High Frame Rates 2.0
Project Supervision: Prof. Stefan Grandinetti
The HFR-technology is not entirely new from a scientific point of view. But it was rarely applied in a major cinema production until today. In contrast, high spatial resolutions like 4K or higher, as well as the High Dynamic Range-technology (HDR) became a more commonly used standard in production. Because of these higher standards and the increased quality of the image, the limitations of a standard time base (24 fps temporal resolution) became more perceivable, and therefore an issue.
We can expect the number of productions using High Resolution-HDR-deliverables to increase over the next years. The high dynamic range technology generates higher peak luminance levels, which emphasizes unpleasant side effects like flicker and judder. HFR resolves the issues caused by low temporal resolution.
It significantly improves the accurate reproduction of fast movements i.e. of action scenes by capturing more frames per time unit (120 fps). The subjective perception of spatial details and sharpness increases with higher frame rates, because motion blur and opacity decrease. This also is an advantage to the human perception of 3D images by providing a much larger amount of image information and detail. Artifacts, such as judder or a lack of sharpness are generally reduced.
This new method also creates a new set of issues. Many viewers praise the evident technical improvements, whereas others reject the appearance of higher temporal resolutions. The authenticity and believability of the picture impacts the subjective perception of the viewer. HFR, in its early stage of development, suggests assotiations to realism, and it reminds viewers i.e. to TV-reportages. Parts of the audience used to aesthetics of traditional 24fps theatrical projection miss the "touch of magic".
Further research into HFR requires us to investigate various technical solutions regarding their impact on the aestetics of the images, that allows filmmakers to control the artistic quality of their work, take advatage of higher technological standards, while avoiding presently evident pitfalls.
A few key aspects of our research are:
How can the use of the aestetics of HFR be utilized as a aesthetic benefit and serve as a creative tool for fiction and documentary films and time based media in general?
We like to learn and analyse the emotional impact and credibility of using variable frame rates. This could mean preserving traditional aethetics for drama scenes on one hand, and to improve image quality in fast paced action scenes on the other hand.
Can we control the aesthetics of the image by changing the framerate of image regions, even individual pixels, to improve image perception?
What are the aquisition and postproduction tools and processes needed to grade various HFR-speeds and shape the perception of motion for storytelling purposes?
Will the audience get used to HFR, just like they had to adjust to grainless images in the emerging aera of digital cinematography?
How does HFR influence the rythm of images, mise-en-scene, camera movement, such as camera tracking, panning speed, and the overall perception of motion and depth of field? What are the creative restrictions or opportunities for camera direction and storytelling?
Does too much visual information impact our ability to process meaning, and reflect about the story? Do we still have room to get emotionally involved?
These are only a few important questions we like to investigate in future research projects.
Our current state of research includes the following projects
The 3D Stereo shortfilm ZWISCHENRÄUME was produced and screened with HFR 48 fps per eye, and filmed on two ARRI Alexa-M using a mirror-rig.
This project was realised as part of the course Studioproduction Film, and in cooperation with ARRI, Stereotech, and Pure4see. The results showed an overall improvement of the S3D-effect and a better imagequality.
Unbowed
The shortfilm UNBOWED was produced and screened with HFR 120 fps, and filmed on a single ARRI Amira.
This project was realised as part of the course Studioproduction Film, and in cooperation with ARRI and RealD. This film establishes a good baseline for further research in postproduction applications, authenticity and visual attention. UNBOWED was screened in full lenght on the SMPTE annual conference in Los Angelos in October 2016 by our consulting and sponsoring partner Tony Davis (RealD).
Abstract for the SMPTE presentation
It's all Backwards: Rethinking Frame Rate and Temporal Fidelity in a Cinema Workflow
As filmmakers move to high dynamic range, high frame rates, stereoscopic presentations, and wide international distribution, the temporal representation of motion is becoming increasingly frayed. A combination of high input frame rate, a refined post processing pipeline with temporal resampling moved to the front of the workflow instead of the back, and careful image reproduction can transform the entire look and feel of motion. This preserves the filmmaker's intent in any viewing environment. The presented footage resulting from this alternate post and distribution workflow results demonstrate the inherent benefits.
Speakers: Anthony Davis
Bachelor Thesis by Aaron Kuder
In his Bachelor Thesis Creation and Comparison of Cinematic Reference Footage filmed at Standard Frame Rate and High Frame Rate and its Impact on the Perception of Motion Aaron Kuder (DoP of the shortfilm UNBOWED) generated a set of HFR reference scenes with 192 fps and created two deliverables with two frame rates from this basis in order to make simultaneous comparisons in a 120 fps DCP in projection:
- 120 fps from 192 fps (frame sampling)
- 24 fps (every frame 5 times)
Get your HdM-HFR-2017 download credentials here.
Abstract
A film can be recorded and played back in standard frame rate (SFR) at 24 frames per second or in high frame rate (HFR) with higher temporal resolution. As part of this thesis, cinematic test footage is created and compared in the light of perception of motion. Higher temporal resolution reduces motion artefacts. In HFR motion is being perceived in higher detail, at slower speed and under certain circumstances unnatural. By synthesizing the shutter in post-production, the HFR footage can be modified in a creative way. The created test footage can furthermore be used for future scientific and empirical studies.
