A few years ago, a prominent manufacturer of studio support equipment did something unusual: it went to NAB with an experienced broadcast camera operator to discuss a part of live production that’s invisible when done well. Following a driven golf ball is one of the trickiest tasks in any branch of film and TV, demanding not only skill but equipment which allows people to exhibit that skill.
That’s something which becomes very clear whenever there’s technological change. Consider the earliest attempts at 4K broadcasting. The F65, which had been built very specifically as a tool for single camera drama, was occasionally seen rigged for live work. Sony had done a lot of work to make that practical, particularly using its SLog3 brightness encoding which was designed to make drama more gradeable but functions ably as a vehicle for effective live HDR. Still, it wasn’t quite ideal. F65, fine though it is, was not designed for broadcast, and neither were compatible lenses.
This matters. When a baseball is struck, it changes direction by about two hundred miles per hour in something comfortably under a millisecond. In a world constantly excited by single-camera drama, it’s easy to overlook the people who are required to keep cameras aimed accurately at that sort of motion. NHK and (principally, among others) Hitachi have spent much of the last decade exhibiting their preparations to broadcast the recent Olympics in 8K, not because there weren’t 8K cameras that could conceivably have been lashed-up to an outside broadcast truck but because those 8K cameras, and their lenses, would have been about as much fun to operate as a ton of wet cement.
On paper, it might seem as if the technology is pretty mature. After all, Canon’s EOS-R5 camera, particularly in combination with the Atomos Ninja V+, is capable of shooting high quality 8K material in an affordable package that’ll happily ride even a medium-sized drone. The problem is, that isn’t a broadcast camera package. It uses a sensor so big that creating broadcast lenses to cover it would be an extremely expensive exercise in frustration and be almost impossible to keep in focus when following that sharply-accelerating baseball.
8K Cameras for Broadcast
The earliest 8K broadcast cameras, which date back to the early-to-mid 2010s in IBC’s future zone, were a Hitachi-NHK collaboration. 7680 by 4320 pixels at 120 frames per second was an impressive enough specification to offset the odd wedge shape and the unconventional lens setup. By the time Sharp announced the 8C-B60A in 2017, things had improved: the camera, with a recording system developed in collaboration with Astrodesign, was approximately the size and shape of a conventional portable broadcast camera.
But a pattern was emerging. Sharp’s camera has a PL lens mount, limiting compatibility with conventional broadcast lenses, for the simple reason that only PL-mount lenses project a large enough image to cover its sensor. The reasons smaller sensors tend to have poorer noise, dynamic range and sensitivity than larger ones of the same pixel count are fairly well known – make the sensitive elements smaller and it’s simply less likely that a photon will hit any one of them. Reduce the sensitivity, and we have to apply more amplification to achieve the same image brightness, compromising noise performance.
That’s why cameras such as F65 were pressed into service to shoot live sports. For a while, though, companies such as Grass Valley have been able to offer highly competent 4K cameras of conventional broadcast layout. With 2/3” sensors, as in broadcast cameras going all the way back to widescreen standard definition, they’re compatible with all the world’s favourite support and optical equipment. Higher-resolution cameras demand a lot from lenses, of course; Canon and Fujifilm have been quick to market broadcast zooms promoted as particularly suitable for high-resolution work, but the general layout and behaviour of modern 4K cameras is familiar to experienced operators.
Conventional Broadcast Lens
But we couldn’t, at least initially, make 4K broadcast cameras with sufficiently small sensors to work well with conventional broadcast-style lenses and operate as operators would like them to. Then, after a while, we could. Is there any chance we can do the same for 8K cameras? Well, despite the fact that this has been worked on since, presumably, the early 2010s, not so far.
We’ve seen that Sharp made the decision to use a single Super-35mm sensor in the 8C-B60A. The same is true of Ikegami’s much more current SHK-810, as well as the Hitachi SK-UHD8060B, a dockable camera-recorder which will be very familiar to experienced broadcast people in terms of its overall layout. Announced in April last year, the camera has been described as “the company’s fifth generation,” although that count might include the early designs which appeared at IBC as essentially technology demonstrators.
The Hitachi and Ikegami cameras have both been shown in use with lenses from Fujifilm’s Cabrio range, particularly the Fujinon ZK4.7×19 (that is, 19-90mm) zoom. It’s a truly superb lens, with even the pickiest single-camera people increasingly willing to accept that zooms at this level are acceptable for demanding big-screen work, let alone run-and-gun field production. With its servo grip installed, it’s also easy to mistake the 19-90, or indeed Canon’s competing CN7×17 KAS S, for a conventional broadcast lens.
