Report - Inaugural Bill Lovell Memorial Lecture - History of ARRI Alexa
The BBC Media Centre board room was full to overflowing on 25 September 2013 as the UK section remembered Bill Lovell who passed away suddenly at a Church meeting in March this year.
Chris Johns, Section Chairman, welcomed Bill’s family and his friends from SMPTE ,the Guild of Television Cameramen and many others who knew him along with other members and in particular, students from Westminster University.
Over 80 people had come to mark Bill’s achievements in our industries.
Bill graduated in electrical engineering from Imperial College London before joining the BBC at its Ealing film in West London, rising to be the Head of technical services before it closed its doors in 1995. He moved to ARRI becoming the product manager for the then new Electronic Cameras section – and was a driving force in the new venture that culminated in the world famous Alexa and the brand new AMIRA launched at IBC2013.
Chris said Bill was a SMPTE man – wedded to the principles of its Three Pillars –
Members – Bill was so much a People person encouraging all he met to support and join, Education – he had the way of explaining things very simply, clearly and quietly
and Standards with his work in setting the standards for cameras and imagery .
Bill supported the UK Section during the first year of operation since it reformed, he would be at almost every meeting - talking but more importantly listening to people we met, sitting towards the back and then during Questions and Answers, you would see his hand go up, and when called to speak, produced the most searching and apposite questions, and a unique view point.
Franz Kraus, Managing Director of ARRI, started the lecture with a retrospective montage of Bill at work during events around the world, talking with people, showing his enthusiasm for his work and the work of others. He also remembered Bill’s role in the company and his input to the building of the new Arri electronic cameras – with insights to what features should be included or excluded to get the camera’s operation to be as perfect as any operator or DoP would expect from equipment made by Arri. More importantly he was instrumental in bringing a then immature technology to application, always giving advice not selling, pointing out the current limitations in the technology and listening to what the users needed.
Franz remembers being asked by the UK Managing director of ARRI media if he would support taking on a Video department alongside the existing Film and lighting departments in order to learn about the technology and to see what the business model might be, if and when it was right for ARRI’s business. He said he was told of a very knowledgeable person, coming from the BBC who understood both film and video and that this was very important for ARRI who had not yet ventured into video beyond laser film recording and scanners.
He recalled talking with Bill about how the first electronic cameras would be not be as good at a 35mm film but would be better than 16mm – hence the names D20 and D21 – somewhere in the middle! He went on to admit though that for the creative and marketing people, the name Alexa was a much better proposition than calling Arri’s next generation camera the “D35” which is what the engineers would have called it!
Franz Kraus then moved to his presentation on to "The Theatrical Experience -Considerations for the Cinema of the Future”. His aim was to show the features which distinguished Cinema from Television, as it was the translation of these features into a future-orientated cinema system in phases that was at the heart of preserving Cinema, its screens content and production of Movies as separate art to that for television.
But much of what will happen in acquisition for the cinema will transfer to television, an area where it was a delight to discuss and learn from Bill.
He began by considering and demonstrating how the cinema had progressed with the fidelity of audio matching human perception, from the 100 to 8Khz analogue Optical Mono covering a 40dB range of sound pressure (like AM radio) though Analogue Dolby, Digital and culminating in DCP 24 Bits with its 20-20KHz frequency range and a dynamic range running from the threshold of hearing to the threshold of pain! Then coming to the now with Dolby ATMOS ™ using objects anywhere in a 3D sound stage. Thus Cinema audio has reached a level of perfection, to a point where the humans’ perception can barely distinguish more detail than a modern audio systems can provide, making further improvements almost impossible!
But this is not true in for pictures - where a natural scene may be greater than 20 f-stops, a contrast ratio of a million to one – and our eyes with fixed aperture cover 13 f-stops or ten thousand to one but with adaption this could encompass 30 f-stops or 108 to 1. He illustrated this with typical test scenes where the 15.5 f-stop range of the scene was well reproduced by the Alexa with its 14.5 f-stop range – some 23 thousand to one. But this is still some way away from the natural scene- there was always a compromise.
We see in colour as well as black & white and he outlined the three levels of our perception – Photopic/daytime, Mesopic/twilight (which is still not fully understood) and Scotopic/Night vision. He pointed out that the majority of Cinematic Projection at 48Cd/m2 was in the Mesopic range and most of the 3D projection at 12 Cd/m2, was well in this range which has little colour sensitivity. He also pointed out that the 500 to 1 contrast ratio of the Projector was degraded to about 100 to 1 by refection and other interference such as safety lighting etc.
