing high-end video camera lenses, excellent stills lenses of all formats, and top quality integrated camera and lens systems in a number of formats under its own brand name and in collaboration with Hasselblad.
Add all that to Fujifilm’s achievements in making stills and movie film and the Japanese company has what it takes to produce one of the most sought-after Super 35 movie production cameras, if it wishes.
Fujifilm’s best first step would be to take on board Mr Leeming’s firmware and other suggestions, below, consult on a range of suitable lenses and commence work on an even more video-capable X-T2 successor, the X-T3.
Fujifilm X-T2 Suggested Improvements, by Paul Leeming
Having tested two X-T2 cameras now, there are several fairly glaring omissions which would need to be added and/or fixed in order to present a credible filmmaking camera to the community.
1. White Balance in Video Mode
There is currently no way to set a custom white balance while in video mode, whereby you use a spectrally neutral white or grey reference card to balance out the RGB channels, then save that balance to one of the C1, C2 or C3 colour slots. Given that maximum dynamic range and colour tonality in 8bit 4:2:0 depends on extracting the fullest range out of each RGB channel, it is imperative that a Custom White Balance be easily achievable in any video mode, ideally able to be mapped to a Custom Function button so that a shooter can hit the function, auto white balance in video mode, and have the subsequent slot kept in operation until otherwise updated or changed.
To accurately judge where overexposure occurs, there is a need for zebras being active in all video modes. These zebras should ideally be adjustable, such that the user can set, for example, 70% IRE, and have it accurately reflected on screen. To judge overexposure, 100% IRE or 100% zebras should show where the clipping point is occuring in real time, so that the user can adjust iris, ISO or shutter as required to reduce exposure to the point where it is no longer clipping. For log-based shooting using F-log, the zebras should still accurately indicate where the clipping point is, even if it occurs before 100% IRE (for example, 79% IRE).
3. DCI 4K
The camera is very close to achieving this already. DCI 4K is 4096 x 2160 x 24.00fps. I’ve tested two different cameras and one had 24.00fps and 23.98fps listed separately; the other did not (different firmware I guess). Both of these however only allowed 3840 x 2160, not 4096 x 2160 as defined by DCI 4K. Having used the DCI 4K feature of the Panasonic GH4 camera previously to match cinema standards fully, the addition of 256 pixels of width plus the 24.00fps framerate would set the X-T2 in a rarified class of filmmaker cameras which actually support the full DCI 4K spec, as only a couple currently support that combination. Supporting the DCI 4K spec with the X-T2’s Super 35 size sensor would actually push it over the GH4 in terms of sensor size and capability, since the GH4 is a M43 sensor only.
The histogram does not show up in 4K shooting modes that I can see, but it should, along with the zebras mentioned above. Between the two, the filmmaker can easily judge exposure and adjust to maximise dynamic range without clipping anything in the shot. It needs to be a selectable option in all video modes.
5. Unlimited Recording Time
There are two things here that need addressing – first is the 10 minute limit without the added battery grip, and the second is the 30 minute limit even with the grip attached.
First off, there are basically no other cameras that limit their recording to under 30 minutes without needing an additional expense of a battery grip, which also adds bulk and weight to the camera. If this is really a hardware problem with not having enough power, then offer an option such as external power through a dummy battery or USB power input.
Second is the 30-minute artificial limitation to avoid taxing the camera as a video camera. This is purely arbitrary and given that most filmmakers often use their cameras for documentaries and interviews, etcetera, should be something that can be offered as an option, or at least done by region so that in the US, for example, you can sell the unlimited recording model (such as the Panasonic GH4 again, whose US and GH4R models offer unlimited recording time, limited only by the SD card or power running out).
6. F-Log Preview
Currently, I cannot view F-Log colourimetry until hitting record, which causes all sorts of issues with external monitoring using LUTs, etcetera. F-Log needs to show up in the colour space it natively uses, at all times, not only when hitting record.
7. F-Log Internal Recording
I am very aware of the potential issues for 8 bit 4:2:0 log recording to cause unwanted artifacts (see the GH4 for what NOT to do here). However, it should be something the user can choose to enable or not, for testing purposes and for those who don’t have an external recorder handy.
8. F-Log 10bit HDMI Output
The GH4 offers 10bit 4:2:2 HDMI output in full DCI 4K, and has done since it was released over two years ago. This should be the minimum standard going into 2017, and would definitely raise the X-T2’s filmmaker credibility.
There are probably more things I could suggest with further testing of the camera, but getting most of the above things fixed would go a long way to putting the X-T2 close to the top of the mirrorless APS-C camera pile for early 2017.
You have probably noticed I refer to the Panasonic GH4 a lot. That’s because it is the most well-designed consumer-facing mirrorless camera I’ve used and tested in depth (and the GH5 looks to best it in some significant ways in early 2017). It has ergonomic controls, a good monitor and lots of other features which make it easy to use on film sets. It really should be the minimum to aspire to in terms of all of the above, for the X-T2 and future cameras going forward.
Closing with my ultimate wishlist for a mirrorless camera in early 2017, in case you want to blow the industry wide open…
- Full Frame sensor with no video mode crop.
- DCI 4K (4096 x 2160 x 24.00fps) recording internally to 10bit 4:2:2 in whatever format/media works best.
- Rolling shutter of less than 14ms.
- 13-14 stops of dynamic range minimum.
- 60fps maximum framerate using DCI 4K at full 10bit 4:2:2 internally.
- Zebras, histogram, false colour mode to judge exposure.
- No internal recording time limitation.
- External power using either a dummy battery with cable or USB-C power delivery spectrally.
- Fully rotating monitor like that on the Panasonic GH4.
Feel free to contact me with feedback or questions, as I would dearly like to see this camera become a real alternative to the GH4 and soon GH5, given its larger sensor and potential colour rendition.
Hope this helps!
Director & Cinematographer
Visceral Psyche Films