It can work well in that role, but it really isn’t a conventional broadcast lens. The ZK4.7×19 is a 4.7:1 zoom achieving a uniform T2.9 and weighing 2.8 kilos with its drive unit. For comparison, in the broadcast world, the 4K-capable Fujinon UA18×5.5 BERD is lighter, faster, wider and longer (even taking into account the smaller sensor for which it is designed), has an extender if we want to go even further, and is only about an inch longer physically. They’re almost exactly the same price.
So it’s not hard to see why the world pushes for 2/3” broadcast cameras. If we want a big-box studio lens, which we often do, both Fujifilm and Canon have longer PL-mount cinema zooms which can be pressed into service to follow distant golf balls, but it’s not clear whether that approach will ever really be fully satisfactory to the sort of people who are most capable of operating shots like that.
The compromise picked by at least part of the industry seems to have been the 1.25-inch, three-chip solution first seen in Sony’s UHC-8300. As with many standards which are pegged to historic customs which seem almost nonsensical in the present day, no part of the sensors in the 8300 actually measure an inch and a quarter on any dimension; it’s a manner of expressing sensor size which goes back to the days of tube cameras and the dimensions of the vacuum tubes used to create their sensors.
Three Sensor Progression
Rest assured, though, that the 1.25-inch standard has considerably bigger sensors than common 2/3-inch designs, and compared to the PL-mount options it also has three such sensors, and so, depending on the exact design of the optical system, up to three times as much sensor area with which to capture photons. It still can’t take conventional 2/3-inch lenses, at least without adaptors, but releases from both Canon and Fujifilm suggest that this design approach allows for lens designs that can look and feel much like conventional broadcast equipment.
Canon’s first contributions were the 7×10.7 KAS S and UHD-Digisuper 51, a 51×5.5mm lens reaching out to 280mm. Soon came a 10×16 portable zoom. Fujifilm now offers the HP7.5×8.5-SM, a short zoom clearly intended for fast-paced handheld work. There’s the HP11×22.5-SM and the 66×15.2, which is practically an astronomical telescope with a maximum focal length of 1000mm.
Still, there is some evidence that the 1.25-inch approach is not quite universally accepted. Fujifilm also offers the SK20×35-ESM, unusual because it’s a studio box lens in PL mount. It offers a reasonably wide 35mm short end and a considerable 700mm maximum reach, making it suitable for sports and concerts. The new 1.25-inch systems will naturally have different behaviour in terms of field of view and depth of field than a 2/3-inch equivalent would, and a PL lens like the SK20 even more different behaviour, but they should make 8K field production feel much like HD field production.
Everyday 8K broadcasting requires a lot of equipment other than camera bodies and lenses. Hitachi’s SK-UHD8240 is an 8K, 240fps high-speed camera of just the type that live sports needs, and a myriad of ancillary equipment from live wireless links to broadcast encoders also exists. Still, it’s early days, and far from clear whether 8K in general will overcome the questions over necessity and usefulness. Devices such as Panasonic’s AK-SHB800GJ region-of-interest camera, which derives up to five HD images from an 8K camera fitted with a wide lens with pan, tilt and zoom control, are perhaps a more immediate application for very high resolution cameras.
Perhaps one day it will be possible to make reasonable 2/3-inch 8K cameras, or perhaps PL mount lenses for broadcast will become normal; in the meantime 8K broadcast work is starting to feel much less like a special event.
Broadcast Bridge Survey
You might also like...
Filmmaking is now an artform with a long history, and that means a lot of received knowledge. Things like primes being better than zooms, and characterful primes best of all. The public attitude to that orthodoxy, though, is starting to…
Creating an LED light that looks reasonable, both on camera and to the eye, is no longer that difficult.
If 2020 was considered a disruptive one for the television production community, 2021 was a year where trial and error and the lessons learned became real-world REMI deployments to keep live sports and entertainment content on the air. Production studios too learned…
2021 saw the end of a British lighting institution with the demise of Photon Beard. The company has existed since the founding of R. R. Beard Limited, founded in 1882 by a man who reveled in the fantastic name of Robert Royou…
Planning for any kind of live TV broadcasting starts with a ‘what-if?’ list. What if the power source fails? What if a key production person gets sick or hurt? What if broadband internet access becomes unstable? What are the chances for…