A modern TV set however has an emissivity typically of 200cd/m2 and a contrast range of close to 500 to 1 working across the transition between Mesopic and Photopic vision and high Dynamic Range displays could probably attain a fifteen thousand to one contrast ratio and 400 Cd/m2. He demonstrated how the brightness of an image gives immediately a greater reality with more saturated colours and feeling of more detail.
But the detail depends on our ability to resolve detail; human visual acuity is about 1 minute of arc which for a typical cinema 2k image is equivalent to sitting at 3 times screen height or about 1.5 times the screen height for a cinema 4k image. Using the example of the 20m screen ARRI cinema in Munich, three times the screen height was under half the seats and the 1.5 time the screen height was just a few seats in the front row! No one sits in the front row in cinemas though as the audience prefer the back rows (for a number of reasons, as ARRI sponsored university studies have shown!).
But the resolution of the static image in the whole signal chain depends on what is thrown away in compression systems and although there is a very noticeable difference between the original images shot and displayed in 4k rather than 2k, when passed though the current DCI JPEG 2000 compression standard, the difference is not that obvious. He demonstrated with images showing the difference between what was fed to the projector and the original.
Franz Kraus pointed out that we do not go to the cinema to see static images, film is about the moving image, and thus the frame rate of thse systems has a great effect on how we see the image.
He showed many examples of motion blur at different pans and frame rate and static resolution and commented that 48Hz picture were not life like! Something television viewers had known for over 60 years! Work carried out by ARRI with 96Hz, 48Hz and 24Hz images shows that even 96Hz is not fast enough for realistic motion capture. We need to have the option of higher frame rates in the cinema which is a good subject for further discussion.
He summed up by pointing out the dominant factor of perceived image quality in cinemas is screen brightness and contrast as this benefits the audience no matter where they sit or what genre or style of the content. In the camera, image capture capabilities have to surpass the capability of the screen, in dynamic range and additional higher spatial resolution so that audiences will perceive detail from very large image screens and that current DCP compression limits the cinema to 4k.
So what for the cinema of the future as different from television? Franz Kraus postulated that television is a medium which tolerates distraction and interruption, that there are 4 billion TV sets in use worldwide, that a wide range of programmes and it has a successful and growing business model. Television is also under pressure to innovate technically and perhaps will merge with IT. Taking all this into account - TV can exist without the cinema!
He looked at future for television to include future growth with new formats for “on demand” and IP and that the first screen will get bigger and higher quality while the second and third screens will give more content and new business models.
He pointed out the contrast with cinema, that it is a medium demands concentration, that there are about 140 thousand screens worldwide, that about 50 films each year make 80% of the box office revenue, that there is a good business model in Hollywood but perhaps nowhere else and that there is stagnation except in a few regions and little convergence with IT.
Thus cinema needs TV as most of the money is not made by theatric distribution but by other means of distribution and merchandising.
What may be the future of cinema? A few premier cinemas with “Digital 70mm” enhanced experience or an immersive "next gen Imax" experience and can there be any new business models?
Franz Kraus summaries the Digital 70mm enhanced experience as being another story of significant length for another time with video and sound being shared by large groups with concentration and no distractions. Thus the only things that make the theatre attractive are the moving picture on the screen and the surround sound where the aspect ratio is appropriate for the story and the image must be brighter and with a greater contrast ratio and a larger colour space , and if greater reality is required, higher frame rates as well.
Where the audience experiences is actually to be part of the action requires 3D sound giving the same experience in every seat, the screen larger than the field of view with good resolution even as high as 16k, with real world peak brightens of about 1000 cd/m2, a contrast range of about the same as human capabilities say 15 f- stops and the colour space matching visual capability.
And of course a much higher frame rate and may be holographic techniques. And like some Korean cinemas, a 4D effect with seat motion, heat & cold and wind!
Franz Kraus closed his main presentation by reminding us that there needs to be a cinema Labs where new ideas can be tried out. Ideally these need to be accessible in Hollywood, Europe and China so that the researching of the technical possibilities of the new cinema can be carried out in a meaningful way. This would require industry consortia to get the state of the art performance as the creative possibilities of novel innovative cinema technology were explored. This should be under the guidance of the academy (AMPAS) or similar organisation.
