In april I put together a piece of technical work based around the controversial new HFR (High Frame Rate) technologies in filmmaking. The goal was to start looking at this problem with a focus on how the newly revived technology might help us tell better stories. I found many people talking about whether they liked it or hated it, debating over corporate financial motivations, and getting very heated over the 'science' of it, but there wasn't a lot of people attempting to research and discuss the really important stuff.
What do frame rates have to do with telling stories?
What actually happens when you put HFR images in front of audiences? How does their relationship to the story change? How does it change the way they connect? How could we use this information to our advantage?
I decided to put together a test comparing two different frame rates for people to watch and get feedback from. I used a pre-existing story sequence and attempted to match it as identically as possible in both formats. Using an existing story gave me a good control to compare them based on connection and engagement with a story, rather than simply observing how 'smooth' or 'not smooth' the motion is. It also provided the opportunity to read the effect against the original work.
The film is the beautiful opening sequence of 'Brick (2005)', by Rian Johnson (Director), and Steve Yedlin (Director of Photography), if you haven't seen it, I recommend you check it out. The frame rates used are 25p and 50p (very slightly different from the 24 and 48 standards) with a 180 degree shutter angle.
Below are the two versions, but note some browsers and computers may have trouble playing them back correctly since the technology is still fairly new. I learnt from my testing it was difficult to get a faithful representation on many configurations. I recomend you update to the latest version of Chrome or Internet Explorer.
The best way to view these though is by downloading the zip containing the full 1080p videos Here.
See how you get on! The versions aren't labeled, and I invite your discussion below. Can you tell which is which? What effect does it have on your experience of the sequence? What are your thoughts and experiences with the technology?
The process of displaying this work presented some challenges. I set up a viewing station where the films could be comfortably viewed in isolation, but also attracting public attention so it could be seen by a good spectrum of people.
The system was an automated web application that randomised the viewing order of the videos so people couldn't compare results amongst themselves, and to fix the problem where order could affect preference outcomes.
Viewers then filled out short forms asking basic information such as identifying which version was HFR, which they prefered, why, and which seemed to be the most cinematic to them.
Despite using my own high spec tower PC to display the tests, issues arose when attempting to output to the HD TV. The same test that looked great on my HD monitor, seemed to have difficulty faithfully displaying the correct framerate on the TV. This is an area I'm still researching and welcome ideas about.
The area I set up in seemed to engage people slowly, and I managed to gather about 40 responses, but many of them were incomplete or not very detailed, leading to some questions around how to engage with a wider audience and stimulate the right kind of conversations.
Regardless of the challenges, the data I managed to collect presented some interesting insights.
The preference results were very close, the 25p version winning by a very small margin.
About half of the feedback mentioned that the HFR version felt a little 'weird'. Specifics included the feeling of the footage playing too fast, an uncomfortable feeling, a 'plasticy' look, and an unnatural feeling. Out of the people that prefered the 25p version, the words 'peacefull' and 'relaxing' were used often, with many actually suggesting that it looked smoother and more seamless to them.
Nearly all of the people who prefered the 50p version simply commented that it was 'smoother', 'less jittery', 'better', and 'nicer'. Most of these responses were only one word responses.
Only a couple of people incorrectly identified the different versions or stated they couldn't tell them apart.
From the feedback, parts of the HFR version that seemed to work 'best' were closeups on the characters face, and other small details. The wide shot seemed to break the engagement due to it having more distracting elements in the scene. In HFR the eye tends to wander away and take note of the ducks floating around in the water etc, whereas in the standard version, we sit more with the stillness of the scene and stay far more acutely focused on his relationship to his situation.
High frame rate (HFR) filmmaking is a Filmic Medium. Not a long awaited upgrade held back due to technological progress.
These tests identify framerate can't really be seen as 'more is better'. They point out areas where standard films cannot simply be 'converted' into ones with more frames, without some consideration into what effect that is having on an audience. If HFR is going to take off, we must start focusing on what the medium offers us as artists, and what it does not. Where does it undo our potential, and intentions? Where does it strengthen them? What can it teach us about other film mediums? The reality is we can talk about things like brain & eye perception, suspension of disbelief, and other dense theories until the very end of time, but until we start testing it, and putting it in front of people, we can't be sure what effect its going to have.
Douglas Trumbull put it very well - "What do you do when you have a medium that's more powerful, or more visceral or more lifelike?"