Finally he then went on to tell us about the new ARRI AMIRA, a name for a princess, launched during BC 2013. He mentioned that many had commented that this product goes back to ARRIs roots, the classic 16mm Arri camera for documentaries operated by a single person.
It is the same imaging system as the Alexa but with different ergonomics and a weight balance for shoulder use and that it’s ready to shoot out of the box. It was first shown only two weeks ago and we are listening to the reaction. If it needs to make a tired politician look good we can grade in the camera and have a 3D LUT.
It has more modern electronics to run at up to 200Hz and more horsepower than is currently available – we are making it a upgradable platform
We see it being used in documentary and factual shooting for television but also for corporate films, trailers etc. as well as some budget cinema production, but we are not compromising ARRI quality.
His presentation ended with the Promotional video for AMIRA.
Andy Quested – Head of Technology HD and UHD at the BBC chaired the questions and discussions.
There was a lively discussion covering why the cinema is unlikely to be replaced by the audience sitting in individual Helmets giving all the immersion that they needed – because that attending a cinema is a social experience.
There was obviously a lot of interest in the AMIRA with questions on its OLED viewfinder and the need for adjustable metameric correction through to the choice of lens mounts and the capability of its electronics. Franz Kraus pointed out that there were many things that the AMIRA could do, that it was the basic model that had enough functionality to make it more than useful at a potential price point but also has a means of upgrading either permanently or by hiring the function for one shoot. There was a need to get the business model right.
Lenses were a special case where he hopes that ARRI users already know that the glassware on the front of a camera makes a difference to the captured image but if there is a good lens on a different mount; Arri would also look at that.
As the ALEXA and AMIRA have the same sensor there is no problem matching a shoot using both cameras. His final comment was that the electronic cameras were not very good at doing speed ramping at acquisition – but with higher frames rates it can be done in POST.
As expected there was a lot of debate about higher frame rates – John Zubrzycki, Principal Researcher at BBC R&D pointed out that the human vision system has no inherent frame rate but if it was able to track motion that mention needed to be sharp however if it could not track motion there as an advantage in motion blur. Peter Wilson talked about thinking about sampling at a rate which is fast enough to resolve detail or slow enough to blur the image – oversampling or under sampling – but this means having a variable frame rate camera and entire broadcasting systems.
Andy Quested, who had been working with the EBU tests on High frame rate said that like ARRI’s result the EBU had seen that there was a great difference between the 50/60 Hz frame rates and something at 100/120 but that still was not enough and although the official view was that the industry required something higher that 120Hz the perceived improvements may stop at frame rate which may be in excess of 200 Hz. There was still a lot of work to do.
Andy Quested In thanked Franz Kraus for a very informative and thought provoking lecture which was to the high standards and clarity that we had expected from everything Bill did. He recounted his first meeting with Bill at the BBC Television Film Studios where Andy as a trainee spent weeks remaking Nagra microphone cables when the boom swinger sound recordist decided to walk in opposite. Many years later an as Head of Technology for HD at the BBC, Andy had such fruitful relationship with Bill whose visions listening power and insight we have heard about tonight and carries on in the range of ARRI cameras.
As a Final remembrance of Bill, Roderick Snell, founder of Snell and Wilcox and a great innovator, recounted Bill’s role when he was with the BBC in the early 1990s and Snell and Wilcox were leading a Europe collaborative project developing a new system which was neither Film or Television working at higher fares rates and spatial resolution than High Definition. Bill’s experience practicality and attention to detail shone through, as did ARRIs ability to make test material on film as there was no reliable way of capturing images electronically then. This work was the precursor of what we have now in higher resolution and high frame rate cameras. Despite the great progress over the past 15 plus years, there are still elements of work from that project which have not yet been seen in products. So there is still a lot to do to carry on the work started by Bill Lovell.
Chris Johns closed the meeting with renewed thanks to ARRI and Franz Kraus for his presentations, the BBC Academy for providing the venue & organisation and SMPTE for the refreshments and to the family and friends of Bill Lovell.
The meeting took the opportunity of meeting each other and getting “hands on” with the AMIRA camera on its first public showing after IBC 2013.
Report by Peter Weitzel and Andy Quested